From gael.varoquaux at normalesup.org Sat Jan 2 02:58:48 2010 From: gael.varoquaux at normalesup.org (Gael Varoquaux) Date: Sat, 2 Jan 2010 08:58:48 +0100 Subject: [IPython-dev] [Numpy-discussion] [SciPy-dev] Announcing toydist, improving distribution and packaging situation In-Reply-To: <5b8d13221001011832p5ca4875dj13f04ba7ee61dd41@mail.gmail.com> References: <5b8d13220912280603p7221a264o875b0d5e74a5404@mail.gmail.com> <5b8d13220912300816s12c934adh4abdd6d703f8928f@mail.gmail.com> <5b8d13221001011832p5ca4875dj13f04ba7ee61dd41@mail.gmail.com> Message-ID: <20100102075848.GA17293@phare.normalesup.org> On Sat, Jan 02, 2010 at 11:32:00AM +0900, David Cournapeau wrote: > [snip] > - supporting different variants of the same package in the > dependency graph at install time > [snip] > The second issue is more challenging. It complicates the dependency > handling quite a bit, and may cause difficult situations to happen at > dependency resolution time. This becomes particularly messy if you mix > packages you build yourself with packages grabbed from a repository. I > wonder if there is a simpler solution which would give a similar > feature set. AFAICT, in Debian, the same feature is given via virtual packages: you would have: python-matplotlib python-matplotlib-basemap for instance. It is interesting to note that the same source package may be used to generate both binary, end-user, packages. And happy new year! Ga?l From gael.varoquaux at normalesup.org Sun Jan 3 06:11:53 2010 From: gael.varoquaux at normalesup.org (Gael Varoquaux) Date: Sun, 3 Jan 2010 12:11:53 +0100 Subject: [IPython-dev] [Numpy-discussion] [matplotlib-devel] Announcing toydist, improving distribution and packaging situation In-Reply-To: <961fa2b41001030305mddd301fp416a2fe23fc11568@mail.gmail.com> References: <5b8d13220912280603p7221a264o875b0d5e74a5404@mail.gmail.com> <64ddb72c0912290527s1143efc7g3efe93936ca5de5@mail.gmail.com> <5b8d13220912290634u5902a6bag33ddb8a15a93406b@mail.gmail.com> <961fa2b41001030305mddd301fp416a2fe23fc11568@mail.gmail.com> Message-ID: <20100103111153.GB24770@phare.normalesup.org> On Sun, Jan 03, 2010 at 03:05:54AM -0800, Nathaniel Smith wrote: > What I do -- and documented for people in my lab to do -- is set up > one virtualenv in my user account, and use it as my default python. (I > 'activate' it from my login scripts.) The advantage of this is that > easy_install (or pip) just works, without any hassle about permissions > etc. This should be easier, but I think the basic approach is sound. > "Integration with the package system" is useless; the advantage of > distribution packages is that distributions can provide a single > coherent system with consistent version numbers across all packages, > etc., and the only way to "integrate" with that is to, well, get the > packages into the distribution. That works because either you use packages that don't have much hard-core compiled dependencies, or these are already installed. Think about installing VTK or ITK this way, even something simpler such as umfpack. I think that you would loose most of your users. In my lab, I do lose users on such packages actually. Beside, what you are describing is possible without package isolation, it is simply the use of a per-user local site-packages, which now semi automatic in python2.6 using the '.local' directory. I do agree that, in a research lab, this is a best practice. Ga?l From p.schellart at gmail.com Mon Jan 4 04:41:57 2010 From: p.schellart at gmail.com (Pim Schellart) Date: Mon, 04 Jan 2010 11:41:57 +0200 Subject: [IPython-dev] Some minor issues with iPython 0.11 on Python 2.6.4 Message-ID: <4B41B7E5.7040201@gmail.com> Hi Everyone, I just wanted to mention some minor problems with IPython 0.11 running on Python 2.6.4. The first (and most important) issue is a bug (reported as bug 488061 on launchpad). If you use list=!command you get deprecation warnings because the popen command is deprecated in favor of the subprocess module. I have submitted a patch for this bug but until now of now it is unassigned. The other issues are minor. I would like tab autocomplete to behave the same way as the shell (e.g. bash). The current behavior lists all files as options when I hit tab to complete the name of the parent directory. The current default of asking the user to list or not when to many files are present violates the principle of minimal input that all unix programs share (e.g. rm is default instead of rm -i) which slows down a typical workflow. Also the distinction between two different types of behavior depending on the number of files in a directory is confusing. Furthermore when there are not so many files present and the completion shows the list immedately it scrolls (the output of) your previous commands off the screen (which means you often have to scroll back to see their results). I usually tend to have thousands of data files in my work directories and so I have to type 'n' whenever I change directories which is quite annoying :). So in conclusion I would like tab to not complete past a '/' and only list all possible completions with a second hit of the tab key. The final issue is that the current IPython puts an empty line before each prompt. While this might improve readability in some cases more often it is a waste of screen space. In particular when using clear the first line of the terminal is blank. I would prefer again to stick to the default of most shells and remove the empty line. Please take these remarks as mere suggestions. In general I like IPython a lot and I hope to see it evolve into an ever more useful tool. Kind regards, Pim Schellart -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3660 bytes Desc: S/MIME Cryptographic Signature URL: From laramichaels1978 at yahoo.com Wed Jan 6 12:16:48 2010 From: laramichaels1978 at yahoo.com (Lara Michaels) Date: Wed, 6 Jan 2010 09:16:48 -0800 (PST) Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? Message-ID: <985948.17135.qm@web111810.mail.gq1.yahoo.com> hi everyone, I have watched the excellent video tutorial on debugging with ipython, but I cannot get a hang of how to use it to step through code. I am using the trick of inserting 1/0 in my code where I want the debugger to come up. Then I get pdb (as expected), but if I try the n(ext) or s(tep) commands I am immediately dropped back onto the "normal" ipython shell. Is there a way to step through code without importing pdb and using set_trace()? thank you for any help lara -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.kern at gmail.com Wed Jan 6 12:27:39 2010 From: robert.kern at gmail.com (Robert Kern) Date: Wed, 06 Jan 2010 11:27:39 -0600 Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? In-Reply-To: <985948.17135.qm@web111810.mail.gq1.yahoo.com> References: <985948.17135.qm@web111810.mail.gq1.yahoo.com> Message-ID: On 2010-01-06 11:16 AM, Lara Michaels wrote: > hi everyone, > > I have watched the excellent video tutorial on debugging with ipython, > but I cannot get a hang of how to use it to step through code. I am > using the trick of inserting 1/0 in my code where I want the debugger to > come up. > > Then I get pdb (as expected), but if I try the n(ext) or s(tep) commands > I am immediately dropped back onto the "normal" ipython shell. You cannot step past an exception. > Is there a way to step through code without importing pdb and using > set_trace()? Why not use set_trace() if you are already inserting exceptions using 1/0? import pdb;pdb.set_trace() Use your editor's macro facilities to make this easy. -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco From laramichaels1978 at yahoo.com Thu Jan 7 10:32:55 2010 From: laramichaels1978 at yahoo.com (Lara Michaels) Date: Thu, 7 Jan 2010 07:32:55 -0800 (PST) Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? In-Reply-To: Message-ID: <4980.48416.qm@web111808.mail.gq1.yahoo.com> Hi Robert, > > Is there a way to step through code without importing pdb and using > > set_trace()? > > Why not use set_trace() if you are already inserting exceptions using 1/0? > > > import pdb;pdb.set_trace() But when in ipython I import or %run code with such a set_trace() call? I get thrown into the normal pdb, not the nicer iPython version that I find much more usable. How can I get into ipython's debugger in a way that I can step through code? thanks! -l -------------- next part -------------- An HTML attachment was scrubbed... URL: From robert.kern at gmail.com Thu Jan 7 10:41:11 2010 From: robert.kern at gmail.com (Robert Kern) Date: Thu, 07 Jan 2010 09:41:11 -0600 Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? In-Reply-To: <4980.48416.qm@web111808.mail.gq1.yahoo.com> References: <4980.48416.qm@web111808.mail.gq1.yahoo.com> Message-ID: On 2010-01-07 09:32 AM, Lara Michaels wrote: > Hi Robert, > > > > > Is there a way to step through code without importing pdb and using > > > set_trace()? > > > > Why not use set_trace() if you are already inserting exceptions using > 1/0? > > > > > > import pdb;pdb.set_trace() > > But when in ipython I import or %run code with such a set_trace() call I > get thrown into the normal pdb, not the nicer iPython version that I > find much more usable. > > How can I get into ipython's debugger in a way that I can step through code? I believe someone is working on adding this ability to 0.11. -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco From laramichaels1978 at yahoo.com Thu Jan 7 10:47:40 2010 From: laramichaels1978 at yahoo.com (Lara Michaels) Date: Thu, 7 Jan 2010 07:47:40 -0800 (PST) Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? In-Reply-To: Message-ID: <3013.25340.qm@web111802.mail.gq1.yahoo.com> Ah, ok, many thanks! lara --- On Thu, 1/7/10, Robert Kern wrote: From: Robert Kern Subject: Re: [IPython-dev] [newbie] stepping through code using the 1/0 trick? To: ipython-dev at scipy.net Date: Thursday, January 7, 2010, 3:41 PM On 2010-01-07 09:32 AM, Lara Michaels wrote: > Hi Robert, > > >? > > Is there a way to step through code without importing pdb and using >? > > set_trace()? >? > >? > Why not use set_trace() if you are already inserting exceptions using > 1/0? >? > >? > >? > import pdb;pdb.set_trace() > > But when in ipython I import or %run code with such a set_trace() call I > get thrown into the normal pdb, not the nicer iPython version that I > find much more usable. > > How can I get into ipython's debugger in a way that I can step through code? I believe someone is working on adding this ability to 0.11. -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma ? that is made terrible by our own mad attempt to interpret it as though it had ? an underlying truth." ???-- Umberto Eco _______________________________________________ IPython-dev mailing list IPython-dev at scipy.org http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An HTML attachment was scrubbed... URL: From jorgen.stenarson at bostream.nu Thu Jan 7 14:05:02 2010 From: jorgen.stenarson at bostream.nu (=?ISO-8859-1?Q?J=F6rgen_Stenarson?=) Date: Thu, 07 Jan 2010 20:05:02 +0100 Subject: [IPython-dev] Running tests Message-ID: <4B46305E.5010405@bostream.nu> Hi what is the recommended way of running tests on trunk now? I tried: test/runtests.py from inside ipython with %run, didn't work IPython/testing/iptest called as iptest all, didn't work nosetests Seems to work but has 19 errors and 5 failures I'd like to see ipy_stock_completers resurrected. I can look into it but is there some example of how an extension should be converted from the old to the new style? /J?rgen From fperez.net at gmail.com Thu Jan 7 15:46:09 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 7 Jan 2010 12:46:09 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B46305E.5010405@bostream.nu> References: <4B46305E.5010405@bostream.nu> Message-ID: Hey Jorgen, On Thu, Jan 7, 2010 at 11:05 AM, J?rgen Stenarson wrote: > > what is the recommended way of running tests on trunk now? > > I tried: > test/runtests.py ? ? ? ?from inside ipython with %run, didn't work > IPython/testing/iptest ?called as iptest all, didn't work > nosetests ? ? ? ? ? ? ? Seems to work but has 19 errors and 5 failures My trunk-dev branch: https://code.launchpad.net/~fdo.perez/ipython/trunk-dev is more test-friendly than trunk right now. Once reviewed (feel free to pitch in, I'd appreciate it!), we'll merge it into trunk with whatever changes are needed. Still, with that branch, the best way to run the tests is just to type at the command line ./iptest from the IPython/scripts directory. We'll need to make this easier to run later though, at least something like python -c "import IPython;IPython.test()" should just work. Testing *inside* ipython itself will always be tricky to make work because of recursiveness issues, but at least the above is doable. > I'd like to see ipy_stock_completers resurrected. I can look into it > but is there some example of how an extension should be converted from > the old to the new style? If you look in IPython/extensions, you'll see two there that work, pretty.py is probably a good place to start. Cheers, f From fperez.net at gmail.com Fri Jan 8 18:42:11 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 8 Jan 2010 15:42:11 -0800 Subject: [IPython-dev] Unifying command-line conventions for 0.11+: single-dash OK? Message-ID: Hi folks, we currently have in our scripts a mix of -option and --option, and we're thinking of unifying things as we move forward. Now that we're using argparse everywhere, it's very easy to choose what we like, and my take on this is that the extra - at the beginning is just unnecessary typing. For old code that used getopt it was necessary, because getopt interprets -xyz as -x -y -z, but we don't have that restriction. So the question is: does anyone object to using -option (with -option-with-more-than-one-word for long ones) as our spelling for all command-line flags? This may require some adjustment of aliases and scripts you may have, but it seems worth cleaning up now as we do other adjustments for 0.11 an onwards. Cheers, f ps - our recent work on this front actually resulted in a small argparse patch: http://groups.google.com/group/argparse-devs/browse_thread/thread/2c03e4b3d28a3203 so I hope we'll be not only carrying fully updated argparse from upstream, but that ultimately (when argparse makes it into python itself) we'll be able to use the standard one without modifications, as our fixes may well be there. From robert.kern at gmail.com Fri Jan 8 19:09:39 2010 From: robert.kern at gmail.com (Robert Kern) Date: Fri, 08 Jan 2010 18:09:39 -0600 Subject: [IPython-dev] Unifying command-line conventions for 0.11+: single-dash OK? In-Reply-To: References: Message-ID: On 2010-01-08 17:42 PM, Fernando Perez wrote: > Hi folks, > > we currently have in our scripts a mix of -option and --option, and > we're thinking of unifying things as we move forward. Now that we're > using argparse everywhere, it's very easy to choose what we like, and > my take on this is that the extra - at the beginning is just > unnecessary typing. For old code that used getopt it was necessary, > because getopt interprets -xyz as -x -y -z, but we don't have that > restriction. I prefer the -o/--option standard, but mostly because it *is* standard. I implicitly associate -option with old/unprofessional software that hacked up its own argument parsing because it could. -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco From fperez.net at gmail.com Fri Jan 8 19:13:24 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 8 Jan 2010 16:13:24 -0800 Subject: [IPython-dev] Unifying command-line conventions for 0.11+: single-dash OK? In-Reply-To: References: Message-ID: On Fri, Jan 8, 2010 at 4:06 PM, Steve Steiner (listsin) wrote: > I have an objection in that this would be the only program that I would have that *didn't* follow the --multiple-words-get-double-dash convention. > > I use argparse too and find myself disambiguating the short commands (e.g. -l for link vs. -ls for list) but always stick to the --double-dashes for the long version. > > I think it would be very confusing, and "unique" in a very bad way, to drop that convention for the sake of saving a single dash's worth of typing. > Fair enough (I just saw R. Kern's comment in the same direction). I did some more checking of other programs, and indeed -dash is rather rare. Points taken. Good thing I asked for feedback before touching any code :) Nothing to revert. We'll leave support for -pylab, -*thread simply because those are very widely used already, and even listed in books in print (like the recent MPL book: http://www.packtpub.com/matplotlib-python-development/book). But --pylab and --*thread are also there, and all new options will be -- then. Thanks for the feedback! Cheers, f From fperez.net at gmail.com Fri Jan 8 22:24:36 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 8 Jan 2010 19:24:36 -0800 Subject: [IPython-dev] [newbie] stepping through code using the 1/0 trick? In-Reply-To: References: <4980.48416.qm@web111808.mail.gq1.yahoo.com> Message-ID: On Thu, Jan 7, 2010 at 7:41 AM, Robert Kern wrote: > I believe someone is working on adding this ability to 0.11. I'm not sure about actual progress made on this yet, but I've added it to this existing pdb-related bug so we don't lose track of it: https://bugs.launchpad.net/ipython/+bug/375764 Cheers, f From fperez.net at gmail.com Sat Jan 9 01:37:35 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 8 Jan 2010 22:37:35 -0800 Subject: [IPython-dev] Some minor issues with iPython 0.11 on Python 2.6.4 In-Reply-To: <4B41B7E5.7040201@gmail.com> References: <4B41B7E5.7040201@gmail.com> Message-ID: Hi Pim, On Mon, Jan 4, 2010 at 1:41 AM, Pim Schellart wrote: > I just wanted to mention some minor problems with IPython 0.11 running on > Python 2.6.4. > The first (and most important) issue is a bug (reported as bug 488061 on > launchpad). > If you use list=!command you get deprecation warnings because the popen > command is deprecated in favor of the subprocess module. > I have submitted a patch for this bug but until now of now it is unassigned. Now applied, thanks for the fix! > The other issues are minor. > I would like tab autocomplete to behave the same way as the shell (e.g. > bash). The current behavior lists all files as options when I hit tab to > complete the name of the parent directory. > The current default of asking the user to list or not when to many files are > present violates the principle of minimal input that all unix programs share > (e.g. rm is default instead of rm -i) which slows down a typical workflow. > Also the distinction between two different types of behavior depending on > the number of files in a directory is confusing. > Furthermore when there are not so many files present and the completion > shows the list immedately it scrolls (the output of) your previous commands > off the screen (which means you often have to scroll back to see their > results). > I usually tend to have thousands of data files in my work directories and so > I have to type 'n' whenever I change directories which is quite annoying :). > So in conclusion I would like tab to not complete past a '/' and only list > all possible completions with a second hit of the tab key. Yes, this is annoying, listed here as bug already: https://bugs.launchpad.net/ipython/+bug/299821 We're now making good progress on trunk, and my hope is that it will make it easier for others to contribute further, so that we can get more fixes coming... But this is indeed annoying, thanks for reminding us (I might have a quick go at it if turns out to be an easy fix). > The final issue is that the current IPython puts an empty line before each > prompt. > While this might improve readability in some cases more often it is a waste > of screen space. > In particular when using clear the first line of the terminal is blank. > I would prefer again to stick to the default of most shells and remove the > empty line. This you can already control, it's the prompt separator variable in your config file. With our shiny new config system, simply put in your ~/.ipython/ipython_config.py file: c = get_config() c.InteractiveShell.separate_in = '' and presto, you're done! > Please take these remarks as mere suggestions. > In general I like IPython a lot and I hope to see it evolve into an ever > more useful tool. Thanks for the kind words! Cheers, f From jorgen.stenarson at bostream.nu Sat Jan 9 10:01:20 2010 From: jorgen.stenarson at bostream.nu (=?ISO-8859-1?Q?J=F6rgen_Stenarson?=) Date: Sat, 09 Jan 2010 16:01:20 +0100 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> Message-ID: <4B489A40.4010700@bostream.nu> Fernando Perez skrev 2010-01-07 21:46: > Hey Jorgen, > > On Thu, Jan 7, 2010 at 11:05 AM, J?rgen Stenarson > wrote: >> >> what is the recommended way of running tests on trunk now? >> >> I tried: >> test/runtests.py from inside ipython with %run, didn't work >> IPython/testing/iptest called as iptest all, didn't work >> nosetests Seems to work but has 19 errors and 5 failures > > My trunk-dev branch: > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > > is more test-friendly than trunk right now. Once reviewed (feel free > to pitch in, I'd appreciate it!), we'll merge it into trunk with > whatever changes are needed. Still, with that branch, the best way to > run the tests is just to type at the command line > > ./iptest > This does not seem to work on windows. I just made a branch of your trunk-dev branch and tried to run the tests. You can see the results below. In case your swedish is a bit rusty;-) it complains that iptest is not a command or batchfile. Any ideas on what to do? /J?rgen C:\python\external\ipython-perez\IPython>python testing\iptest.py ***************************************************************************** IPython test group: IPython.utils c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando, program eller kommandofil. ***************************************************************************** IPython test group: IPython.scripts c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.lib c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.frontend c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.extensions c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.core c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.config c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** IPython test group: IPython.testing c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett internt kommando, externt kommando,program eller kommandofil. ***************************************************************************** Ran 8 test groups in 1.266s ERROR - 8 out of 8 test groups failed. ---------------------------------------- Runner failed: IPython.lib You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.lib ---------------------------------------- Runner failed: IPython.scripts You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.scripts ---------------------------------------- Runner failed: IPython.utils You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.utils ---------------------------------------- Runner failed: IPython.frontend You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.frontend ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.extensions ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.core ---------------------------------------- Runner failed: IPython.config You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.config ---------------------------------------- Runner failed: IPython.testing You may wish to rerun this one individually, with: c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.testing C:\python\external\ipython-perez\IPython> From ellisonbg.net at gmail.com Sat Jan 9 13:38:05 2010 From: ellisonbg.net at gmail.com (Brian Granger) Date: Sat, 9 Jan 2010 10:38:05 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B489A40.4010700@bostream.nu> References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> Message-ID: <6ce0ac131001091038p355905ccr77d2ab81cc6fb013@mail.gmail.com> Jorgen, OK, this should be easy to fix as this is a common problem we have on Windows (finding executables). Could you file a bug report for this so we can keep track of it? Thanks, Brian On Sat, Jan 9, 2010 at 7:01 AM, J?rgen Stenarson < jorgen.stenarson at bostream.nu> wrote: > Fernando Perez skrev 2010-01-07 21:46: > > Hey Jorgen, > > > > On Thu, Jan 7, 2010 at 11:05 AM, J?rgen Stenarson > > wrote: > >> > >> what is the recommended way of running tests on trunk now? > >> > >> I tried: > >> test/runtests.py from inside ipython with %run, didn't work > >> IPython/testing/iptest called as iptest all, didn't work > >> nosetests Seems to work but has 19 errors and 5 failures > > > > My trunk-dev branch: > > > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > > > > is