From dave@spacehotel.com Wed Jan 1 01:22:26 2003 From: dave@spacehotel.com (Dave White) Date: Tue, 31 Dec 2002 17:22:26 -0800 Subject: [Pythonmac-SIG] httplib slow on MacOSX 10.2.3? Message-ID: <3E1242D2.6030206@spacehotel.com> Hello, I was moving over some python apps from linux to my new 1ghz powerbook and was running a few to see how fast they run. Normal calculations and threads ran fast, a little faster than linux (intel) mhz for mhz. However a load testing app I have that uses httplib runs incredibly slow, about 1/100th the speed when the app runs on a similarly configured linux machine. I've narrowed it down to the calls to httplib, tested this on 2 macs running Jaguar with very similar results. Here's the code I use, basically calling this repeatedly via multiple threads to simulate a heavy load on a web server: conn = httplib.HTTPConnection("www.python.org") conn.request("GET", "/index.html") r1 = conn.getresponse() conn.close() I tried replacing the code above with pow(999,999) to test just the threads and mac os x is really fast, it really seems that something is funky with mac os x and httplib. Has anyone experienced this? I'm using the stock python 2.2 that ships with jaguar. cheers, Dave White From jhrsn+@pitt.edu Thu Jan 2 16:46:58 2003 From: jhrsn+@pitt.edu (James Harrison) Date: Thu, 02 Jan 2003 11:46:58 -0500 Subject: [Pythonmac-SIG] Problem building 2.3a1 on OSX Message-ID: I downloaded the new 2.3a1 distribution this morning and tried a build (first time for me with 2.3). I built as a user with admin privileges (didn't use sudo). I followed the instructions for the framework build; ./configure --enable-framework and make seemed to work OK. 201 tests were OK with 28 skipped (1 unexpected on darwin: test_bz2). However, make frameworkinstall seemed to have some problems: ... /usr/bin/install -c -s ./python.exe /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/2.3/R esources/Python.app/Contents/MacOS/python ln -fsn ../../../../English.lproj/Documentation /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/2.3/R esources/Python.app/Contents/Resources/English.lproj/Documentation /usr/bin/install -c -d /Library/Frameworks/../../usr/local/bin /usr/bin/install: chmod 755 /Library/Frameworks/../../usr/local/bin: Operation not permitted ln -fs /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/2.3/b in/python /Library/Frameworks/../../usr/local/bin/python ln: /Library/Frameworks/../../usr/local/bin/python: Permission denied make[1]: *** [installunixtools] Error 1 make: *** [frameworkinstallunixtools] Error 2 It appears that the framework was installed in /Library/Frameworks as intended. There is a link in /usr/local/bin that correctly starts 2.3a1 and basic functionality is available. The unix tools do not appear to have installed in /usr/local/bin. There is a MacPython-2.3 folder in /Applications, but it contains only BuildApplet and PythonLauncher. PythonLauncher appears to work. I do have a basic install of fink on the system, though I haven't used it to install software yet (I seen the comment about potential fink problems in the docs). Any comments are appreciated. Thanks, Jim Harrison University of Pittsburgh From altis@semi-retired.com Thu Jan 2 20:04:47 2003 From: altis@semi-retired.com (Kevin Altis) Date: Thu, 2 Jan 2003 12:04:47 -0800 Subject: [Pythonmac-SIG] two-level namespaces enabled? Message-ID: Bill Bumgarner has been porting some of my PythonCard sample code to work under PyObjC. Yesterday I sent him the life sample code cut down to just the pure Python calculation of each generation. Basically, the "grid" is just a dictionary of (x, y) tuples with values of either 0 or 1. No Mac libs, floating point, disk i/o, etc. is involved in the loop that I timed. Anyway, I ran it with both the Jaguar Python (/usr/bin/python) and the MachoPython 2.2.1-4 disk image (/usr/local/bin/python) at: http://sourceforge.net/project/showfiles.php?group_id=10718 and there was a 10% speed difference on the same box. I asked Bill for an explanation and he suggested... > From: Bill Bumgarner > > On Thursday, Jan 2, 2003, at 00:10 US/Eastern, Kevin Altis wrote: > > I was a bit surprised when I ran the cmdLife.py program on my Mac. > > /usr/bin/python (Apple) takes 38.8 seconds pretty consistently and > > /usr/local/bin/python (2.2.1 framework build) takes roughly 43.3. Not > > sure what that means, but it is a 10% difference. > > In any case, this is likely because the Framework build is not built > w/two level namespaces enabled and, as such, is causing an extra > indirection when dispatching calls into other frameworks [or something > like that]. > > The Apple build definitely had two-level namespaces enabled and is also > a static build, so no indirection is involved. I can send the code to anyone that is interested, but I bet it would probably be more useful to simply run Python through its test suite and see how the numbers compare on both versions in case this would have any impact on the 2.2.2 and/or 2.3 builds people are working on. ka --- Kevin Altis altis@semi-retired.com http://radio.weblogs.com/0102677/ From altis@semi-retired.com Thu Jan 2 20:21:05 2003 From: altis@semi-retired.com (Kevin Altis) Date: Thu, 2 Jan 2003 12:21:05 -0800 Subject: [Pythonmac-SIG] RE: [wxPython-mac] two-level namespaces enabled? In-Reply-To: Message-ID: Sorry I forgot to include this bit when I edited the original message > From: Bill Bumgarner > > If you have OS9 installed, try the same test under MacPython under > OS9-- if it is noticeably faster, then the test on OS X is getting > nailed by an unfortunate bit o' implementation in the Mach ABI, I > believe. I don't have OS9 installed, so I wasn't able to do the test. ka > -----Original Message----- > From: Kevin Altis [mailto:altis@semi-retired.com] > Sent: Thursday, January 02, 2003 12:05 PM > To: pythonmac-sig@python.org > Cc: Wxpython-Mac > Subject: [wxPython-mac] two-level namespaces enabled? > > > Bill Bumgarner has been porting some of my PythonCard sample code to work > under PyObjC. Yesterday I sent him the life sample code cut down > to just the > pure Python calculation of each generation. Basically, the "grid" > is just a > dictionary of (x, y) tuples with values of either 0 or 1. No Mac libs, > floating point, disk i/o, etc. is involved in the loop that I timed. > > Anyway, I ran it with both the Jaguar Python (/usr/bin/python) and the > MachoPython 2.2.1-4 disk image (/usr/local/bin/python) at: > > http://sourceforge.net/project/showfiles.php?group_id=10718 > > and there was a 10% speed difference on the same box. I asked Bill for an > explanation and he suggested... > > > From: Bill Bumgarner > > > > On Thursday, Jan 2, 2003, at 00:10 US/Eastern, Kevin Altis wrote: > > > I was a bit surprised when I ran the cmdLife.py program on my Mac. > > > /usr/bin/python (Apple) takes 38.8 seconds pretty consistently and > > > /usr/local/bin/python (2.2.1 framework build) takes roughly 43.3. Not > > > sure what that means, but it is a 10% difference. > > > > In any case, this is likely because the Framework build is not built > > w/two level namespaces enabled and, as such, is causing an extra > > indirection when dispatching calls into other frameworks [or something > > like that]. > > > > The Apple build definitely had two-level namespaces enabled and is also > > a static build, so no indirection is involved. > > I can send the code to anyone that is interested, but I bet it would > probably be more useful to simply run Python through its test > suite and see > how the numbers compare on both versions in case this would have > any impact > on the 2.2.2 and/or 2.3 builds people are working on. > > ka > --- > Kevin Altis > altis@semi-retired.com > http://radio.weblogs.com/0102677/ > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: wxPython-mac-unsubscribe@lists.wxwindows.org > For additional commands, e-mail: wxPython-mac-help@lists.wxwindows.org > > From jhrsn@pitt.edu Thu Jan 2 20:21:29 2003 From: jhrsn@pitt.edu (Jim Harrison) Date: Thu, 02 Jan 2003 15:21:29 -0500 Subject: [Pythonmac-SIG] Re: Problem building 2.3a1 on OSX & suggestions In-Reply-To: Message-ID: on 1/2/03 11:46 AM, James Harrison at jhrsn+@pitt.edu wrote: > However, make frameworkinstall seemed to have some problems: > > ... > /usr/bin/install: chmod 755 /Library/Frameworks/../../usr/local/bin: > Operation not permitted > ln -fs > /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/2.3/b > in/python /Library/Frameworks/../../usr/local/bin/python > ln: /Library/Frameworks/../../usr/local/bin/python: Permission denied > make[1]: *** [installunixtools] Error 1 > make: *** [frameworkinstallunixtools] Error 2 > > The unix tools do not appear to have installed in /usr/local/bin. There is a > MacPython-2.3 folder in /Applications, but it contains only BuildApplet and > PythonLauncher. PythonLauncher appears to work. Sorry for the false alarm. Python apparently builds at least mostly correctly even though the errors above are reported, and the unix tools (python and pythonw) do appear in /usr/local/bin. Here are some comments/suggestions that I hope will be useful to others just getting started with 2.3: 1. Though it is in the README file if you look around, it may not be obvious to some that IDLE and the PythonIDE will not build with the 2.3a1 distribution unless you add Tkinter/AquaTk and Waste separately. This is easy: Get Waste 2.1b1 from http://www.merzwaren.com/bin/waste/waste-21b1.sit and unstuff it. This yields a disk image (.dmg). Make a folder inside the main Python-2.3a1 folder named 'waste' and copy the contents of the disk image into it. This allows building the PythonIDE. Get TclTkAqua-8.4.1-Jaguar.dmg (for Jaguar) from http://sourceforge.net/project/showfiles.php?group_id=10894&release_id=11786 8 and run the installer. This allows building IDLE and other TkInter apps. Do both of the above (assuming you want to build both IDEs) before building Python. 2. Do a framework build according to the instructions in Python-2.3a1/Mac/OSX/README. 3. This should work, yielding both IDEs in the /Applications/MacPython folder, though if you're like me (10.2.3, Tibook), you might get a couple of errors at the end of the process. I don't yet know the significance of this. 4. IDLE doesn't look so good at first. I recommend editing its configuration files, which are inside IDLE (IDLE.app, actually). I used BBEdit's "Open Hidden" feature to open the IDLE/Contents/Resources/idlelib/ directory and edit config-unix.txt to change the font from Courier to ProFont and change the font size; and config.txt to change the default window size. I don't think the config-mac.txt file does anything in OSX. These edits make IDLE a bit more usable. Now if we could only do something about the ugly black "active text box" rectangle that surrounds the text editing areas in IDLE. Jim Harrison Univ. of Pittsburgh. From jhrsn+@pitt.edu Thu Jan 2 20:36:37 2003 From: jhrsn+@pitt.edu (James Harrison) Date: Thu, 02 Jan 2003 15:36:37 -0500 Subject: [Pythonmac-SIG] IDLE/Tk loss-of-focus bug Message-ID: Just a quick bug report. I'm sure this must not be new. When the IDLE tool tips for built-in functions appear, they "steal" the focus from text entry area. You're dead in the water until you mouse-click on the text area. Reproducible. Just type 'range(' and see what happens. OSX 10.2.3, Python 2.3a1, Tibook. Jim Harrison Univ. of Pittsburgh From Jack.Jansen@oratrix.com Thu Jan 2 21:11:18 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Thu, 2 Jan 2003 22:11:18 +0100 Subject: [Pythonmac-SIG] two-level namespaces enabled? In-Reply-To: Message-ID: On donderdag, jan 2, 2003, at 21:04 Europe/Amsterdam, Kevin Altis wrote: > [...] I asked Bill for an > explanation and he suggested... > >> From: Bill Bumgarner >> >> On Thursday, Jan 2, 2003, at 00:10 US/Eastern, Kevin Altis wrote: >>> I was a bit surprised when I ran the cmdLife.py program on my Mac. >>> /usr/bin/python (Apple) takes 38.8 seconds pretty consistently and >>> /usr/local/bin/python (2.2.1 framework build) takes roughly 43.3. Not >>> sure what that means, but it is a 10% difference. >> >> In any case, this is likely because the Framework build is not built >> w/two level namespaces enabled and, as such, is causing an extra >> indirection when dispatching calls into other frameworks [or something >> like that]. Bill, could you explain this, please? Why would building without two-level namespaces cause extra indirection? I'll also look into building with two-level namespace, there was a reason originally why I didn't do it for the framework build, but it may well have been solved in the mean time (I think it was the environ problem). Still, I'd like to know why flat namespaces could cause a significant drop in performance... -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From bbum@mac.com Thu Jan 2 21:20:39 2003 From: bbum@mac.com (bbum@mac.com) Date: Thu, 2 Jan 2003 16:20:39 -0500 Subject: [Pythonmac-SIG] two-level namespaces enabled? In-Reply-To: Message-ID: <0A930D73-1E98-11D7-8111-000393877AE4@mac.com> Now would be the time to enable my vacation message.... I should have put extra emphasis on [or something like that]. Unfortunately, I'm on a reallllyyysssllloooowwwwconnnecttion in the Cayman islands [connection is unfortunate, location is everything]. Let me dig a bit-- I believe I may have all the resources here to research this enough for a decent explanation (and I'm curious, myself)... the explanation may very well be I misheard something. In any case, figuring out why the framework build is slower by 10% across the boards than the Apple build would certainly be useful. I'll see what I can come up with ... more fun that reading a bad novel. :-) b.bum On Thursday, Jan 2, 2003, at 16:11 US/Eastern, Jack Jansen wrote: > On donderdag, jan 2, 2003, at 21:04 Europe/Amsterdam, Kevin Altis > wrote: >>> From: Bill Bumgarner >>> On Thursday, Jan 2, 2003, at 00:10 US/Eastern, Kevin Altis wrote: >>>> I was a bit surprised when I ran the cmdLife.py program on my Mac. >>>> /usr/bin/python (Apple) takes 38.8 seconds pretty consistently and >>>> /usr/local/bin/python (2.2.1 framework build) takes roughly 43.3. >>>> Not >>>> sure what that means, but it is a 10% difference. >>> >>> In any case, this is likely because the Framework build is not built >>> w/two level namespaces enabled and, as such, is causing an extra >>> indirection when dispatching calls into other frameworks [or >>> something >>> like that]. > > Bill, > could you explain this, please? Why would building without two-level > namespaces cause extra indirection? I'll also look into building with > two-level namespace, there was a reason originally why I didn't do it > for the framework build, but it may well have been solved in the mean > time (I think it was the environ problem). Still, I'd like to know why > flat namespaces could cause a significant drop in performance... From tony@lownds.com Thu Jan 2 21:29:10 2003 From: tony@lownds.com (Tony Lownds) Date: Thu, 2 Jan 2003 13:29:10 -0800 Subject: [Pythonmac-SIG] IDLE/Tk loss-of-focus bug In-Reply-To: References: Message-ID: Hi Jim, That bug is fixed in IDLEfork alpha 1, which is scheduled to go into Python 2.3a2. In the meantime, I suggest disabling the CallTips extension. Go into config.txt and add a line below the [CallTips] section: [CallTips] enable=0 Its not nearly as cool as CallTips, but try using ParenMatch by adding enable=1 in the same file: [ParenMatch] enable=1 -Tony At 3:36 PM -0500 1/2/03, James Harrison wrote: >Just a quick bug report. I'm sure this must not be new. When the IDLE tool >tips for built-in functions appear, they "steal" the focus from text entry >area. You're dead in the water until you mouse-click on the text area. > >Reproducible. Just type 'range(' and see what happens. > >OSX 10.2.3, Python 2.3a1, Tibook. > >Jim Harrison >Univ. of Pittsburgh > > >_______________________________________________ >Pythonmac-SIG maillist - Pythonmac-SIG@python.org >http://mail.python.org/mailman/listinfo/pythonmac-sig From skip@pobox.com Thu Jan 2 21:46:36 2003 From: skip@pobox.com (Skip Montanaro) Date: Thu, 2 Jan 2003 15:46:36 -0600 Subject: [Pythonmac-SIG] two-level namespaces enabled? In-Reply-To: <0A930D73-1E98-11D7-8111-000393877AE4@mac.com> References: <0A930D73-1E98-11D7-8111-000393877AE4@mac.com> Message-ID: <15892.45884.733000.841644@montanaro.dyndns.org> bbum> Unfortunately, I'm on a reallllyyysssllloooowwwwconnnecttion in bbum> the Cayman islands [connection is unfortunate, location is bbum> everything]. ... bbum> I'll see what I can come up with ... more fun that reading a bad bbum> novel. :-) So