more test-friendly than trunk right now. Once reviewed (feel free > > to pitch in, I'd appreciate it!), we'll merge it into trunk with > > whatever changes are needed. Still, with that branch, the best way to > > run the tests is just to type at the command line > > > > ./iptest > > > This does not seem to work on windows. I just made a branch of your > trunk-dev branch and tried to run the tests. You can see the results > below. In case your swedish is a bit rusty;-) it complains that iptest > is not a command or batchfile. > > Any ideas on what to do? > > /J?rgen > > C:\python\external\ipython-perez\IPython>python testing\iptest.py > > ***************************************************************************** > IPython test group: IPython.utils > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando, program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.scripts > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.lib > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.frontend > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.extensions > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.core > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.config > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > ***************************************************************************** > IPython test group: IPython.testing > c:\python\external\ipython-perez\IPython\scripts\iptest ?r inte ett > internt kommando, externt kommando,program eller kommandofil. > > > ***************************************************************************** > Ran 8 test groups in 1.266s > > ERROR - 8 out of 8 test groups failed. > ---------------------------------------- > Runner failed: IPython.lib > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.lib > > ---------------------------------------- > Runner failed: IPython.scripts > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.scripts > > ---------------------------------------- > Runner failed: IPython.utils > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.utils > > ---------------------------------------- > Runner failed: IPython.frontend > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.frontend > > ---------------------------------------- > Runner failed: IPython.extensions > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v > IPython.extensions > > ---------------------------------------- > Runner failed: IPython.core > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.core > > ---------------------------------------- > Runner failed: IPython.config > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.config > > ---------------------------------------- > Runner failed: IPython.testing > You may wish to rerun this one individually, with: > c:\python\external\ipython-perez\IPython\scripts\iptest -v IPython.testing > > > C:\python\external\ipython-perez\IPython> > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ellisonbg.net at gmail.com Sat Jan 9 13:40:26 2010 From: ellisonbg.net at gmail.com (Brian Granger) Date: Sat, 9 Jan 2010 10:40:26 -0800 Subject: [IPython-dev] Unifying command-line conventions for 0.11+: single-dash OK? In-Reply-To: References: Message-ID: <6ce0ac131001091040n6a3d4992yd87b19e341d81fc1@mail.gmail.com> On Fri, Jan 8, 2010 at 4:13 PM, Fernando Perez wrote: > On Fri, Jan 8, 2010 at 4:06 PM, Steve Steiner (listsin) > wrote: > > I have an objection in that this would be the only program that I would > have that *didn't* follow the --multiple-words-get-double-dash convention. > > > > I use argparse too and find myself disambiguating the short commands > (e.g. -l for link vs. -ls for list) but always stick to the --double-dashes > for the long version. > > > > I think it would be very confusing, and "unique" in a very bad way, to > drop that convention for the sake of saving a single dash's worth of typing. > > > > Fair enough (I just saw R. Kern's comment in the same direction). I > did some more checking of other programs, and indeed -dash is rather > rare. Points taken. > Yes, I looked around some too at various projects, and the --dash is more common, so I think it is a good idea of stick to that. > Good thing I asked for feedback before touching any code :) Nothing to > revert. > > We'll leave support for > > -pylab, -*thread > I think this too is a good idea. > simply because those are very widely used already, and even listed in > books in print (like the recent MPL book: > http://www.packtpub.com/matplotlib-python-development/book). But > --pylab and --*thread are also there, and all new options will be -- > then. > > Thanks for the feedback! > > Cheers, Brian > Cheers, > > f > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ellisonbg.net at gmail.com Sat Jan 9 13:47:01 2010 From: ellisonbg.net at gmail.com (Brian Granger) Date: Sat, 9 Jan 2010 10:47:01 -0800 Subject: [IPython-dev] RFC on policy: setting object attributes? In-Reply-To: <20091230220223.GB14951@phare.normalesup.org> References: <20091230220223.GB14951@phare.normalesup.org> Message-ID: <6ce0ac131001091047l2421d6eav4aeb39356dc2a834@mail.gmail.com> I agree that putting attributes in class is a good idea. Cheers, Brian On Wed, Dec 30, 2009 at 2:02 PM, Gael Varoquaux < gael.varoquaux at normalesup.org> wrote: > On Wed, Dec 30, 2009 at 02:00:00PM -0800, Fernando Perez wrote: > > Opinions? For now I'm going to go with putting things in the class as > > I need them, but I'd like to hear people's thoughts so we have a > > policy we're happy with and good reasons to choose it. > > I am fully with you. > > Ga?l > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Sat Jan 9 20:45:34 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 17:45:34 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B489A40.4010700@bostream.nu> References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> Message-ID: On Sat, Jan 9, 2010 at 7:01 AM, J?rgen Stenarson wrote: > This does not seem to work on windows. I just made a branch of your > trunk-dev branch and tried to run the tests. You can see the results below. > In case your swedish is a bit rusty;-) it complains that iptest is not a > command or batchfile. > > Any ideas on what to do? > Could you try updating? I've done a ton of work today on the test suite, and my last commit makes the pretty.py extension tests run and pass on my system: maqroll[scripts]> ./iptest IPython.extensions ---------------------------------------------------------------------- Ran 4 tests in 0.475s OK Let me know if it works any better for you... If it does, use pretty.py and test_pretty as references. Cheers, f From fperez.net at gmail.com Sat Jan 9 22:08:47 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 19:08:47 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch Message-ID: Hi folks, The good news is that things are looking very promising. The code in this branch (review link): https://code.launchpad.net/~fdo.perez/ipython/trunk-dev/+merge/16695 now passes a full test suite, runs and behaves very well as far as I can tell, and has a lot of internal improvements as well as some very nice user-facing things (new config system, run-time support for guis and matplotlib, ...). But I've been coding like a maniac, and there's a LOT in there, so it would be really useful if we could get more eyes on this thing. I know Brian has been pounding on it, but it would be great if others could pitch in to give him a hand as well. Don't be intimidated by the size of the diff, even if you just test it on your system and report, that's *very* useful to us. If you have a launchpad account you can add your review with the above link, but if not just reply on this thread, any feedback you provide will be beneficial. Once this gets merged in, we'll be looking at putting IPython out for broader testing as well, and then we can make a release plan for 0.11. Thanks! f From fperez.net at gmail.com Sat Jan 9 22:31:28 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 19:31:28 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: References: Message-ID: On Sat, Jan 9, 2010 at 7:08 PM, Fernando Perez wrote: > But I've been coding like a maniac, and there's a LOT in there, so it > would be really useful if we could get more eyes on this thing. ?I > know Brian has been pounding on it, but it would be great if others > could pitch in to give him a hand as well. ?Don't be intimidated by > the size of the diff, even if you just test it on your system and > report, that's *very* useful to us. ?If you have a launchpad account > you can add your review with the above link, but if not just reply on > this thread, any feedback you provide will be beneficial. In order to make this as easy as possible for anyone willing to help, I've uploaded source (zip and tgz) as well as an egg and a win32 installer: http://ipython.scipy.org/dist/testing/ Thanks for any help! f From jdh2358 at gmail.com Sat Jan 9 22:38:46 2010 From: jdh2358 at gmail.com (John Hunter) Date: Sat, 9 Jan 2010 21:38:46 -0600 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: References: Message-ID: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> On Sat, Jan 9, 2010 at 9:31 PM, Fernando Perez wrote: > On Sat, Jan 9, 2010 at 7:08 PM, Fernando Perez wrote: > >> But I've been coding like a maniac, and there's a LOT in there, so it >> would be really useful if we could get more eyes on this thing. ?I >> know Brian has been pounding on it, but it would be great if others >> could pitch in to give him a hand as well. ?Don't be intimidated by >> the size of the diff, even if you just test it on your system and >> report, that's *very* useful to us. ?If you have a launchpad account >> you can add your review with the above link, but if not just reply on >> this thread, any feedback you provide will be beneficial. > > In order to make this as easy as possible for anyone willing to help, > I've uploaded source (zip and tgz) as well as an egg and a win32 > installer: > > http://ipython.scipy.org/dist/testing/ > > Thanks for any help! I did a bzr clone lp:~fdo.perez/ipython/trunk-dev on an ubuntu 9.10 box, and when I try a python setup.py build I get the following failure:: msierig at pinchiepie:~/ipython$ python setup.py build Traceback (most recent call last): File "setup.py", line 32, in from IPython.utils.genutils import target_update File "/home/msierig/ipython/IPython/__init__.py", line 41, in from IPython.core.iplib import InteractiveShell File "/home/msierig/ipython/IPython/core/iplib.py", line 47, in from IPython.core.magic import Magic File "/home/msierig/ipython/IPython/core/magic.py", line 59, in from IPython.testing import decorators as testdec File "/home/msierig/ipython/IPython/testing/decorators.py", line 68, in from ipunittest import ipdoctest, ipdocstring File "/home/msierig/ipython/IPython/testing/ipunittest.py", line 54, in from . import globalipapp ImportError: cannot import name globalipapp Must be missing something obvious.... JDHcd From fperez.net at gmail.com Sat Jan 9 22:40:51 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 19:40:51 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> References: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> Message-ID: On Sat, Jan 9, 2010 at 7:38 PM, John Hunter wrote: > > > > Must be missing something obvious.... Well, I am :) I just forgot to commit the file! Do a bzr pull again, it should be there now. Thanks! f From jdh2358 at gmail.com Sat Jan 9 22:53:07 2010 From: jdh2358 at gmail.com (John Hunter) Date: Sat, 9 Jan 2010 21:53:07 -0600 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: References: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> Message-ID: <88e473831001091953y2c18c001xd980232e640120e6@mail.gmail.com> On Sat, Jan 9, 2010 at 9:40 PM, Fernando Perez wrote: > On Sat, Jan 9, 2010 at 7:38 PM, John Hunter wrote: >> >> >> >> Must be missing something obvious.... > > Well, I am :) ?I just forgot to commit the file! > > Do a bzr pull again, it should be there now. > > Thanks! OK, that worked. The following worked for me using svn mpl and your ipy branch (line 6 generated a plot)-- nice work! msierig at pinchiepie:~$ PYTHONPATH=~/dev/lib/python2.6/site-packages/ ~/dev/bin/ipython Python 2.6.4 (r264:75706, Dec 7 2009, 18:45:15) Type "copyright", "credits" or "license" for more information. IPython 0.11.bzr.r1219 -- An enhanced Interactive Python. ? -> Introduction and overview of IPython's features. %quickref -> Quick reference. help -> Python's own help system. object? -> Details about 'object'. ?object also works, ?? prints more. In [1]: 1+1 Out[1]: 2 In [2]: plot([1,2,3]) --------------------------------------------------------------------------- NameError Traceback (most recent call last) /home/msierig/ in () NameError: name 'plot' is not defined In [4]: help pylab ------> help(pylab) --------------------------------------------------------------------------- NameError Traceback (most recent call last) /home/msierig/ in () NameError: name 'pylab' is not defined In [5]: %pylab Welcome to pylab, a matplotlib-based Python environment [backend: GTKAgg]. For more information, type 'help(pylab)'. In [6]: plot([1,2,3]) /home/msierig/dev/lib/python2.6/site-packages/matplotlib/backends/backend_gtk.py:621: DeprecationWarning: Use the new widget gtk.Tooltip self.tooltips = gtk.Tooltips() Out[6]: [] In [7]: matplotlib.rcParams['backend'] Out[7]: 'GTKAgg' In [8]: From fperez.net at gmail.com Sat Jan 9 22:57:27 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 19:57:27 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: <88e473831001091953y2c18c001xd980232e640120e6@mail.gmail.com> References: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> <88e473831001091953y2c18c001xd980232e640120e6@mail.gmail.com> Message-ID: On Sat, Jan 9, 2010 at 7:53 PM, John Hunter wrote: > OK, that worked. ?The following worked for me using svn mpl and your > ipy branch (line 6 generated a plot)-- nice work! Great, thanks! This is on ubuntu, right? > msierig at pinchiepie:~$ PYTHONPATH=~/dev/lib/python2.6/site-packages/ > ~/dev/bin/ipython > Python 2.6.4 (r264:75706, Dec ?7 2009, 18:45:15) > Type "copyright", "credits" or "license" for more information. > > IPython 0.11.bzr.r1219 -- An enhanced Interactive Python. > ? ? ? ? ? -> Introduction and overview of IPython's features. > %quickref -> Quick reference. > help ? ? ?-> Python's own help system. > object? ? -> Details about 'object'. ?object also works, ?? prints more. > > In [1]: 1+1 > Out[1]: 2 > > In [2]: plot([1,2,3]) > --------------------------------------------------------------------------- > NameError ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Traceback (most recent call last) > > /home/msierig/ in () > > NameError: name 'plot' is not defined > > In [4]: help pylab > ------> help(pylab) > --------------------------------------------------------------------------- > NameError ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Traceback (most recent call last) > > /home/msierig/ in () > > NameError: name 'pylab' is not defined BTW, since it's a magic, typing pylab? will give you info (help doesn't know anything about magics). Cheers, f From jdh2358 at gmail.com Sat Jan 9 23:12:03 2010 From: jdh2358 at gmail.com (John Hunter) Date: Sat, 9 Jan 2010 22:12:03 -0600 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: References: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> <88e473831001091953y2c18c001xd980232e640120e6@mail.gmail.com> Message-ID: <88e473831001092012v48653208n32f670747a7c2586@mail.gmail.com> On Sat, Jan 9, 2010 at 9:57 PM, Fernando Perez wrote: > On Sat, Jan 9, 2010 at 7:53 PM, John Hunter wrote: >> OK, that worked. ?The following worked for me using svn mpl and your >> ipy branch (line 6 generated a plot)-- nice work! > > Great, thanks! ?This is on ubuntu, right? Yes Karmic (9.10) >> NameError: name 'pylab' is not defined > > BTW, since it's a magic, typing > > pylab? > will give you info (help doesn't know anything about magics). Ahh that is good to know. I wasn't aware that "help" in ipython did not have the extra functionality of "?" and was wondering why the magic "pylab" had no docs before loaded.... Thanks for the clarification. I was also able to test this with the default ubuntu 9.10 mpl and it worked fine. Backend switching does not appear to work with "%pylab qt" and then "%pylab wx" or "%pylab tk" but I had read in a previous thread that there should be some limited support for backend switching.... What is the status here? msierig at pinchiepie:~$ PYTHONPATH=~/dev/lib/python2.6/site-packages/ ~/dev/bin/ipython Python 2.6.4 (r264:75706, Dec 7 2009, 18:45:15) Type "copyright", "credits" or "license" for more information. IPython 0.11.bzr.r1219 -- An enhanced Interactive Python. ? -> Introduction and overview of IPython's features. %quickref -> Quick reference. help -> Python's own help system. object? -> Details about 'object'. ?object also works, ?? prints more. In [1]: %pylab qt Welcome to pylab, a matplotlib-based Python environment [backend: Qt4Agg]. For more information, type 'help(pylab)'. In [2]: plot([1,2,3]) Out[2]: [] In [3]: %pylab tk /usr/lib/pymodules/python2.6/matplotlib/__init__.py:835: UserWarning: This call to matplotlib.use() has no effect because the the backend has already been chosen; matplotlib.use() must be called *before* pylab, matplotlib.pyplot, or matplotlib.backends is imported for the first time. if warn: warnings.warn(_use_error_msg) Welcome to pylab, a matplotlib-based Python environment [backend: TkAgg]. For more information, type 'help(pylab)'. In [4]: figure() Out[4]: In [5]: matplotlib.rcParams['backend'] Out[5]: 'Qt4Agg' In [6]: JDH From fperez.net at gmail.com Sun Jan 10 01:05:28 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 22:05:28 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: <88e473831001092012v48653208n32f670747a7c2586@mail.gmail.com> References: <88e473831001091938u34ca064ak1dccbeaaa5775925@mail.gmail.com> <88e473831001091953y2c18c001xd980232e640120e6@mail.gmail.com> <88e473831001092012v48653208n32f670747a7c2586@mail.gmail.com> Message-ID: 2010/1/9 John Hunter : > > I was also able to test this with the default ubuntu 9.10 mpl and it > worked fine. ?Backend switching does not appear to work with "%pylab > qt" and then "%pylab wx" or "%pylab tk" but I had read in a previous > thread that there should be some limited support for backend > switching.... ?What is the status here? This particular ball is on your court: http://sourceforge.net/mailarchive/forum.php?thread_name=db6b5ecc0912302040i2580ddf1g454c8eeaa017b50d%40mail.gmail.com&forum_name=matplotlib-devel Basically, just comment out in backends/__init__.py:use() this: if 'matplotlib.backends' in sys.modules: if warn: warnings.warn(_use_error_msg) return And uncomment in IPython/core/pylabtools.py these two lines: # XXX For now leave this commented out, but depending on discussions with # mpl-dev, we may be able to allow interactive switching... #import matplotlib.pyplot #matplotlib.pyplot.switch_backend(backend) Then, you should be able to switch backends. We'll need to experiment to figure out which ones work and which don't, and eventually we may leave some of this available to end users. Let me know how it goes, in my experience, qt4/wx seem to switch OK, tk and gtk lock up. Cheers, f From fperez.net at gmail.com Sun Jan 10 02:03:50 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 9 Jan 2010 23:03:50 -0800 Subject: [IPython-dev] Ipython 0.10 pygtk main loop In-Reply-To: <290ea1b00912070830p53f0c068x6cec00450a1cc5a9@mail.gmail.com> References: <290ea1b00912070752j616ae6a9m2b55d38816556ea3@mail.gmail.com> <20091207160623.GA6519@phare.normalesup.org> <290ea1b00912070830p53f0c068x6cec00450a1cc5a9@mail.gmail.com> Message-ID: Hi Michele, On Mon, Dec 7, 2009 at 8:30 AM, Michele Mattioni wrote: > Hi Ga?l, > > I tried that but it seems the backend is correct. > There is a strange disclaimer which is printend when ipython is started with > the pylab (or gthread) options as shown on this bug: > > https://bugs.launchpad.net/ipython/+bug/270856 > > Can be that the case? See my comments there... Ideally, what we want to know is whether this works with 0.11, where the gui support is much cleaner. If you can test (currently with my trunk-dev branch or the downloads at http://ipython.scipy.org/dist/testing/) and let us know how it goes it would be great. I tried but on my Ubuntu 9.10 system your code segfaults immediately (without ipython at all): maqroll[lib]> python ~/Desktop/hellovisual.py Segmentation fault so I can't verify here... Cheers, f From eudoxos at arcig.cz Sun Jan 10 05:42:34 2010 From: eudoxos at arcig.cz (=?UTF-8?Q?V=C3=A1clav_=C5=A0milauer?=) Date: Sun, 10 Jan 2010 02:42:34 -0800 (PST) Subject: [IPython-dev] ipython sphinx directive In-Reply-To: References: <88e473830911061424y35e0d87bxc84eaa0763cd6d8c@mail.gmail.com> <20091107095102.GA2068@phare.normalesup.org> <88e473830911070721i22622c93g62ad2bd9c671eefc@mail.gmail.com> <88e473830911081018t312c61dbt1caac66eee6b5dcd@mail.gmail.com> <88e473830911081022q2e2af3a9i78df1a4213746dcc@mail.gmail.com> <88e473830912130636tab7caa9k6daf950c412dba56@mail.gmail.com> Message-ID: <27097074.post@talk.nabble.com> Fernando Perez wrote: > > I've added this to a new branch I just made: > > https://code.launchpad.net/~ipython-dev/ipython/0.10.1 > I found a problem with rc_override due to fact that EmbeddedSphinxShell is instantiated at import time. This patch works around the issue (diff'd against the branch you mention): --- ipython_directive.py.orig 2010-01-10 11:20:40.993799492 +0100 +++ ipython_directive.py 2010-01-10 11:28:29.863710962 +0100 @@ -25,6 +25,7 @@ prompt_in2=" .\D..", prompt_out=" -> [\#]:" ) + id.reconfig_shell() import ipython_console_highlighting as ich ich.IPythonConsoleLexer.input_prompt= @@ -213,7 +214,7 @@ argv, self.user_ns, self.user_glocal_ns, embedded=True, #shell_class=IPython.Shell.InteractiveShell, shell_class=MatplotlibShell, - rc_override = dict(colors = 'NoColor'), **rc_override) + rc_override = dict(colors = 'NoColor', **rc_override)) self.input = '' self.output = '' @@ -402,6 +403,13 @@ # ipython_directive itself. shell = EmbeddedSphinxShell() +def reconfig_shell(): + """Called after setting module-level variables to re-instantiate + with the set values (since shell is instantiated first at import-time + when module variables have default values)""" + global shell + shell = EmbeddedSphinxShell() + def ipython_directive(name, arguments, options, content, lineno, content_offset, block_text, state, state_machine, -- View this message in context: http://old.nabble.com/ipython-sphinx-directive-tp26239546p27097074.html Sent from the IPython - Development mailing list archive at Nabble.com. From fperez.net at gmail.com Sun Jan 10 11:05:27 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 10 Jan 2010 08:05:27 -0800 Subject: [IPython-dev] ipython sphinx directive In-Reply-To: <27097074.post@talk.nabble.com> References: <88e473830911061424y35e0d87bxc84eaa0763cd6d8c@mail.gmail.com> <20091107095102.GA2068@phare.normalesup.org> <88e473830911070721i22622c93g62ad2bd9c671eefc@mail.gmail.com> <88e473830911081018t312c61dbt1caac66eee6b5dcd@mail.gmail.com> <88e473830911081022q2e2af3a9i78df1a4213746dcc@mail.gmail.com> <88e473830912130636tab7caa9k6daf950c412dba56@mail.gmail.com> <27097074.post@talk.nabble.com> Message-ID: Hi Vaclav, 2010/1/10 V?clav ?milauer : > --- ipython_directive.py.orig ? 2010-01-10 11:20:40.993799492 +0100 > +++ ipython_directive.py ? ? ? ?2010-01-10 11:28:29.863710962 +0100 Could you please resend your patch as an attachment? It was likely mangled by a mail client (yours or mine, I don't know), because it doesn't apply against the current branch: maqroll[sphinxext]> patch -p0 < ../../directive.patch patching file ipython_directive.py Hunk #1 FAILED at 25. Hunk #2 FAILED at 214. Hunk #3 succeeded at 403 with fuzz 1. patch unexpectedly ends in middle of line 2 out of 3 hunks FAILED -- saving rejects to file ipython_directive.py.rej patch unexpectedly ends in middle of line just do bzr diff > filename.patch and send me that so I can apply it. Thanks! f From fperez.net at gmail.com Sun Jan 10 12:55:24 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 10 Jan 2010 09:55:24 -0800 Subject: [IPython-dev] ipython sphinx directive In-Reply-To: <1263143370.3103.44.camel@flux> References: <88e473830911061424y35e0d87bxc84eaa0763cd6d8c@mail.gmail.com> <88e473830911081018t312c61dbt1caac66eee6b5dcd@mail.gmail.com> <88e473830911081022q2e2af3a9i78df1a4213746dcc@mail.gmail.com> <88e473830912130636tab7caa9k6daf950c412dba56@mail.gmail.com> <27097074.post@talk.nabble.com> <1263143370.3103.44.camel@flux> Message-ID: 2010/1/10 V?clav ?milauer : > here it is (I was pasting that into nabble web interface, which might be > the cause; I always fear that if I reply to list message from mail > client, the list manager will not pickup the thread and start new one > instead). It is patch just for that one file. Thanks! Applied: http://bazaar.launchpad.net/~ipython-dev/ipython/0.10.1/revision/1221 Cheers, f From fperez.net at gmail.com Sun Jan 10 13:51:03 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 10 Jan 2010 10:51:03 -0800 Subject: [IPython-dev] A request for help reviewing/testing a branch In-Reply-To: References: Message-ID: Hi all, Updated builds are up at http://ipython.scipy.org/dist/testing/ with source, eggs and win32 binaries from the current state of the branch. To the best of my knowledge, everything works on 2.5 and 2.6, with perhaps one failure for the pretty.py extension in 2.5 I haven't had time to track down. But I'm done for now, I'm sorely behind on much other work, so I'll wait for feedback/review on this before proposing any plans for 0.11. Cheers, f From jorgen.stenarson at bostream.nu Mon Jan 11 14:36:40 2010 From: jorgen.stenarson at bostream.nu (=?ISO-8859-1?Q?J=F6rgen_Stenarson?=) Date: Mon, 11 Jan 2010 20:36:40 +0100 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> Message-ID: <4B4B7DC8.8080206@bostream.nu> Fernando Perez skrev 2010-01-10 02:45: > On Sat, Jan 9, 2010 at 7:01 AM, J?rgen Stenarson > wrote: >> This does not seem to work on windows. I just made a branch of your >> trunk-dev branch and tried to run the tests. You can see the results below. >> In case your swedish is a bit rusty;-) it complains that iptest is not a >> command or batchfile. >> >> Any ideas on what to do? >> > > Could you try updating? I've done a ton of work today on the test > suite, and my last commit makes the pretty.py extension tests run and > pass on my system: > > maqroll[scripts]> ./iptest IPython.extensions > ---------------------------------------------------------------------- > Ran 4 tests in 0.475s > > OK > > Let me know if it works any better for you... If it does, use > pretty.py and test_pretty as references. > > Cheers, > > f > I often run IPython/testing/iptest.py as a script but this is not possible now due to a relative import: C:\python\external\ipython-perez\IPython>python testing\iptest.py Traceback (most recent call last): File "testing\iptest.py", line 39, in from . import globalipapp ValueError: Attempted relative import in non-package Using my usual trick of just adding a py extension to iptest in python26/Scripts so I can use iptest at the prompt I get a similar failure when trying to run all tests. It looks like iptest tries to launch a new iptest for every test group. /J?rgen C:\python\external\ipython-perez\IPython>iptest ***************************************************************************** IPython test group: IPython.utils python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.scripts python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.lib python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.frontend python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.extensions python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.core python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.config python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** IPython test group: IPython.testing python: can't open file 'C:\Python26\lib\site-packages\IPython\scripts\iptest': [Errno 2] No such file or directory ***************************************************************************** Ran 8 test groups in 2.860s ERROR - 8 out of 8 test groups failed. ---------------------------------------- Runner failed: IPython.lib You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.lib ---------------------------------------- Runner failed: IPython.scripts You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.scripts ---------------------------------------- Runner failed: IPython.utils You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.utils ---------------------------------------- Runner failed: IPython.frontend You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.frontend ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.extension s ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.core ---------------------------------------- Runner failed: IPython.config You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.config ---------------------------------------- Runner failed: IPython.testing You may wish to rerun this one individually, with: python C:\Python26\lib\site-packages\IPython\scripts\iptest -v IPython.testing /J?rgen From fperez.net at gmail.com Mon Jan 11 15:27:34 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 11 Jan 2010 12:27:34 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B4B7DC8.8080206@bostream.nu> References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> Message-ID: Hi Jorgen, On Mon, Jan 11, 2010 at 11:36 AM, J?rgen Stenarson wrote: > I often run IPython/testing/iptest.py as a script but this is not possible > now due to a relative import: > > C:\python\external\ipython-perez\IPython>python testing\iptest.py > Traceback (most recent call last): > ?File "testing\iptest.py", line 39, in > ? ?from . import globalipapp > ValueError: Attempted relative import in non-package Argh, yes. This is unfortunately only fixed for 2.6: http://www.python.org/dev/peps/pep-0366/ We might just add that boilerplate if __name__ == "__main__" and __package__ is None: __package__ = "expected.package.name" protected by a version check, so your approach continues to work. I've started using absolute imports because they will become default eventually, and they do provide some other benefits (see pep if curious). For the most part they are innocuous, but this little glitch will need addressing, thanks. Now tracked: https://bugs.launchpad.net/ipython/+bug/506112 > Using my usual trick of just adding a py extension to iptest in > python26/Scripts so I can use iptest at the prompt I get a similar failure > when trying to run all tests. It looks like iptest tries to launch a new > iptest for every test group. > > > /J?rgen > > C:\python\external\ipython-perez\IPython>iptest > ***************************************************************************** > IPython test group: IPython.utils > python: can't open file > 'C:\Python26\lib\site-packages\IPython\scripts\iptest': > [Errno 2] No such file or directory Mmmh, if you can track the problem, let me know. Otherwise I'll try to get to a Windows VM as soon as possible, but it may take me a couple of days. Question: had you installed ipython, or where you running from a source tree? Cheers, f From fperez.net at gmail.com Mon Jan 11 15:52:24 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 11 Jan 2010 12:52:24 -0800 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> Message-ID: On Mon, Jan 11, 2010 at 12:27 PM, Fernando Perez wrote: > > I've started using absolute imports because they will become default > eventually, and they do provide some other benefits (see pep if > curious). ?For the most part they are innocuous, but this little > glitch will need addressing, thanks. ?Now tracked: > > https://bugs.launchpad.net/ipython/+bug/506112 > Actually it looks like I'm going to need some help here. I tried changing the bottom to: # Relative imports and running a module as a script collide. For Python2.6, # there's a fix: if sys.version_info[:2] >= (2,6): if __name__ == "__main__" and __package__ is None: __package__ = "IPython.testing" main() # For 2.5 unfortunately this just doesn't work. Our iptest script provides an # entry point though... else: if __name__ == '__main__': main() But trying to run it as a script, with or without -m, still fails: uqbar[testing]> python -m iptest.py Traceback (most recent call last): File "/usr/lib/python2.6/runpy.py", line 104, in _run_module_as_main loader, code, fname = _get_module_details(mod_name) File "/usr/lib/python2.6/runpy.py", line 79, in _get_module_details loader = get_loader(mod_name) File "/usr/lib/python2.6/pkgutil.py", line 456, in get_loader return find_loader(fullname) File "/usr/lib/python2.6/pkgutil.py", line 466, in find_loader for importer in iter_importers(fullname): File "/usr/lib/python2.6/pkgutil.py", line 422, in iter_importers __import__(pkg) File "iptest.py", line 39, in from . import globalipapp ValueError: Attempted relative import in non-package uqbar[testing]> python iptest.py Traceback (most recent call last): File "iptest.py", line 39, in from . import globalipapp ValueError: Attempted relative import in non-package uqbar[testing]> I'm obviously missing something in the explanation of http://www.python.org/dev/peps/pep-0366/ Can someone enlighten me? Otherwise I'll just back off the absolute imports for that module, it may not be worth the hassle for now (at least until we understand how to do it right). Thanks! f From jorgen.stenarson at bostream.nu Mon Jan 11 16:13:06 2010 From: jorgen.stenarson at bostream.nu (=?ISO-8859-1?Q?J=F6rgen_Stenarson?=) Date: Mon, 11 Jan 2010 22:13:06 +0100 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> Message-ID: <4B4B9462.1020103@bostream.nu> Fernando Perez skrev 2010-01-11 21:27: > Question: had you installed ipython, or where you running from a source tree? > I have tried both approaches with the same result. /J?rgen From fperez.net at gmail.com Wed Jan 13 05:19:44 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Jan 2010 02:19:44 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help Message-ID: Hi folks, I've managed to make enough progress on this machinery that I think we can soon start looking at what's needed to call it 0.11. But we could really use some help with review... We're trying to really do a good job of looking at each other's work, but I'm afraid I've made things a bit painful for Brian, who lately has taken the brunt of this. But reviewing and critiquing each other's work is the best way to improve the quality of the project, and I really think we have something very good coming along... Right now the merge request for my trunk-dev branch: https://code.launchpad.net/~fdo.perez/ipython/trunk-dev at https://code.launchpad.net/~fdo.perez/ipython/trunk-dev/+merge/16695 shows a diff of ~11400 lines, the result of very little sleep in about 12 days. There's simply no way that anyone can possibly review all that line by line, but some testing and overall feedback would be very welcome, from any of you. Very importantly, our test status is, as shown in r1322: As of this revision, we're down to 0 errors on linux and 1 on windows (that appears twice). The windows error is tracked here: https://bugs.launchpad.net/ipython/+bug/506839 And I'm pretty sure the windows error will be trivial for Brian to fix, and it only shows if you have twisted. I ran the test suite with only py2.6.4, pyreadline and nose, and got 0 errors on windows. I then installed EPD 6 and ran ipython from the source tree, with the results above. So we're in pretty good shape tests-wise. Furthermore, I've documented the test system, so that from now on we never again have an excuse not to blanket our code with tests: http://ipython.scipy.org/doc/bzr.r1321/development/testing.html In order to make it as easy as possible for anyone interested to test, in addition to the branch, here are various self-contained downloads, including a win32 binary (the ones with bzr.r1321 in the title, the nightlies are from trunk): http://ipython.scipy.org/dist/testing/ At this point, though we do have some regressions and code that hasn't been ported over, there are enough improvements both user-facing and structural, that I think we can begin thinking about 0.11. But for that, we need some eyes on this recent coding blitz... Thanks for any feedback! f From fperez.net at gmail.com Wed Jan 13 11:40:09 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Jan 2010 08:40:09 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <12aaa0811001130656kd5d5226la4c873d0839ceeb5@mail.gmail.com> References: <12aaa0811001130656kd5d5226la4c873d0839ceeb5@mail.gmail.com> Message-ID: Hi, Thanks for the feedback! Comments below. On Wed, Jan 13, 2010 at 6:56 AM, Mikhail Terekhov wrote: > Hi, > > I've downloaded ipython-0.11.bzr.r1321.tar.gz and run > > cd ipython-0.11.bzr.r1321/IPython/scripts > ./iptest -v IPython.utils > > as suggested in the new testing documentation and got an exception: > > Failure: AttributeError ('module' object has no attribute 'utils') ... ERROR > > ====================================================================== > ERROR: Failure: AttributeError ('module' object has no attribute 'utils') > ---------------------------------------------------------------------- > Traceback (most recent call last): > ?File "/usr/lib64/python2.5/site-packages/nose/loader.py", line 368, > in loadTestsFromName > ? ?module = resolve_name(addr.module) > ?File "/usr/lib64/python2.5/site-packages/nose/util.py", line 344, in > resolve_name > ? ?obj = getattr(obj, part) > AttributeError: 'module' object has no attribute 'utils' > > ---------------------------------------------------------------------- > Ran 1 test in 0.001s > > FAILED (errors=1) This looks like it's picking up an older version of IPython. If you simply start ipython, you should see the new version number, do you get that? Something like maqroll[~]> ipython Python 2.6.4 (r264:75706, Dec 7 2009, 18:45:15) Type "copyright", "credits" or "license" for more information. IPython 0.11.bzr.r1321 -- An enhanced Interactive Python. > Then I tried ./iptest -v and got two failures: > ====================================================================== > FAIL: Test that prompts correctly generate after %run > ---------------------------------------------------------------------- > Traceback (most recent call last): > ?File "/usr/lib64/python2.5/site-packages/nose/case.py", line 183, in runTest > ? ?self.test(*self.arg) > ?File "/usr/lib64/python2.5/site-packages/IPython/tests/test_magic.py", > line 250, in test_prompts > ? ?nt.assert_equals(p2[:3], '...') > AssertionError: '\x01\x1b[' != '...' >>> ?raise self.failureException, \ > ? ? ? ? ?(None or '%r != %r' % ('\x01\x1b[', '...')) > > > ====================================================================== > FAIL: Test that object's __del__ methods are called on exit. > ---------------------------------------------------------------------- > Traceback (most recent call last): > ?File "/usr/lib64/python2.5/site-packages/nose/case.py", line 183, in runTest > ? ?self.test(*self.arg) > ?File "/usr/lib64/python2.5/site-packages/IPython/tests/test_magic.py", > line 72, in test_obj_del > ? ?nt.assert_equals(out,'obj_del.py: object A deleted') > AssertionError: '\x1b[?1034hobj_del.py: object A deleted' != > 'obj_del.py: object A deleted' >>> ?raise self.failureException, \ > ? ? ? ? ?(None or '%r != %r' % ('\x1b[?1034hobj_del.py: object A > deleted', 'obj_del.py: object A deleted')) > > > ---------------------------------------------------------------------- > Ran 148 tests in 2.476s > > FAILED (SKIP=13, failures=2) > obj_del.py: object A deleted Mmh, I just realized that iptest is a bit brittle against 'naked' calls with flags but no module arguments. Try without the -v, just ./iptest and let us know how it goes. I'm a bit puzzled that it's showing those prompt control sequences there, we start our test instances in no color mode... > BTW I run python 2.5.2 ?on SuSE 11.0 x86-64 Thanks again for the info, this is very useful. It's exactly the kind of feedback we need. Cheers, f From jdh2358 at gmail.com Wed Jan 13 11:56:29 2010 From: jdh2358 at gmail.com (John Hunter) Date: Wed, 13 Jan 2010 10:56:29 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: Message-ID: <88e473831001130856v2bafd965yf348e52a1d0e2444@mail.gmail.com> On Wed, Jan 13, 2010 at 4:19 AM, Fernando Perez wrote: > Hi folks, > > I've managed to make enough progress on this machinery that I think we > can soon start looking at what's needed to call it 0.11. ?But we could What minimum version of python are you targeting? I just attempted to install at work on python2.4 and got johnh at udesktop191:ipython> python setup.py install --prefix=~/dev Traceback (most recent call last): File "setup.py", line 33, in ? from IPython.utils.genutils import target_update File "/home/udesktop191/johnh/python/bzr/ipython/IPython/__init__.py", line 41 from .config.loader import Config ^ SyntaxError: invalid syntax From fperez.net at gmail.com Wed Jan 13 12:00:15 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Jan 2010 09:00:15 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <88e473831001130856v2bafd965yf348e52a1d0e2444@mail.gmail.com> References: <88e473831001130856v2bafd965yf348e52a1d0e2444@mail.gmail.com> Message-ID: Hey, On Wed, Jan 13, 2010 at 8:56 AM, John Hunter wrote: > > What minimum version of python are you targeting? ?I just attempted to > install at work on python2.4 and got 2.5 :( if sys.version[0:3] < '2.5': raise ImportError('Python Version 2.5 or above is required for IPython.') See: https://bugs.launchpad.net/ipython/+bug/505090 Sorry, but 2.4 is getting way too long in the tooth, and for us the only realistic way to think about 3.x compatibility is to start ditching the older versions. We simply don't have the manpower to maintain a codebase from 2.4 to 3.1... Cheers, f From fperez.net at gmail.com Wed Jan 13 12:09:25 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Jan 2010 09:09:25 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <12aaa0811001130836o31fa1bq5df3699cc274cf5b@mail.gmail.com> References: <12aaa0811001130656kd5d5226la4c873d0839ceeb5@mail.gmail.com> <12aaa0811001130836o31fa1bq5df3699cc274cf5b@mail.gmail.com> Message-ID: Hi, It looks like we crossed emails :) On Wed, Jan 13, 2010 at 8:36 AM, Mikhail Terekhov wrote: > I didn't notice that the failures above are in the IPython-0.10 > installed system wide. Yup :) > That means that I tested the wrong version. Probably the instructions > for testing > should be modified or may be the iptest script could be modified to > take care of setting > PYTHONPATH correctly? ?So, to be sure, I uninstalled 0.10 and rerun the tests. > Obviously I have to set PYTHONPATH manually to do that: Either that or you have to install this download in full, so it's findable. But yes, I need to make those instructions clearer, this is now: https://bugs.launchpad.net/ipython/+bug/507078 > PYTHONPATH=/home/terekhov/python/ipython/ipython-0.11.bzr.r1321 ./iptest > > and get: > .... > ***************************************************************************** > Ran 10 test groups in 35.004s > > ERROR - 1 out of 10 test groups failed. > ---------------------------------------- > Runner failed: IPython.core > You may wish to rerun this one individually, with: > python /home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/scripts/iptest > -v IPython.core Great. This is actually pretty good... > As suggested ?I hen run: > > PYTHONPATH=/home/terekhov/python/ipython/ipython-0.11.bzr.r1321 > ./iptest -v IPython.core > > and get: > ... > > ====================================================================== > FAIL: Test that object's __del__ methods are called on exit. > ---------------------------------------------------------------------- > Traceback (most recent call last): > ?File "/usr/lib64/python2.5/site-packages/nose/case.py", line 183, in runTest > ? ?self.test(*self.arg) > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/testing/decorators.py", > line 225, in skipper_func > ? ?return f(*args, **kwargs) > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/core/tests/test_run.py", > line 160, in test_obj_del > ? ?tt.ipexec_validate(self.fname, 'object A deleted') > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/testing/tools.py", > line 288, in ipexec_validate > ? ?nt.assert_equals(out.strip(), expected_out.strip()) > AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >>> ?raise self.failureException, \ > ? ? ? ? ?(None or '%r != %r' % ('\x1b[?1034hobject A deleted', > 'object A deleted')) > > > ====================================================================== > FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass > ---------------------------------------------------------------------- > Traceback (most recent call last): > ?File "/usr/lib64/python2.5/site-packages/nose/case.py", line 183, in runTest > ? ?self.test(*self.arg) > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/testing/decorators.py", > line 225, in skipper_func > ? ?return f(*args, **kwargs) > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/core/tests/test_run.py", > line 174, in test_tclass > ? ?tt.ipexec_validate(self.fname, out) > ?File "/home/terekhov/python/ipython/ipython-0.11.bzr.r1321/IPython/testing/tools.py", > line 288, in ipexec_validate > ? ?nt.assert_equals(out.strip(), expected_out.strip()) > AssertionError: "\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: > ['C-second']\ntclass.py: deleting object: C-first" != "ARGV 1-: > ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: > C-first" >>> ?raise self.failureException, \ > ? ? ? ? ?(None or '%r != %r' % ("\x1b[?1034hARGV 1-: > ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: > C-first", "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: > deleting object: C-first")) > > > ---------------------------------------------------------------------- > Ran 94 tests in 1.301s > > FAILED (SKIP=1, failures=2) > > BTW a lot of debugging noise is printed to stdout during the tests > which somewhat masks > the real errors. Yes, unfortunately I'm not sure how to improve that yet... I don't quite know where it's coming from, and with all the stdout black magic that doctest does (this isn't nose's fault, it's doctest itself), it's really hard to control stdout. > Out of curiosity I've run: > > PYTHONPATH=/home/terekhov/python/ipython/ipython-0.11.bzr.r1321 ./iptest -v > > ?(notice the '-v') and got two more failures! I.e. 'iptest' and > 'iptest -v' give different number of failures: Ignore 'iptest -v' by itself, that's a bug now: https://bugs.launchpad.net/ipython/+bug/507079 Basically what's reliable is ./iptest # no other options or ./iptest [any nose option] IPython.something But I need to fix that, thanks a lot for your feedback! Cheers, f From fperez.net at gmail.com Wed Jan 13 14:35:40 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 13 Jan 2010 11:35:40 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B4B9462.1020103@bostream.nu> References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> <4B4B9462.1020103@bostream.nu> Message-ID: On Mon, Jan 11, 2010 at 1:13 PM, J?rgen Stenarson wrote: >> > I have tried both approaches with the same result. BTW, try the new code I posted yesterday (if you're running from branch, just bzr pull on it). It should be a lot better, bug me if not. f From warren.weckesser at enthought.com Wed Jan 13 17:25:48 2010 From: warren.weckesser at enthought.com (Warren Weckesser) Date: Wed, 13 Jan 2010 16:25:48 -0600 Subject: [IPython-dev] Digging into IPython Message-ID: <4B4E486C.5070808@enthought.com> Hi, I'm starting to dig into the ipython code, with the eventual goal of updating some of the code in the Enthought Tool Suite to make better ipython plugins. I have a lot of questions, but I'll start with a simple one. What is wrong with the following code? ----- from IPython.iplib import InteractiveShell ip = InteractiveShell("my_name") ipapi = ip.getapi() print "\nnames:" print ipapi.IP.user_ns.keys() print "\ncalling runlines()" ipapi.runlines("a = 'abc'\nx = 999") print "\nnames:" print ipapi.IP.user_ns.keys() ----- I am using version 0.10. When I run this code, I get the attached crash report; the exception that it reports is "AttributeError: 'InteractiveShell' object has no attribute 'log'". My second question: I would like to learn more about the topics here: http://ipython.scipy.org/doc/stable/html/development/notification_blueprint.html Are there any working examples of the use cases described in that page? Thanks, Warren -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: IPython_crash_report.txt URL: From warren.weckesser at enthought.com Wed Jan 13 17:30:23 2010 From: warren.weckesser at enthought.com (Warren Weckesser) Date: Wed, 13 Jan 2010 16:30:23 -0600 Subject: [IPython-dev] Digging into IPython In-Reply-To: <4B4E486C.5070808@enthought.com> References: <4B4E486C.5070808@enthought.com> Message-ID: <4B4E497F.8050204@enthought.com> Oops, I attached the wrong crash report. The correct one is attached here. Warren Warren Weckesser wrote: > Hi, > > I'm starting to dig into the ipython code, with the eventual goal of > updating some of the code in the Enthought Tool Suite to make better > ipython plugins. I have a lot of questions, but I'll start with a > simple one. What is wrong with the following code? > > ----- > from IPython.iplib import InteractiveShell > > ip = InteractiveShell("my_name") > > ipapi = ip.getapi() > > print "\nnames:" > print ipapi.IP.user_ns.keys() > > print "\ncalling runlines()" > ipapi.runlines("a = 'abc'\nx = 999") > > print "\nnames:" > print ipapi.IP.user_ns.keys() > ----- > > I am using version 0.10. When I run this code, I get the attached > crash report; the exception that it reports is "AttributeError: > 'InteractiveShell' object has no attribute 'log'". > > > My second question: I would like to learn more about the topics here: > > http://ipython.scipy.org/doc/stable/html/development/notification_blueprint.html > > Are there any working examples of the use cases described in that page? > > Thanks, > > Warren > > ------------------------------------------------------------------------ > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: IPython_crash_report.txt URL: From gokhansever at gmail.com Wed Jan 13 17:48:52 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Wed, 13 Jan 2010 16:48:52 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: Message-ID: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> Here is one test comes from me: ================================================================================ Platform : Linux-2.6.31.9-174.fc12.i686.PAE-i686-with-fedora-12-Constantine Python : ('CPython', 'tags/r262', '71600') IPython : 0.11.bzr.r1321 ================================================================================ bzr branch lp:~fdo.perez/ipython/trunk-dev python setupegg.py develop python -c "import IPython; IPython.test()" ---------------------------------------------------------------------- Ran 66 tests in 0.142s OK (SKIP=6) ***************************************************************************** Ran 9 test groups in 7.364s ERROR - 3 out of 9 test groups failed. ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python /usr/bin/iptest -v IPython.core ---------------------------------------- Runner failed: IPython.kernel You may wish to rerun this one individually, with: python /usr/bin/iptest -v IPython.kernel ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: python /usr/bin/iptest -v IPython.extensions and the details: *python /usr/bin/iptest -v IPython.core* ====================================================================== FAIL: Test that object's __del__ methods are called on exit. ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", line 160, in test_obj_del tt.ipexec_validate(self.fname, 'object A deleted') File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 288, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >> raise self.failureException, \ (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A deleted')) ====================================================================== FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", line 174, in test_tclass tt.ipexec_validate(self.fname, out) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 288, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: "\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" != "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" >> raise self.failureException, \ (None or '%r != %r' % ("\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first", "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first")) ---------------------------------------------------------------------- Ran 102 tests in 1.003s FAILED (SKIP=1, failures=2) *python /usr/bin/iptest -v IPython.kernel* Failure: AttributeError ('module' object has no attribute 'kernel') ... ERROR ====================================================================== ERROR: Failure: AttributeError ('module' object has no attribute 'kernel') ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/loader.py", line 353, in loadTestsFromName module = resolve_name(addr.module) File "/usr/lib/python2.6/site-packages/nose/util.py", line 310, in resolve_name obj = getattr(obj, part) AttributeError: 'module' object has no attribute 'kernel' ---------------------------------------------------------------------- Ran 1 test in 0.001s FAILED (errors=1) *python /usr/bin/iptest -v IPython.extensions* ====================================================================== FAIL: IPython.extensions.tests.test_pretty.TestPrettyInteractively.test_printers ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/extensions/tests/test_pretty.py", line 101, in test_printers tt.ipexec_validate(self.fname, ipy_out) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 288, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '\x1b[?1034hA()\nB()\n\n' != 'A()\nB()\n\n' >> raise self.failureException, \ (None or '%r != %r' % ('\x1b[?1034hA()\nB()\n\n', 'A()\nB()\n\n')) ---------------------------------------------------------------------- Ran 4 tests in 0.286s FAILED (failures=1) Overall it works nice in my system. However I couldn't get it loading my previous ipy_user_conf.py file? Is there a setting for this? Ahaa, here is one catch (some extra outputs in whos listing --which I was going to ask if it is possible to list the command that we used to construct a object next the Data/Info column in whos listing? Sometimes I search my history to see how I constructed an object. That would be useful to have it listed on a whos listing. In [1]: a = range(10) In [2]: b = [i*i for i in range(5)] In [4]: c = {"x":3, "y":5} In [5]: whos Variable Type Data/Info ----------------------------------------- In InputList ['\n', u'a = range(10)\n'<...>thon().magic("whos ")\n'] Out dict {} a list [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] b list [0, 1, 4, 9, 16] c dict {'y': 5, 'x': 3} get_ipython instancemethod l('ipython.component0')>> i int 4 On Wed, Jan 13, 2010 at 4:19 AM, Fernando Perez wrote: > Hi folks, > > I've managed to make enough progress on this machinery that I think we > can soon start looking at what's needed to call it 0.11. But we could > really use some help with review... We're trying to really do a good > job of looking at each other's work, but I'm afraid I've made things a > bit painful for Brian, who lately has taken the brunt of this. But > reviewing and critiquing each other's work is the best way to improve > the quality of the project, and I really think we have something very > good coming along... > > Right now the merge request for my trunk-dev branch: > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > > at > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev/+merge/16695 > > shows a diff of ~11400 lines, the result of very little sleep in about > 12 days. There's simply no way that anyone can possibly review all > that line by line, but some testing and overall feedback would be very > welcome, from any of you. Very importantly, our test status is, as > shown in r1322: > > As of this revision, we're down to 0 errors on linux and 1 on windows (that > appears twice). The windows error is tracked here: > > https://bugs.launchpad.net/ipython/+bug/506839 > > And I'm pretty sure the windows error will be trivial for Brian to > fix, and it only shows if you have twisted. I ran the test suite with > only py2.6.4, pyreadline and nose, and got 0 errors on windows. I > then installed EPD 6 and ran ipython from the source tree, with the > results above. > > So we're in pretty good shape tests-wise. Furthermore, I've > documented the test system, so that from now on we never again have an > excuse not to blanket our code with tests: > > http://ipython.scipy.org/doc/bzr.r1321/development/testing.html > > In order to make it as easy as possible for anyone interested to test, > in addition to the branch, here are various self-contained downloads, > including a win32 binary (the ones with bzr.r1321 in the title, the > nightlies are from trunk): > > http://ipython.scipy.org/dist/testing/ > > At this point, though we do have some regressions and code that hasn't > been ported over, there are enough improvements both user-facing and > structural, that I think we can begin thinking about 0.11. But for > that, we need some eyes on this recent coding blitz... > > Thanks for any feedback! > > f > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jorgen.stenarson at bostream.nu Thu Jan 14 13:05:23 2010 From: jorgen.stenarson at bostream.nu (=?ISO-8859-1?Q?J=F6rgen_Stenarson?=) Date: Thu, 14 Jan 2010 19:05:23 +0100 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> <4B4B9462.1020103@bostream.nu> Message-ID: <4B4F5CE3.5020704@bostream.nu> Fernando Perez skrev 2010-01-13 20:35: > On Mon, Jan 11, 2010 at 1:13 PM, J?rgen Stenarson > wrote: >>> >> I have tried both approaches with the same result. > > BTW, try the new code I posted yesterday (if you're running from > branch, just bzr pull on it). It should be a lot better, bug me if > not. > > f > There is something strange going on here. I just installed the latest trunk with setup.py install If I just use iptest then I get a bunch of tests run with a very verbose output i.e. one line for every test. With iptest -q I get lots of errors and failures it seems not all testgroups are used. Using the verbosity setting also generates lots of errors. The output of iptest IPython.core looks strange: ............................................................................S... >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> None ........ERROR: File `"c:\docume~1\jstenar\lokala~1\temp\tmpjbspu3.py".py` not fo und. .ERROR: File `"c:\docume~1\jstenar\lokala~1\temp\tmpaawkae.py".py` not found. .ERROR: File `"c:\docume~1\jstenar\lokala~1\temp\tmp88arfl.py".py` not found. .S.S.ERROR: File `"c:\docume~1\jstenar\lokala~1\temp\tmpjjry9e.py".py` not found . . ---------------------------------------------------------------------- Ran 99 tests in 0.454s OK (SKIP=3) Is it possible to run the tests without installing? /J?rgen From fperez.net at gmail.com Thu Jan 14 18:07:02 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 14 Jan 2010 15:07:02 -0800 Subject: [IPython-dev] Digging into IPython In-Reply-To: <4B4E486C.5070808@enthought.com> References: <4B4E486C.5070808@enthought.com> Message-ID: Hi Warren, On Wed, Jan 13, 2010 at 2:25 PM, Warren Weckesser wrote: > I'm starting to dig into the ipython code, with the eventual goal of > updating some of the code in the Enthought Tool Suite to make better ipython > plugins. ?I have a lot of questions, but I'll start with a simple one. ?What > is wrong with the following code? > Since you're just starting out and this is for new work for ETS, let's get off the ground with the new machinery instead. The new api isn't fully baked out yet, but it's much, much nicer in many ways, the functionality is better (in particular for ETS, the GUI support is better), we have the traits-'lite' code we talked about with Eric Jones at Scipy'09 in place now, etc. The branch I've been posting about is in review, so you can work off of it (and any feedback from you will be very useful): https://code.launchpad.net/~fdo.perez/ipython/trunk-dev And if you want docs/binaries: * the docs for the in-testing ipython are at: http://ipython.scipy.org/doc/bzr.r1321/ * and downloads in various formats are at: http://ipython.scipy.org/dist/testing/ * Note: the files with .bzr.r1321 in the names are the relevant ones for today's call, NOT the 'nightly' ones. Please have a go off this code instead, where your time and ours will be much more productively spent than on 0.10, whose APIs are now deprecated. Cheers, f From fperez.net at gmail.com Thu Jan 14 18:23:38 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 14 Jan 2010 15:23:38 -0800 Subject: [IPython-dev] Running tests In-Reply-To: <4B4F5CE3.5020704@bostream.nu> References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> <4B4B9462.1020103@bostream.nu> <4B4F5CE3.5020704@bostream.nu> Message-ID: Hey Jorgen, On Thu, Jan 14, 2010 at 10:05 AM, J?rgen Stenarson wrote: > Fernando Perez skrev 2010-01-13 20:35: >> >> On Mon, Jan 11, 2010 at 1:13 PM, J?rgen Stenarson >> ?wrote: >>>> >>> I have tried both approaches with the same result. >> >> BTW, try the new code I posted yesterday (if you're running from >> branch, just bzr pull on it). ?It should be a lot better, bug me if >> not. >> >> f >> > There is something strange going on here. > > I just installed the latest trunk with setup.py install > > If I just use iptest then I get a bunch of tests run with a very verbose > output i.e. one line for every test. > > With iptest -q I get lots of errors and failures it seems not all testgroups > are used. > > Using the verbosity setting also generates lots of errors. Yes, right now iptest handles options badly, I need to fix that: https://bugs.launchpad.net/ipython/+bug/507079 > The output of iptest IPython.core looks strange: > > ............................................................................S... >>>> I'm totally stumped by that verbosity, I have *no clue* where it's coming from, though I've looked. Tracked now: https://bugs.launchpad.net/ipython/+bug/507705 Those ERROR messages aren't affecting the final result, but I still want them gone: https://bugs.launchpad.net/ipython/+bug/507706 > Is it possible to run the tests without installing? Yes, I was running them yesterday without installing. In Windows, what I did was to add to my PYTHONPATH (Control Panel -> Sysstem-> Advanced -> Environment Variables) the source download directory, and to my PATH C:\Python26\Scripts. With those two things, I went to the IPython/scripts directory, and in a console would type python iptest or python iptest -v IPython.core etc. Note that right now, until I fix the bug abov, you should ONLY run python iptest # without ANY OPTIONS! or python iptest -any-option-you-want IPython.something.something python iptest -any-option-you-want IPython.something.something:somefunc Both of these were working fine for me. I'm also curious, in addition to test-suite problems, is it working OK for you? Cheers, f https://bugs.launchpad.net/ipython/+bug/507078 From warren.weckesser at enthought.com Thu Jan 14 18:27:09 2010 From: warren.weckesser at enthought.com (Warren Weckesser) Date: Thu, 14 Jan 2010 17:27:09 -0600 Subject: [IPython-dev] Digging into IPython In-Reply-To: References: <4B4E486C.5070808@enthought.com> Message-ID: <4B4FA84D.2070305@enthought.com> Fernando Perez wrote: > Hi Warren, > > On Wed, Jan 13, 2010 at 2:25 PM, Warren Weckesser > wrote: > >> I'm starting to dig into the ipython code, with the eventual goal of >> updating some of the code in the Enthought Tool Suite to make better ipython >> plugins. I have a lot of questions, but I'll start with a simple one. What >> is wrong with the following code? >> >> > > Since you're just starting out and this is for new work for ETS, let's > get off the ground with the new machinery instead. Yes, good idea, especially if the 0.10 API is already deprecated. > The new api isn't > fully baked out yet, but it's much, much nicer in many ways, the > functionality is better (in particular for ETS, the GUI support is > better), we have the traits-'lite' code we talked about with Eric > Jones at Scipy'09 in place now, etc. > > The branch I've been posting about is in review, so you can work off > of it (and any feedback from you will be very useful): > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > > And if you want docs/binaries: > > * the docs for the in-testing ipython are at: > http://ipython.scipy.org/doc/bzr.r1321/ > * and downloads in various formats are at: > http://ipython.scipy.org/dist/testing/ > * Note: the files with .bzr.r1321 in the names are the relevant > ones for today's call, NOT the 'nightly' ones. > > > Please have a go off this code instead, where your time and ours will > be much more productively spent than on 0.10, whose APIs are now > deprecated. > > Thanks. I'll be experimenting with this over the next week or so. Expect more questions. :) Warren > Cheers, > > f > From fperez.net at gmail.com Thu Jan 14 18:49:51 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 14 Jan 2010 15:49:51 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> Message-ID: Hi Gokhan, On Wed, Jan 13, 2010 at 2:48 PM, G?khan Sever wrote: > Here is one test comes from me: > > ================================================================================ > Platform???? : > Linux-2.6.31.9-174.fc12.i686.PAE-i686-with-fedora-12-Constantine > Python?????? : ('CPython', 'tags/r262', '71600') > IPython????? : 0.11.bzr.r1321 > ================================================================================ Great, thanks! > > bzr branch lp:~fdo.perez/ipython/trunk-dev > python setupegg.py develop > > python -c "import IPython; IPython.test()" > > ---------------------------------------------------------------------- > Ran 66 tests in 0.142s > > OK (SKIP=6) > > ***************************************************************************** > Ran 9 test groups in 7.364s > > ERROR - 3 out of 9 test groups failed. > ---------------------------------------- > Runner failed: IPython.core > You may wish to rerun this one individually, with: > python /usr/bin/iptest -v IPython.core > > ---------------------------------------- > Runner failed: IPython.kernel > You may wish to rerun this one individually, with: > python /usr/bin/iptest -v IPython.kernel > > ---------------------------------------- > Runner failed: IPython.extensions > You may wish to rerun this one individually, with: > python /usr/bin/iptest -v IPython.extensions > > > and the details: > > > python /usr/bin/iptest -v IPython.core > ====================================================================== > FAIL: Test that object's __del__ methods are called on exit. > ---------------------------------------------------------------------- > Traceback (most recent call last): > ? File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest > ??? self.test(*self.arg) > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", > line 225, in skipper_func > ??? return f(*args, **kwargs) > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", > line 160, in test_obj_del > ??? tt.ipexec_validate(self.fname, 'object A deleted') > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", > line 288, in ipexec_validate > ??? nt.assert_equals(out.strip(), expected_out.strip()) > AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >>>? raise self.failureException, \ > ????????? (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A > deleted')) Wow, this is weird. It's producing color escapes for the prompts, which I don't see here and have no idea how they are happening, because I'm initializing the internal instance in nocolor mode. Very bizarre. Could you do me a favor? Try the attached patch and see if that helps, I'm a bit stumped. > ====================================================================== > FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass same error as above > python /usr/bin/iptest -v IPython.kernel > Failure: AttributeError ('module' object has no attribute 'kernel') ... > ERROR Huh? That doesn't make any sense. Are you sure you're running off the branch? I don't use 'setup.py develop', so I'm not sure it works correctly. I simply have in my PYTHONPATH a symlink to my source ipython directory, so that import IPython loads the right version. Could you verify that you are getting the right version? > ====================================================================== > ERROR: Failure: AttributeError ('module' object has no attribute 'kernel') > ---------------------------------------------------------------------- > Traceback (most recent call last): > ? File "/usr/lib/python2.6/site-packages/nose/loader.py", line 353, in > loadTestsFromName > ??? module = resolve_name(addr.module) > ? File "/usr/lib/python2.6/site-packages/nose/util.py", line 310, in > resolve_name > ??? obj = getattr(obj, part) > AttributeError: 'module' object has no attribute 'kernel' > > ---------------------------------------------------------------------- > Ran 1 test in 0.001s > > FAILED (errors=1) > > > python /usr/bin/iptest -v IPython.extensions > ====================================================================== > FAIL: > IPython.extensions.tests.test_pretty.TestPrettyInteractively.test_printers Same again as before.. > Overall it works nice in my system. However I couldn't get it loading my > previous ipy_user_conf.py file? Is there a setting for this? The config system is different now: http://ipython.scipy.org/doc/bzr.r1321/config/overview.html#config-overview > Ahaa, here is one catch (some extra outputs in whos listing --which I was > going to ask if it is possible to list the command that we used to construct > a object next the Data/Info column in whos listing? > > Sometimes I search my history to see how I constructed an object. That would > be useful to have it listed on a whos listing. There is no way to know this in python in general, since multiple names can be bound to one object, assignments can happen in many different ways (code can write to namespace dicts directly, etc). But you can write a little extension of yours that walks the In input history list looking for things that look like variable assignments. It won't be very robust, but could be useful. If you end up liking it, put it up on the cookbook and we can ship it later as an optional tool, once you've stress-tested it. Cheers, f -------------- next part -------------- A non-text attachment was scrubbed... Name: testtools.diff Type: text/x-patch Size: 939 bytes Desc: not available URL: From gokhansever at gmail.com Thu Jan 14 20:53:27 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Thu, 14 Jan 2010 19:53:27 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> Message-ID: <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> On Thu, Jan 14, 2010 at 5:49 PM, Fernando Perez wrote: > Hi Gokhan, > > On Wed, Jan 13, 2010 at 2:48 PM, G?khan Sever > wrote: > > Here is one test comes from me: > > > > > ================================================================================ > > Platform : > > Linux-2.6.31.9-174.fc12.i686.PAE-i686-with-fedora-12-Constantine > > Python : ('CPython', 'tags/r262', '71600') > > IPython : 0.11.bzr.r1321 > > > ================================================================================ > > Great, thanks! > > > > bzr branch lp:~fdo.perez/ipython/trunk-dev > > python setupegg.py develop > > > > python -c "import IPython; IPython.test()" > > > > ---------------------------------------------------------------------- > > Ran 66 tests in 0.142s > > > > OK (SKIP=6) > > > > > ***************************************************************************** > > Ran 9 test groups in 7.364s > > > > ERROR - 3 out of 9 test groups failed. > > ---------------------------------------- > > Runner failed: IPython.core > > You may wish to rerun this one individually, with: > > python /usr/bin/iptest -v IPython.core > > > > ---------------------------------------- > > Runner failed: IPython.kernel > > You may wish to rerun this one individually, with: > > python /usr/bin/iptest -v IPython.kernel > > > > ---------------------------------------- > > Runner failed: IPython.extensions > > You may wish to rerun this one individually, with: > > python /usr/bin/iptest -v IPython.extensions > > > > > > and the details: > > > > > > python /usr/bin/iptest -v IPython.core > > ====================================================================== > > FAIL: Test that object's __del__ methods are called on exit. > > ---------------------------------------------------------------------- > > Traceback (most recent call last): > > File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in > runTest > > self.test(*self.arg) > > File > > > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", > > line 225, in skipper_func > > return f(*args, **kwargs) > > File > > > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", > > line 160, in test_obj_del > > tt.ipexec_validate(self.fname, 'object A deleted') > > File > > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", > > line 288, in ipexec_validate > > nt.assert_equals(out.strip(), expected_out.strip()) > > AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' > >>> raise self.failureException, \ > > (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A > > deleted')) > > Wow, this is weird. It's producing color escapes for the prompts, > which I don't see here and have no idea how they are happening, > because I'm initializing the internal instance in nocolor mode. Very > bizarre. > > Could you do me a favor? Try the attached patch and see if that helps, > I'm a bit stumped. > > > ====================================================================== > > FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass > > same error as above > > > python /usr/bin/iptest -v IPython.kernel > > Failure: AttributeError ('module' object has no attribute 'kernel') ... > > ERROR > > Huh? That doesn't make any sense. > > Are you sure you're running off the branch? I don't use 'setup.py > develop', so I'm not sure it works correctly. I simply have in my > PYTHONPATH a symlink to my source ipython directory, so that > > import IPython > > loads the right version. Could you verify that you are getting the > right version? > I did "python setup.py install" after re-pulling your branch. In [2]: IPython.