get a better novel and put the work away. :-) In Chicago, where it's beginning to look a lot like winter and I'm about to search for those tickets to Puerto Vallarta... -- Skip Montanaro - skip@pobox.com http://www.musi-cal.com/ http://www.mojam.com/ From jhrsn+@pitt.edu Thu Jan 2 22:08:30 2003 From: jhrsn+@pitt.edu (James Harrison) Date: Thu, 02 Jan 2003 17:08:30 -0500 Subject: [Pythonmac-SIG] Further comment re 2.3a1 install Message-ID: > /usr/bin/install: chmod 755 /Library/Frameworks/../../usr/local/bin: > Operation not permitted > ln -fs > /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/2.3/b > in/python /Library/Frameworks/../../usr/local/bin/python > ln: /Library/Frameworks/../../usr/local/bin/python: Permission denied > make[1]: *** [installunixtools] Error 1 > make: *** [frameworkinstallunixtools] Error 2 As it turned out, these errors were blocking the installation of pythonw. It was necessary for me to 'sudo make frameworkinstall' to get the install to run error-free. The initial configure and make didn't required sudo, of course. pyw files using TkInter run fine with PythonLauncher now. Should this be added to the build instructions or is my setup just strange? Jim Harrison Univ. of Pittsburgh From kgunn@wgen.net Thu Jan 2 22:43:56 2003 From: kgunn@wgen.net (Kenneth M. Gunn) Date: Thu, 2 Jan 2003 17:43:56 -0500 Subject: [Pythonmac-SIG] Build question: PythonCarbonCore and SSL Message-ID: <15C9A96157D35F479D95B02D1E2C3E330FD471@cookiemonster.wgenhq.net> I am wondering if anyone has experience building PythonCarbonCore with = SSL support. (Unlike the PC, which has the socket module in a separate = library, MacPython builds the socket module right into the main = library.) I've had a bear of a time getting it all to just compile, let = alone link and work. It appears that the Mac section of OpenSSL (v0.9.6h is the version I've = been working with) hasn't received much love in a while. The project = file refers to source files that ceased to exist a few versions ago, and = the platform-specific defines don't always address the Mac. Particular = areas which are tricky include functions that want to use terminal I/O, = which naturally doesn't make to much sense on the (non-OSX) Mac. The libraries I've been using are MacPython-2.2.1, OpenSSL-0.9.6h, and = the modified GUSI2 library available from the MacPython website. I would do this all with MachO or Unix, which people seem to have = experience and luck with, but I am in the land of education where MacOS = preX is going strong, and probably will be for quite some time. Any guidance or comments are welcome. Thanks. -k From bbum@mac.com Thu Jan 2 22:47:34 2003 From: bbum@mac.com (bbum@mac.com) Date: Thu, 2 Jan 2003 17:47:34 -0500 Subject: [Pythonmac-SIG] two-level namespaces enabled? In-Reply-To: Message-ID: <2F412F74-1EA4-11D7-9B09-000393877AE4@mac.com> On Thursday, Jan 2, 2003, at 16:11 US/Eastern, Jack Jansen wrote: > Bill, > could you explain this, please? Why would building without two-level > namespaces cause extra indirection? I'll also look into building with > two-level namespace, there was a reason originally why I didn't do it > for the framework build, but it may well have been solved in the mean > time (I think it was the environ problem). Still, I'd like to know why > flat namespaces could cause a significant drop in performance... It isn't simply the namespaces that is causing the difference in performance. I used the DYLD_FORCE_FLAT_NAMESPACE to force the Python build included with OS X to execute in a flat namespace and there was no change in the performance. So, I misremembered, misread, and/or am lacking in clue. To rectify my cluelessness, I did some testing. Not exactly the most scientific of testing, but I was careful not to actually *do* anything on the machine as the tests were running and I let each test run about a dozen times just to make sure things were consistent. Short answer: calls from the main executable into dynamically linked code is slower than non-dynamically linked code. Not by much, but it can really add up over the lifespan of an app. In the output below, case '1:' is a call to a simple function (see below) linked in a framework. Case '2:' is a call to the same function [renamed] that is statically linked into the executable. 2003-01-02 17:30:12.327 foo[9579] 1: 8.642446 2003-01-02 17:30:12.329 foo[9579] 2: 6.734165 2003-01-02 17:30:29.401 foo[9580] 1: 8.977464 2003-01-02 17:30:29.403 foo[9580] 2: 6.754741 2003-01-02 17:30:45.818 foo[9581] 1: 8.462819 2003-01-02 17:30:45.820 foo[9581] 2: 6.706616 The main source file: #import #import void baz(int i) { static int total; total = total + i; if (total == 100) fprintf(stderr, "Foo bar baz\n"); } int main (int argc, const char * argv[]) { int i; NSTimeInterval s, l, e; s = [NSDate timeIntervalSinceReferenceDate]; for(i=0; i<100000000; i++) bar(i); l = [NSDate timeIntervalSinceReferenceDate]; for(i=0; i<100000000; i++) baz(i); e = [NSDate timeIntervalSinceReferenceDate]; NSLog(@"1: %f", l - s); NSLog(@"2: %f", e - l); return 0; } bar() is exactly like baz(), but in a framework. Switching the order of the test does not affect output. Of course, this would only explain the performance difference Kevin is seeing if the python binary effectively contains the interpreter loop within itself and constantly makes calls into the Python framework. Seems unlikely, but I honestly didn't look. Now, if the cost is high for even within a framework, that would explain it... and it does. Wow. That hurts. 2003-01-02 17:42:20.903 foo[9691] 1: 7.284420 2003-01-02 17:42:20.906 foo[9691] 2: 6.735255 2003-01-02 17:42:38.152 foo[9692] 1: 7.334371 2003-01-02 17:42:38.155 foo[9692] 2: 6.752173 2003-01-02 17:42:55.392 foo[9693] 1: 7.282018 2003-01-02 17:42:55.395 foo[9693] 2: 6.757783 In this case, I moved the loop into the framework (and added a loop in the main executable for parity). #import #import void baz(int i) { static int total; total = total + i; if (total == 100)fprintf(stderr, "Foo bar baz\n"); } void dobaz() { int i; for(i=0; i<100000000; i++) baz(i); } int main (int argc, const char * argv[]) { NSTimeInterval s, l, e; s = [NSDate timeIntervalSinceReferenceDate]; dobar(); l = [NSDate timeIntervalSinceReferenceDate]; dobaz(); e = [NSDate timeIntervalSinceReferenceDate]; NSLog(@"1: %f", l - s); NSLog(@"2: %f", e - l); return 0; } -- and the framework -- #import void bar(int i) { static int total; total = total + i; if (total == 100) fprintf(stderr, "Foo bar baz\n"); } void dobar() { int i; for(i=0; i<100000000; i++) bar(i); } From oussoren@cistron.nl Fri Jan 3 09:32:39 2003 From: oussoren@cistron.nl (Ronald Oussoren) Date: Fri, 3 Jan 2003 10:32:39 +0100 Subject: [Pythonmac-SIG] Building a framework install from CVS? Message-ID: <4D2C03A1-1EFE-11D7-8024-0003931CFE24@cistron.nl> Hi, Building a framework-install using a current CVS version (last update: this morning) fails due to a missing 'MACFS' module. There is a 'macfs' module that contains at least some of the symbols that are expected to be in MACFS. /Library/Frameworks/../../Library/Frameworks/Python.framework/Versions/ 2.3/Resources/Python.app/Contents/MacOS/python ./Mac/scripts/BuildApplet.py \ --output /Library/Frameworks/../../Applications/MacPython-2.3/BuildApplet.app \ ./Mac/scripts/BuildApplet.py 'import macfsn' failed; use -v for traceback Traceback (most recent call last): File "./Mac/scripts/BuildApplet.py", line 17, in ? import buildtools File "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/plat- mac/buildtools.py", line 10, in ? import MACFS ImportError: No module named MACFS I haven't tried the 2.3a1 release or a non-framework build yet. Ronald From just@letterror.com Fri Jan 3 09:37:58 2003 From: just@letterror.com (Just van Rossum) Date: Fri, 3 Jan 2003 10:37:58 +0100 Subject: [Pythonmac-SIG] Building a framework install from CVS? In-Reply-To: <4D2C03A1-1EFE-11D7-8024-0003931CFE24@cistron.nl> Message-ID: Ronald Oussoren wrote: > 'import macfsn' failed; use -v for traceback This I saw as well: there's an macfs.so left over from a previous build, you should remove it my hand (macfs is now a Python module). I don't think this is related to the other problem, though... Just From oussoren@cistron.nl Fri Jan 3 10:31:06 2003 From: oussoren@cistron.nl (Ronald Oussoren) Date: Fri, 3 Jan 2003 11:31:06 +0100 Subject: [Pythonmac-SIG] Building a framework install from CVS? In-Reply-To: Message-ID: <77B4F190-1F06-11D7-8F1C-0003931CFE24@cistron.nl> On Friday, Jan 3, 2003, at 10:37 Europe/Amsterdam, Just van Rossum wrote: > Ronald Oussoren wrote: > >> 'import macfsn' failed; use -v for traceback > > This I saw as well: there's an macfs.so left over from a previous > build, > you should remove it my hand (macfs is now a Python module). > > I don't think this is related to the other problem, though... After removing the previous install the CVS version installed just fine. Thanks, Ronald From Jack.Jansen@cwi.nl Fri Jan 3 10:11:55 2003 From: Jack.Jansen@cwi.nl (Jack Jansen) Date: Fri, 3 Jan 2003 11:11:55 +0100 Subject: [Pythonmac-SIG] Building a framework install from CVS? In-Reply-To: Message-ID: On Friday, Jan 3, 2003, at 10:37 Europe/Amsterdam, Just van Rossum wrote: > Ronald Oussoren wrote: > >> 'import macfsn' failed; use -v for traceback > > This I saw as well: there's an macfs.so left over from a previous > build, > you should remove it my hand (macfs is now a Python module). > > I don't think this is related to the other problem, though... Yes, it could very well be related. The new "macfs" module, after being imported, also registers itself in sys.modules as "MACFS". The old one clearly won't do that. For 2.3a2 I will try to get rid of all MACFS references. -- Jack Jansen, , http://www.cwi.nl/~jack If I can't dance I don't want to be part of your revolution -- Emma Goldman From kevino@tulane.edu Fri Jan 3 14:59:01 2003 From: kevino@tulane.edu (Kevin Ollivier) Date: Fri, 3 Jan 2003 09:59:01 -0500 Subject: [Pythonmac-SIG] Building Mac extensions for Apple Python Message-ID: Hi all, I've taken some time out to try out the new MacPython extensions to Apple's included Python, but I'm getting errors when trying to build. I'm on a TiBook 1 GHz running 10.2.3 and I do have the MachOPython 2.2 framework install for wxPython support. I'm wondering if this could be interfering in some way, and if so, should I remove the framework MachOPython completely or build the latest version? Following the steps in README.jaguar and running make -f Makefile.jaguar gives the following output: /usr/bin/install -c -d -m 755 /./Applications/MacPython-OSX-2.2 /usr/bin/install -c -d -m 755 /./Applications/MacPython-OSX-2.2/python-additions /usr/bin/install -c -d -m 755 /./Applications/MacPython-OSX-2.2/python-additions/lib-dynload /usr/bin/python ../../Mac/OSX/setup.jaguar.py install --install-lib=/./Applications/MacPython-OSX-2.2/python-additions/lib- dynload running install running build running build_ext building 'OverrideFrom23._AE' extension gcc -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -no-cpp-precomp -I../../Mac/Include -I/usr/include/python2.2 -c ../../Mac/Modules/ae/_AEmodule.c -o build/temp.darwin-6.3-Power Macintosh-2.2/_AEmodule.o ../../Mac/Modules/ae/_AEmodule.c:922: `PyObject_Del' undeclared here (not in a function) ../../Mac/Modules/ae/_AEmodule.c:922: initializer element is not constant ../../Mac/Modules/ae/_AEmodule.c:922: (near initialization for `AEDesc_Type.tp_free') error: command 'gcc' failed with exit status 1 Any help/advice would be greatly appreciated! Thanks, Kevin From Jack.Jansen@oratrix.com Fri Jan 3 20:37:04 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Fri, 3 Jan 2003 21:37:04 +0100 Subject: [Pythonmac-SIG] Building Mac extensions for Apple Python In-Reply-To: Message-ID: <1ECE6725-1F5B-11D7-8857-000A27B19B96@oratrix.com> On vrijdag, jan 3, 2003, at 15:59 Europe/Amsterdam, Kevin Ollivier wrote: > Hi all, > > I've taken some time out to try out the new MacPython extensions to > Apple's included Python, but I'm getting errors when trying to build. Kevin, I seriously broke this last week, when I modified a large part of the source tree to move most of the Mac modules to Lib/plat-mac (from the original Mac/Lib location). So, for the time being I've removed Makefile.jaguar and setup.jaguar.py. As the schedule was extremely tight (I started on this less than 48 hours before 2.3a1 came out) I had no time to keep them functional. Although the breakage you see (_AE failing to build) is caused by an earlier mod of mine it is difficult to fix right now. I will try to revive the MacPython-OSX-on-top-of-Python-2.2 (especially now I know that you are interested too, I could use a hand!), but it will have to wait. First I have to get MacPython-OS9-2.3a1 out the door, and that will be another few days. If you are willing to revive things, here's what needs to be done: - Check out the last versions of Makefile.jaguar and setup.jaguar.py (checking out with a date of 28-dec-2002 should work) - Check out the previous version of Makefile, compare to the current version and work out which bits I removed (mainly the copying of $(srcdir)/Mac/Lib to the destination). That functionality will have to go into Makefile.jaguar, but with $(srcdir)/Lib/plat-mac as the source. - Then you'll have to find out where the bug you're seeing is coming from. PyObject_Del is declared for python 2.2 (in /usr/include/python2.2/objimpl.h) so I really have no idea why it would fail... -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From kevino@tulane.edu Fri Jan 3 21:52:22 2003 From: kevino@tulane.edu (Kevin Ollivier) Date: Fri, 3 Jan 2003 16:52:22 -0500 Subject: [Pythonmac-SIG] Building Mac extensions for Apple Python In-Reply-To: <1ECE6725-1F5B-11D7-8857-000A27B19B96@oratrix.com> Message-ID: Hi Jack, Thanks for your help! I am also clueless as to why PyObject_Del would be causing that error message. Do those extensions get built with a framework build as well? I did some playing around (removing the extensions that were not building and changing $(scripts)/Lib/Mac to its new home), but I'm getting import errors for the mac modules during the import_BuildApplet phase. It cannot find import macfsn and EasyDialogs, which are of course in the CVS but the path is probably not set properly. Are these supposed to be copied over to /Applications/MacPython-OSX-2.3 before this portion of setup is run? However, in the process of troubleshooting all of this I came across what I was really looking for - buildbundle.py. =) I thought I needed to install the Jaguar extensions to do this, but now I see that I don't have to. I've been getting ready to distribute the final release of a wxPython app I have been working on and I wanted to make it look more like a Mac application. Now my wxPython application is running as a full-fledged .app bundle. =) Runs great - the only thing I have yet to figure out is how to hook into the About and Preferences menus of the app, but that may be a wxPython issue. Thanks, Kevin On Friday, January 3, 2003, at 03:37 PM, Jack Jansen wrote: > > On vrijdag, jan 3, 2003, at 15:59 Europe/Amsterdam, Kevin Ollivier > wrote: > >> Hi all, >> >> I've taken some time out to try out the new MacPython extensions to >> Apple's included Python, but I'm getting errors when trying to build. > > Kevin, > I seriously broke this last week, when I modified a large part of the > source tree to move most of the Mac modules to Lib/plat-mac (from the > original Mac/Lib location). So, for the time being I've removed > Makefile.jaguar and setup.jaguar.py. As the schedule was extremely > tight (I started on this less than 48 hours before 2.3a1 came out) I > had no time to keep them functional. > > Although the breakage you see (_AE failing to build) is caused by an > earlier mod of mine it is difficult to fix right now. > > I will try to revive the MacPython-OSX-on-top-of-Python-2.2 > (especially now I know that you are interested too, I could use a > hand!), but it will have to wait. First I have to get > MacPython-OS9-2.3a1 out the door, and that will be another few days. > > If you are willing to revive things, here's what needs to be done: > - Check out the last versions of Makefile.jaguar and setup.jaguar.py > (checking out with a date of 28-dec-2002 should work) > - Check out the previous version of Makefile, compare to the current > version and work out which bits I removed (mainly the copying of > $(srcdir)/Mac/Lib to the destination). That functionality will have to > go into Makefile.jaguar, but with $(srcdir)/Lib/plat-mac as the > source. > - Then you'll have to find out where the bug you're