__version__ Out[2]: '0.11.bzr.r1321' applied your patch. Test runs produced the same errors. > > > ====================================================================== > > ERROR: Failure: AttributeError ('module' object has no attribute > 'kernel') > > ---------------------------------------------------------------------- > > Traceback (most recent call last): > > File "/usr/lib/python2.6/site-packages/nose/loader.py", line 353, in > > loadTestsFromName > > module = resolve_name(addr.module) > > File "/usr/lib/python2.6/site-packages/nose/util.py", line 310, in > > resolve_name > > obj = getattr(obj, part) > > AttributeError: 'module' object has no attribute 'kernel' > > > > ---------------------------------------------------------------------- > > Ran 1 test in 0.001s > > > > FAILED (errors=1) > > > > > > python /usr/bin/iptest -v IPython.extensions > > ====================================================================== > > FAIL: > > > IPython.extensions.tests.test_pretty.TestPrettyInteractively.test_printers > > Same again as before.. > > > Overall it works nice in my system. However I couldn't get it loading my > > previous ipy_user_conf.py file? Is there a setting for this? > > The config system is different now: > > http://ipython.scipy.org/doc/bzr.r1321/config/overview.html#config-overview > Thanks for the pointer :) > > > Ahaa, here is one catch (some extra outputs in whos listing --which I was > > going to ask if it is possible to list the command that we used to > construct > > a object next the Data/Info column in whos listing? > > > > Sometimes I search my history to see how I constructed an object. That > would > > be useful to have it listed on a whos listing. > > There is no way to know this in python in general, since multiple > names can be bound to one object, assignments can happen in many > different ways (code can write to namespace dicts directly, etc). But > you can write a little extension of yours that walks the In input > history list looking for things that look like variable assignments. > It won't be very robust, but could be useful. If you end up liking > it, put it up on the cookbook and we can ship it later as an optional > tool, once you've stress-tested it. > > Nice assignment. I will work on once I have some more free time. > Cheers, > > f > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Thu Jan 14 21:02:39 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Thu, 14 Jan 2010 18:02:39 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> Message-ID: On Thu, Jan 14, 2010 at 5:53 PM, G?khan Sever wrote: > I did "python setup.py install" after re-pulling your branch. > > In [2]: IPython.__version__ > Out[2]: '0.11.bzr.r1321' > > applied your patch. Test runs produced the same errors. OK, I'm seriously stumped. I think we have a Fedora 12 box around here, I'll try to test on it and see if I get them too. I have no clue what's going on, it's truly weird. One more thing to try, shouldn't matter but I'm willing to test wild ideas. Move your ~/.ipython dir to some other name for a minute, and try the test again, so ipython can't find any of your user defaults. Let's see if that helps, if it does it would indicate we're picking up user info we shouldn't at test time. Cheers, f From gokhansever at gmail.com Thu Jan 14 21:21:46 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Thu, 14 Jan 2010 20:21:46 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> Message-ID: <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> On Thu, Jan 14, 2010 at 8:02 PM, Fernando Perez wrote: > On Thu, Jan 14, 2010 at 5:53 PM, G?khan Sever > wrote: > > I did "python setup.py install" after re-pulling your branch. > > > > In [2]: IPython.__version__ > > Out[2]: '0.11.bzr.r1321' > > > > applied your patch. Test runs produced the same errors. > > OK, I'm seriously stumped. I think we have a Fedora 12 box around > here, I'll try to test on it and see if I get them too. I have no > clue what's going on, it's truly weird. > > One more thing to try, shouldn't matter but I'm willing to test wild > ideas. Move your ~/.ipython dir to some other name for a minute, and > try the test again, so ipython can't find any of your user defaults. > Let's see if that helps, if it does it would indicate we're picking up > user info we shouldn't at test time. > The same. I can give it try on a Fedora 10 with Python 2.5.2 later. > > Cheers, > > f > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Fri Jan 15 05:04:00 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 15 Jan 2010 02:04:00 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> Message-ID: On Thu, Jan 14, 2010 at 6:21 PM, G?khan Sever wrote: > > The same. Argh... > I can give it try on a Fedora 10 with Python 2.5.2 later. Please do pull first, I fixed a few things that could help... Thanks f From fperez.net at gmail.com Fri Jan 15 05:05:25 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 15 Jan 2010 02:05:25 -0800 Subject: [IPython-dev] Running tests In-Reply-To: References: <4B46305E.5010405@bostream.nu> <4B489A40.4010700@bostream.nu> <4B4B7DC8.8080206@bostream.nu> <4B4B9462.1020103@bostream.nu> <4B4F5CE3.5020704@bostream.nu> Message-ID: On Thu, Jan 14, 2010 at 3:23 PM, Fernando Perez wrote: > > Note that right now, until I fix the bug abov, you should ONLY run > > python iptest ?# without ANY OPTIONS! > > or > > python iptest -any-option-you-want ? IPython.something.something > python iptest -any-option-you-want ? IPython.something.something:somefunc > OK, this isn't true anymore. I also added at the top-level a simple iptest.py script, so you should be able to update the branch and type python iptest.py Let me know if we fare any better... Thanks, f From fperez.net at gmail.com Fri Jan 15 14:22:09 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 15 Jan 2010 11:22:09 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> Message-ID: On Thu, Jan 14, 2010 at 6:02 PM, Fernando Perez wrote: > OK, I'm seriously stumped. ?I think we have a Fedora 12 box around > here, I'll try to test on it and see if I get them too. ?I have no > clue what's going on, it's truly weird. OK, it's fc11, but I do see the problem. I can't fix it now, but I can at least reproduce it. That's a start. Stay tuned. Cheers, f From gokhansever at gmail.com Fri Jan 15 14:35:31 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Fri, 15 Jan 2010 13:35:31 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> Message-ID: <49d6b3501001151135v6e34dceay96152d543656ddd8@mail.gmail.com> On Thu, Jan 14, 2010 at 8:21 PM, G?khan Sever wrote: > > > On Thu, Jan 14, 2010 at 8:02 PM, Fernando Perez wrote: > >> On Thu, Jan 14, 2010 at 5:53 PM, G?khan Sever >> wrote: >> > I did "python setup.py install" after re-pulling your branch. >> > >> > In [2]: IPython.__version__ >> > Out[2]: '0.11.bzr.r1321' >> > >> > applied your patch. Test runs produced the same errors. >> >> OK, I'm seriously stumped. I think we have a Fedora 12 box around >> here, I'll try to test on it and see if I get them too. I have no >> clue what's going on, it's truly weird. >> >> One more thing to try, shouldn't matter but I'm willing to test wild >> ideas. Move your ~/.ipython dir to some other name for a minute, and >> try the test again, so ipython can't find any of your user defaults. >> Let's see if that helps, if it does it would indicate we're picking up >> user info we shouldn't at test time. >> > > The same. > > I can give it try on a Fedora 10 with Python 2.5.2 later. > > There is a FC10 in our lab, but I don't have the root-pass. Any ideas how to install without being root? Also there are RHEL 4, 5 machines but you drop Python 2.3 and 2.4 support, so those are no avail. > > >> >> Cheers, >> >> f >> > > > > -- > G?khan > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Fri Jan 15 16:07:47 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 15 Jan 2010 13:07:47 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001151135v6e34dceay96152d543656ddd8@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> <49d6b3501001151135v6e34dceay96152d543656ddd8@mail.gmail.com> Message-ID: On Fri, Jan 15, 2010 at 11:35 AM, G?khan Sever wrote: > > There is a FC10 in our lab, but I don't have the root-pass. Any ideas how to > install without being root? No need to install anything anymore. Yesterday I added a new iptest.py script, so just cd /your/ipython/path bzr pull python iptest.py That should run the test suite 100% from the locally downloaded tree, regardless of what's in the system. Cheers, f From gokhansever at gmail.com Sat Jan 16 12:17:04 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Sat, 16 Jan 2010 11:17:04 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> <49d6b3501001151135v6e34dceay96152d543656ddd8@mail.gmail.com> Message-ID: <49d6b3501001160917t5d757484y96bc2e79d7546742@mail.gmail.com> On Fri, Jan 15, 2010 at 3:07 PM, Fernando Perez wrote: > On Fri, Jan 15, 2010 at 11:35 AM, G?khan Sever > wrote: > > > > There is a FC10 in our lab, but I don't have the root-pass. Any ideas how > to > > install without being root? > > No need to install anything anymore. Yesterday I added a new > iptest.py script, so just > > cd /your/ipython/path > bzr pull > python iptest.py > > That should run the test suite 100% from the locally downloaded tree, > regardless of what's in the system. > > Cheers, > > f > Fernando, It seems like the new test script doesn't run IPython.kernel. I only get errors from IPython.core and IPython.extensions. kernel is not listed in the output. -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mattions at gmail.com Sat Jan 16 15:08:44 2010 From: mattions at gmail.com (Michele Mattioni) Date: Sat, 16 Jan 2010 20:08:44 +0000 Subject: [IPython-dev] Ipython 0.10 pygtk main loop In-Reply-To: References: <290ea1b00912070752j616ae6a9m2b55d38816556ea3@mail.gmail.com> <20091207160623.GA6519@phare.normalesup.org> <290ea1b00912070830p53f0c068x6cec00450a1cc5a9@mail.gmail.com> Message-ID: <290ea1b01001161208p2f23fc33g2aad22c1ab39be5@mail.gmail.com> Hi Fernando, I changed the backend for my project and I don't use anymore visual but mayavi. I can't test anymore the old code. I'm sorry. Best, Michele. On Sun, Jan 10, 2010 at 7:03 AM, Fernando Perez wrote: > Hi Michele, > > On Mon, Dec 7, 2009 at 8:30 AM, Michele Mattioni wrote: >> Hi Ga?l, >> >> I tried that but it seems the backend is correct. >> There is a strange disclaimer which is printend when ipython is started with >> the pylab (or gthread) options as shown on this bug: >> >> https://bugs.launchpad.net/ipython/+bug/270856 >> >> Can be that the case? > > See my comments there... Ideally, what we want to know is whether this > works with 0.11, where the gui support is much cleaner. ?If you can > test (currently with my trunk-dev branch or the downloads at > http://ipython.scipy.org/dist/testing/) and let us know how it goes it > would be great. > > I tried but on my Ubuntu 9.10 system your code segfaults immediately > (without ipython at all): > > maqroll[lib]> python ~/Desktop/hellovisual.py > Segmentation fault > > so I can't verify here... > > Cheers, > > f > From fperez.net at gmail.com Sat Jan 16 16:47:40 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 16 Jan 2010 13:47:40 -0800 Subject: [IPython-dev] Ipython 0.10 pygtk main loop In-Reply-To: <290ea1b01001161208p2f23fc33g2aad22c1ab39be5@mail.gmail.com> References: <290ea1b00912070752j616ae6a9m2b55d38816556ea3@mail.gmail.com> <20091207160623.GA6519@phare.normalesup.org> <290ea1b00912070830p53f0c068x6cec00450a1cc5a9@mail.gmail.com> <290ea1b01001161208p2f23fc33g2aad22c1ab39be5@mail.gmail.com> Message-ID: On Sat, Jan 16, 2010 at 12:08 PM, Michele Mattioni wrote: > > I changed the backend for my project and I don't use anymore visual but mayavi. > I can't test anymore the old code. > I'm sorry. No worries, in fact what we really want to know is that the new support should work great with mayavi. In all my testing, using ipython --pylab wx I can then load mayavi and both pylab and mayavi.mlab coexist happily, but let us know if you see any glitches. Cheers, f From fperez.net at gmail.com Sat Jan 16 18:27:31 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 16 Jan 2010 15:27:31 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001160917t5d757484y96bc2e79d7546742@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001141753w125b55e1q680eafcae7768c99@mail.gmail.com> <49d6b3501001141821h48249d77jf722fe0dec42b1d1@mail.gmail.com> <49d6b3501001151135v6e34dceay96152d543656ddd8@mail.gmail.com> <49d6b3501001160917t5d757484y96bc2e79d7546742@mail.gmail.com> Message-ID: On Sat, Jan 16, 2010 at 9:17 AM, G?khan Sever wrote: > It seems like the new test script doesn't run IPython.kernel. I only get > errors from IPython.core and IPython.extensions. kernel is not listed in the > output. Please update and try again, I think I fixed this. Thanks for the feedback! Cheers, f From fperez.net at gmail.com Sat Jan 16 19:27:31 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 16 Jan 2010 16:27:31 -0800 Subject: [IPython-dev] Updated build of trunk-dev available Message-ID: Hi all, thanks to the feedback I've been getting, I have been able to fix a bunch of small nits with the testing machinery (many thanks to Gokhan and Mikhail!), so here's a new build for those of you willing to give it a try: http://ipython.scipy.org/dist/testing/ This has self-contained downloads of this branch as of r1341, so you can run from bzr as well: https://code.launchpad.net/~fdo.perez/ipython/trunk-dev In short, now once you download (whether a build or with bzr) you should be able to run: python iptest.py As far as I know, this passes 100% on ubuntu 32/64 bits, Fedora 11 and Windows without Twisted. On Windows with twisted we have one bug left (that causes 6 identical failures) but it's easy to fix and already being tracked, it will be fixed before release. Updated docs are in: http://ipython.scipy.org/doc/trunk-dev In particular, details on testing: http://ipython.scipy.org/doc/trunk-dev/development/testing.html Thanks for any feedback you can provide us, from test suite results to comments on using it. If you are willing to go one step further and help us with the code review at: https://code.launchpad.net/~fdo.perez/ipython/trunk-dev/+merge/16695 That would be enormously appreciated. Best regards, f From fperez.net at gmail.com Sat Jan 16 19:42:03 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sat, 16 Jan 2010 16:42:03 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: Message-ID: On Sat, Jan 16, 2010 at 4:27 PM, Fernando Perez wrote: > > As far as I know, this passes 100% on ubuntu 32/64 bits, Fedora 11 and > Windows without Twisted. And I forgot to add, also passes on OSX (tested without Twisted though). Keep in mind that you do need to have nose installed for the test suite to run. Cheers, f From satra at mit.edu Sat Jan 16 20:21:08 2010 From: satra at mit.edu (Satrajit Ghosh) Date: Sat, 16 Jan 2010 20:21:08 -0500 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: Message-ID: <83351d261001161721te13143bie47bb7060adeeb6@mail.gmail.com> on osx snow leopard 64-bit kernel. (ipythonenv)unwarped:pysoft satra$ python /software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/iptest.pyc IPython.core .......................................................................S..............F.F.. ====================================================================== FAIL: Test that object's __del__ methods are called on exit. ---------------------------------------------------------------------- Traceback (most recent call last): File "/Library/Python/2.6/site-packages/nose-0.11.1-py2.6.egg/nose/case.py", line 183, in runTest self.test(*self.arg) File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/core/tests/test_run.py", line 160, in test_obj_del tt.ipexec_validate(self.fname, 'object A deleted') File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/tools.py", line 290, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '' != 'object A deleted' >> raise self.failureException, \ (None or '%r != %r' % ('', 'object A deleted')) ====================================================================== FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass ---------------------------------------------------------------------- Traceback (most recent call last): File "/Library/Python/2.6/site-packages/nose-0.11.1-py2.6.egg/nose/case.py", line 183, in runTest self.test(*self.arg) File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/core/tests/test_run.py", line 174, in test_tclass tt.ipexec_validate(self.fname, out) File "/software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/tools.py", line 290, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '' != "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" >> raise self.failureException, \ (None or '%r != %r' % ('', "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first")) ---------------------------------------------------------------------- Ran 94 tests in 0.249s FAILED (SKIP=1, failures=2) -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists_ravi at lavabit.com Sat Jan 16 21:51:57 2010 From: lists_ravi at lavabit.com (Ravi) Date: Sat, 16 Jan 2010 21:51:57 -0500 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: Message-ID: <201001162151.57608.lists_ravi@lavabit.com> On Saturday 16 January 2010 19:27:31 Fernando Perez wrote: > Thanks for any feedback you can provide us, from test suite results to > comments on using it. On up-to-date Fedora core 12, I needed to remove the '--traverse-namespace' option from iptest.py. After that, here's the output: [ipython-0.11.bzr.r1340]$ python iptest.py ********************************************************************** IPython test group: IPython.config ......... ---------------------------------------------------------------------- Ran 10 tests in 0.012s OK ********************************************************************** IPython test group: IPython.core ......................................................................S..............F.F.. ====================================================================== FAIL: Test that object's __del__ methods are called on exit. ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File ".../ipython-0.11.bzr.r1340/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File ".../ipython-0.11.bzr.r1340/IPython/core/tests/test_run.py", line 160, in test_obj_del tt.ipexec_validate(self.fname, 'object A deleted') File ".../ipython-0.11.bzr.r1340/IPython/testing/tools.py", line 290, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >> raise self.failureException, \ (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A deleted')) ====================================================================== FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File ".../ipython-0.11.bzr.r1340/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File ".../ipython-0.11.bzr.r1340/IPython/core/tests/test_run.py", line 174, in test_tclass tt.ipexec_validate(self.fname, out) File ".../ipython-0.11.bzr.r1340/IPython/testing/tools.py", line 290, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: "\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C- second']\ntclass.py: deleting object: C-first" != "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" >> raise self.failureException, \ (None or '%r != %r' % ("\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first", "ARGV 1-: ['C- first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first")) ---------------------------------------------------------------------- Ran 93 tests in 0.803s FAILED (SKIP=1, failures=2) ********************************************************************** IPython test group: IPython.extensions ---------------------------------------------------------------------- Ran 1 test in 0.003s OK ********************************************************************** IPython test group: IPython.frontend .. ---------------------------------------------------------------------- Ran 3 tests in 0.235s OK ********************************************************************** IPython test group: IPython.lib .... ---------------------------------------------------------------------- Ran 5 tests in 0.005s OK ********************************************************************** IPython test group: IPython.scripts ---------------------------------------------------------------------- Ran 0 tests in 0.000s OK ********************************************************************** IPython test group: IPython.testing .......................SSS.......S..S......................S ---------------------------------------------------------------------- Ran 65 tests in 0.141s OK (SKIP=6) ********************************************************************** IPython test group: IPython.utils .........................SS..SSS............................S.S............................................................ ---------------------------------------------------------------------- Ran 125 tests in 0.160s OK (SKIP=7) ********************************************************************** IPython test group: IPython.gui ---------------------------------------------------------------------- Ran 0 tests in 0.001s OK ********************************************************************** Ran 9 test groups in 7.621s ERROR - 1 out of 9 test groups failed. ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python .../build/ipython-0.11.bzr.r1340/IPython/testing/iptest.pyc IPython.core Hope this helps. Ravi From gokhansever at gmail.com Sat Jan 16 22:07:04 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Sat, 16 Jan 2010 21:07:04 -0600 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: Message-ID: <49d6b3501001161907x4b30f5c0k77d3187cd31278b0@mail.gmail.com> On Sat, Jan 16, 2010 at 6:27 PM, Fernando Perez wrote: > Hi all, > > thanks to the feedback I've been getting, I have been able to fix a > bunch of small nits with the testing machinery (many thanks to Gokhan > and Mikhail!) > Aside from the main testing issue, I have a few ideas to share: "Testing and development in SciPy ecosystem" For the upcoming SciPy conference, this would be a great presentation title for those who hasn't ever touched the code repositories and only living off from the official releases. Because many tools have their own software development system. IPython is on Launchpad, Matplotlib is on Sourceforge, Numpy and SciPy have their own system along with the ETS. Not only the code management systems different but also interactions with users and developers (e.g. different bug tracking systems, code reviewing etc...) I am not a computer scientist in profession but it would be good to take some more steps to promote these --clear result of open-source development habits to show the melting boundaries of being a user and a developer. -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From gokhansever at gmail.com Sat Jan 16 23:12:05 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Sat, 16 Jan 2010 22:12:05 -0600 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <201001162151.57608.lists_ravi@lavabit.com> References: <201001162151.57608.lists_ravi@lavabit.com> Message-ID: <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> On Sat, Jan 16, 2010 at 8:51 PM, Ravi wrote: > On Saturday 16 January 2010 19:27:31 Fernando Perez wrote: > > Thanks for any feedback you can provide us, from test suite results to > > comments on using it. > > On up-to-date Fedora core 12, I needed to remove the '--traverse-namespace' > option from iptest.py. After that, here's the output: > I did the same. I get an additional failure comparing to yours on my FC12: ********************************************************************** Ran 8 test groups in 7.642s ERROR - 2 out of 8 test groups failed. ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python /home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/iptest.py IPython.core ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: python /home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/iptest.py IPython.extensions Interestingly, tests for IPython.kernel is still notting running. -------------- next part -------------- An HTML attachment was scrubbed... URL: From gokhansever at gmail.com Sun Jan 17 00:48:01 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Sat, 16 Jan 2010 23:48:01 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> Message-ID: <49d6b3501001162148v57b31ae7n26fdce6609f33a9b@mail.gmail.com> On Thu, Jan 14, 2010 at 5:49 PM, Fernando Perez wrote: > Hi Gokhan, > > On Wed, Jan 13, 2010 at 2:48 PM, G?khan Sever > wrote: > > > Ahaa, here is one catch (some extra outputs in whos listing --which I was > > going to ask if it is possible to list the command that we used to > construct > > a object next the Data/Info column in whos listing? > > > > Sometimes I search my history to see how I constructed an object. That > would > > be useful to have it listed on a whos listing. > > There is no way to know this in python in general, since multiple > names can be bound to one object, assignments can happen in many > different ways (code can write to namespace dicts directly, etc). But > you can write a little extension of yours that walks the In input > history list looking for things that look like variable assignments. > It won't be very robust, but could be useful. If you end up liking > it, put it up on the cookbook and we can ship it later as an optional > tool, once you've stress-tested it. > > Cheers, > > f > Here is my idea to implement this following your %history suggestion: 1- Look in the history in the reverse order to obtain the latest entry of the same variable assignment. Each element listed in the history output is a separate string pulled out of a list. 2- Compare the "string".split("=")[0] with each variable name in %whos call. 3-Write the output in the 4th column of the whos listing (using a one of the names: "Assignment", "Command", "History"): Variable Type Data/Info Assignment -------------------------------------------------- a int 5 a=5 n float 0.0 n = math.sin(0) Data/Info column usually produces longer lines this might cause some ugly looking output. Could I get some suggestions? Thanks, -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mattions at gmail.com Sun Jan 17 04:35:03 2010 From: mattions at gmail.com (Michele Mattioni) Date: Sun, 17 Jan 2010 09:35:03 +0000 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: Message-ID: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> On Sun, Jan 17, 2010 at 12:27 AM, Fernando Perez wrote: > Hi all, > > thanks to the feedback I've been getting, I have been able to fix a > bunch of small nits with the testing machinery (many thanks to Gokhan > and Mikhail!), so here's a new build for those of you willing to give > it a try: > > http://ipython.scipy.org/dist/testing/ > > This has self-contained downloads of this branch as of r1341, so you > can run from bzr as well: > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > Hi, I tried to test the qt integration but it seems it doesn't work properly. The ipython doesn't release the prompt as it should (an as it happens with 0.10) I attach a rough script (and adaptation of Mayavi qt embedding), just for testing purpose. The expect behaviour is to have the prompt back as usual, ready to process input. I started it with python ipython --gui qt I'm missing an option? Cheers, Michele. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: qt_embedding.py Type: application/octet-stream Size: 4090 bytes Desc: not available URL: From fperez.net at gmail.com Sun Jan 17 04:44:11 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 01:44:11 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> Message-ID: On Sun, Jan 17, 2010 at 1:35 AM, Michele Mattioni wrote: > I tried to test the qt integration but it seems it doesn't work properly. > > The ipython doesn't release the prompt as it should (an as it happens with > 0.10) > > I attach a rough script (and adaptation of Mayavi qt embedding), just for > testing purpose. Mmh, unfortunately I can't seem to run your script, because I get a traceback: In [2]: run qt_embedding.py Warning: Unable to import the qt4 backend for pyface due to traceback: Traceback (most recent call last): File "/usr/lib/python2.6/dist-packages/enthought/pyface/toolkit.py", line 40, in _init_toolkit __import__(be + 'init') ImportError: No module named qt4.init Info: Unable to import any backend (qt4) for pyface; using the 'null' toolkit instead. /usr/lib/python2.6/dist-packages/enthought/traits/ui/toolkit.py:128: UserWarning: Unable to import the 'qt4' backend for traits UI; using the 'null' toolkit instead. "using the 'null' toolkit instead." % toolkit_name ) --------------------------------------------------------------------------- ImportError Traceback (most recent call last) /home/fperez/Desktop/qt_embedding.py in () 17 Int, Dict 18 from enthought.traits.ui.api import View, Item ---> 19 from enthought.mayavi.core.ui.api import MayaviScene, MlabSceneModel, \ 20 SceneEditor 21 from enthought.mayavi import mlab ImportError: No module named api This is on ubuntu karmic. Do I need to run from SVN Enthought for this example? It would be great if we could have a smaller example illustrating your problem that didn't depend on all of Enthought's machinery, because we really want to make sure things work OK with Qt. I've tried simpler tests using matplotlib with qt, and those seem OK so far. One thing to try: 1. Start ipython *without* --gui 2. At the ipython prompt, type gui qt 3. Try to run your script. Then, repeat, but with 'gui -a qt' instead. Let us know if it produces any difference at all. If you have a simpler non-mayavi example, send it our way. Thanks for the feedback! From fperez.net at gmail.com Sun Jan 17 06:19:33 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 03:19:33 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <83351d261001161721te13143bie47bb7060adeeb6@mail.gmail.com> References: <83351d261001161721te13143bie47bb7060adeeb6@mail.gmail.com> Message-ID: On Sat, Jan 16, 2010 at 5:21 PM, Satrajit Ghosh wrote: > on osx snow leopard 64-bit kernel. > > (ipythonenv)unwarped:pysoft satra$ python > /software/pysoft/ipythonenv/lib/python2.6/site-packages/ipython-0.11.bzr.r1340-py2.6.egg/IPython/testing/iptest.pyc > IPython.core > .......................................................................S..............F.F.. > ====================================================================== > FAIL: Test that object's __del__ methods are called on exit. OK, I've been able to reproduce this, though I don't know how to fix it yet. Nose behaves differently depending on whether running from an installed copy or from the source dir. Odd. But thanks for the feedback, it's super useful. We still have to figure out the Qt issues, but it seems on the rest of the test suite we're converging on only a few recalcitrant failures that are fortunately the same. I hope with a little more work tomorrow to be able to fix them. Stay tuned. Cheers, f From mattions at gmail.com Sun Jan 17 07:39:22 2010 From: mattions at gmail.com (Michele Mattioni) Date: Sun, 17 Jan 2010 12:39:22 +0000 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> Message-ID: <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> On Sun, Jan 17, 2010 at 9:44 AM, Fernando Perez wrote: > On Sun, Jan 17, 2010 at 1:35 AM, Michele Mattioni > wrote: > > I tried to test the qt integration but it seems it doesn't work properly. > > > > The ipython doesn't release the prompt as it should (an as it happens > with > > 0.10) > > > > I attach a rough script (and adaptation of Mayavi qt embedding), just for > > testing purpose. > > Mmh, unfortunately I can't seem to run your script, because I get a > traceback: > > In [2]: run qt_embedding.py > Warning: Unable to import the qt4 backend for pyface due to traceback: > Traceback (most recent call last): > File > "/usr/lib/python2.6/dist-packages/enthought/pyface/toolkit.py", > line 40, in _init_toolkit > __import__(be + 'init') > ImportError: No module named qt4.init > > Info: Unable to import any backend (qt4) for pyface; using the 'null' > toolkit instead. > > /usr/lib/python2.6/dist-packages/enthought/traits/ui/toolkit.py:128: > UserWarning: Unable to import the 'qt4' backend for traits UI; using > the 'null' toolkit instead. > "using the 'null' toolkit instead." % toolkit_name ) > --------------------------------------------------------------------------- > ImportError Traceback (most recent call last) > > /home/fperez/Desktop/qt_embedding.py in () > 17 Int, Dict > 18 from enthought.traits.ui.api import View, Item > ---> 19 from enthought.mayavi.core.ui.api import MayaviScene, > MlabSceneModel, \ > 20 SceneEditor > 21 from enthought.mayavi import mlab > > ImportError: No module named api > > > This is on ubuntu karmic. Do I need to run from SVN Enthought for this > example? > > Yes, the latest mayavi api for qt integration are only in the latest SVN Enthought > It would be great if we could have a smaller example illustrating your > problem that didn't depend on all of Enthought's machinery, because we > really want to make sure things work OK with Qt. I've tried simpler > tests using matplotlib with qt, and those seem OK so far. > I tested with only qt and it works. > One thing to try: > > 1. Start ipython *without* --gui > 2. At the ipython prompt, type > gui qt > 3. Try to run your script. > Then, repeat, but with 'gui -a qt' instead. Let us know if it > produces any difference at all. > > No difference. > If you have a simpler non-mayavi example, send it our way. > > It seems it's mayavi related, I'm adding the mayavi list to this mail. Cheers, Michele. > Thanks for the feedback! > -------------- next part -------------- An HTML attachment was scrubbed... URL: From scorpion032 at gmail.com Sun Jan 17 09:11:30 2010 From: scorpion032 at gmail.com (Lakshman Prasad) Date: Sun, 17 Jan 2010 19:41:30 +0530 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> Message-ID: Hi, I tested it on Ubuntu Karmic Python 2.6, The r1350 (ipython-0.11.bzr.r1340-py2.6.) and there seemed to be 1 failure in IPython.core: - FAIL: Test that object's __del__ methods are called on exit. (The other failure seems to be related) Here is the complete traceback: http://gist.github.com/279386 Here is the traceback of the relevant section: http://gist.github.com/279387 Also, is there any quick reference (not necessarily explained and/or detailed), of all the new features/fixes in 0.11 Thanks. On Sun, Jan 17, 2010 at 6:09 PM, Michele Mattioni wrote: > > > On Sun, Jan 17, 2010 at 9:44 AM, Fernando Perez wrote: > >> On Sun, Jan 17, 2010 at 1:35 AM, Michele Mattioni >> wrote: >> > I tried to test the qt integration but it seems it doesn't work >> properly. >> > >> > The ipython doesn't release the prompt as it should (an as it happens >> with >> > 0.10) >> > >> > I attach a rough script (and adaptation of Mayavi qt embedding), just >> for >> > testing purpose. >> >> Mmh, unfortunately I can't seem to run your script, because I get a >> traceback: >> >> In [2]: run qt_embedding.py >> Warning: Unable to import the qt4 backend for pyface due to traceback: >> Traceback (most recent call last): >> File >> "/usr/lib/python2.6/dist-packages/enthought/pyface/toolkit.py", >> line 40, in _init_toolkit >> __import__(be + 'init') >> ImportError: No module named qt4.init >> >> Info: Unable to import any backend (qt4) for pyface; using the 'null' >> toolkit instead. >> >> /usr/lib/python2.6/dist-packages/enthought/traits/ui/toolkit.py:128: >> UserWarning: Unable to import the 'qt4' backend for traits UI; using >> the 'null' toolkit instead. >> "using the 'null' toolkit instead." % toolkit_name ) >> >> --------------------------------------------------------------------------- >> ImportError Traceback (most recent call >> last) >> >> /home/fperez/Desktop/qt_embedding.py in () >> 17 Int, Dict >> 18 from enthought.traits.ui.api import View, Item >> ---> 19 from enthought.mayavi.core.ui.api import MayaviScene, >> MlabSceneModel, \ >> 20 SceneEditor >> 21 from enthought.mayavi import mlab >> >> ImportError: No module named api >> >> >> This is on ubuntu karmic. Do I need to run from SVN Enthought for this >> example? >> >> > > Yes, the latest mayavi api for qt integration are only in the latest SVN > Enthought > > > >> It would be great if we could have a smaller example illustrating your >> problem that didn't depend on all of Enthought's machinery, because we >> really want to make sure things work OK with Qt. I've tried simpler >> tests using matplotlib with qt, and those seem OK so far. >> > > > I tested with only qt and it works. > > > >> One thing to try: >> >> 1. Start ipython *without* --gui >> 2. At the ipython prompt, type >> gui qt >> 3. Try to run your script. >> > > Then, repeat, but with 'gui -a qt' instead. Let us know if it >> produces any difference at all. >> >> > No difference. > > >> If you have a simpler non-mayavi example, send it our way. >> >> > It seems it's mayavi related, I'm adding the mayavi list to this mail. > > Cheers, > Michele. > > >> Thanks for the feedback! >> > > > _______________________________________________ > IPython-dev mailing list > IPython-dev at scipy.org > http://mail.scipy.org/mailman/listinfo/ipython-dev > > -- Regards, Lakshman becomingguru.com lakshmanprasad.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From satra at mit.edu Sun Jan 17 11:03:41 2010 From: satra at mit.edu (Satrajit Ghosh) Date: Sun, 17 Jan 2010 11:03:41 -0500 Subject: [IPython-dev] ipython+virtualenv Message-ID: <83351d261001170803k2a77bae9s2705218b6ed3706e@mail.gmail.com> hi, i installed ipython in a virtual environment. however when i launch it picks up the system ipython even though the executable listed by path is the virtual one. when i explicitly call the virtual one it starts the virtual one. http://gist.github.com/279433 does iptest check to see that the correct version of ipython is being imported? if not, a check might be good. cheers, satra -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Sun Jan 17 19:02:16 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 16:02:16 -0800 Subject: [IPython-dev] ipython+virtualenv In-Reply-To: <83351d261001170803k2a77bae9s2705218b6ed3706e@mail.gmail.com> References: <83351d261001170803k2a77bae9s2705218b6ed3706e@mail.gmail.com> Message-ID: Hey Satra, On Sun, Jan 17, 2010 at 8:03 AM, Satrajit Ghosh wrote: > > i installed ipython in a virtual environment. however when i launch it picks > up the system ipython even though the executable listed by path is the > virtual one. when i explicitly call the virtual one it starts the virtual > one. > > http://gist.github.com/279433 > > does iptest check to see that the correct version of ipython is being > imported? if not, a check might be good. If you update again, I think I've fixed most of that, which incidentally was related to some of the failures you and others were seeing. But I don't understand what you're seeing there though: that running `which X` and X produce different results is odd. Unfortunately I can't test with virtualenvs, because setuptools/virtualenv seems to have broken on my system: maqroll[~]> mkvirtualenv --no-site-packages ipythonenv New python executable in ipythonenv/bin/python Please make sure you remove any previous custom paths from your /home/fperez/.pydistutils.cfg file. Installing setuptools...................... Complete output from command ipythonenv/bin/python -c "#!python \"\"\"Bootstrap setuptoo... " /usr/lib/python2.6/dist-packag...6.egg: Downloading http://pypi.python.org/packages/2.6/s/setuptools/setuptools-0.6c9-py2.6.egg Traceback (most recent call last): File "", line 270, in File "", line 204, in main File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/setuptools/__init__.py", line 2, in File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/setuptools/extension.py", line 2, in File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/setuptools/dist.py", line 6, in File "/usr/lib/python2.6/dist-packages/support-files/setuptools-0.6c9-py2.6.egg/setuptools/command/sdist.py", line 4, in File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/pkg_resources.py", line 2582, in add_activation_listener(lambda dist: dist.activate()) File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/pkg_resources.py", line 640, in subscribe callback(dist) File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/pkg_resources.py", line 2582, in add_activation_listener(lambda dist: dist.activate()) File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/pkg_resources.py", line 2131, in activate map(declare_namespace, self._get_metadata('namespace_packages.txt')) File "/home/fperez/usr/opt/virtualenv/setuptools-0.6c9-py2.6.egg/pkg_resources.py", line 1735, in declare_namespace __import__(parent) ImportError: No module named enthought ---------------------------------------- ...Installing setuptools...done. Traceback (most recent call last): File "/home/fperez/usr/opt/bin/virtualenv", line 8, in load_entry_point('virtualenv==1.3.3', 'console_scripts', 'virtualenv')() File "/usr/lib/python2.6/dist-packages/virtualenv.py", line 420, in main unzip_setuptools=options.unzip_setuptools) File "/usr/lib/python2.6/dist-packages/virtualenv.py", line 507, in create_environment install_setuptools(py_executable, unzip=unzip_setuptools) File "/usr/lib/python2.6/dist-packages/virtualenv.py", line 295, in install_setuptools cwd=cwd) File "/usr/lib/python2.6/dist-packages/virtualenv.py", line 481, in call_subprocess % (cmd_desc, proc.returncode)) OSError: Command ipythonenv/bin/python -c "#!python \"\"\"Bootstrap setuptoo... " /usr/lib/python2.6/dist-packag...6.egg failed with error code 1 Now you can use workon to activate your new virtualenv #### The old venvs I have all work, but I can't seem to make new ones anymore. I don't really have time right now to debug setuptools itself, unfortunately. Cheers, f From fperez.net at gmail.com Sun Jan 17 19:09:41 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 16:09:41 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> Message-ID: Hi Michele, On Sun, Jan 17, 2010 at 4:39 AM, Michele Mattioni wrote: > Yes, the latest mayavi api for qt integration are only in the latest SVN > Enthought Ah, OK. I don't have that right now, but eventually this will be very high priority, it's just that first we need to get things right with the stable releases of the various tools. I did check with stable mayavi (wx) and pylab with wx backend, and those two do coexist happily on the new branch. >> It would be great if we could have a smaller example illustrating your >> problem that didn't depend on all of Enthought's machinery, because we >> really want to make sure things work OK with Qt. ?I've tried simpler >> tests using matplotlib with qt, and those seem OK so far. >> > > I tested with only qt and it works. > > >> >> One thing to try: >> >> 1. Start ipython *without* --gui >> 2. At the ipython prompt, type >> gui qt >> 3. Try to run your script. >> >> >> Then, repeat, but with 'gui -a qt' instead. ?Let us know if ?it >> produces any difference at all. >> > > No difference. OK, thanks for this feedback, we can track this further here: https://bugs.launchpad.net/ipython/+bug/508953 Cheers, f From fperez.net at gmail.com Sun Jan 17 19:12:20 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 16:12:20 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> Message-ID: On Sun, Jan 17, 2010 at 6:11 AM, Lakshman Prasad wrote: > I tested it on Ubuntu Karmic Python 2.6, The r1350 > (ipython-0.11.bzr.r1340-py2.6.) and there seemed to be 1 failure in > IPython.core: > > FAIL: Test that object's __del__ methods are called on exit. These should now be fixed with the recent updates I've made, let me know how it goes. > Also, is there any quick reference (not necessarily explained and/or > detailed), of all the new features/fixes in 0.11 Kind of. This is what we have so far: http://ipython.scipy.org/doc/trunk-dev/whatsnew/development.html I still need to document all the work I've done recently better, though the updates to testing have been documented here: http://ipython.scipy.org/doc/trunk-dev/development/testing.html Cheers, f From fperez.net at gmail.com Sun Jan 17 19:17:54 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 16:17:54 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> Message-ID: Hi all, On Sat, Jan 16, 2010 at 8:12 PM, G?khan Sever wrote: > I did the same. I get an additional failure comparing to yours on my FC12: I'd appreciate it if you let me know whether the last round of updates fixes these problems. I have now tested on ubuntu 9.10 (32 and 64 bit), Fedora 11 64 bit, OSX 10.6 and Windows XP, both from the source dir with python iptest.py and from an installed version, and except for the known win32 'job' failure, all of these pass in all forms. Here are some sample reports: ********************************************************************** Test suite completed for system with the following information: IPython version: 0.11.bzr.r1346 BZR revision : 1346 Platform info : os.name -> posix, sys.platform -> darwin : Darwin-10.2.0-i386-64bit Python info : 2.6.1 (r261:67515, Jul 7 2009, 23:51:51) [GCC 4.2.1 (Apple Inc. build 5646)] Running from an installed IPython: False Tools and libraries available at test time: curses objc twisted zope.interface Tools and libraries NOT available at test time: foolscap gobject gtk pexpect wx wx.aui Ran 8 test groups in 9.159s Status: OK ********************************************************************** Test suite completed for system with the following information: IPython version: 0.11.bzr.r1346 BZR revision : 1346 Platform info : os.name -> posix, sys.platform -> linux2 : Linux-2.6.30.9-99.fc11.x86_64-x86_64-with-fedora-11-Leonidas Python info : 2.6 (r26:66714, Nov 3 2009, 17:33:18) [GCC 4.4.1 20090725 (Red Hat 4.4.1-2)] Running from an installed IPython: True Tools and libraries available at test time: curses gobject gtk wx wx.aui Tools and libraries NOT available at test time: foolscap objc pexpect twisted zope.interface Ran 9 test groups in 15.685s Status: OK ********************************************************************** Test suite completed for system with the following information: IPython version: 0.11.bzr.r1346 BZR revision : 1346 Platform info : os.name -> posix, sys.platform -> linux2 : Linux-2.6.31-17-generic-x86_64-with-Ubuntu-9.10-karmic Python info : 2.6.4 (r264:75706, Dec 7 2009, 18:43:55) [GCC 4.4.1] Running from an installed IPython: False Tools and libraries available at test time: curses foolscap gobject gtk pexpect twisted wx zope.interface Tools and libraries NOT available at test time: objc wx.aui Ran 11 test groups in 38.313s Status: OK etc... Please paste this little report at the end, it will help us quite a bit understand better how your environment looks. > Interestingly, tests for IPython.kernel is still notting running. Also let us know about this... I've uploaded updated builds at the usual place: http://ipython.scipy.org/dist/testing/ Cheers, f From fperez.net at gmail.com Sun Jan 17 20:27:14 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 17:27:14 -0800 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: <49d6b3501001162148v57b31ae7n26fdce6609f33a9b@mail.gmail.com> References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001162148v57b31ae7n26fdce6609f33a9b@mail.gmail.com> Message-ID: On Sat, Jan 16, 2010 at 9:48 PM, G?khan Sever wrote: > 1- Look in the history in the reverse order to obtain the latest entry of > the same variable assignment. Each element listed in the history output is a > separate string pulled out of a list. > > 2- Compare the "string".split("=")[0] with each variable name in %whos call. > > 3-Write the output in the 4th column of the whos listing (using a one of the > names: "Assignment", "Command", "History"): > > Variable?? Type??? Data/Info??? Assignment > -------------------------------------------------- > a????????????? int??????? 5?????????????? a=5 > n????????????? float???? 