seeing is coming > from. PyObject_Del is declared for python 2.2 (in > /usr/include/python2.2/objimpl.h) so I really have no idea why it > would fail... > -- > - Jack Jansen > http://www.cwi.nl/~jack - > - If I can't dance I don't want to be part of your revolution -- Emma > Goldman - > > From deleeuw@stat.ucla.edu Sat Jan 4 20:03:40 2003 From: deleeuw@stat.ucla.edu (Jan de Leeuw) Date: Sat, 4 Jan 2003 12:03:40 -0800 Subject: [Pythonmac-SIG] Rpy in various interfaces Message-ID: <9EB93632-201F-11D7-9D87-000393BB6D36@stat.ucla.edu> R is a statistics environment. All of R can be compiled into a dylib, and the Rpy package on Sourceforge allows you to do R from Python. R has various graphics devices for its plotting etc, there is pdf(), x11(), eps(), jpeg() etc, which open these devices and make them the current device. Now I am interested in the situation where the device is X11. If I run python (latest CVS build of 2.3a) from the terminal I can say from rpy import * r.x11() and my device opens, a graphics window appears, and I can plot. In fact I dont even have to say r.x11() explicitly, because from the terminal that's the default device (if there is a X server running). If I give the same commands from interactive mode in PythonIDE it crashes. On the other hand from rpy import * r.x11(display='gifi.stat.ucla.edu:0.0') works fine. If I give either of these commands from IDLE, it crashes. Now this does not surprise me, because there can be all kind of nasty interactions between plotting modules and event loops. === Jan de Leeuw; Professor and Chair, UCLA Department of Statistics; Editor: Journal of Multivariate Analysis, Journal of Statistical Software US mail: 9432 Boelter Hall, Box 951554, Los Angeles, CA 90095-1554 phone (310)-825-9550; fax (310)-206-5658; email: deleeuw@stat.ucla.edu homepage: http://gifi.stat.ucla.edu ------------------------------------------------------------------------ ------------------------- No matter where you go, there you are. --- Buckaroo Banzai http://gifi.stat.ucla.edu/sounds/nomatter.au ------------------------------------------------------------------------ ------------------------- From rus20376@salemstate.edu Sat Jan 4 20:35:47 2003 From: rus20376@salemstate.edu (Adam Russell) Date: Sat, 04 Jan 2003 15:35:47 -0500 Subject: [Pythonmac-SIG] complete ui examples Message-ID: <3E1745A3.1030303@salemstate.edu> The examples of Macintosh GUIs in the distribution s a bit light. However, I cannot find more complete example anywhere on the web>:( . Can someone point me to some more larger scale Macintosh UI examples? Something beyond showing a dialog box. An application with a main application window with several controls perhaps....... Thanks!! From deleeuw@stat.ucla.edu Sat Jan 4 23:20:48 2003 From: deleeuw@stat.ucla.edu (Jan de Leeuw) Date: Sat, 4 Jan 2003 15:20:48 -0800 Subject: [Pythonmac-SIG] Just checking .. Message-ID: <289666B8-203B-11D7-BB3E-000393BB6D36@stat.ucla.edu> It seems that wxPython now installs with no problems (except for the glcanvas extension) using MacPython-2.3a from CVS ? The demo runs fine. It looks as if there is no more need for MachoPython. Correct ? There is a framework build (with Rpy, wxPython, Numeric, numarray, and pyobjc) at http://gifi.stat.ucla.edu/pub -- it seems to do most of the things it is supposed to do. === Jan de Leeuw; Professor and Chair, UCLA Department of Statistics; Editor: Journal of Multivariate Analysis, Journal of Statistical Software US mail: 9432 Boelter Hall, Box 951554, Los Angeles, CA 90095-1554 phone (310)-825-9550; fax (310)-206-5658; email: deleeuw@stat.ucla.edu homepage: http://gifi.stat.ucla.edu ------------------------------------------------------------------------ ------------------------- No matter where you go, there you are. --- Buckaroo Banzai http://gifi.stat.ucla.edu/sounds/nomatter.au ------------------------------------------------------------------------ ------------------------- From Jack.Jansen@oratrix.com Sun Jan 5 00:24:24 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sun, 5 Jan 2003 01:24:24 +0100 Subject: [Pythonmac-SIG] complete ui examples In-Reply-To: <3E1745A3.1030303@salemstate.edu> Message-ID: <0AC06DCF-2044-11D7-BF12-000A27B19B96@oratrix.com> On zaterdag, jan 4, 2003, at 21:35 Europe/Amsterdam, Adam Russell wrote: > The examples of Macintosh GUIs in the distribution s a bit light. > However, I cannot find more complete example anywhere on the web>:( . > Can someone point me to some more larger scale Macintosh UI examples? > Something beyond showing a dialog box. An application with a main > application window with several controls perhaps....... The next thing up is really the IDE itself. But you should note that the IDE (and the underlying W and Framework packages) are built with Carbon, and Carbon could be considered obsolete nowadays (at least for building GUIs). You may want to have a look at PyObjC (you'll find it on sourceforge), which is a complete bridge between Python and Cocoa (or any other ObjC package, but for GUI development Cocoa is the interesting one). -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From rus20376@salemstate.edu Sun Jan 5 03:11:02 2003 From: rus20376@salemstate.edu (Adam Russell) Date: Sat, 04 Jan 2003 22:11:02 -0500 Subject: [Pythonmac-SIG] error with reading fields of dialog Message-ID: <3E17A246.8070708@salemstate.edu> This is a multi-part message in MIME format. --------------050400010406020006020105 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit I get the following error when I run the attached script. Thisis really simple code!! What is going on here? Being pretty new to python I need to turn to the experts....... Any ideas? Thanks!! Traceback (most recent call last): File "clash:Desktop Folder:python_dev:ftp:RabbitTransfer.py", line 99, in ? main() File "clash:Desktop Folder:python_dev:ftp:RabbitTransfer.py", line 25, in main RabbitTransfer() File "clash:Desktop Folder:python_dev:ftp:RabbitTransfer.py", line 34, in __init__ self.mainloop() File "clash:Applications (Mac OS 9):Python 2.2.1:Mac:Lib:FrameWork.py", line 166, in mainloop self.do1event(mask, wait) File "clash:Applications (Mac OS 9):Python 2.2.1:Mac:Lib:FrameWork.py", line 184, in do1event if self.do_dialogevent(event): File "clash:Applications (Mac OS 9):Python 2.2.1:Mac:Lib:FrameWork.py", line 244, in do_dialogevent self._windows[window].do_itemhit(item, event) File "clash:Desktop Folder:python_dev:ftp:RabbitTransfer.py", line 80, in do_itemhit self.onConnect() File "clash:Desktop Folder:python_dev:ftp:RabbitTransfer.py", line 83, in onConnect tp, h, rect=self.dlg.GetDialogItem(ITEM_SERVER) AttributeError: 'NoneType' object has no attribute 'GetDialogItem' --------------050400010406020006020105 Content-Type: multipart/appledouble; boundary="------------ad020303000002020401010202"; x-mac-type="72737263"; x-mac-creator="52534544"; name="RabbitTransfer.rsrc" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="RabbitTransfer.rsrc" --------------ad020303000002020401010202 Content-Type: application/applefile Content-Transfer-Encoding: base64 AAUWBwACAAAAAAAAAAAAAAAAAAAAAAAAAAUAAAADAAAAVgAAABMAAAAJAAAAaQAAACAAAAAI AAAAiQAAABAAAAAEAAAAmQAAAAAAAAACAAAAmQAAA9VSYWJiaXRUcmFuc2Zlci5yc3JjcnNy Y1JTRUQBAAAAAQEAAAAAAAAAAAAAAAAAAAAAAAAC54MkBanbe0ttDAAFqhh2AAABAAAAAuwA AAHsAAAA6bwAAAACdAK0gVfBwAFnUEKnLy4ACC8GTq0JqhAfZj5CDmRuc2xvb2t1cC5yc3Jj cHlyZW5jZXN4aW5nc3J0aHMCAHJzcmNSU0VEAQAABAAMAAAAAAAAAAAAAAAAAAAAAAAArp5d GQAAAAAAAAKZ8E6tCZodX/7jUkdpCL5u/t5vAP8YfgFCZz8sARBOrQiqPV//+r5u//puej8s ARBIbv/wPwdOrQiyJqwBeEJnPywBEC8u//BOrQi6OB8mrv7sSkRmTD18AAH+3EJnPywBEC8u //BOrQi6PV/+2jou/txgIkKnPywBEC8u//A/PAABTgAAACYARAAwAPQBZwAFAQABAAAAAAAC ABFSZW1vdGUgQ29ubmVjdGlvbgAAABUAYABsARYBeAABAQAAAAAAAAACAQAAAABPQ0sKIT/l MjAABAAAAAAAAAAAuLzXtLd6dyQAAAAAAAAAAAAAAAAAAAAACmphY2tqYW5zZW4AAAAQZG5z bG9va3VwLTIucnNyYwAAAAAAAAAAAIYABgAAAAAAJwBhADkBLBAAAAAAAABJAGAAWwErEAAA AAAAAJUA4QCpARsEB0Nvbm5lY3QAAAAAAAAmAAgAOABbiApTZXJ2ZXJOYW1lAAAAAABIAAkA WgBViAhVc2VybmFtZQAAAAAAbABgAH4BKxAAAAAAAABqAAkAfABViAhQYXNzd29yZAAAAD4A AgAAAAAAlgBxAKoAqwQCT0sAAAAAAAoAcQBKALGgAgWXAAAAAABPAF0AXgC+iA5SYWJiaXRU cmFuc2ZlcgAAACYAQAAQAT8B/AAEAQABAAAAAAACAg5QSUNUIFJlc291cmNlcwAAAAAAAFwA AwAAAAAACgCqABoBDgcCBZcAAAAAANIAFADnAGoEDFJlY3YgRmlsZShzKQAAAAAA0gC0AOcB CgQMU2VuZCBGaWxlKHMpAAAAAADSAVQA5wGqBAdSZWZyZXNoKQAAAQAAAALsAAAB7AAAAOkR z6aEA2YAAAAcAIoAAkRJVEwAAgAaRExPRwACAD5ja2lkAAAAYgIAACsAAACWAAAAAAIBAEoA AAEgEc+ToAICADkAAAGMEc+sjAIAABkAAAAAEc+slAIBAAAAAAAqEc+Q3AICAFAAAAFiEc+j xACAAAoAAABDEc+s0AlBYm91dCBib3gOUHJvamVjdG9yIERhdGERQ29ubmVjdGlvbiBEaWFs b2cNY29ubmVjdERpYWxvZxBSZW1vdGUgRmlsZSBMaXN0BUFib3V0DkJyb3dzZXIgd2luZG93 --------------ad020303000002020401010202 Content-Type: application/octet-stream; name="RabbitTransfer.rsrc" Content-Transfer-Encoding: base64 Content-Disposition: inline; filename="RabbitTransfer.rsrc" --------------ad020303000002020401010202-- --------------050400010406020006020105 Content-Type: text/plain; x-mac-type="54455854"; x-mac-creator="50797468"; name="RabbitTransfer.py" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="RabbitTransfer.py" import FrameWork import EasyDialogs from Carbon import Res from Carbon import Dlg from Carbon import Qd from Carbon import Win from Carbon import Controls from Carbon import List import sys import socket import string import macresource # # Definitions for our resources ID_CONNECTION=512 ID_ABOUT=513 ITEM_SERVER=1 ITEM_USERNAME=2 ITEM_PASSWD=6 ITEM_CONNECT=3 def main(): macresource.need("DLOG", ID_CONNECTION, "RabbitTransfer.rsrc") RabbitTransfer() class RabbitTransfer(FrameWork.Application): "Application class for RabbitTransfer" def __init__(self): # First init menus, etc. FrameWork.Application.__init__(self) # Finally, go into the event loop self.mainloop() def makeusermenus(self): self.filemenu = m = FrameWork.Menu(self.menubar, "File") self.quititem = FrameWork.MenuItem(m, "Quit", "Q", self.quit) self.remotemenu = r = FrameWork.Menu(self.menubar, "Remote") self.connectitem=FrameWork.MenuItem(r,"Connect","C",self.connect) self.disconnectitem=FrameWork.MenuItem(r,"Disconnect","D",self.disconnect) self.refreshitem=FrameWork.MenuItem(r,"Refresh","R",self.refresh) def quit(self, *args): self._quit() def connect(self, *args): self.connection_dialog = ConnectionDialog(self) self.connection_dialog.open(ID_CONNECTION) def disconnect(self, *args): self._quit() def refresh(self, *args): self._quit() def do_about(self, *args): f = Dlg.GetNewDialog(ID_ABOUT, -1) while 1: n = Dlg.ModalDialog(None) if n == 1: return class ConnectionDialog(FrameWork.DialogWindow): "Connection dialog window for RabbitTransfer" def __init__(self, parent): FrameWork.DialogWindow.__init__(self, parent) self.parent = parent def open(self, id): self.id=id dlg = Dlg.GetNewDialog(id, -1) FrameWork.DialogWindow.open(self,ID_CONNECTION) self.dlg.SetDialogDefaultItem(ITEM_CONNECT) def do_itemhit(self, item, event): if item==ITEM_CONNECT: self.close() self.onConnect() def onConnect(self): tp, h, rect=self.dlg.GetDialogItem(ITEM_SERVER) """txt = Dlg.GetDialogItemText(h) print txt tp, h, rect = self.dlg.GetDialogItem(ITEM_USERNAME) txt = Dlg.GetDialogItemText(h) print txt tp, h, rect = self.dlg.GetDialogItem(ITEM_PASSWD) txt = Dlg.GetDialogItemText(h) print txt""" def do_close(self): self.close() main() --------------050400010406020006020105-- From robrogers@mac.com Sun Jan 5 08:38:31 2003 From: robrogers@mac.com (Robert Rogers) Date: Sun, 5 Jan 2003 03:38:31 -0500 Subject: [Pythonmac-SIG] Python 2.1 dies when Executing Zope z2.py Message-ID: <11C5376E-2089-11D7-AA02-000A27AB5498@mac.com> > Hi, > > I'm trying to get Zope up on my 'Mac os X' client. Unfortunately, I > can't tell if the problem lies with my python 2.1 build. > > Specifically, when I run the start script, which is basically 'python > z2.py', python dies after generating this error message: > > dyld: /usr/local/bin/python.exe Undefined symbols: > _PyType_IsSubtype > Trace/BPT trap > > I'm a python newbie (my plans were to learn python while programming > zope :-) ), so I have no idea what this means. > > Any help in elucidating this matter would be greatly appreciated. > > Fyi, I also got some warnings after running make install saying: > > warning: install: modules installed to > '/usr/local/lib/python2.1/lib-dynload/', which is not in Python's > module search path (sys.path) -- you'll have to change the search path > yourself > > Which i did through the ENV variable PYTHONPATH. > > Also my running of make test also showed inclusive results, a number > of objects were skipped > > Cheers, > Rob > > Notes: I'm on OS X 10.2.3 with fink installed for some unix tools. > > ------------------------------ > Rob Rogers > 415.238.2972 > robrogers@mac.com > ------------------------------ > > From python-docs@python.org Sun Jan 5 09:33:13 2003 From: python-docs@python.org (python-docs) Date: Sun, 5 Jan 2003 10:33:13 +0100 (added by postmaster@dns-oi.com) Subject: [Pythonmac-SIG] For information on suggesting changes. Message-ID: <3D946F64000630F4@antivirusgw.dns-oi.com> (added by postmaster@dns-oi.com) ------=_NextPartTM-000-38c398fc-093a-47c3-a862-b2ecfd1b7404 Content-Type: multipart/alternative; boundary=AD17450M9eoh9082984I6oWJBRcP79h68t955 --AD17450M9eoh9082984I6oWJBRcP79h68t955 Content-Type: text/html; Content-Transfer-Encoding: quoted-printable --AD17450M9eoh9082984I6oWJBRcP79h68t955 --AD17450M9eoh9082984I6oWJBRcP79h68t955 Content-Type: application/octet-stream; name=node80.html Content-Transfer-Encoding: base64 Content-ID: PCFET0NUWVBFIGh0bWwgUFVCTElDICItLy9XM0MvL0RURCBIVE1MIDQuMCBUcmFuc2l0aW9u YWwvL0VOIj4NCjxodG1sPg0KPGhlYWQ+DQo8dGl0bGU+Qi4gSGlzdG9yeSBhbmQgTGljZW5z ZTwvdGl0bGU+DQo8TUVUQSBOQU1FPSJkZXNjcmlwdGlvbiIgQ09OVEVOVD0iQi4gSGlzdG9y eSBhbmQgTGljZW5zZSI+DQo8TUVUQSBOQU1FPSJrZXl3b3JkcyIgQ09OVEVOVD0iYXBpIj4N CjxNRVRBIE5BTUU9InJlc291cmNlLXR5cGUiIENPTlRFTlQ9ImRvY3VtZW50Ij4NCjxNRVRB IE5BTUU9ImRpc3RyaWJ1dGlvbiIgQ09OVEVOVD0iZ2xvYmFsIj4NCjxtZXRhIGh0dHAtZXF1 aXY9IkNvbnRlbnQtVHlwZSIgY29udGVudD0idGV4dC9odG1sOyBjaGFyc2V0PWlzby04ODU5 LTEiPg0KPGxpbmsgcmVsPSJTVFlMRVNIRUVUIiBocmVmPSJhcGkuY3NzIj4NCjxsaW5rIHJl bD0iZmlyc3QiIGhyZWY9ImFwaS5odG1sIj4NCjxsaW5rIHJlbD0iY29udGVudHMiIGhyZWY9 ImNvbnRlbnRzLmh0bWwiIHRpdGxlPSJDb250ZW50cyI+DQo8bGluayByZWw9ImluZGV4IiBo cmVmPSJnZW5pbmRleC5odG1sIiB0aXRsZT0iSW5kZXgiPg0KPExJTksgUkVMPSJuZXh0IiBo cmVmPSJnZW5pbmRleC5odG1sIj4NCjxMSU5LIFJFTD0icHJldmlvdXMiIGhyZWY9InJlcG9y dGluZy1idWdzLmh0bWwiPg0KPExJTksgUkVMPSJ1cCIgSFJFRj0iYXBpLmh0bWwiPg0KPExJ TksgUkVMPSJuZXh0IiBIUkVGPSJub2RlODEuaHRtbCI+DQo8L2hlYWQ+DQo8Ym9keT4NCjxE SVYgQ0xBU1M9Im5hdmlnYXRpb24iPg0KPHRhYmxlIGFsaWduPSJjZW50ZXIiIHdpZHRoPSIx MDAlIiBjZWxscGFkZGluZz0iMCIgY2VsbHNwYWNpbmc9IjIiPg0KPHRyPg0KPHRkPjxBIGhy ZWY9InJlcG9ydGluZy1idWdzLmh0bWwiPjxpbWcgc3JjPSIuLi9pY29ucy9wcmV2aW91cy5n aWYiDQogIGJvcmRlcj0iMCIgaGVpZ2h0PSIzMiINCiAgYWx0PSJQcmV2aW91cyBQYWdlIiB3 aWR0aD0iMzIiPjwvQT48L3RkPg0KPHRkPjxBIEhSRUY9ImFwaS5odG1sIj48aW1nIHNyYz0i Li4vaWNvbnMvdXAuZ2lmIg0KICBib3JkZXI9IjAiIGhlaWdodD0iMzIiDQogIGFsdD0iVXAg T25lIExldmVsIiB3aWR0aD0iMzIiPjwvQT48L3RkPg0KPHRkPjxBIEhSRUY9Im5vZGU4MS5o dG1sIj48aW1nIHNyYz0iLi4vaWNvbnMvbmV4dC5naWYiDQogIGJvcmRlcj0iMCIgaGVpZ2h0 PSIzMiINCiAgYWx0PSJOZXh0IFBhZ2UiIHdpZHRoPSIzMiI+PC9BPjwvdGQ+DQo8dGQgYWxp Z249ImNlbnRlciIgd2lkdGg9IjEwMCUiPlB5dGhvbi9DIEFQSSBSZWZlcmVuY2UgTWFudWFs PC90ZD4NCjx0ZD48QSBocmVmPSJjb250ZW50cy5odG1sIj48aW1nIHNyYz0iLi4vaWNvbnMv Y29udGVudHMuZ2lmIg0KICBib3JkZXI9IjAiIGhlaWdodD0iMzIiDQogIGFsdD0iQ29udGVu dHMiIHdpZHRoPSIzMiI+PC9BPjwvdGQ+DQo8dGQ+PGltZyBzcmM9Ii4uL2ljb25zL2JsYW5r LmdpZiINCiAgYm9yZGVyPSIwIiBoZWlnaHQ9IjMyIg0KICBhbHQ9IiIgd2lkdGg9IjMyIj48 L3RkPg0KPHRkPjxBIGhyZWY9ImdlbmluZGV4Lmh0bWwiPjxpbWcgc3JjPSIuLi9pY29ucy9p bmRleC5naWYiDQogIGJvcmRlcj0iMCIgaGVpZ2h0PSIzMiINCiAgYWx0PSJJbmRleCIgd2lk dGg9IjMyIj48L0E+PC90ZD4NCjwvdHI+PC90YWJsZT4NCjxiIGNsYXNzPSJuYXZsYWJlbCI+ UHJldmlvdXM6PC9iPiA8YSBjbGFzcz0ic2VjdHJlZiIgaHJlZj0icmVwb3J0aW5nLWJ1Z3Mu aHRtbCI+QS4gUmVwb3J0aW5nIEJ1Z3M8L0E+DQo8YiBjbGFzcz0ibmF2bGFiZWwiPlVwOjwv Yj4gPGEgY2xhc3M9InNlY3RyZWYiIEhSRUY9ImFwaS5odG1sIj5QeXRob24vQyBBUEkgUmVm ZXJlbmNlIE1hbnVhbDwvQT4NCjxiIGNsYXNzPSJuYXZsYWJlbCI+TmV4dDo8L2I+IDxhIGNs YXNzPSJzZWN0cmVmIiBIUkVGPSJub2RlODEuaHRtbCI+Qi4xIEhpc3Rvcnkgb2YgdGhlPC9B Pg0KPGJyPjxocj4NCjwvRElWPg0KPCEtLUVuZCBvZiBOYXZpZ2F0aW9uIFBhbmVsLS0+DQoN CjxIMT48QSBOQU1FPSJTRUNUSU9OMDAxNDAwMDAwMDAwMDAwMDAwMDAwMCI+DQpCLiBIaXN0 b3J5IGFuZCBMaWNlbnNlPC9BPg0KPC9IMT4NCg0KPHA+PGhyPg0KPCEtLVRhYmxlIG9mIENo aWxkLUxpbmtzLS0+DQo8QSBOQU1FPSJDSElMRF9MSU5LUyI+PFNUUk9ORz5TdWJzZWN0aW9u czwvU1RST05HPjwvYT4NCg0KPFVMIENMQVNTPSJDaGlsZExpbmtzIj4NCjxMST48QSBocmVm PSJub2RlODEuaHRtbCI+Qi4xIEhpc3Rvcnkgb2YgdGhlIHNvZnR3YXJlPC9hPg0KPExJPjxB IGhyZWY9Im5vZGU4Mi5odG1sIj5CLjIgVGVybXMgYW5kIGNvbmRpdGlvbnMgZm9yIGFjY2Vz c2luZyBvciBvdGhlcndpc2UgdXNpbmcgUHl0aG9uPC9hPg0KPC91bD4NCjwhLS1FbmQgb2Yg VGFibGUgb2YgQ2hpbGQtTGlua3MtLT4NCg0KPERJViBDTEFTUz0ibmF2aWdhdGlvbiI+DQo8 cD48aHI+DQo8dGFibGUgYWxpZ249ImNlbnRlciIgd2lkdGg9IjEwMCUiIGNlbGxwYWRkaW5n PSIwIiBjZWxsc3BhY2luZz0iMiI+DQo8dHI+DQo8dGQ+PEEgaHJlZj0icmVwb3J0aW5nLWJ1 Z3MuaHRtbCI+PGltZyBzcmM9Ii4uL2ljb25zL3ByZXZpb3VzLmdpZiINCiAgYm9yZGVyPSIw IiBoZWlnaHQ9IjMyIg0KICBhbHQ9IlByZXZpb3VzIFBhZ2UiIHdpZHRoPSIzMiI+PC9BPjwv dGQ+DQo8dGQ+PEEgSFJFRj0iYXBpLmh0bWwiPjxpbWcgc3JjPSIuLi9pY29ucy91cC5naWYi DQogIGJvcmRlcj0iMCIgaGVpZ2h0PSIzMiINCiAgYWx0PSJVcCBPbmUgTGV2ZWwiIHdpZHRo PSIzMiI+PC9BPjwvdGQ+DQo8dGQ+PEEgSFJFRj0ibm9kZTgxLmh0bWwiPjxpbWcgc3JjPSIu Li9pY29ucy9uZXh0LmdpZiINCiAgYm9yZGVyPSIwIiBoZWlnaHQ9IjMyIg0KICBhbHQ9Ik5l eHQgUGFnZSIgd2lkdGg9IjMyIj48L0E+PC90ZD4NCjx0ZCBhbGlnbj0iY2VudGVyIiB3aWR0 aD0iMTAwJSI+UHl0aG9uL0MgQVBJIFJlZmVyZW5jZSBNYW51YWw8L3RkPg0KPHRkPjxBIGhy ZWY9ImNvbnRlbnRzLmh0bWwiPjxpbWcgc3JjPSIuLi9pY29ucy9jb250ZW50cy5naWYiDQog IGJvcmRlcj0iMCIgaGVpZ2h0PSIzMiINCiAgYWx0PSJDb250ZW50cyIgd2lkdGg9IjMyIj48 L0E+PC90ZD4NCjx0ZD48aW1nIHNyYz0iLi4vaWNvbnMvYmxhbmsuZ2lmIg0KICBib3JkZXI9 IjAiIGhlaWdodD0iMzIiDQogIGFsdD0iIiB3aWR0aD0iMzIiPjwvdGQ+DQo8dGQ+PEEgaHJl Zj0iZ2VuaW5kZXguaHRtbCI+PGltZyBzcmM9Ii4uL2ljb25zL2luZGV4LmdpZiINCiAgYm9y ZGVyPSIwIiBoZWlnaHQ9IjMyIg0KICBhbHQ9IkluZGV4IiB3aWR0aD0iMzIiPjwvQT48L3Rk Pg0KPC90cj48L3RhYmxlPg0KPGIgY2xhc3M9Im5hdmxhYmVsIj5QcmV2aW91czo8L2I+IDxh IGNsYXNzPSJzZWN0cmVmIiBocmVmPSJyZXBvcnRpbmctYnVncy5odG1sIj5BLiBSZXBvcnRp bmcgQnVnczwvQT4NCjxiIGNsYXNzPSJuYXZsYWJlbCI+VXA6PC9iPiA8YSBjbGFzcz0ic2Vj dHJlZiIgSFJFRj0iYXBpLmh0bWwiPlB5dGhvbi9DIEFQSSBSZWZlcmVuY2UgTWFudWFsPC9B Pg0KPGIgY2xhc3M9Im5hdmxhYmVsIj5OZXh0OjwvYj4gPGEgY2xhc3M9InNlY3RyZWYiIEhS RUY9Im5vZGU4MS5odG1sIj5CLjEgSGlzdG9yeSBvZiB0aGU8L0E+DQo8aHI+DQo8c3BhbiBj bGFzcz0icmVsZWFzZS1pbmZvIj5SZWxlYXNlIDIuMi4xLCBkb2N1bWVudGF0aW9uIHVwZGF0 ZWQgb24gQXByaWwgMTAsIDIwMDIuPC9zcGFuPg0KPC9ESVY+DQo8IS0tRW5kIG9mIE5hdmln YXRpb24gUGFuZWwtLT4NCjxBRERSRVNTPg0KU2VlIDxpPjxhIGhyZWY9ImFib3V0Lmh0bWwi PkFib3V0IHRoaXMgZG9jdW1lbnQuLi48L2E+PC9pPiBmb3IgaW5mb3JtYXRpb24gb24gc3Vn Z2VzdGluZyBjaGFuZ2VzLg0KPC9BRERSRVNTPg0KPC9CT0RZPg0KPC9IVE1MPg0K --AD17450M9eoh9082984I6oWJBRcP79h68t955-- ------=_NextPartTM-000-38c398fc-093a-47c3-a862-b2ecfd1b7404 Content-Type: text/plain; name="InterScan_SafeStamp.txt" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="InterScan_SafeStamp.txt" ****** Message from InterScan E-Mail VirusWall NT ****** ** WARNING! Attached file src.pif contains: WORM_KLEZ.H virus Attempted to clean the file but it is not cleanable. It has been deleted. Message : Virus détecté ***************** End of message *************** ------=_NextPartTM-000-38c398fc-093a-47c3-a862-b2ecfd1b7404-- From deleeuw@stat.ucla.edu Sun Jan 5 17:30:54 2003 From: deleeuw@stat.ucla.edu (Jan de Leeuw) Date: Sun, 5 Jan 2003 09:30:54 -0800 Subject: [Pythonmac-SIG] Rpy in IDLE and PythonIDE Message-ID: <715A1490-20D3-11D7-9038-000393BB6D36@stat.ucla.edu> This is peculiar. If I "from rpy import *" in IDLE it starts leaking memory right away (with ProcessViewer open, one sees it going), but no such thing happens in PythonIDE. It is known that Rpy has leaks, but why doesn't it happen in PythonIDE, only in IDLE ? For what it's worth: there also does not seem to be any leakage in pycrust/pyshell when importing and working with rpy. === Jan de Leeuw; Professor and Chair, UCLA Department of Statistics; Editor: Journal of Multivariate Analysis, Journal of Statistical Software US mail: 9432 Boelter Hall, Box 951554, Los Angeles, CA 90095-1554 phone (310)-825-9550; fax (310)-206-5658; email: deleeuw@stat.ucla.edu homepage: http://gifi.stat.ucla.edu ------------------------------------------------------------------------ ------------------------- No matter where you go, there you are. --- Buckaroo Banzai http://gifi.stat.ucla.edu/sounds/nomatter.au ------------------------------------------------------------------------ ------------------------- From Jack.Jansen@oratrix.com Sun Jan 5 22:43:45 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sun, 5 Jan 2003 23:43:45 +0100 Subject: [Pythonmac-SIG] Python 2.1 dies when Executing Zope z2.py In-Reply-To: <11C5376E-2089-11D7-AA02-000A27AB5498@mac.com> Message-ID: <25DD052C-20FF-11D7-B774-000A27B19B96@oratrix.com> On zondag, jan 5, 2003, at 09:38 Europe/Amsterdam, Robert Rogers wrote: >> Hi, >> >> I'm trying to get Zope up on my 'Mac os X' client. Unfortunately, I >> can't tell if the problem lies with my python 2.1 build. >> >> Specifically, when I run the start script, which is basically 'python >> z2.py', python dies after generating this error message: >> >> dyld: /usr/local/bin/python.exe Undefined symbols: >> _PyType_IsSubtype >> Trace/BPT trap >> [...] >> warning: install: modules installed to >> '/usr/local/lib/python2.1/lib-dynload/', which is not in Python's >> module search path (sys.path) -- you'll have to change the search >> path yourself I think you are using two different version of Python: the Python 2.1 you installed yourself into /usr/local/bin and the Apple-supplied Python 2.2 in /usr/bin/python. My guess is that you have used /usr/bin/python to build Zope and are now using /usr/local/bin/python to run it. You must use the exact same python to do the build as you will use to run Zope (because the ABI that the Python core presents to extension modules may change between releases). -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From tony@lownds.com Mon Jan 6 04:41:53 2003 From: tony@lownds.com (Tony Lownds) Date: Sun, 5 Jan 2003 20:41:53 -0800 Subject: [Pythonmac-SIG] Building Mac extensions for Apple Python In-Reply-To: <1ECE6725-1F5B-11D7-8857-000A27B19B96@oratrix.com> References: <1ECE6725-1F5B-11D7-8857-000A27B19B96@oratrix.com> Message-ID: At 9:37 PM +0100 1/3/03, Jack Jansen wrote: >PyObject_Del is declared for python 2.2 (in >/usr/include/python2.2/objimpl.h) so I really have no idea why it >would fail... I think the problem is that PyObject_Del is defined in python2.2 as a macro with arguments: #define PyObject_Del(op) _PyObject_Del((PyObject *)(op)) So, it can't be used to provide a function pointer for the AEDesc_Type struct. Fixing that one module isn't hard: diff -u -r1.18 _AEmodule.c --- ../Modules/ae/_AEmodule.c 23 Dec 2002 23:16:20 -0000 1.18 +++ ../Modules/ae/_AEmodule.c 6 Jan 2003 04:27:23 -0000 @@ -875,8 +875,10 @@ return self; } -#define AEDesc_tp_free PyObject_Del - +static void AEDesc_tp_free(PyObject *self) +{ + PyObject_Del(self); +} PyTypeObject AEDesc_Type = { PyObject_HEAD_INIT(NULL) But aren't those modules generated by something? -Tony From gherman@darwin.in-berlin.de Tue Jan 7 08:18:21 2003 From: gherman@darwin.in-berlin.de (Dinu Gherman) Date: Tue, 7 Jan 2003 09:18:21 +0100 Subject: [Pythonmac-SIG] Re: [Image-SIG] Fixes for PIL enhancers In-Reply-To: <3E1A13E0.4010307@onsitetech.com> Message-ID: <954F5AA6-2218-11D7-85A4-00039345C610@darwin.in-berlin.de> Robb Shecter: > Any suggestions (or more Enhancer implementations!) are appreciated. Well, I was experimenting with automated enhancers for photographs and played a bit with the effects of shifting an image's histogram by some constant offset. (The code below works, but I'm not sure if this is fully PIL-mantra-compliant?! ;-) The existing Brightness enhancer is different since it multiplies pixel values by a factor and the histogram changes its width slight- ly. I reckon that it is probably hopeless to try one's hands on fully automated brightness/contrast enhancement of photographs without some AI component... But maybe I should try some non-linear ver- sions, too, before saaying that... class Shift(ImageEnhance._Enhance): "An enhancer that shifts the histogram of an image by some offset." def __init__(self, im): self.im =3D im def enhance(self, offset): return self.im.point(lambda i: i+offset) Regards, Dinu -- Dinu C. Gherman ...................................................................... "Dans la vie, il n'y a pas de solutions. Il y a des forces en marche: il faut les cr=E9er, et les solutions suivent." (Antoine de Saint-Exup=E9ry) From gherman@darwin.in-berlin.de Tue Jan 7 15:58:06 2003 From: gherman@darwin.in-berlin.de (Dinu Gherman) Date: Tue, 7 Jan 2003 16:58:06 +0100 Subject: [Pythonmac-SIG] Re: [Image-SIG] Fixes for PIL enhancers In-Reply-To: <954F5AA6-2218-11D7-85A4-00039345C610@darwin.in-berlin.de> Message-ID: I wrote: > Well, I was experimenting with automated enhancers for photographs > and played a bit with the effects of shifting an image's histogram > by some constant offset. (The code below works, but I'm not sure if > this is fully PIL-mantra-compliant?! ;-) [...] Sorry, wrong list...! :-/ Dinu -- Dinu C. Gherman ...................................................................... "Of course America had often been discovered before Columbus, but it had always been hushed up." (Oscar Wilde) From Larry.A.Meyn@nasa.gov Tue Jan 7 20:12:19 2003 From: Larry.A.Meyn@nasa.gov (Larry Meyn) Date: Tue, 07 Jan 2003 12:12:19 -0800 Subject: [Pythonmac-SIG] Off Topic: X11 for OS X Message-ID: <530A0CB1-227C-11D7-8F4F-000A277E7E9E@nasa.gov> This is off topic, but it may be of interest, there is now an Apple version of X11 available at http://www.apple.com/macosx/x11/ From bob@mastersofbranding.com Tue Jan 7 21:16:19 2003 From: bob@mastersofbranding.com (Bob Ippolito) Date: Tue, 7 Jan 2003 16:16:19 -0500 Subject: [Pythonmac-SIG] Off Topic: X11 for OS X In-Reply-To: <530A0CB1-227C-11D7-8F4F-000A277E7E9E@nasa.gov> Message-ID: <43B41711-2285-11D7-83A6-000A95686CD8@mastersofbranding.com> Yeah, I just started playing with it, it doesn't come with anything special other than the quartz_wm and window server.. however it does seem to have excellent OpenGL support! As far as UI goes, you can cmd-tilde though windows, and there's a window list in the menubar.. but the window list isn't in the contextual menu of the X11 dock item. X11 windows don't show up in the dock unless they're minimized. It looks like it has FreeType in there with working support (it does include a couple ttf fonts); I can't confirm this, but I'm pretty sure it works. It has all the standard XFree86 fonts as well. Fonts are not antialiased, and you don't get your OS X fonts. It does NOT have DPS support, though it has a couple of the applications sitting around in there. I wonder if it's dirt from the NeXT days or if they plan on adding this feature? It's kind of freaky if you run xeyes, it has a transparent background, with dropshadows on the eyes! It uses the X11 window's visible to do drop shadows, not the rectangle. Clipboard support works *perfectly* as far as I can tell (for text.. converts rich text on OS X to plain text for X11), but drag & drop doesn't work either way (nor should it, really). xmag doesn't seem to work (shows up all black).. probably something to do with being rootless? xman says no manual pages found that's about all I could come up with.. without trying to compile anything. On Tuesday, Jan 7, 2003, at 15:12 America/New_York, Larry Meyn wrote: > This is off topic, but it may be of interest, there is now an Apple > version of X11 available at http://www.apple.com/macosx/x11/ From Larry.A.Meyn@nasa.gov Mon Jan 6 20:47:57 2003 From: Larry.A.Meyn@nasa.gov (Larry Meyn) Date: Mon, 6 Jan 2003 12:47:57 -0800 Subject: [Pythonmac-SIG] Build Problem - libWASTE.a is out of date Message-ID: <22D7CF68-21B8-11D7-886E-00306546DFB6@nasa.gov> I've tried the last few day to Python with PythonIDE, but it always complains of a missing waste extension module. I downloaded the waste files from http://www.merzwaren.com/bin/waste/waste-21b1.sit and installed them where instructed, but it doesn't seem to work. I do get the following message while running make: gcc -Wl,-F. -bundle -framework Python build/temp.darwin-6.3-Power_Macintosh-2.3/wastemodule.o build/temp.darwin-6.3-Power_Macintosh-2.3/WEObjectHandlers.o build/temp.darwin-6.3-Power_Macintosh-2.3/WETabHooks.o build/temp.darwin-6.3-Power_Macintosh-2.3/WETabs.o -L../waste/Static Libraries -L/Library/Frameworks/Python.framework/Versions/2.3/lib -L/usr/local/lib -lWASTE -o build/lib.darwin-6.3-Power_Macintosh-2.3/waste.so -framework Carbon ld: table of contents for archive: ../waste/Static Libraries/libWASTE.a is out of date; rerun ranlib(1) (can't load from it) Any suggestions? Larry Meyn From Jack.Jansen@oratrix.com Tue Jan 7 22:04:01 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Tue, 7 Jan 2003 23:04:01 +0100 Subject: [Pythonmac-SIG] Build Problem - libWASTE.a is out of date In-Reply-To: <22D7CF68-21B8-11D7-886E-00306546DFB6@nasa.gov> Message-ID: On maandag, jan 6, 2003, at 21:47 Europe/Amsterdam, Larry Meyn wrote: > ld: table of contents for archive: ../waste/Static > Libraries/libWASTE.a is out of date; rerun ranlib(1) (can't load from > it) Yes, just follow the instructions and run ranlib ../waste/Static Libraries/libWASTE.a There's also some system libraries that will have this problem, I know libbz2.a does. For these you should do "sudo ranlib ....../libxxxx.a". -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From Jack.Jansen@oratrix.com Tue Jan 7 23:18:21 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Wed, 8 Jan 2003 00:18:21 +0100 Subject: [Pythonmac-SIG] MacPython-OS9 2.3a1 available Message-ID: <50246499-2296-11D7-AEFC-000A27B19B96@oratrix.com> MacPython-OS9 2.3a1 (the successor to MacPython 2.2) is available via http://www.cwi.nl/~jack/macpython.html and you are all cordially invited to try it out! Actually: if a Python on OS8.6 or OS9 or a CFM-Python on OSX is important to you'd better test it, I've moved almost completely to OSX so this distribution has seen very little testing, and the same will be true for the next releases. As usual I will wait another day before I send this note to a wider audience, so please report showstopper bugs ASAP. Here are main the Mac-specific changes, see :Misc:NEWS in the distribution for more: - The current naming convention for Python on the Macintosh is that MacPython refers to the unix-based OSX-only version, and MacPython-OS9 refers to the CFM-based version that runs on both OS9 and OSX. - All MacPython-OS9 functionality is now available in an OSX unix build, including the Carbon modules, the IDE, OSA support, etc. A lot of this will only work correctly in a framework build, though, because you cannot talk to the window manager unless your application is run from a .app bundle. There is a command line tool "pythonw" that runs your script with an interpreter living in such a .app bundle, this interpreter should be used to run any Python script using the window manager (including Tkinter or wxPython scripts). - Most of Mac/Lib has moved to Lib/plat-mac, which is again used both in MacPython-OSX and MacPython-OS9. The only modules remaining in Mac/Lib are specifically for MacPython-OS9 (CFM support, preference resources, etc). - MacPython-OS9 is now Carbon-only, so it runs on Mac OS 9 or Mac OS X and possibly on Mac OS 8.6 with the right CarbonLib installed, but not on earlier releases. - All the Carbon classes are now PEP253 compliant, meaning that you can subclass them from Python. Most of the attributes have gone, you should now use the accessor function call API, which is also what Apple's documentation uses. Some attributes such as grafport.visRgn are still available for convenience. - New Carbon modules File (implementing the APIs in Files.h and Aliases.h) and Folder (APIs from Folders.h). The old macfs builtin module is gone, and replaced by a Python wrapper around the new modules. - New Carbon modules Help and AH give access to the Carbon Help Manager. There are hooks in the IDE to allow accessing the Python documentation (and Apple's Carbon and Cocoa documentation) through the Help Viewer. See Mac/OSX/README for converting the Python documentation to a Help Viewer comaptible form and installing it. - OSA support has been redesigned and the generated Python classes now mirror the inheritance defined by the underlying OSA classes. - MacPython no longer maps both \r and \n to \n on input for any text file. This feature has been replaced by universal newline support (PEP278). - The default encoding for Python sourcefiles in MacPython-OS9 is no longer mac-roman (or whatever your local Mac encoding was but "ascii", like on other platforms. If you really need sourcefiles with Mac characters in them you can change this in site.py. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From Jack.Jansen@oratrix.com Tue Jan 7 23:19:44 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Wed, 8 Jan 2003 00:19:44 +0100 Subject: [Pythonmac-SIG] MacPython-OS9 2.3a1 available Message-ID: <8150A91A-2296-11D7-AEFC-000A27B19B96@oratrix.com> MacPython-OS9 2.3a1 (the successor to MacPython 2.2) is available via http://www.cwi.nl/~jack/macpython.html and you are all cordially invited to try it out! Actually: if a Python on OS8.6 or OS9 or a CFM-Python on OSX is important to you'd better test it, I've moved almost completely to OSX so this distribution has seen very little testing, and the same will be true for the next releases. As usual I will wait another day before I send this note to a wider audience, so please report showstopper bugs ASAP. Here are main the Mac-specific changes, see :Misc:NEWS in the distribution for more: - The current naming convention for Python on the Macintosh is that MacPython refers to the unix-based OSX-only version, and MacPython-OS9 refers to the CFM-based version that runs on both OS9 and OSX. - All MacPython-OS9 functionality is now available in an OSX unix build, including the Carbon modules, the IDE, OSA support, etc. A lot of this will only work correctly in a framework build, though, because you cannot talk to the window manager unless your application is run from a .app bundle. There is a command line tool "pythonw" that runs your script with an interpreter living in such a .app bundle, this interpreter should be used to run any Python script using the window manager (including Tkinter or wxPython scripts). - Most of Mac/Lib has moved to Lib/plat-mac, which is again used both in MacPython-OSX and MacPython-OS9. The only modules remaining in Mac/Lib are specifically for MacPython-OS9 (CFM support, preference resources, etc). - MacPython-OS9 is now Carbon-only, so it runs on Mac OS 9 or Mac OS X and possibly on Mac OS 8.6 with the right CarbonLib installed, but not on earlier releases. - All the Carbon classes are now PEP253 compliant, meaning that you can subclass them from Python. Most of the attributes have gone, you should now use the accessor function call API, which is also what Apple's documentation uses. Some attributes such as grafport.visRgn are still available for convenience. - New Carbon modules File (implementing the APIs in Files.h and Aliases.h) and Folder (APIs from Folders.h). The old macfs builtin module is gone, and replaced by a Python wrapper around the new modules. - New Carbon modules Help and AH give access to the Carbon Help Manager. There are hooks in the IDE to allow accessing the Python documentation (and Apple's Carbon and Cocoa documentation) through the Help Viewer. See Mac/OSX/README for converting the Python documentation to a Help Viewer comaptible form and installing it. - OSA support has been redesigned and the generated Python classes now mirror the inheritance defined by the underlying OSA classes. - MacPython no longer maps both \r and \n to \n on input for any text file. This feature has been replaced by universal newline support (PEP278). - The default encoding for Python sourcefiles in MacPython-OS9 is no longer mac-roman (or whatever your local Mac encoding was but "ascii", like on other platforms. If you really need sourcefiles with Mac characters in them you can change this in site.py. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From mday@mac.com Wed Jan 8 00:51:51 2003 From: mday@mac.com (Mark Day) Date: Tue, 7 Jan 2003 16:51:51 -0800 Subject: [Pythonmac-SIG] Build Problem - libWASTE.a is out of date In-Reply-To: <22D7CF68-21B8-11D7-886E-00306546DFB6@nasa.gov> Message-ID: <600A8270-22A3-11D7-BF13-00039354009A@mac.com> On Monday, January 6, 2003, at 12:47 PM, Larry Meyn wrote: > I've tried the last few day to Python with PythonIDE, but it always > complains of a missing waste extension module. I downloaded the waste > files from http://www.merzwaren.com/bin/waste/waste-21b1.sit and > installed them where instructed, but it doesn't seem to work. I do > get the following message while running make: > ld: table of contents for archive: ../waste/Static > Libraries/libWASTE.a is out of date; rerun ranlib(1) (can't load from > it) As the error message suggests, you need to run ranlib to fix up the archive. So: ranlib ../waste/Static Libraries/libWASTE.a should do it. -Mark From just@letterror.com Wed Jan 8 08:58:27 2003 From: just@letterror.com (Just van Rossum) Date: Wed, 8 Jan 2003 09:58:27 +0100 Subject: [Pythonmac-SIG] Off Topic: X11 for OS X In-Reply-To: <43B41711-2285-11D7-83A6-000A95686CD8@mastersofbranding.com> Message-ID: Bob Ippolito wrote: > Fonts are not antialiased [ ... ] I'm not surre whether this is still true, but I believe there is a fundamental limitation in X11 that makes anti-aliasing impossible. Just From bob@redivi.com Wed Jan 8 13:39:00 2003 From: bob@redivi.com (Bob Ippolito) Date: Wed, 8 Jan 2003 08:39:00 -0500 Subject: [Pythonmac-SIG] Off Topic: X11 for OS X In-Reply-To: Message-ID: <8B4BEC5B-230E-11D7-9724-000A95686CD8@redivi.com> On Wednesday, Jan 8, 2003, at 03:58 America/New_York, Just van Rossum wrote: > Bob Ippolito wrote: > >> Fonts are not antialiased [ ... ] > > I'm not surre whether this is still true, but I believe there is a > fundamental limitation in X11 that makes anti-aliasing impossible. Nah, it's not as true as it used to be.. I think part/most of it was because all the fonts in X11 were bitmap fonts, but they're not [necessarily] anymore. A quick example from Google: http://www.openbsd.org/faq/truetype.html -bob From joanca@typerware.com Thu Jan 9 10:07:19 2003 From: joanca@typerware.com (Joancarles =?iso-8859-1?Q?Casas=EDn?=) Date: Thu, 9 Jan 2003 11:07:19 +0100 Subject: [Pythonmac-SIG] Buildapplication Message-ID: I'm having a "syntax error" when trying to build an application using "BuildApplication" (! :) while the same script builds an applet without any problem. I'm also building application from another simple old script w/o problems... I'm using Python 2.2.2 under OSX.2.2 Any clue? Many thanks, Joancarles From just@letterror.com Thu Jan 9 10:19:39 2003 From: just@letterror.com (Just van Rossum) Date: Thu, 9 Jan 2003 11:19:39 +0100 Subject: [Pythonmac-SIG] Buildapplication In-Reply-To: Message-ID: Joancarles Casas=EDn wrote: > I'm having a "syntax error" when trying to build an application using > "BuildApplication" (!=20 > :) > while the same script builds an applet without any problem. I'm also > building application from another simple old script w/o problems... > I'm using Python 2.2.2 under OSX.2.2 Check whether the file causing the syntaxerror ends with a newline. I suppose BuildApplication should add a newline when compiling sources. Let me know if it helps. Just From joanca@typerware.com Thu Jan 9 10:42:55 2003 From: joanca@typerware.com (Joancarles =?iso-8859-1?Q?Casas=EDn?=) Date: Thu, 9 Jan 2003 11:42:55 +0100 Subject: [Pythonmac-SIG] Buildapplication In-Reply-To: Message-ID: On 09/01/03 at 11:19, Just van Rossum wrote: > Check whether the file causing the syntaxerror ends with a newline. I > suppose BuildApplication should add a newline when compiling sources. > Let me know if it helps. Absolutely! The file ended with a newline + tab Many thanks Just Groet, Joancarles From p.oberndoerfer@urheberrecht.org Thu Jan 9 17:58:27 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Thu, 09 Jan 2003 18:58:27 +0100 Subject: [Pythonmac-SIG] termios Message-ID: Just a short question (I hope): Will there be a chance to talk to a Keyspan USB to Serial adapter via termios from MacPython(-OSX) on a box running MacOS X (10.1.5)? Thanks, Pascal From bob@redivi.com Thu Jan 9 22:00:17 2003 From: bob@redivi.com (Bob Ippolito) Date: Thu, 9 Jan 2003 17:00:17 -0500 Subject: [Pythonmac-SIG] termios In-Reply-To: Message-ID: On Thursday, Jan 9, 2003, at 12:58 America/New_York, Pascal Oberndoerfer wrote: > Will there be a chance to talk to a Keyspan USB to Serial adapter via > termios from MacPython(-OSX) on a box running MacOS X (10.1.5)? I don't know about termios directly, but I have successfully used the exact same adapter with MacOS X 10.2.2 and 10.2.3 with the pyserial package ( http://pyserial.sourceforge.net/ ) and I developed a Twisted ( http://www.twistedmatrix.com/ ) serial transport that most definitely works with that particular adapter on MacOS X, Jython (w/ appropriate class libraries), and Linux (x86). It does not currently work with Windows, though (in Twisted, pyserial supports windows just fine). -bob From p.oberndoerfer@urheberrecht.org Thu Jan 9 22:23:44 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Thu, 09 Jan 2003 23:23:44 +0100 Subject: [Pythonmac-SIG] mac-roman and PythonIDE Message-ID: Hmm... Seems like the PythonIDE (under MacPython 2.3a1 [*not* OS9]) lost it= s ability to display mac-roman characters like =E4, =F6, =FC, or '...' (i.e. true ellipsis) from the scripts menu if they occur inside script names. Anybody else seeing this? Pascal From just@letterror.com Thu Jan 9 22:40:47 2003 From: just@letterror.com (Just van Rossum) Date: Thu, 9 Jan 2003 23:40:47 +0100 Subject: [Pythonmac-SIG] mac-roman and PythonIDE In-Reply-To: Message-ID: Pascal Oberndoerfer wrote: > Hmm... Seems like the PythonIDE (under MacPython 2.3a1 [*not* OS9]) > lost its ability to display mac-roman characters like =E4, =F6, =FC, or > '...' (i.e. true ellipsis) from the scripts menu if they occur inside > script names. >=20 > Anybody else seeing this? Strange, it works for me. Just From skip@pobox.com Thu Jan 9 22:38:39 2003 From: skip@pobox.com (Skip Montanaro) Date: Thu, 9 Jan 2003 16:38:39 -0600 Subject: [Pythonmac-SIG] mac-roman and PythonIDE In-Reply-To: References: Message-ID: <15901.63983.688510.213940@montanaro.dyndns.org> Pascal> Hmm... Seems like the PythonIDE (under MacPython 2.3a1 [*no= t* Pascal> OS9]) lost its ability to display mac-roman characters like= =E4, Pascal> =F6, =FC, or '...' (i.e. true ellipsis) from the scripts me= nu if Pascal> they occur inside script names. I thought I saw a note here or on the python-checkins list that Jack ha= d switched the default encoding to ascii. If you want/need mac-roman I t= hink you'll need to edit site.py. Skip From Larry.A.Meyn@nasa.gov Thu Jan 9 22:57:23 2003 From: Larry.A.Meyn@nasa.gov (Larry Meyn) Date: Thu, 9 Jan 2003 14:57:23 -0800 Subject: [Pythonmac-SIG] mac-roman and PythonIDE In-Reply-To: <15901.63983.688510.213940@montanaro.dyndns.org> Message-ID: --Apple-Mail-4--578666147 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; delsp=yes; charset=ISO-8859-1; format=flowed This has been reported as Bug item #665101 https://sourceforge.net/tracker/=20 ?func=3Ddetail&atid=3D105470&aid=3D665101&group_id=3D5470 I'll add your comments. Larry On Thursday, January 9, 2003, at 02:38 PM, Skip Montanaro wrote: > > Pascal> Hmm... Seems like the PythonIDE (under MacPython 2.3a1 =20 > [*not* > Pascal> OS9]) lost its ability to display mac-roman characters =20 > like =E4, > Pascal> =F6, =FC, or '...' (i.e. true ellipsis) from the scripts = menu =20 > if > Pascal> they occur inside script names. > > I thought I saw a note here or on the python-checkins list that Jack =20= > had > switched the default encoding to ascii. If you want/need mac-roman I =20= > think > you'll need to edit site.py. > > Skip > > _______________________________________________ > Pythonmac-SIG maillist - Pythonmac-SIG@python.org > http://mail.python.org/mailman/listinfo/pythonmac-sig > > Larry Meyn Aerospace Operations Modeling Office M/S 210-10 Phone: (650) 604-5038 NASA Ames Research Center Fax: (650) 604-0222 Moffett Field, CA 94035-1000 E-mail: Larry.A.Meyn@nasa.gov E-Fax: (425) 944-5526 sent via e-mail --Apple-Mail-4--578666147 Content-Transfer-Encoding: quoted-printable Content-Type: text/enriched; charset=ISO-8859-1 This has been reported as Bug item #665101 = https://sourceforge.net/tracker/?func=3Ddetail&atid=3D105470&aid=3D665101&= group_id=3D5470 I'll add your comments. Larry On Thursday, January 9, 2003, at 02:38 PM, Skip Montanaro wrote: Pascal> Hmm... Seems like the PythonIDE (under MacPython 2.3a1 [*not* Pascal> OS9]) lost its ability to display mac-roman characters like =E4, Pascal> =F6, =FC, or '...' (i.e. true ellipsis) from the scripts = menu if Pascal> they occur inside script names. I thought I saw a note here or on the python-checkins list that Jack had switched the default encoding to ascii. If you want/need mac-roman I think you'll need to edit site.py. Skip _______________________________________________ Pythonmac-SIG maillist - Pythonmac-SIG@python.org http://mail.python.org/mailman/listinfo/pythonmac-sig CourierLarry Meyn Aerospace Operations Modeling Office M/S 210-10 Phone: (650) 604-5038 NASA Ames Research Center Fax: (650) 604-0222 Moffett Field, CA 94035-1000 E-mail: Larry.A.Meyn@nasa.gov E-Fax: (425) 944-5526 sent via e-mail --Apple-Mail-4--578666147-- From just@letterror.com Thu Jan 9 23:24:58 2003 From: just@letterror.com (Just van Rossum) Date: Fri, 10 Jan 2003 00:24:58 +0100 Subject: [Pythonmac-SIG] mac-roman and PythonIDE In-Reply-To: Message-ID: Pascal Oberndoerfer wrote: > Hmm... Seems like the PythonIDE (under MacPython 2.3a1 [*not* OS9]) > lost its ability to display mac-roman characters like =E4, =F6, =FC, or > '...' (i.e. true ellipsis) from the scripts menu if they occur inside > script names. >=20 > Anybody else seeing this? Not with the OS9-compatible MacPython 2.3a1, but I did see it in a framework build. I've fixed it in CVS by explicitly interpreting the output of os.listdir() as utf-8, and added a tiny bit of unicode support to the Menu stuff in FrameWork.py. Just From p.oberndoerfer@urheberrecht.org Thu Jan 9 17:32:30 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Thu, 09 Jan 2003 18:32:30 +0100 Subject: [Pythonmac-SIG] MacPython-OS9 2.3a1 available In-Reply-To: <8150A91A-2296-11D7-AEFC-000A27B19B96@oratrix.com> Message-ID: Jack Jansen at Jack.Jansen@oratrix.com: > MacPython-OS9 2.3a1 (the successor to MacPython 2.2) is available via > http://www.cwi.nl/~jack/macpython.html and you are all cordially > invited to try it out! Jack, after installing on two OS 8.6 (7600/G3) and one 9.2.2 systems (TiBook) I ran into some (and the same) problems. Doing the usual from test import regrtest; regrtest.main() this is what happened: - test_import.py left the file longlist.py inside the MacPython-OS9 folder. I thought it should be removed already after "compiling" the .pyc equivalent? - The interpreter stopped after threaded-import. Cursor is blinking, interpreter is *not* terminated. Moving the window results in a error 3 message. Pascal BTW: contrary(?) to the README socket takes a while but seems to work ... From p.oberndoerfer@urheberrecht.org Fri Jan 10 22:20:41 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Fri, 10 Jan 2003 23:20:41 +0100 Subject: [Pythonmac-SIG] termios In-Reply-To: Message-ID: Bob Ippolito at bob@redivi.com: > I don't know about termios directly, but I have successfully used the > exact same adapter with MacOS X 10.2.2 and 10.2.3 with the pyserial > package ( http://pyserial.sourceforge.net/ ) and I developed a Twisted Bob, thanks again. It works very, very nice! Just one more question: is it true that I have to open a _named_ serial connection? I used "/dev/tty.USA19QW[12]1P1.1" depending on the USB port (no. 1 or 2) used. Attempts through /dev/cuaa[0-9] or /dev/ttyp[0-9] failed all. So if I don't know the device name in advance, all I can do is loop through "/dev/*" doing a RegEx or similar? Thanks. Pascal From bob@redivi.com Fri Jan 10 23:05:28 2003 From: bob@redivi.com (Bob Ippolito) Date: Fri, 10 Jan 2003 18:05:28 -0500 Subject: [Pythonmac-SIG] termios In-Reply-To: Message-ID: <0270CAF9-24F0-11D7-B784-000A95686CD8@redivi.com> On Friday, Jan 10, 2003, at 17:20 America/New_York, Pascal Oberndoerfer wrote: > Bob Ippolito at bob@redivi.com: > >> I don't know about termios directly, but I have successfully used the >> exact same adapter with MacOS X 10.2.2 and 10.2.3 with the pyserial >> package ( http://pyserial.sourceforge.net/ ) and I developed a Twisted > > Bob, > > thanks again. It works very, very nice! Just one more question: > is it true that I have to open a _named_ serial connection? I used > "/dev/tty.USA19QW[12]1P1.1" depending on the USB port (no. 1 or 2) > used. > Attempts through /dev/cuaa[0-9] or /dev/ttyp[0-9] failed all. So if > I don't know the device name in advance, all I can do is loop through > "/dev/*" doing a RegEx or similar? I'm sure there's a way to enumerate which serial ports are available, as VirtualPC does it (shows modem, irda, and the keyspan if it's connected on my laptop). I don't know how to do it. I'm pretty sure that the string USA19QW* is specific to that USB device, and another manufacturer would use a different string. I think you'd be fine getting away with just going through /dev/cua* and see which ones you can open in the meantime. -bob From p.oberndoerfer@urheberrecht.org Sat Jan 11 10:25:23 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Sat, 11 Jan 2003 11:25:23 +0100 Subject: [Pythonmac-SIG] termios (now slightly OT) In-Reply-To: <0270CAF9-24F0-11D7-B784-000A95686CD8@redivi.com> Message-ID: Bob Ippolito at bob@redivi.com: > I'm sure there's a way to enumerate which serial ports are available, > as VirtualPC does it (shows modem, irda, and the keyspan if it's > connected on my laptop). I don't know how to do it. I'm pretty sure > that the string USA19QW* is specific to that USB device, and another > manufacturer would use a different string. > > I think you'd be fine getting away with just going through /dev/cua* > and see which ones you can open in the meantime. What they might have in common is the "dot-syntax". I go through /dev/tty.