0.0???????????? n = math.sin(0) > > Data/Info column usually produces longer lines this might cause some ugly > looking output. > > Could I get some suggestions? Well, sounds like you have a plan, go ahead and implement it :) Some pointers: - get the global ipython object via get_ipython() - its .user_ns is your user namespace. - in there, the 'In' variable contains your inputs. I'd just add this as a normal function initially loaded in your startup file. Once it's working the way you want it, you just need a few lines to expose it as a magic for convenience. Cheers, f From gokhansever at gmail.com Mon Jan 18 00:59:09 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Sun, 17 Jan 2010 23:59:09 -0600 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> Message-ID: <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> On Sun, Jan 17, 2010 at 6:17 PM, Fernando Perez wrote: > Hi all, > > On Sat, Jan 16, 2010 at 8:12 PM, G?khan Sever > wrote: > > I did the same. I get an additional failure comparing to yours on my > FC12: > > I'd appreciate it if you let me know whether the last round of updates > fixes these problems. I have now tested on ubuntu 9.10 (32 and 64 > bit), Fedora 11 64 bit, OSX 10.6 and Windows XP, both from the source > dir with > > python iptest.py > > and from an installed version, and except for the known win32 'job' > failure, all of these pass in all forms. Here are some sample > reports: > > ********************************************************************** > Test suite completed for system with the following information: > IPython version: 0.11.bzr.r1346 > BZR revision : 1346 > Platform info : os.name -> posix, sys.platform -> darwin > : Darwin-10.2.0-i386-64bit > Python info : 2.6.1 (r261:67515, Jul 7 2009, 23:51:51) > [GCC 4.2.1 (Apple Inc. build 5646)] > > Running from an installed IPython: False > > Tools and libraries available at test time: > curses objc twisted zope.interface > > Tools and libraries NOT available at test time: > foolscap gobject gtk pexpect wx wx.aui > > Ran 8 test groups in 9.159s > > Status: > OK > > > ********************************************************************** > Test suite completed for system with the following information: > IPython version: 0.11.bzr.r1346 > BZR revision : 1346 > Platform info : os.name -> posix, sys.platform -> linux2 > : > Linux-2.6.30.9-99.fc11.x86_64-x86_64-with-fedora-11-Leonidas > Python info : 2.6 (r26:66714, Nov 3 2009, 17:33:18) > [GCC 4.4.1 20090725 (Red Hat 4.4.1-2)] > > Running from an installed IPython: True > > Tools and libraries available at test time: > curses gobject gtk wx wx.aui > > Tools and libraries NOT available at test time: > foolscap objc pexpect twisted zope.interface > > Ran 9 test groups in 15.685s > > Status: > OK > > ********************************************************************** > Test suite completed for system with the following information: > IPython version: 0.11.bzr.r1346 > BZR revision : 1346 > Platform info : os.name -> posix, sys.platform -> linux2 > : Linux-2.6.31-17-generic-x86_64-with-Ubuntu-9.10-karmic > Python info : 2.6.4 (r264:75706, Dec 7 2009, 18:43:55) > [GCC 4.4.1] > > Running from an installed IPython: False > > Tools and libraries available at test time: > curses foolscap gobject gtk pexpect twisted wx zope.interface > > Tools and libraries NOT available at test time: > objc wx.aui > > Ran 11 test groups in 38.313s > > Status: > OK > > > etc... > > Please paste this little report at the end, it will help us quite a > bit understand better how your environment looks. > Direct test from after pulling your dev branch, I still get the same errors as before. Only 8 groups tested comparing to your 9. This is Fedora 12. ********************************************************************** Test suite completed for system with the following information: IPython version: 0.11.bzr.r1346 BZR revision : 1346 Platform info : os.name -> posix, sys.platform -> linux2 : Linux-2.6.31.9-174.fc12.i686.PAE-i686-with-fedora-12-Constantine Python info : 2.6.2 (r262:71600, Aug 21 2009, 12:22:21) [GCC 4.4.1 20090818 (Red Hat 4.4.1-6)] Running from an installed IPython: False Tools and libraries available at test time: curses gobject gtk pexpect Tools and libraries NOT available at test time: foolscap objc twisted wx wx.aui zope.interface Ran 8 test groups in 6.615s Status: ERROR - 2 out of 8 test groups failed. ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python /home/gsever/Desktop/python-repo/ipython-dev/iptest.py IPython.core ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: python /home/gsever/Desktop/python-repo/ipython-dev/iptest.py IPython.extensions NOTE: In [3]: import IP here tab-complete doesn't work for IPython itself. Isn't it weird? Does it work on your system? It is still the same after I installed it. The same errors and no kernel test run. > > Interestingly, tests for IPython.kernel is still notting running. > > Also let us know about this... > > I've uploaded updated builds at the usual place: > > http://ipython.scipy.org/dist/testing/ > > Cheers, > > f > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Mon Jan 18 01:02:51 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 22:02:51 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> Message-ID: Hey, On Sun, Jan 17, 2010 at 9:59 PM, G?khan Sever wrote: > > NOTE: > > In [3]: import IP > > here tab-complete doesn't work for IPython itself. Isn't it weird? Does it > work on your system? No, that's a known regression: the completion extensions haven't been ported to the new API yet, so only basic completion is working right now. Your first contribution on this front for 0.11 would make lots of people happy :) > It is still the same after I installed it. The same errors and no kernel > test run. OK, that makes sense now, given this: Tools and libraries NOT available at test time: foolscap objc twisted wx wx.aui zope.interface Without foolscap, twisted and zope.interface, the kernel can't be tested. If you install those and try again, the kernel tests can run. BTW, could you paste again the current error tracebacks you're getting so I see them exactly now? Cheers, f From gokhansever at gmail.com Mon Jan 18 01:21:14 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Mon, 18 Jan 2010 00:21:14 -0600 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> Message-ID: <49d6b3501001172221i4d83f432u681f7cea8f658677@mail.gmail.com> On Mon, Jan 18, 2010 at 12:02 AM, Fernando Perez wrote: > Hey, > > On Sun, Jan 17, 2010 at 9:59 PM, G?khan Sever > wrote: > > > > NOTE: > > > > In [3]: import IP > > > > here tab-complete doesn't work for IPython itself. Isn't it weird? Does > it > > work on your system? > > No, that's a known regression: the completion extensions haven't been > ported to the new API yet, so only basic completion is working right > now. Your first contribution on this front for 0.11 would make lots > of people happy :) > I always complain about this tab-completion issue, don't I ? :D IPython is a very complete and complex project to delve in to learn the nitty-gritty details of Python. Well I need to advance on my MSc degree first to be able to move on as a PhD student. That means, I will most likely keep my "complaining" "testing" and "feature-requesting" role for a while :) > > > It is still the same after I installed it. The same errors and no kernel > > test run. > > OK, that makes sense now, given this: > > Tools and libraries NOT available at test time: > foolscap objc twisted wx wx.aui zope.interface > > Without foolscap, twisted and zope.interface, the kernel can't be tested. > > If you install those and try again, the kernel tests can run. > > BTW, could you paste again the current error tracebacks you're getting > so I see them exactly now? > Here it comes all of the optional requirements installed except objc (Although I installed gcc-objc and libobjc): ********************************************************************** Test suite completed for system with the following information: IPython version: 0.11.bzr.r1346 BZR revision : 1346 Platform info : os.name -> posix, sys.platform -> linux2 : Linux-2.6.31.9-174.fc12.i686.PAE-i686-with-fedora-12-Constantine Python info : 2.6.2 (r262:71600, Aug 21 2009, 12:22:21) [GCC 4.4.1 20090818 (Red Hat 4.4.1-6)] Running from an installed IPython: True Tools and libraries available at test time: curses foolscap gobject gtk pexpect twisted wx wx.aui zope.interface Tools and libraries NOT available at test time: objc Ran 11 test groups in 32.203s Status: ERROR - 2 out of 11 test groups failed. ---------------------------------------- Runner failed: IPython.core You may wish to rerun this one individually, with: python /usr/bin/iptest IPython.core ---------------------------------------- Runner failed: IPython.extensions You may wish to rerun this one individually, with: python /usr/bin/iptest IPython.extensions [gsever at ccn Desktop]$ python /usr/bin/iptest IPython.core ................................................................................S..............F.F.. ====================================================================== FAIL: Test that object's __del__ methods are called on exit. ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", line 160, in test_obj_del tt.ipexec_validate(self.fname, 'object A deleted') File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 312, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >> raise self.failureException, \ (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A deleted')) ====================================================================== FAIL: IPython.core.tests.test_run.TestMagicRunSimple.test_tclass ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", line 174, in test_tclass tt.ipexec_validate(self.fname, out) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 312, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: "\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" != "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first" >> raise self.failureException, \ (None or '%r != %r' % ("\x1b[?1034hARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first", "ARGV 1-: ['C-first']\nARGV 1-: ['C-second']\ntclass.py: deleting object: C-first")) ---------------------------------------------------------------------- Ran 102 tests in 0.779s FAILED (SKIP=1, failures=2) [gsever at ccn Desktop]$ python /usr/bin/iptest IPython.extensions ..F. ====================================================================== FAIL: IPython.extensions.tests.test_pretty.TestPrettyInteractively.test_printers ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest self.test(*self.arg) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", line 225, in skipper_func return f(*args, **kwargs) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/extensions/tests/test_pretty.py", line 101, in test_printers tt.ipexec_validate(self.fname, ipy_out) File "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", line 312, in ipexec_validate nt.assert_equals(out.strip(), expected_out.strip()) AssertionError: '\x1b[?1034hA()\nB()\n\n' != 'A()\nB()\n\n' >> raise self.failureException, \ (None or '%r != %r' % ('\x1b[?1034hA()\nB()\n\n', 'A()\nB()\n\n')) ---------------------------------------------------------------------- Ran 4 tests in 0.288s FAILED (failures=1) > > Cheers, > > f > -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Mon Jan 18 01:33:14 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 22:33:14 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: <49d6b3501001172221i4d83f432u681f7cea8f658677@mail.gmail.com> References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> <49d6b3501001172221i4d83f432u681f7cea8f658677@mail.gmail.com> Message-ID: Hey, On Sun, Jan 17, 2010 at 10:21 PM, G?khan Sever wrote: > I always complain about this tab-completion issue, don't I ? :D IPython is a > very complete and complex project to delve in to learn the nitty-gritty > details of Python. Well I need to advance on my MSc degree first to be able > to move on as a PhD student. That means, I will most likely keep my > "complaining" "testing" and "feature-requesting" role for a while :) No worries, this is actually very useful, and we appreciate it greatly. > Here it comes all of the optional requirements installed except objc > (Although I installed gcc-objc and libobjc): No problem, objc really means the python bindings, which are normally only available on OSX. So you're fine. > FAIL: Test that object's __del__ methods are called on exit. > ---------------------------------------------------------------------- > Traceback (most recent call last): > ? File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest > ??? self.test(*self.arg) > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", > line 225, in skipper_func > ??? return f(*args, **kwargs) > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", > line 160, in test_obj_del > ??? tt.ipexec_validate(self.fname, 'object A deleted') > ? File > "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", > line 312, in ipexec_validate > ??? nt.assert_equals(out.strip(), expected_out.strip()) > AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >>>? raise self.failureException, \ > ????????? (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A > deleted')) OK, I'm officially stumped. I've run the tests on Ubuntu 32, Ubuntu64, Fedora11/64, OSX 10.6 and Windows XP, and I can't reproduce this. I'm downloading a Fedora12 ISO and will try to build a VirtualBox installation of it, but if anyone beats me to the solution I'd be grateful, as I need to switch gears to other pressing matters right now. Gokhan, in the meantime, just try to use this new ipython and let us know in general use how it behaves, if things work well or not, etc. File bugs at will :) Any feedback is much appreciated. Cheers, f From fperez.net at gmail.com Mon Jan 18 02:51:34 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 17 Jan 2010 23:51:34 -0800 Subject: [IPython-dev] What do we do about repository formats??? Message-ID: Hi folks, I needed to test how things would go with a temporary merge of trunk, my trunk-dev and Dav's branch: https://code.launchpad.net/~ipython-contrib/ipython/traitlets-rename which is essentially everything we have in flight right now, but it seems we have a problem. Dav started with a more updated bzr on his system, so his branch uses a new repo format, and I can't simply keep it in the same shared repo as the others, merge, etc: maqroll[repo]> bzr branch lp:~ipython-contrib/ipython/traitlets-rename bzr: ERROR: KnitPackRepository('file:///home/fperez/ipython/repo/.bzr/repository/') is not compatible with RemoteRepository(bzr+ssh://bazaar.launchpad.net/~ipython-contrib/ipython/traitlets-rename/.bzr/) different rich-root support There are tons of reports about this, here's one example: https://bugs.launchpad.net/bzr/+bug/450147 I tried running upgrade on my local repo, and now I can pull Dav's branch, but I can't push anymore: maqroll[0.11-tmp]> bzr push --remember lp:~fdo.perez/ipython/0.11-tmp Using default stacking branch /~ipython-dev/ipython/trunk at lp-64843152:///~fdo.perez/ipython bzr: ERROR: KnitPackRepository('lp-64843152:///~ipython-dev/ipython/trunk/.bzr/repository') is not compatible with CHKInventoryRepository('lp-64843152:///~fdo.perez/ipython/0.11-tmp/.bzr/repository') different rich-root support Now I could issue lp commands and upgrade the repos in lp itself, but that's going to cascade to everyone else, so I don't want to do that without further feedback. It's nuts that bzr would make a change like this with such a nasty cascading effect... Ideas from the experts? (I know, I want to switch to git too so we can get off the mess that is bzr and its never ending stream of weird problems, but that's probably too disruptive for now, so let's stick to solutions that use bzr...) Thanks, f ps - it's not Dav's fault; anyone new contributor starting fresh would end up in this situation. From gokhansever at gmail.com Mon Jan 18 03:25:53 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Mon, 18 Jan 2010 02:25:53 -0600 Subject: [IPython-dev] Closing in on 0.11? We need some review/feedback help In-Reply-To: References: <49d6b3501001131448x7bfc8a17nfc3fa71a224acd97@mail.gmail.com> <49d6b3501001162148v57b31ae7n26fdce6609f33a9b@mail.gmail.com> Message-ID: <49d6b3501001180025q711d6026r9710af66b3807dc3@mail.gmail.com> On Sun, Jan 17, 2010 at 7:27 PM, Fernando Perez wrote: > On Sat, Jan 16, 2010 at 9:48 PM, G?khan Sever > wrote: > > 1- Look in the history in the reverse order to obtain the latest entry of > > the same variable assignment. Each element listed in the history output > is a > > separate string pulled out of a list. > > > > 2- Compare the "string".split("=")[0] with each variable name in %whos > call. > > > > 3-Write the output in the 4th column of the whos listing (using a one of > the > > names: "Assignment", "Command", "History"): > > > > Variable Type Data/Info Assignment > > -------------------------------------------------- > > a int 5 a=5 > > n float 0.0 n = math.sin(0) > > > > Data/Info column usually produces longer lines this might cause some ugly > > looking output. > > > > Could I get some suggestions? > > Well, sounds like you have a plan, go ahead and implement it :) > > Some pointers: > > - get the global ipython object via get_ipython() > - its .user_ns is your user namespace. > - in there, the 'In' variable contains your inputs. > > I'd just add this as a normal function initially loaded in your > startup file. Once it's working the way you want it, you just need a > few lines to expose it as a magic for convenience. > > Cheers, > f > I placed the following a few lines of code into magic_whos() in magic.py: cmdlist = {} for cmd in self.get_ipython().user_ns['In']: for vn in varnames: if cmd.split("=")[0].rstrip() == vn: cmdlist[vn] = str(cmd.strip()) print cmdlist Here is a mini demo: In [1]: a=5 In [2]: whos {'a': 'a=5'} Variable Type Data/Info ---------------------------- a int 5 In [3]: a = abs(1-2+3-4+5-6+7-8+9-10) In [4]: whos {'a': 'a = abs(1-2+3-4+5-6+7-8+9-10)'} Variable Type Data/Info ---------------------------- a int 5 We need a smart and clear space for this new column. Putting under each variable makes it look uglier. Putting next to Data/Info on a new column titled "Command" needs some smart length checking. Alternatively, the outputs could be embedded into the Data/Info fields directly. -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From gael.varoquaux at normalesup.org Mon Jan 18 08:37:30 2010 From: gael.varoquaux at normalesup.org (Gael Varoquaux) Date: Mon, 18 Jan 2010 14:37:30 +0100 Subject: [IPython-dev] [MayaVi-users] Updated build of trunk-dev available In-Reply-To: <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> Message-ID: <20100118133730.GK27552@phare.normalesup.org> On Sun, Jan 17, 2010 at 12:39:22PM +0000, Michele Mattioni wrote: > Mmh, unfortunately I can't seem to run your script, because I get a > traceback: > In [2]: run qt_embedding.py > Warning: Unable to import the qt4 backend for pyface due to traceback: > Traceback (most recent call last): > ? ? ? ? ?File > "/usr/lib/python2.6/dist-packages/enthought/pyface/toolkit.py", > line 40, in _init_toolkit > ? ? ? ? ? ?__import__(be + 'init') > ? ? ? ?ImportError: No module named qt4.init > Info: Unable to import any backend (qt4) for pyface; using the 'null' > toolkit instead. > Yes, the latest mayavi api for qt integration are only in the latest SVN > Enthought I am a bit surprised. Fernando, could it be that you don't have the QT backend installed for Traits? It will be named something like 'python-traitsbackendqt'. Ga?l From fperez.net at gmail.com Mon Jan 18 17:54:59 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 18 Jan 2010 14:54:59 -0800 Subject: [IPython-dev] [MayaVi-users] Updated build of trunk-dev available In-Reply-To: <20100118133730.GK27552@phare.normalesup.org> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> <20100118133730.GK27552@phare.normalesup.org> Message-ID: Hey, On Mon, Jan 18, 2010 at 5:37 AM, Gael Varoquaux wrote: > > I am a bit surprised. Fernando, could it be that you don't have the QT > backend installed for Traits? It will be named something like > 'python-traitsbackendqt'. > Thanks, that was it! I didn't realize I hadn't installed enough of the traits packages on my Karmic box, sorry... Now I do get a different error though: In [3]: run qt_embedding.py --------------------------------------------------------------------------- ImportError Traceback (most recent call last) /home/fperez/Desktop/qt_embedding.py in () 17 Int, Dict 18 from enthought.traits.ui.api import View, Item ---> 19 from enthought.mayavi.core.ui.api import MayaviScene, MlabSceneModel, \ 20 SceneEditor 21 from enthought.mayavi import mlab ImportError: No module named api Is it possible to have a version of your script that runs with the mayavi shipped in Karmic? Because now with Gael's note and your script, I was able to test things interactively: In [1]: import os In [2]: os.environ['ETS_TOOLKIT'] = 'qt4' In [3]: gui qt In [4]: from enthought.mayavi import mlab In [5]: mlab.test_molecule () Out[5]: (, , , ) In [6]: mlab.figure() Out[6]: In [7]: mlab.test_surf () Out[7]: In [8]: mlab.test_surf () Out[8]: In [9]: pylab qt Welcome to pylab, a matplotlib-based Python environment [backend: Qt4Agg]. For more information, type 'help(pylab)'. In [10]: plt.figure() Out[10]: In [11]: plot(sin(linspace(0,2*pi,200)**2) ....: ) Out[11]: [] In [12]: And everything is fine: mayavi windows open up, their qt dialogs work, I can then load pylab with the qt backend and it works, etc. It's *really* cool, in fact. So I'd love to be able to reproduce your bug, so we can track it down to either something we need to fix in ipython or something that can be fixed upstream in mayavi... Cheers, f From gael.varoquaux at normalesup.org Mon Jan 18 17:56:40 2010 From: gael.varoquaux at normalesup.org (Gael Varoquaux) Date: Mon, 18 Jan 2010 23:56:40 +0100 Subject: [IPython-dev] [MayaVi-users] Updated build of trunk-dev available In-Reply-To: References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> <20100118133730.GK27552@phare.normalesup.org> Message-ID: <20100118225640.GD1450@phare.normalesup.org> On Mon, Jan 18, 2010 at 02:54:59PM -0800, Fernando Perez wrote: > /home/fperez/Desktop/qt_embedding.py in () > 17 Int, Dict > 18 from enthought.traits.ui.api import View, Item > ---> 19 from enthought.mayavi.core.ui.api import MayaviScene, MlabSceneModel, \ > 20 SceneEditor > 21 from enthought.mayavi import mlab > ImportError: No module named api Yes, that's new. You can just replace the api imports with the ones in: https://svn.enthought.com/enthought/browser/Mayavi/trunk/enthought/mayavi/core/ui/api.py HTH, Ga?l From fperez.net at gmail.com Mon Jan 18 18:18:03 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 18 Jan 2010 15:18:03 -0800 Subject: [IPython-dev] [MayaVi-users] Updated build of trunk-dev available In-Reply-To: <20100118225640.GD1450@phare.normalesup.org> References: <290ea1b01001170135v7ecc7e13t425c830613b809f1@mail.gmail.com> <290ea1b01001170439r22d78dc2r24a23fca8e3a33b3@mail.gmail.com> <20100118133730.GK27552@phare.normalesup.org> <20100118225640.GD1450@phare.normalesup.org> Message-ID: On Mon, Jan 18, 2010 at 2:56 PM, Gael Varoquaux wrote: > > Yes, that's new. You can just replace the api imports with the ones in: > > https://svn.enthought.com/enthought/browser/Mayavi/trunk/enthought/mayavi/core/ui/api.py > > HTH, Awesome, many thanks. At least now I can reproduce the problem here as well, I've updated the bug report so anyone can download the test script: https://bugs.launchpad.net/ipython/+bug/508953 This is indeed a pretty serious regression, so I made it critical for 0.11. But I don't know what the problem is, so help from anyone would be very, very welcome. I can't work on this right now, unfortunately. Cheers, f From fperez.net at gmail.com Wed Jan 20 04:25:46 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Wed, 20 Jan 2010 01:25:46 -0800 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> <49d6b3501001172221i4d83f432u681f7cea8f658677@mail.gmail.com> Message-ID: On Sun, Jan 17, 2010 at 10:33 PM, Fernando Perez wrote: > >> FAIL: Test that object's __del__ methods are called on exit. >> ---------------------------------------------------------------------- >> Traceback (most recent call last): >> ? File "/usr/lib/python2.6/site-packages/nose/case.py", line 182, in runTest >> ??? self.test(*self.arg) >> ? File >> "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/decorators.py", >> line 225, in skipper_func >> ??? return f(*args, **kwargs) >> ? File >> "/home/gsever/Desktop/python-repo/ipython-dev/IPython/core/tests/test_run.py", >> line 160, in test_obj_del >> ??? tt.ipexec_validate(self.fname, 'object A deleted') >> ? File >> "/home/gsever/Desktop/python-repo/ipython-dev/IPython/testing/tools.py", >> line 312, in ipexec_validate >> ??? nt.assert_equals(out.strip(), expected_out.strip()) >> AssertionError: '\x1b[?1034hobject A deleted' != 'object A deleted' >>>>? raise self.failureException, \ >> ????????? (None or '%r != %r' % ('\x1b[?1034hobject A deleted', 'object A >> deleted')) > > OK, I'm officially stumped. > > I've run the tests on Ubuntu 32, Ubuntu64, Fedora11/64, OSX 10.6 and > Windows XP, and I can't reproduce this. > > I'm downloading a Fedora12 ISO and will try to build a VirtualBox > installation of it, but if anyone beats me to the solution I'd be > grateful, as I need to switch gears to other pressing matters right > now. Just to udpate you Gokhan, I've installed the FC12 virtual machine (VirtualBox rocks, btw), and I can see the problem. It's really odd that it's there on fc12 but not on fc11, and I don't know yet where it comes from. But being able to reproduce it locally, I should be able to get to the bottom of this before 0.11 is out. Thanks for the detailed reporting! f From Nicolas.Rougier at loria.fr Wed Jan 20 04:45:26 2010 From: Nicolas.Rougier at loria.fr (Nicolas Rougier) Date: Wed, 20 Jan 2010 10:45:26 +0100 Subject: [IPython-dev] launch_new_instance and thread Message-ID: <1263980726.3017.26.camel@sulfur> Hello, I've a simple question related to IPython.ipapi.launch_new_instance. In order to have an interactive console using OpenGL and GLUT, I have to start an ipython shell within its own thread and then, when the user closes the main GLUT window, I need to close the ipython shell as well from the GL shell but I don't know how to do it (currently I use a join but it leaves the console in a weird state). Is there a proper way to do that ? Or do I need to start the IPython shell in another way ? Nicolas From gokhansever at gmail.com Wed Jan 20 11:31:21 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Wed, 20 Jan 2010 10:31:21 -0600 Subject: [IPython-dev] Updated build of trunk-dev available In-Reply-To: References: <201001162151.57608.lists_ravi@lavabit.com> <49d6b3501001162012t5f7adebel264c487fc4e24a1a@mail.gmail.com> <49d6b3501001172159g76436d6va5b4af7552c7368f@mail.gmail.com> <49d6b3501001172221i4d83f432u681f7cea8f658677@mail.gmail.com> Message-ID: <49d6b3501001200831q1eaf8291h75abb7e6d7d4fa82@mail.gmail.com> On Wed, Jan 20, 2010 at 3:25 AM, Fernando Perez wrote: > Just to udpate you Gokhan, I've installed the FC12 virtual machine > (VirtualBox rocks, btw), and I can see the problem. It's really odd > that it's there on fc12 but not on fc11, and I don't know yet where it > comes from. But being able to reproduce it locally, I should be able > to get to the bottom of this before 0.11 is out. > > Thanks for the detailed reporting! > > f > Fernando, Glad to see that you are very persistent not to leave any testing failure behind. We can ask the issue to RedHat staff and other Fedora Python maintainers to get more insight on the issue. There is a dedicated support for Python on Fedora as well. They are planning to ship Python 2.6 and Python 3.x in the next release. -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From vivainio at gmail.com Wed Jan 20 16:46:21 2010 From: vivainio at gmail.com (Ville M. Vainio) Date: Wed, 20 Jan 2010 23:46:21 +0200 Subject: [IPython-dev] What do we do about repository formats??? In-Reply-To: References: Message-ID: <46cb515a1001201346y6e10c58ar58a99141fffde0c4@mail.gmail.com> On Mon, Jan 18, 2010 at 9:51 AM, Fernando Perez wrote: > Now I could issue lp commands and upgrade the repos in lp itself, but > that's going to cascade to everyone else, so I don't want to do that > without further feedback. We did it for Leo and it didn't cause problems. People just needed to get a fresh copy from launchpad and merge their changes there. Remote repository upgrade did take quite a bit of time, though (possibly hours?). > It's nuts that bzr would make a change like this with such a nasty > cascading effect... It's actually quite simple to get over this; I agree that launchpad admins could have made a little bit more noise about this change. > Ideas from the experts? (I know, I want to switch to git too so we can > get off the mess that is bzr and its never ending stream of weird > problems, but that's probably too disruptive for now, so let's stick > to solutions that use bzr...) Judging by my experience with git, I don't think it would be much of an improvement. Mercurial, on the other hand... -- Ville M. Vainio http://tinyurl.com/vainio From gokhansever at gmail.com Fri Jan 22 16:20:21 2010 From: gokhansever at gmail.com (=?UTF-8?Q?G=C3=B6khan_Sever?