* and get the same result as VirtualPC or Zterm: crw-rw-rw- 1 root wheel 8, 2 Jan 11 11:11 tty.IrDA-IrCOMMch-b crw-rw-rw- 1 root wheel 8, 4 Jan 11 11:14 tty.USA19QW11P1.1 crw-rw-rw- 1 root wheel 8, 0 Jan 11 11:11 tty.modem Works with /dev/cu.* also. Thanks. Pascal From gherman@darwin.in-berlin.de Sat Jan 11 10:56:25 2003 From: gherman@darwin.in-berlin.de (Dinu Gherman) Date: Sat, 11 Jan 2003 11:56:25 +0100 Subject: [Pythonmac-SIG] Python-Events iCalendar!? Message-ID: <53FAA350-2553-11D7-B79A-00039345C610@darwin.in-berlin.de> Slightly off-topic, but... I wonder if we could maintain an iCalendar related to Python events like this experimental one, but managed in some more collaborative fashion and/or without necessarily depending on .Mac? http://ical.mac.com/dinu/Python-Events webcal://ical.mac.com/dinu/Python-Events.ics I tried adding URLs to individual events, but they totally mess-up the HTML table in the published web calendar (first link above)... Dinu -- Dinu C. Gherman ...................................................................... "A whole generation has grown up with the idea that it is normal for them to have no freedom." (Richard M. Stallman) From Dan@Grassi.org Sat Jan 11 18:06:41 2003 From: Dan@Grassi.org (Dan Grassi) Date: Sat, 11 Jan 2003 13:06:41 -0500 Subject: [Pythonmac-SIG] termios In-Reply-To: Message-ID: <6FD0275B-258F-11D7-92B5-0030658F8A58@grassi.org> On Friday, January 10, 2003, at 05:20 PM, Pascal Oberndoerfer wrote: > is it true that I have to open a _named_ serial connection? I used > "/dev/tty.USA19QW[12]1P1.1" depending on the USB port (no. 1 or 2) > used. > Attempts through /dev/cuaa[0-9] or /dev/ttyp[0-9] failed all. So if > I don't know the device name in advance, all I can do is loop through > "/dev/*" doing a RegEx or similar? The answer is to use the IORegistry to find a device. Search for IOSerialBSDClient with a key of IOSerialBSCClientTyle with a value of IORS232SerialSteeam. Then get the value of the IODialinDevice or IOCalloutDevice which are respectively (on my system) /dev/tty.USA191P1.1 and dev/cu.USA191P1.1. Play with IORegistryExplorer to get a feel for it. :-) The calls to use are in Core Foundation (CF) and are "C" code based. AFAIK there is no python wrapper for these calls. I would be happy to provide the "C" code if someone would provide the python wrapper. Dan From amk@amk.ca Sun Jan 12 01:54:43 2003 From: amk@amk.ca (A.M. Kuchling) Date: Sat, 11 Jan 2003 20:54:43 -0500 Subject: [Pythonmac-SIG] Python-Events iCalendar!? In-Reply-To: <53FAA350-2553-11D7-B79A-00039345C610@darwin.in-berlin.de>; from gherman@darwin.in-berlin.de on Sat, Jan 11, 2003 at 11:56:25AM +0100 References: <53FAA350-2553-11D7-B79A-00039345C610@darwin.in-berlin.de> Message-ID: <20030111205443.A4628@amk.ca> On Sat, Jan 11, 2003 at 11:56:25AM +0100, Dinu Gherman wrote: >Slightly off-topic, but... I wonder if we could maintain an iCalendar >related to Python events like this experimental one, but managed in >some more collaborative fashion and/or without necessarily depending >on .Mac? A natural place for it would be on python.org. Generating iCal data isn't very difficult; I can post a simple example if that would be useful. --amk From gherman@darwin.in-berlin.de Sun Jan 12 14:20:27 2003 From: gherman@darwin.in-berlin.de (Dinu Gherman) Date: Sun, 12 Jan 2003 15:20:27 +0100 Subject: [Pythonmac-SIG] Python-Events iCalendar!? In-Reply-To: <20030111205443.A4628@amk.ca> Message-ID: A.M. Kuchling: > Dinu Gherman wrote: >> Slightly off-topic, but... I wonder if we could maintain an iCalendar >> related to Python events like this experimental one, but managed in >> some more collaborative fashion and/or without necessarily depending >> on .Mac? > > A natural place for it would be on python.org. Generating iCal data > isn't very difficult; I can post a simple example if that would be > useful. Yes, the format stuff is simple, but the question is how to maintain one (or more) such calendars by a group of people? In other words, can such a file on python.org be maintained by any calendar "sub- scriber" with special rights (using iCal)? Otherwise the calendar file needs to be created programatically from other data (which just distributes further the issue of maintaining that data). Dinu -- Dinu C. Gherman ...................................................................... "I value the friend who for me finds time on his calendar, but I cherish the friend who for me does not consult his calendar." (Robert Brault) From p.oberndoerfer@urheberrecht.org Sun Jan 12 15:23:37 2003 From: p.oberndoerfer@urheberrecht.org (Pascal Oberndoerfer) Date: Sun, 12 Jan 2003 16:23:37 +0100 Subject: [Pythonmac-SIG] termios In-Reply-To: <6FD0275B-258F-11D7-92B5-0030658F8A58@grassi.org> Message-ID: Dan Grassi at dan@grassi.org: > Play with IORegistryExplorer to get a feel for it. :-) Didn't even know it existed! Thanks. > The calls to use are in Core Foundation (CF) and are "C" code based. > AFAIK there is no python wrapper for these calls. I would be happy to > provide the "C" code if someone would provide the python wrapper. I would certainly try to help, but I am afraid I don't have the slightest clue what it takes... Any advice and I'll do my best. Pascal From Jack.Jansen@oratrix.com Sun Jan 12 21:20:03 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sun, 12 Jan 2003 22:20:03 +0100 Subject: [Pythonmac-SIG] termios In-Reply-To: <6FD0275B-258F-11D7-92B5-0030658F8A58@grassi.org> Message-ID: <9D3CC4EA-2673-11D7-B27C-000A27B19B96@oratrix.com> On zaterdag, jan 11, 2003, at 19:06 Europe/Amsterdam, Dan Grassi wrote: > The calls to use are in Core Foundation (CF) and are "C" code based. > AFAIK there is no python wrapper for these calls. I would be happy to > provide the "C" code if someone would provide the python wrapper. Anything that is in CF but not wrapped in the Carbon.CF module should be considered a bug. I know there's still quite a bit missing, but if people put bug reports up on sourceforge that will influence the order in which I add stuff to the CF module. It would help a lot if you mentioned the data structure and/or some of the routines you need. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From berkowit@silcom.com Mon Jan 13 16:47:09 2003 From: berkowit@silcom.com (Paul Berkowitz) Date: Mon, 13 Jan 2003 08:47:09 -0800 Subject: [Pythonmac-SIG] [OT] Someone's spreading klez out there Message-ID: My ISP has a very good virus-and-spam server detection. For some time, I've been regularly getting notices about messages from the PythonMac list containing klez.h. Of course there's no telling from whom the messages are really coming, but it's someone who receives PythonMac and has everyone's email address available for distribution on his or her Windows computer, probably using Outlook Express Windows. The two current messages containing klez (which I can't read even on the virus-detector website) are "[Pythonmac-SIG] For Information on suggestions" claiming to be from python-docs@python.org sent on 01-05-2003, and an earlier "[Pythonmac-SIG] Have a good New year" claiming to be from jph@emilia.engr.sgi.com on 12-31-2002. I could safely have them delivered to my Mac, but anyone who ever checks email to their subscribed account from Windows should be careful. One or more people on the list do not have adequate virus protection on their Windows machines. -- Paul Berkowitz From jonah@urban.net Mon Jan 13 17:42:48 2003 From: jonah@urban.net (Jonah Crawford) Date: Mon, 13 Jan 2003 12:42:48 -0500 Subject: [Pythonmac-SIG] import errors... Message-ID: <6E2C280A-271E-11D7-8D66-0003934273FA@urban.net> I compiled both the PyXML package and the pygoogle packages on my 10.2.3 box. When I import any of the modules it first traces to an import error then when I try again it executes the import. What have I done improperly during the compiling that it doesn't load correctly the first time around? Any suggestions? thanks in advance -j ---->>> import SOAP ----Traceback (most recent call last): ---- File "", line 1, in ? ---- File "/usr/bin/Python-2.2.1/Lib/SOAP.py", line 89, in ? ---- import cgi ---- File "/usr/bin/Python-2.2.1/Lib/cgi.py", line 39, in ? ---- import urllib ---- File "/usr/bin/Python-2.2.1/Lib/urllib.py", line 26, in ? ---- import socket ---- File "/usr/bin/Python-2.2.1/Lib/socket.py", line 41, in ? ---- from _socket import * ----ImportError: No module named _socket ---->>> import SOAP ---->>> From lee.list@joramo.com Tue Jan 14 15:11:02 2003 From: lee.list@joramo.com (Lee Joramo) Date: Tue, 14 Jan 2003 08:11:02 -0700 Subject: [Pythonmac-SIG] Sending emails via localhost Message-ID: <655F8D22-27D2-11D7-9986-003065D5B57C@joramo.com> I am trying to send mail from python's smtplib via the localhost, and get an error: socket.error: (61, 'Connection refused') a quick search of the net indicates that Mac OS X 10.2 does not have a localhost SMTP server running by default. I want python to be able to send mail to local accounts for administrative notices and I want to be able to test web development code while my powerbook is not on the net. I suspect that I need to learn how to setup sendmail or some other mail server on my system, but I was hoping that other mac python users may have some experience to share. Thanks Lee Joramo www.joramo.com From mwh@python.net Tue Jan 14 15:39:17 2003 From: mwh@python.net (Michael Hudson) Date: 14 Jan 2003 15:39:17 +0000 Subject: [Pythonmac-SIG] Sending emails via localhost In-Reply-To: Lee Joramo's message of "Tue, 14 Jan 2003 08:11:02 -0700" References: <655F8D22-27D2-11D7-9986-003065D5B57C@joramo.com> Message-ID: <2mk7h7lp0a.fsf@starship.python.net> Lee Joramo writes: > I am trying to send mail from python's smtplib via the localhost, and > get an error: > > socket.error: (61, 'Connection refused') > > a quick search of the net indicates that Mac OS X 10.2 does not have a > localhost SMTP server running by default. I want python to be able to > send mail to local accounts for administrative notices and I want to be > able to test web development code while my powerbook is not on the net. > > I suspect that I need to learn how to setup sendmail or some other mail > server on my system, but I was hoping that other mac python users may > have some experience to share. Mac OS X has /usr/bin/mail... sendmail seems to be installed, but as you say not started by default. Cheers, M. -- Those who have deviant punctuation desires should take care of their own perverted needs. -- Erik Naggum, comp.lang.lisp From francois.granger@free.fr Tue Jan 14 16:30:27 2003 From: francois.granger@free.fr (Fran=?ISO-8859-1?B?5w==?=ois Granger) Date: Tue, 14 Jan 2003 17:30:27 +0100 Subject: [Pythonmac-SIG] Sending emails via localhost In-Reply-To: <655F8D22-27D2-11D7-9986-003065D5B57C@joramo.com> Message-ID: on 14/01/03 16:11, Lee Joramo at lee.list@joramo.com wrote: > I suspect that I need to learn how to setup sendmail or some other mail > server on my system, but I was hoping that other mac python users may > have some experience to share. Sendmail is available but not started as default. It config seems to be a nightmare ;-) Simple explaations available all over the net: http://www.netneurotic.de/mac/configuring.html http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&group=comp.mail.sendmail Or http://minilien.com/?xdBHAFZRKT -- Le courrier est un moyen de communication. Les gens devraient se poser des questions sur les implications politiques des choix (ou non choix) de leurs outils et technologies. Pour des courriers propres : -- From webmaster@pferdemarkt.ws Wed Jan 15 12:33:05 2003 From: webmaster@pferdemarkt.ws (webmaster@pferdemarkt.ws) Date: Wed, 15 Jan 2003 04:33:05 -0800 Subject: [Pythonmac-SIG] Pferdemarkt.ws informiert! Newsletter 01/2003 Message-ID: <200301151233.EAA07222@eagle.he.net> http://www.pferdemarkt.ws Wir sind in 2003 erfolgreich in des neue \"Pferdejahr 2003 gestartet. Für den schnellen Erfolg unseres Marktes möchten wir uns bei Ihnen bedanken. Heute am 15. Januar 2003 sind wir genau 14 Tage Online! Täglich wächst unsere Datenbank um ca. 30 neue Angebote. Stellen auch Sie als Privatperson Ihre zu verkaufenden Pferde direkt und vollkommen Kostenlos ins Internet. Zur besseren Sichtbarmachung Ihrer Angebote können SIe bis zu ein Bild zu Ihrer Pferdeanzeige kostenlos einstellen! Klicken Sie hier um sich direkt einzuloggen http://www.Pferdemarkt.ws Kostenlos Anbieten, Kostenlos Suchen! Direkt von Privat zu Privat! Haben Sie noch Fragen mailto: webmaster@pferdemarkt.ws From cjl@physics.otago.ac.nz Wed Jan 15 19:46:47 2003 From: cjl@physics.otago.ac.nz (Chris Lee) Date: Thu, 16 Jan 2003 08:46:47 +1300 Subject: [Pythonmac-SIG] HDF module Message-ID: <154E7415-28C2-11D7-984D-0003937807FC@physics.otago.ac.nz> Has anyone compiled and installed the HDF module for Mac OS X using either MacPython or Unix Python? I'll do anything to avoid MATLab :) Cheers Chris ################################# Chris Lee Physics Department Otago University PO Box 56 Dunedin New Zealand Phone ++64 3 479 7749 Fax ++64 3 479 0964 ################################# From Jack.Jansen@oratrix.com Wed Jan 15 23:31:51 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Thu, 16 Jan 2003 00:31:51 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. Message-ID: <85ED827A-28E1-11D7-AE69-000A27B19B96@oratrix.com> Folks, I'm thinking of replacing macfs.StandardGetFile() and friends with another set of methods to give similar functionality, but geared more towards the way things are used nowadays. There will of course (for 2.3) be backward-compatible macfs.StandardGetFile() methods and such. First of all I was thinking of moving the functionality to EasyDialogs. The full interface is in Nav anyway, and an easy way to get files for opening or saving sounds like something that should go into EasyDialogs. Second, I was thinking of providing three calls: AskFileForOpen(), AskFileForSave() and AskFolder(). Are those enough? Any thoughts on the names? All will have a couple of optional parameters: - prompt: a string to use as the prompt. - return_type: pass the type of object you want returned here, either a string, an FSSpec or an FSRef. FSRef is not allowed for AskFileForSave(). - typelist: a list of 4-char filetypes that are acceptable (AskFileForOpen only). Do we need another interface, such as extensions? Mimetypes? - default: (AskFileForSave only) the default for the saved file name. - folder: the folder that you want to show the user initially. - preferenceKey: the Nav Services preference key. By passing something here the next time you call the dialog with the same preferenceKey it will be in the same place. Interaction of this with folder needs to be examined. Any more NavServices features that should be exposed, such as automatic conversion of documents? I was thinking of raising an exception if the user cancels, maybe KeyboardInterrupt? Is this a good idea, or should I stick with the current (value, ok) tuple? Return None? -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From diordna@silvernetwork.net Wed Jan 15 23:42:51 2003 From: diordna@silvernetwork.net (Stephen Johnson) Date: Wed, 15 Jan 2003 18:42:51 -0500 Subject: [Pythonmac-SIG] MacPython Carbon GUI Message-ID: <0FB764F2-28E3-11D7-AA23-0003931852EA@silvernetwork.net> Is there one I can download somewhere? Tkinter isn't supported in Carbon. I'm new to Python, but I also program in Java and METAL BASIC. I have MacPython v2.2.1. From aparente@adobe.com Thu Jan 16 00:29:11 2003 From: aparente@adobe.com (Alexandre Parenteau) Date: Wed, 15 Jan 2003 16:29:11 -0800 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <85ED827A-28E1-11D7-AE69-000A27B19B96@oratrix.com> Message-ID: I'd love to be able to filter based on the FSRef or FSSpec. Using the typeList only is less useful, especially on OSX where it is not so widely used. > I was thinking of raising an exception if the user cancels, maybe > KeyboardInterrupt? Is this a good idea, or should I stick with the Raising... I don't know... Don't like that too much for user cancellation... [my 2c in order to make it more difficult] Alex. > Folks, > I'm thinking of replacing macfs.StandardGetFile() and friends with > another set of methods to give similar functionality, but geared more > towards the way things are used nowadays. There will of course (for > 2.3) be backward-compatible macfs.StandardGetFile() methods and such. > > First of all I was thinking of moving the functionality to EasyDialogs. > The full interface is in Nav anyway, and an easy way to get files for > opening or saving sounds like something that should go into EasyDialogs. > > Second, I was thinking of providing three calls: AskFileForOpen(), > AskFileForSave() and AskFolder(). Are those enough? Any thoughts on the > names? All will have a couple of optional parameters: > - prompt: a string to use as the prompt. > - return_type: pass the type of object you want returned here, either a > string, an FSSpec or an FSRef. FSRef is not allowed for > AskFileForSave(). > - typelist: a list of 4-char filetypes that are acceptable > (AskFileForOpen only). Do we need another interface, such as > extensions? Mimetypes? > - default: (AskFileForSave only) the default for the saved file name. > - folder: the folder that you want to show the user initially. > - preferenceKey: the Nav Services preference key. By passing something > here the next time you call the dialog with the same preferenceKey it > will be in the same place. Interaction of this with folder needs to be > examined. > > Any more NavServices features that should be exposed, such as automatic > conversion of documents? > > I was thinking of raising an exception if the user cancels, maybe > KeyboardInterrupt? Is this a good idea, or should I stick with the > current (value, ok) tuple? Return None? > -- > - Jack Jansen > http://www.cwi.nl/~jack - > - If I can't dance I don't want to be part of your revolution -- Emma > Goldman - > > > _______________________________________________ > Pythonmac-SIG maillist - Pythonmac-SIG@python.org > http://mail.python.org/mailman/listinfo/pythonmac-sig > -- Alexandre Parenteau Computer Scientist Core Technologies, AGM Adobe Systems, Inc. 408-536-6166 From just@letterror.com Thu Jan 16 09:09:39 2003 From: just@letterror.com (Just van Rossum) Date: Thu, 16 Jan 2003 10:09:39 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: Message-ID: Alexandre Parenteau wrote: > I'd love to be able to filter based on the FSRef or FSSpec. Or even a plain posix path... > Using the > typeList only is less useful, especially on OSX where it is not so > widely used. Yep. > > I was thinking of raising an exception if the user cancels, maybe > > KeyboardInterrupt? Is this a good idea, or should I stick with the > > Raising... I don't know... Don't like that too much for user > cancellation... Me neither, I don't like that at all... Returning None is the way to go. Just From Jack.Jansen@oratrix.com Thu Jan 16 14:38:33 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Thu, 16 Jan 2003 15:38:33 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: Message-ID: <30603A7C-2960-11D7-8DBE-000A27B19B96@oratrix.com> On donderdag, jan 16, 2003, at 10:09 Europe/Amsterdam, Just van Rossum wrote: > Alexandre Parenteau wrote: > >> I'd love to be able to filter based on the FSRef or FSSpec. > > Or even a plain posix path... How would you like this filtering to be done? I know navservices has various filtering abilities (even down to giving all the filenames to user code) but what would you like? > >> Using the >> typeList only is less useful, especially on OSX where it is not so >> widely used. > > Yep. Agreed, but I want to keep that for backward compatibility. > >>> I was thinking of raising an exception if the user cancels, maybe >>> KeyboardInterrupt? Is this a good idea, or should I stick with the >> >> Raising... I don't know... Don't like that too much for user >> cancellation... > > Me neither, I don't like that at all... Returning None is the way to > go. Sounds good to me. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From just@letterror.com Thu Jan 16 14:45:33 2003 From: just@letterror.com (Just van Rossum) Date: Thu, 16 Jan 2003 15:45:33 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <30603A7C-2960-11D7-8DBE-000A27B19B96@oratrix.com> Message-ID: Jack Jansen wrote: > >> I'd love to be able to filter based on the FSRef or FSSpec. > > > > Or even a plain posix path... > > How would you like this filtering to be done? I know navservices has > various filtering abilities (even down to giving all the filenames to > user code) but what would you like? That's the one: you should be able to provide a callback that will be called with an FSRef (or path or alias, whatever) and should return a bool. I've once tried to do it with Nav directly (but still from Python) and it's hell, so if an intuitive Python interface existed I would be very happy indeed... Just From matsakis@mit.edu Thu Jan 16 15:09:35 2003 From: matsakis@mit.edu (Nick Matsakis) Date: Thu, 16 Jan 2003 10:09:35 -0500 (EST) Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <85ED827A-28E1-11D7-AE69-000A27B19B96@oratrix.com> References: <85ED827A-28E1-11D7-AE69-000A27B19B96@oratrix.com> Message-ID: On Thu, 16 Jan 2003, Jack Jansen wrote: > - typelist: a list of 4-char filetypes that are acceptable > (AskFileForOpen only). Do we need another interface, such as extensions? > Mimetypes? >From day one, Mac OS X has needed a system-wide database that translates between HFS type codes, MIME types, and filename extensions (with the latter one being configurable). I hope that one day we will have this. It would be very cool if Python on the Mac provided the functionality for a programmer to specify the file type(s) as either a MIME type or HFS code and have the system work it out. Though creating and maintaining a database of all file type information is daunting, restricting the scope to just those file types that are MIME standards would be doable (there are only a few dozen standard MIME types). And, if Apple or someone else ever releases a more exhaustive database you can just use that without changing the API. Generally, I'm very much against filename extensions for reasons I won't go into here except to say that there is no centralized registry for guaranteeing them to be unique. However, given the poor treatment that Apple has given this problem, there are a great many files out there whose type information is exclusively available as a filename extension. For this reason, the MacPython APIs should also provide the ability to select based on extension as well. Nick Matsakis From Jack.Jansen@oratrix.com Thu Jan 16 19:47:18 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Thu, 16 Jan 2003 20:47:18 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: Message-ID: <51E94F9C-298B-11D7-8DBE-000A27B19B96@oratrix.com> On donderdag, jan 16, 2003, at 15:45 Europe/Amsterdam, Just van Rossum wrote: >> I know navservices has >> various filtering abilities (even down to giving all the filenames to >> user code) but what would you like? > > That's the one: you should be able to provide a callback that will be > called with an FSRef (or path or alias, whatever) and should return a > bool. I've once tried to do it with Nav directly (but still from > Python) > and it's hell, so if an intuitive Python interface existed I would be > very happy indeed... Hmm, I've had that same experience. Does anyone happen to have any sample code? Also, does anyone know at what point those callbacks will be called? Or, to elaborate on the last question, how should we handle the global interpreter lock, is the only way to keep this safe to just hold the lock through the whole Nav Services interaction? On the other hand: because this is the most general mechanism (albeit the most slow one too) it's probably good to provide it. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From dev@brotsky.com Fri Jan 17 03:19:53 2003 From: dev@brotsky.com (Dan Brotsky) Date: Thu, 16 Jan 2003 19:19:53 -0800 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <51E94F9C-298B-11D7-8DBE-000A27B19B96@oratrix.com> Message-ID: <8BB28C58-29CA-11D7-B781-0003931036B4@brotsky.com> On Thursday, January 16, 2003, at 11:47 AM, Jack Jansen wrote: >> That's the one: you should be able to provide a callback that will be >> called with an FSRef (or path or alias, whatever) and should return a >> bool. I've once tried to do it with Nav directly (but still from >> Python) >> and it's hell, so if an intuitive Python interface existed I would be >> very happy indeed... > > Hmm, I've had that same experience. Does anyone happen to have any > sample code? See apple tech note 2017: http://developer.apple.com/technotes/tn/tn2017.html > Also, does anyone know at what point those callbacks will be called? Depends whether you are using a modal sheet or not. > Or, to elaborate on the last question, how should we handle the > global interpreter lock, is the only way to keep this safe to just > hold the lock through the whole Nav Services interaction? For modal sheets, I would guess yes. For non-modal sheets, we're way out of my depth :^). dan > On the other hand: because this is the most general mechanism (albeit > the most slow one too) it's probably good to provide it. > -- > - Jack Jansen > http://www.cwi.nl/~jack - > - If I can't dance I don't want to be part of your revolution -- Emma > Goldman - > > > _______________________________________________ > Pythonmac-SIG maillist - Pythonmac-SIG@python.org > http://mail.python.org/mailman/listinfo/pythonmac-sig > From Jack.Jansen@cwi.nl Fri Jan 17 16:16:40 2003 From: Jack.Jansen@cwi.nl (Jack Jansen) Date: Fri, 17 Jan 2003 17:16:40 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <85ED827A-28E1-11D7-AE69-000A27B19B96@oratrix.com> Message-ID: <0F7893A4-2A37-11D7-ABDC-0030655234CE@cwi.nl> The calls EasyDialogs.AskFileForOpen(), AskFileForSave() and AskFolder() have been implemented in CVS. Please give them a try. They've been tested only very lightly, so I could use a helping hand here with all the options. No documentation yet, but the main options are as follows (read the source for EasyDialogs.py and Nav.c for details): - wanted: you can pass a type here telling what you want back. FSSpec, FSRef, str or unicode (or subtypes thereof!) are acceptable. - message: the prompt (only ascii, for now) - savedFileName: the default output filename (only ascii, for now) - typeList: a list or tuple of 4-char filetype codes (or the Res.Handle data structure Nav really uses) - defaultLocation: which folder to show the user first (FSSpec, FSRef, pathname and AE are acceptable) The return value is of type "wanted" or None if the user cancelled. If someone can come up with an example for a filterProc use that would be nice. -- Jack Jansen, , http://www.cwi.nl/~jack If I can't dance I don't want to be part of your revolution -- Emma Goldman From just@letterror.com Fri Jan 17 18:37:05 2003 From: just@letterror.com (Just van Rossum) Date: Fri, 17 Jan 2003 19:37:05 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: <0F7893A4-2A37-11D7-ABDC-0030655234CE@cwi.nl> Message-ID: Jack Jansen wrote: > The calls EasyDialogs.AskFileForOpen(), AskFileForSave() and > AskFolder() have been implemented in CVS. Please give them a try. > They've been tested only very lightly, so I could use a helping hand > here with all the options. AskFileForSave() traces back if called with no arguments: Traceback (most recent call last): File "/Users/just/code/tmp/build/Ask.app/Contents/Resources/Ask.py", line 3, in ? print EasyDialogs.AskFileForSave() File "/Users/just/code/python_cvs/Lib/plat-mac/EasyDialogs.py", line 663, in AskFileForSave pardir_fsr = Carbon.File.FSRef(fss) NameError: global name 'fss' is not defined I think the dialogs should be movable; this should work correctly in an eventloop-less app as well in a CarbonEvents-based app. (This may be an incentive for me to convert the IDE to use CarbonEvents...) > No documentation yet, but the main options are as follows (read the > source for EasyDialogs.py and Nav.c for details): > - wanted: you can pass a type here telling what you want back. FSSpec, > FSRef, str or unicode (or subtypes thereof!) are acceptable. > - message: the prompt (only ascii, for now) > - savedFileName: the default output filename (only ascii, for now) > - typeList: a list or tuple of 4-char filetype codes (or the Res.Handle > data structure Nav really uses) > - defaultLocation: which folder to show the user first (FSSpec, FSRef, > pathname and AE are acceptable) > > The return value is of type "wanted" or None if the user cancelled. > > If someone can come up with an example for a filterProc use that would > be nice. Traceback (most recent call last): File "/Users/just/code/tmp/build/Ask.app/Contents/Resources/Ask.py", line 7, in ? print EasyDialogs.AskFileForSave(message="Hello", filterProc=proc) File "/Users/just/code/python_cvs/Lib/plat-mac/EasyDialogs.py", line 651, in AskFileForSave raise Nav.error, arg Nav.error: Unknown DialogOption key: filterProc Which leads me to the next point: I think these functions should have the possible keyword args explicitly set. That way you can more or less figur it out by merely reading EasyDialogs.py... Just From Jack.Jansen@oratrix.com Fri Jan 17 23:21:09 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sat, 18 Jan 2003 00:21:09 +0100 Subject: [Pythonmac-SIG] Replacing macfs.StandardGetFile and friends. In-Reply-To: Message-ID: <5C195E1A-2A72-11D7-9533-000A27B19B96@oratrix.com> On vrijdag, jan 17, 2003, at 19:37 Europe/Amsterdam, Just van Rossum wrote: > AskFileForSave() traces back if called with no arguments: That was an easy one, a typo. Fixed. > I think the dialogs should be movable; this should work correctly in an > eventloop-less app as well in a CarbonEvents-based app. (This may be an > incentive for me to convert the IDE to use CarbonEvents...) Yes, they will be, at some point. I first want to get them working. The reason I want to get them working first is that if we have these we can get rid of FSSpecs in almost all of the core, and that is good for all sorts of things (like international filenames or long filenames). > Traceback (most recent call last): > File "/Users/just/code/tmp/build/Ask.app/Contents/Resources/Ask.py", > line 7, in ? > print EasyDialogs.AskFileForSave(message="Hello", filterProc=proc) > File "/Users/just/code/python_cvs/Lib/plat-mac/EasyDialogs.py", line > 651, in AskFileForSave > raise Nav.error, arg > Nav.error: Unknown DialogOption key: filterProc This appeared to be simple, but it uncovered another bug which crashed Python, which turned out to be missing functionality in Carbon.AE: the AEDesc records passed to the filterproc are borrowed, so Python shouldn't dispose them when disposing the Python object. There's now support for this in Carbon.AE (a C function AEDesc_NewBorrowed, and a Python method Carbon.AE.AEDesc.Autodispose(onoff) if you ever need to change this from Python code. I vaguely remember that I ran in to this problem long ago, in a completely different setting. As you've kept me busy all evening you are now morally obliged to come up with a good example:-) > Which leads me to the next point: I think these functions should have > the possible keyword args explicitly set. That way you can more or less > figur it out by merely reading EasyDialogs.py... Yes, I'll do so later. Initially I thought it would be difficult, but I can use None to mean "don't set this value". -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From aparente@adobe.com Mon Jan 20 01:04:38 2003 From: aparente@adobe.com (Alexandre Parenteau) Date: Sun, 19 Jan 2003 17:04:38 -0800 Subject: [Pythonmac-SIG] Dylib vs framework vs application Message-ID: Hi, I'm still just wondering if in case the framework install was not becoming the dominant form of distribution for mac python, why not having the python library into a dylib form, like TCL does ?! By default, and like Linux does. I tried the several configure options (--enable-shared for example), but I always end-up with a python.exe which doesn't seem to be suitable for the new NSAddImage : const struct mach_header *header = NSAddImage("/Sources/python/dist/src/darwin/python.exe", NSADDIMAGE_OPTION_RETURN_ON_ERROR); Returns the error : "dyld: /Sources/cvsgui/MacCvs/Build/Debug/maccvsX/Contents/MacOS/maccvsX malformed library: /Sources/python/dist/src/dyload/python.exe (not a Mach-O library file, bad