=) Date: Fri, 22 Jan 2010 15:20:21 -0600 Subject: [IPython-dev] isympy doesn't work with the new ipython-dev Message-ID: <49d6b3501001221320x43700e51m914d366c11b7cbf3@mail.gmail.com> Hello, I can't get the isympy working with the Fernando's trunk. I solved one of the missing attribute error but stuck at the other one: [gsever at ccn bin]$ ./isympy Traceback (most recent call last): File "./isympy", line 153, in main() File "./isympy", line 147, in main init_session('ipython', **args) File "/home/gsever/Desktop/python-repo/sympy/sympy/interactive/__init__.py", line 97, in init_session ip = IPython.ipapi.get() AttributeError: 'module' object has no attribute 'ipapi' [gsever at ccn bin]$ ./isympy Traceback (most recent call last): File "./isympy", line 153, in main() File "./isympy", line 147, in main init_session('ipython', **args) File "/home/gsever/Desktop/python-repo/sympy/sympy/interactive/__init__.py", line 106, in init_session ip = init_IPython() File "/home/gsever/Desktop/python-repo/sympy/sympy/interactive/__init__.py", line 61, in init_IPython return IPython.Shell.make_IPython(argv) AttributeError: 'module' object has no attribute 'Shell' Could you give it a try installing sympy from http://code.google.com/p/sympy/ Thanks. -- G?khan -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Fri Jan 22 19:58:06 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Fri, 22 Jan 2010 16:58:06 -0800 Subject: [IPython-dev] isympy doesn't work with the new ipython-dev In-Reply-To: <49d6b3501001221320x43700e51m914d366c11b7cbf3@mail.gmail.com> References: <49d6b3501001221320x43700e51m914d366c11b7cbf3@mail.gmail.com> Message-ID: On Fri, Jan 22, 2010 at 1:20 PM, G?khan Sever wrote: > > I can't get the isympy working with the Fernando's trunk. I solved one of > the missing attribute error but stuck at the other one: No, things have changed a bit. I'm insanely pressed for time right now, the best I suggest is looking at this test code file, around line 130: http://bazaar.launchpad.net/%7Efdo.perez/ipython/trunk-dev/annotate/head%3A/IPython/testing/globalipapp.py#L130 That makes the kind of shell you're looking for, sort of. Sorry not to be able to give you a detailed example right now... cheers, f From ellisonbg.net at gmail.com Sun Jan 24 16:43:42 2010 From: ellisonbg.net at gmail.com (Brian Granger) Date: Sun, 24 Jan 2010 13:43:42 -0800 Subject: [IPython-dev] Support for Windows HPC Server 2008 Message-ID: <6ce0ac131001241343k79226001uda159572326389ee@mail.gmail.com> Hi, I wanted to let everyone know that the upcoming 0.11 release of IPython (now in trunk) has support for Windows HPC Server 2008. What does this mean? Now, IPython's parallel computing capabilties (ipcluster/ipengine/ipcontroller) are fully integrated with the Windows HPC job scheduler. The Windows HPC job scheduler is quite nice (one of the nicer batch systems I have dealt with) and this integration makes it quite easy to use start and stop IPython controllers.engines on a Windows cluster. Here are some links about this: A whitepaper about using IPython on Windows HPC Server 2008: http://ipython.scipy.org/moin/News?action=AttachFile&do=get&target=ipython_winhpc_whitepaper_v1.pdf A short announcement on the IPython website: http://ipython.scipy.org/moin/News?action=show Docs: http://ipython.scipy.org/doc/nightly/html/parallel/parallel_winhpc.html http://ipython.scipy.org/doc/nightly/html/parallel/parallel_demos.html Screencasts on Microsoft's dev network: http://channel9.msdn.com/shows/The+HPC+Show/Open-source-HPC-code-Episode-11-IPython-Grid-Engine-running-on-Windows-HPC-Server-2008/ http://channel9.msdn.com/shows/The+HPC+Show/Open-source-HPC-code-Episode-12-IPython-computes-150-million-digits-of-Pi-in-Parallel/ Microsoft was very generous is funding Brian Granger to work on this. We hope that this is the beginning of more open collaboration between Microsoft and the various open source projects in Python ecosystem. Cheers, Brian -------------- next part -------------- An HTML attachment was scrubbed... URL: From fperez.net at gmail.com Mon Jan 25 01:03:47 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Sun, 24 Jan 2010 22:03:47 -0800 Subject: [IPython-dev] Support for Windows HPC Server 2008 In-Reply-To: <6ce0ac131001241343k79226001uda159572326389ee@mail.gmail.com> References: <6ce0ac131001241343k79226001uda159572326389ee@mail.gmail.com> Message-ID: On Sun, Jan 24, 2010 at 1:43 PM, Brian Granger wrote: > I wanted to let everyone know that the upcoming 0.11 release of IPython (now > in trunk) has support for Windows HPC Server 2008.? What does this mean? > Now, IPython's parallel computing capabilties > (ipcluster/ipengine/ipcontroller) are fully integrated with the Windows HPC > job scheduler.? The Windows HPC job scheduler is quite nice (one of the > nicer batch systems I have dealt with) and this integration makes it quite > easy to use start and stop IPython controllers.engines on a Windows > cluster.? Here are some links about this: In addition to being very happy about this and thanking Microsoft for their support, we'd like to add that we're eager for feedback on users who may have this platform. This is new code, hence its APIs haven't been widely used yet and we'd love to hear of what works and what could be improved before they settle too much. Cheers, f From Nicolas.Rougier at loria.fr Mon Jan 25 05:46:05 2010 From: Nicolas.Rougier at loria.fr (Nicolas Rougier) Date: Mon, 25 Jan 2010 11:46:05 +0100 Subject: [IPython-dev] glumpy, fast opengl visualization Message-ID: <1264416365.376.19.camel@sulfur> Hello, This is an update about glumpy, a fast-OpenGL based numpy visualization. I modified the code such that the only dependencies are PyOpenGL and IPython (for interactive sessions). You will also need matplotlib and scipy for some demos. Sources: hg clone http://glumpy.googlecode.com/hg/ glumpy No installation required, you can run all demos inplace. Homepage: http://code.google.com/p/glumpy/ Nicolas From fperez.net at gmail.com Mon Jan 25 19:11:29 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Mon, 25 Jan 2010 16:11:29 -0800 Subject: [IPython-dev] glumpy, fast opengl visualization In-Reply-To: <1264416365.376.19.camel@sulfur> References: <1264416365.376.19.camel@sulfur> Message-ID: Hi Nicolas, On Mon, Jan 25, 2010 at 2:46 AM, Nicolas Rougier wrote: > > > Hello, > > This is an update about glumpy, a fast-OpenGL based numpy visualization. > I modified the code such that the only dependencies are PyOpenGL and > IPython (for interactive sessions). You will also need matplotlib and > scipy for some demos. > > Sources: hg clone http://glumpy.googlecode.com/hg/ glumpy > No installation required, you can run all demos inplace. > > Homepage: http://code.google.com/p/glumpy/ This is great, and it would be very cool to have it updated to the new code we're now landing in ipython with a much cleaner internal API (finally :) Have you had a chance to look at the code in my trunk-dev branch? https://code.launchpad.net/~fdo.perez/ipython/trunk-dev Brian finished a large review of it and we just had a chance to go over his feedback directly, so there's now one more round of reviews to do (once he applies the changes from our discussion) and this should become trunk very soon. The apis are much cleaner, this is the big cleanup I told you about last year, and now we're getting to the point where having multiple ipython frontends is a very realistic prospect. Unfortunately we won't be able to use your code directly in IPython as it stands, since the GPL provisions in it would require us to GPL all of IPython to make use of any of it directly in IPython. Your code uses iptyhon, numpy, matplotlib and scipy (in some demos), which amounts to hundreds of thousands of lines of code; here are the sloccount outputs from their respective trunks: IPython Totals grouped by language (dominant language first): python: 47357 (99.24%) lisp: 262 (0.55%) sh: 62 (0.13%) objc: 37 (0.08%) Numpy Totals grouped by language (dominant language first): ansic: 152950 (67.19%) python: 73188 (32.15%) cpp: 828 (0.36%) fortran: 298 (0.13%) sh: 156 (0.07%) pascal: 120 (0.05%) f90: 97 (0.04%) Matplotlib Totals grouped by language (dominant language first): python: 83290 (52.64%) cpp: 68212 (43.11%) objc: 4517 (2.85%) ansic: 2149 (1.36%) sh: 69 (0.04%) Scipy Totals grouped by language (dominant language first): cpp: 220149 (48.35%) fortran: 87240 (19.16%) python: 79164 (17.38%) ansic: 68746 (15.10%) sh: 61 (0.01%) Glumpy: Totals grouped by language (dominant language first): python: 3751 (100.00%) We're looking at ~300.000 lines of python alone in these tools. It's unfortunately not realistic for us to consider GPL-ing them in order to incorporate glumpy into the core set; it would be fantastic if you were willing to consider licensing your code under a license that is compatible with the body of work you are building on top of. You are obviously free to choose your license as you see fit, and end users (myself included) will be always able to use glumpy along with ipython, numpy, matplotlib and scipy. So *users* get all of the benefit of your contribution, and for that I am the first to be delighted and grateful that you've put your code out there. But as it stands, your code builds on close to half a million lines of other code which can not benefit back from your contributions. If you consider licensing glumpy to be compatible with ipython, numpy and matplotlib, it would be possible to incorporate your ideas back into those projects: perhaps in some places the right solution would be to fix our own designs to better provide what glumpy needs, in other cases we may find fixes you've made fit better upstream, etc. But this kind of collaboration will not be possible as long as glumpy can benefit from our tools but our codes are not allowed to benefit from glumpy (without changing licenses, which isn't going to happen). I hope you consider this from our perspective and in the most friendly and open manner: I completely respect your right to license your own code as you see fit (I've seen people put out GPL 'projects' that effectively consist of 3 lines that import IPython and make a function call, and that's OK too, and allowed by the license I chose to use). The only reason I ask you is because I think your tool is very interesting, and it would ultimately lead to a much more productive relationship with ipython, numpy and matplotlib if it could be a collaboration instead of a one-way benefit. Best regards, Fernando. From Nicolas.Rougier at loria.fr Tue Jan 26 03:23:47 2010 From: Nicolas.Rougier at loria.fr (Nicolas Rougier) Date: Tue, 26 Jan 2010 09:23:47 +0100 Subject: [IPython-dev] glumpy, fast opengl visualization In-Reply-To: References: <1264416365.376.19.camel@sulfur> Message-ID: <1264494227.2911.6.camel@sulfur> Hi Fernando, I will change the license from GPL to BSD. Concerning the ipython-trunk, I'm not sure where to look exactly within your sources. Currently, I'm using the IPython.ipapi.launch_new_instance to start an ipython shell in its own thread (I need this because OpenGL requires to be in the main thread from what I remember). Furthermore I proxy all gl calls so they are executed within the main thread while standard python operations are made directly from the ipython thread. The only problem so far with this approach is to be able to kill the IPython session cleanly from the GL thread but I did not find the way to do it, is there something like this in the dev branch ? Nicolas On Mon, 2010-01-25 at 16:11 -0800, Fernando Perez wrote: > Hi Nicolas, > > On Mon, Jan 25, 2010 at 2:46 AM, Nicolas Rougier > wrote: > > > > > > Hello, > > > > This is an update about glumpy, a fast-OpenGL based numpy visualization. > > I modified the code such that the only dependencies are PyOpenGL and > > IPython (for interactive sessions). You will also need matplotlib and > > scipy for some demos. > > > > Sources: hg clone http://glumpy.googlecode.com/hg/ glumpy > > No installation required, you can run all demos inplace. > > > > Homepage: http://code.google.com/p/glumpy/ > > This is great, and it would be very cool to have it updated to the new > code we're now landing in ipython with a much cleaner internal API > (finally :) Have you had a chance to look at the code in my trunk-dev > branch? > > https://code.launchpad.net/~fdo.perez/ipython/trunk-dev > > Brian finished a large review of it and we just had a chance to go > over his feedback directly, so there's now one more round of reviews > to do (once he applies the changes from our discussion) and this > should become trunk very soon. The apis are much cleaner, this is the > big cleanup I told you about last year, and now we're getting to the > point where having multiple ipython frontends is a very realistic > prospect. > > Unfortunately we won't be able to use your code directly in IPython as > it stands, since the GPL provisions in it would require us to GPL all > of IPython to make use of any of it directly in IPython. Your code > uses iptyhon, numpy, matplotlib and scipy (in some demos), which > amounts to hundreds of thousands of lines of code; here are the > sloccount outputs from their respective trunks: > > IPython > Totals grouped by language (dominant language first): > python: 47357 (99.24%) > lisp: 262 (0.55%) > sh: 62 (0.13%) > objc: 37 (0.08%) > > > Numpy > Totals grouped by language (dominant language first): > ansic: 152950 (67.19%) > python: 73188 (32.15%) > cpp: 828 (0.36%) > fortran: 298 (0.13%) > sh: 156 (0.07%) > pascal: 120 (0.05%) > f90: 97 (0.04%) > > Matplotlib > Totals grouped by language (dominant language first): > python: 83290 (52.64%) > cpp: 68212 (43.11%) > objc: 4517 (2.85%) > ansic: 2149 (1.36%) > sh: 69 (0.04%) > > Scipy > Totals grouped by language (dominant language first): > cpp: 220149 (48.35%) > fortran: 87240 (19.16%) > python: 79164 (17.38%) > ansic: 68746 (15.10%) > sh: 61 (0.01%) > > Glumpy: > Totals grouped by language (dominant language first): > python: 3751 (100.00%) > > We're looking at ~300.000 lines of python alone in these tools. It's > unfortunately not realistic for us to consider GPL-ing them in order > to incorporate glumpy into the core set; it would be fantastic if you > were willing to consider licensing your code under a license that is > compatible with the body of work you are building on top of. > > You are obviously free to choose your license as you see fit, and end > users (myself included) will be always able to use glumpy along with > ipython, numpy, matplotlib and scipy. So *users* get all of the > benefit of your contribution, and for that I am the first to be > delighted and grateful that you've put your code out there. > > But as it stands, your code builds on close to half a million lines of > other code which can not benefit back from your contributions. If you > consider licensing glumpy to be compatible with ipython, numpy and > matplotlib, it would be possible to incorporate your ideas back into > those projects: perhaps in some places the right solution would be to > fix our own designs to better provide what glumpy needs, in other > cases we may find fixes you've made fit better upstream, etc. > > But this kind of collaboration will not be possible as long as glumpy > can benefit from our tools but our codes are not allowed to benefit > from glumpy (without changing licenses, which isn't going to happen). > > I hope you consider this from our perspective and in the most friendly > and open manner: I completely respect your right to license your own > code as you see fit (I've seen people put out GPL 'projects' that > effectively consist of 3 lines that import IPython and make a function > call, and that's OK too, and allowed by the license I chose to use). > The only reason I ask you is because I think your tool is very > interesting, and it would ultimately lead to a much more productive > relationship with ipython, numpy and matplotlib if it could be a > collaboration instead of a one-way benefit. > > Best regards, > > Fernando. From fperez.net at gmail.com Tue Jan 26 04:02:55 2010 From: fperez.net at gmail.com (Fernando Perez) Date: Tue, 26 Jan 2010 01:02:55 -0800 Subject: [IPython-dev] glumpy, fast opengl visualization In-Reply-To: <1264494227.2911.6.camel@sulfur> References: <1264416365.376.19.camel@sulfur> <1264494227.2911.6.camel@sulfur> Message-ID: Dear Nicolas, On Tue, Jan 26, 2010 at 12:23 AM, Nicolas Rougier wrote: > I will change the license from GPL to BSD. Well, many thanks! I greatly appreciate this gesture, I'm sure it will benefit us all. > Concerning the ipython-trunk, I'm not sure where to look exactly within > your sources. Currently, I'm using the IPython.ipapi.launch_new_instance > to start an ipython shell in its own thread (I need this because OpenGL > requires to be in the main thread from what I remember). Furthermore I > proxy all gl calls so they are executed within the main thread while > standard python operations are made directly from the ipython thread. > > The only problem so far with this approach is to be able to kill the > IPython session cleanly from the GL thread but I did not find the way to > do it, is there something like this in the dev branch ? A few points: - You may want to simply try to do what %Exit does: def magic_Exit(self, parameter_s=''): """Exit IPython without confirmation.""" self.shell.ask_exit() I unfortunately can't verify if this works, because with your current hg code on my Karmic 32-bit laptop, I get this: maqroll[glumpy]> python -c 'import glumpy' Segmentation fault so none of the demos run at all. - Just a couple of days ago, Brian and I started thinking hard about the long-term design requirements of the interactive app for civilized cooperation with other mainloops (like what you need). Let us know if the above basic trick doesn't work, and very soon we'll try to finish up the current reviews in-progress and put up some design docs on these questions, so we can start building up with glumpy as cleanly as possible. Great timing! Last we'd spoken I said it would be a while before things got cleaned up enough to really make sure we weren't wasting your time, but you came back just as things are getting in shape over here. This should be fun :) Best regards, and many thanks again for your consideration of my licensing discussion, f From Nicolas.Rougier at loria.fr Thu Jan 28 03:47:12 2010 From: Nicolas.Rougier at loria.fr (Nicolas Rougier) Date: Thu, 28 Jan 2010 09:47:12 +0100 Subject: [IPython-dev] glumpy, fast opengl visualization In-Reply-To: References: <1264416365.376.19.camel@sulfur> <1264494227.2911.6.camel@sulfur> Message-ID: <1264668432.19467.11.camel@sulfur> Hi, I modified the sources to use ask_exit but I still have problems. In glut (not freeglut), the only way to exit the mainloop is to exec sys.exit() and this is a problem for the IPython shell that seems to be unable to exit cleanly in such a case (I use atexit to catch the exit). It lefts the tty in a weird state (no echo) and I need to save the tty state using termios to restore it properly before exiting. I think this problem is very specific to glut (which is rather old anyway) and I'm not sure there is smarter fix. Nicolas On Tue, 2010-01-26 at 01:02 -0800, Fernando Perez wrote: > Dear Nicolas, > > On Tue, Jan 26, 2010 at 12:23 AM, Nicolas Rougier > wrote: > > I will change the license from GPL to BSD. > > Well, many thanks! I greatly appreciate this gesture, I'm sure it > will benefit us all. > > > Concerning the ipython-trunk, I'm not sure where to look exactly within > > your sources. Currently, I'm using the IPython.ipapi.launch_new_instance > > to start an ipython shell in its own thread (I need this because OpenGL > > requires to be in the main thread from what I remember). Furthermore I > > proxy all gl calls so they are executed within the main thread while > > standard python operations are made directly from the ipython thread. > > > > The only problem so far with this approach is to be able to kill the > > IPython session cleanly from the GL thread but I did not find the way to > > do it, is there something like this in the dev branch ? > > A few points: > > - You may want to simply try to do what %Exit does: > > def magic_Exit(self, parameter_s=''): > """Exit IPython without confirmation.""" > > self.shell.ask_exit() > > I unfortunately can't verify if this works, because with your current > hg code on my Karmic 32-bit laptop, I get this: > > maqroll[glumpy]> python -c 'import glumpy' > Segmentation fault > > so none of the demos run at all. > > - Just a couple of days ago, Brian and I started thinking hard about > the long-term design requirements of the interactive app for civilized > cooperation with other mainloops (like what you need). Let us know if > the above basic trick doesn't work, and very soon we'll try to finish > up the current reviews in-progress and put up some design docs on > these questions, so we can start building up with glumpy as cleanly as > possible. > > > Great timing! Last we'd spoken I said it would be a while before > things got cleaned up enough to really make sure we weren't wasting > your time, but you came back just as things are getting in shape over > here. This should be fun :) > > Best regards, and many thanks again for your consideration of my > licensing discussion, > > f