filetype value)\n" Ref: http://developer.apple.com/techpubs/macosx/ReleaseNotes/TwoLevelNamespaces.h tml Also I was curious if dynload_next.c is in need to be updated with the new APIs from Apple. This all mach-O loader is very confusing to me overall. For those who wonder, I'm still trying over and over the same thing : embed the python into an application. Thanks, Alex. -- Alexandre Parenteau Computer Scientist Core Technologies, AGM Adobe Systems, Inc. 408-536-6166 From hackenyo@proaxis.com Thu Jan 23 18:38:30 2003 From: hackenyo@proaxis.com (Dave Hackenyos) Date: Thu, 23 Jan 2003 10:38:30 -0800 Subject: [Pythonmac-SIG] PYTHONPATH problem Message-ID: Hello, I am using the default installation of python2.2 on MACOSX 10.2 (jaguar). I have set the PYTHONPATH env variable in my shell start up script to a place where I wish to load modules. But python fails to find anything in that place. Running a test as follows: #! /usr/bin/python import sys print sys.path shows that python did not prepend the value of PYTHONPATH to the sys.path information. I am unsure what to do about this problem. I know the command line environment knows about the PYTHONPATH value as follows: (start up script) . . setenv PYTHONPATH = "~/place/where/modules/live" from the command line I echo it properly -->echo $PYTHONPATH -->Users/davehack/place/where/modules/live I run a script "tests.py" which imports a module from the PYTHONPATH path but python complains that the module does not exist. Does anyone know what I may be missing in my environment setup to keep python from using the PYTHONPATH value? Thanks for any help you could give. From Jack.Jansen@oratrix.com Thu Jan 23 21:29:43 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Thu, 23 Jan 2003 22:29:43 +0100 Subject: [Pythonmac-SIG] PYTHONPATH problem In-Reply-To: Message-ID: On donderdag, jan 23, 2003, at 19:38 Europe/Amsterdam, Dave Hackenyos wrote: > Hello, > I am using the default installation of python2.2 on MACOSX 10.2 > (jaguar). I have set the PYTHONPATH env variable in my shell start up > script to a place where I wish to load modules. But python fails to > find anything in that place. Running a test as follows: > > #! /usr/bin/python > import sys > print sys.path > > shows that python did not prepend the value of PYTHONPATH to the > sys.path information. Sounds like something went wrong with setting the PYTHONPATH variable in your startup script. Maybe you're setting it in .login (which is for csh and tcsh) but you're using bash, or the other way around? Try the following two shell commands: echo $PYTHONPATH sh -c 'echo $PYTHONPATH' (Note: single quotes, not double quotes!). If the first one fails you're not setting PYTHONPATH all all (the suggestion above, maybe?). If the first one prints the right value but the second one doesn't then you are setting the value, but you're not exporting it (in bash you need either the command "export PYTHONPATH=blabla" or the 2 commands "set PYTHONPATH=blabla" and "export PYTHONPATH". Just the "set" won't do the trick. > -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From hackenyo@proaxis.com Thu Jan 23 23:58:33 2003 From: hackenyo@proaxis.com (Dave Hackenyos) Date: Thu, 23 Jan 2003 15:58:33 -0800 Subject: [Pythonmac-SIG] PYTHONPATH problem Message-ID: <94728558-2F2E-11D7-B83F-00039344C7C0@proaxis.com> Attn: Jack Jansen Hi, Thanks for responding. I tried the tests you suggested and something else is going on. I am running the default tcsh. My home directory is "Users/davehack". I have PYTHONPATH set to: "~/trollcode/Trolls/source" in a file I call environment.mine called by .tcshrc during the shell start up. The result of: echo $PYTHONPATH is: /Users/davehack/trollcode/Trolls/source which is correct. The result of: sh -c 'echo $PYTHONPATH' is: ~/trollcode/Trolls/source which is also correct. Perhaps there is an explicit shell export command for tcsh that I am unaware of. The interpreter does not show the prepended path in sys.path nor does a script run from the command line containing: #! /usr/bin/python import sys print sys.path In any case, thank you for getting back to me. Dave From wlmyers@stellarscience.com Fri Jan 24 00:24:30 2003 From: wlmyers@stellarscience.com (Willard Myers) Date: Thu, 23 Jan 2003 19:24:30 -0500 Subject: [Pythonmac-SIG] PYTHONPATH problem In-Reply-To: <94728558-2F2E-11D7-B83F-00039344C7C0@proaxis.com> Message-ID: <348BB5A2-2F32-11D7-B5BD-0003930AD59E@stellarscience.com> On Thursday, Jan 23, 2003, at 18:58 US/Eastern, Dave Hackenyos wrote: > Attn: Jack Jansen > Hi, > Thanks for responding. I tried the tests you suggested and something > else is going on. I am running the default tcsh. My home directory is > "Users/davehack". [snip] Try defining PYTHONPATH in ~/.MacOSX/environment.plist This file contains a dictionary of strings. For example, mine looks like this: ... QMAKESPEC macx-pbuilder QTDIR /usr/local/qt ... The structure is managed for you if you use Apple's tool: /Developer/Applications/Property List Editor. Bill From just@letterror.com Fri Jan 24 00:26:21 2003 From: just@letterror.com (Just van Rossum) Date: Fri, 24 Jan 2003 01:26:21 +0100 Subject: [Pythonmac-SIG] PYTHONPATH problem In-Reply-To: <94728558-2F2E-11D7-B83F-00039344C7C0@proaxis.com> Message-ID: Dave Hackenyos wrote: > ~/trollcode/Trolls/source I think you should use an absolute path (without the ~). Check sys.path when you run python -S This disables site.py. site.py removes path items that don't exist (in Python 2.2, not in 2.3), so -S will help you see what was actually inserted in sys.path. Just From Jack.Jansen@cwi.nl Fri Jan 24 09:31:49 2003 From: Jack.Jansen@cwi.nl (Jack Jansen) Date: Fri, 24 Jan 2003 10:31:49 +0100 Subject: [Pythonmac-SIG] PYTHONPATH problem In-Reply-To: <94728558-2F2E-11D7-B83F-00039344C7C0@proaxis.com> Message-ID: On Friday, Jan 24, 2003, at 00:58 Europe/Amsterdam, Dave Hackenyos wrote: > The result of: > sh -c 'echo $PYTHONPATH' > is: > > ~/trollcode/Trolls/source > > which is also correct. For some definition of "correct": yes. For Python not, unfortunately:-) You've done something like setenv PYTHONPATH "~/trollcode/Trolls/source" which causes the ~ to be put into the environment string verbatim, in stead of being expanded immediately. If you give this string to tcsh again it will expand it at that point, and there's no problem. However, only a few shells (and a few other programs) do this expansion, most programs do not do the expansion of ~ to /Users/davehack. So, you should do setenv PYTHONPATH ~/trollcode/Trolls/source to have the tilde expanded there-and-then, and everything should be fine. And if you follow Willard Myers' suggestion of setting the variable in environment.plist in stead of in your shell startup file (this has some advantages: programs that haven't been started via the shell but directly from the finder will also see the environment variable): you should beware that *no* ~ expansion is done in environment.plist, so for PYTHONPATH you should use the full /Users/davehack. > -- Jack Jansen, , http://www.cwi.nl/~jack If I can't dance I don't want to be part of your revolution -- Emma Goldman From hackenyo@proaxis.com Fri Jan 24 17:28:31 2003 From: hackenyo@proaxis.com (Dave Hackenyos) Date: Fri, 24 Jan 2003 09:28:31 -0800 Subject: [Pythonmac-SIG] PYTHONPATH problem Message-ID: <422A8C8E-2FC1-11D7-B83F-00039344C7C0@proaxis.com> Hello all, Thanks to everyone who offered suggestions for my PYTHONPATH problem. It has been resolved. In the end it had more to do with pilot error than stubborn operating systems. Firstly, the tilde was indeed not being properly handled. I need to give the environment variable the complete path. By use of "python -S" I was able to verify that the path was being prepended to the sys.path info. But it still did not work until I realized that the path I had used to set the variable did not have a leading forward slash. This was my (embarrasing) fault. I believe that adding the path as a key/value pair in the plist would also have worked provided I gave it the correct string. But I don't need to try that now. Thanks again, Dave From robin@alldunn.com Fri Jan 24 18:17:20 2003 From: robin@alldunn.com (Robin Dunn) Date: Fri, 24 Jan 2003 10:17:20 -0800 Subject: [Pythonmac-SIG] MacPython-0SX-2.2 Message-ID: <3E318330.5010803@alldunn.com> I noticed some discussion a couple weeks ago of the problems with building MacPython-OSX-2.2 using Python-2.3a1. Has any work gone into CVS yet to fix this? -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Fri Jan 24 18:39:41 2003 From: robin@alldunn.com (Robin Dunn) Date: Fri, 24 Jan 2003 10:39:41 -0800 Subject: [Pythonmac-SIG] unix link to framework python Message-ID: <3E31886D.5000301@alldunn.com> In the 2.3 framework build /usr/local/bin/python is linked to the python inside the framework. I think it would be better if the link was "python2.3" and then there could be a separate link from "python" to "python2.3" as well. This would allow command line (or #!) users to specify either "python2.2" or "python2.3" to get the one they want. I expect that it would also allow installation of the unix links/scripts into /usr/bin... -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From diordna@silvernetwork.net Sat Jan 25 14:10:00 2003 From: diordna@silvernetwork.net (Stephen Johnson) Date: Sat, 25 Jan 2003 09:10:00 -0500 Subject: [Pythonmac-SIG] Okay, I'll try this again... Message-ID: No one bothered to answer last time. Does anyone know of a GUI that works with MacPython carbon? From francois.granger@free.fr Sat Jan 25 18:33:30 2003 From: francois.granger@free.fr (=?iso-8859-1?Q?Fran=E7ois?= Granger) Date: Sat, 25 Jan 2003 19:33:30 +0100 Subject: [Pythonmac-SIG] Okay, I'll try this again... In-Reply-To: References: Message-ID: At 09:10 -0500 25/01/2003, in message [Pythonmac-SIG] Okay, I'll try this again..., Stephen Johnson wrote: >No one bothered to answer last time. > >Does anyone know of a GUI that works with MacPython carbon? I am afraid that the answer is not really good. I am waiting for a stable Match-o-python on OSX to switch to vxPython. -- Recently using MacOSX....... From Jack.Jansen@oratrix.com Sat Jan 25 22:43:59 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sat, 25 Jan 2003 23:43:59 +0100 Subject: [Pythonmac-SIG] MacPython-0SX-2.2 In-Reply-To: <3E318330.5010803@alldunn.com> Message-ID: <7E57580C-30B6-11D7-8C5A-000A27B19B96@oratrix.com> On vrijdag, jan 24, 2003, at 19:17 Europe/Amsterdam, Robin Dunn wrote: > I noticed some discussion a couple weeks ago of the problems with > building MacPython-OSX-2.2 using Python-2.3a1. Has any work gone into > CVS yet to fix this? Not yet. I bascially ripped the support out, after I utterly broke it when I moved most of Mac/Lib to Lib/plat-mac for 2.3a1. It will come back, but I don't know yet when. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From Jack.Jansen@oratrix.com Sat Jan 25 22:44:32 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sat, 25 Jan 2003 23:44:32 +0100 Subject: [Pythonmac-SIG] unix link to framework python In-Reply-To: <3E31886D.5000301@alldunn.com> Message-ID: <9252060E-30B6-11D7-8C5A-000A27B19B96@oratrix.com> On vrijdag, jan 24, 2003, at 19:39 Europe/Amsterdam, Robin Dunn wrote: > In the 2.3 framework build /usr/local/bin/python is linked to the > python inside the framework. I think it would be better if the link > was "python2.3" and then there could be a separate link from "python" > to "python2.3" as well. This would allow command line (or #!) users > to specify either "python2.2" or "python2.3" to get the one they want. > I expect that it would also allow installation of the unix > links/scripts into /usr/bin... Good idea. Could you file a sourceforge bug report, please? -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From Jack.Jansen@oratrix.com Sat Jan 25 22:47:30 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Sat, 25 Jan 2003 23:47:30 +0100 Subject: [Pythonmac-SIG] Okay, I'll try this again... In-Reply-To: Message-ID: On zaterdag, jan 25, 2003, at 15:10 Europe/Amsterdam, Stephen Johnson wrote: > No one bothered to answer last time. > > Does anyone know of a GUI that works with MacPython carbon? There isn't much, really. The best bet is native Carbon (Win, Dlg, Evt, etc), possibly through the Framework or W wrappers. wxPython is rumoured to work, but I have no idea as to the status. Unless you really still have to support MacOS9 machines I would suggest moving to MacPython-OSX. There you have all of the above, PyObjC, wxPython, Tkinter and probably more. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From robin@alldunn.com Mon Jan 27 21:35:36 2003 From: robin@alldunn.com (Robin Dunn) Date: Mon, 27 Jan 2003 13:35:36 -0800 Subject: [Pythonmac-SIG] unix link to framework python In-Reply-To: <9252060E-30B6-11D7-8C5A-000A27B19B96@oratrix.com> References: <9252060E-30B6-11D7-8C5A-000A27B19B96@oratrix.com> Message-ID: <3E35A628.9070906@alldunn.com> Jack Jansen wrote: > > On vrijdag, jan 24, 2003, at 19:39 Europe/Amsterdam, Robin Dunn wrote: > >> In the 2.3 framework build /usr/local/bin/python is linked to the >> python inside the framework. I think it would be better if the link >> was "python2.3" and then there could be a separate link from "python" >> to "python2.3" as well. This would allow command line (or #!) users >> to specify either "python2.2" or "python2.3" to get the one they want. >> I expect that it would also allow installation of the unix >> links/scripts into /usr/bin... > > > Good idea. Could you file a sourceforge bug report, please? Done, #675745 -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Mon Jan 27 21:48:17 2003 From: robin@alldunn.com (Robin Dunn) Date: Mon, 27 Jan 2003 13:48:17 -0800 Subject: [Pythonmac-SIG] unix link to framework python In-Reply-To: <3E35722C.320AA006@noaa.gov> References: <3E31886D.5000301@alldunn.com> <3E35722C.320AA006@noaa.gov> Message-ID: <3E35A921.3080306@alldunn.com> Chris Barker wrote: > Robin Dunn wrote: > >>I think it would be better if the link was >>"python2.3" and then there could be a separate link from "python" to >>"python2.3" as well. This would allow command line (or #!) users to >>specify either "python2.2" or "python2.3" to get the one they want. > > > Here! here! (hear! hear! ?) > > > This is critical. On Linux I have at least three versions of Python > installed at once, and I now have two on OS-X and will be adding more. I > can't experiment with 2.3 unless it is set up this way (I know I can do > it by hand, and probably will) I've made an Installer package for MaxPython-OSX 2.3a1 that has a postinstall script that sets up the links as I described above. You can get it at the wxPython Files page at SF: https://sourceforge.net/project/showfiles.php?group_id=10718 (look near the bottom.) BTW notice the huge file size difference between the 2.3a1 build and my previous 2.2.1 build! That's the result of not including the .pyc/.pyo files and then regenerating them in the postinstall. I havn't yet tried including the Python docs, but will when I get some more time. Right now I'm working on an Installer package for wxPython that works with the Ptyhon 2.3a1 above and also installs app bundles for running the demo, PyShell, and etc. So far so good, I just need to look into how to create the resources in the bundle for setting the Icon... > > We are still missing a step however. If I have it right, you can tell > finder that you want a *.py or *.pyw file opened with (*and run by) the > python interpreter. What would be great is if that file go opened be a > samll utility that looked at the #! line and opened it with the correct > interpreter (I suppose that utility could be a simple shell script, come > to think of it). That way, one could have programs written for various > versions of Python all on one machine, and be able to run them easily > from the GUI. I suppose BuildApplet (if it exists for OS-X) can do the > same thing, but I kind of like the cross-platform elegance of the #! > line. The Windows folks could do the same thing as well... It's not quite the same but if you set the properties such that are using the PythonLauncher from 2.3 then if you hold down the Alt key while double clicking a .py or .pyw file in Finder then you will get an opportunity to specify which python is executed. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From Jack.Jansen@oratrix.com Tue Jan 28 12:21:32 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Tue, 28 Jan 2003 13:21:32 +0100 Subject: [Pythonmac-SIG] unix link to framework python In-Reply-To: <3E35A921.3080306@alldunn.com> Message-ID: <091FD884-32BB-11D7-8301-000A27B19B96@oratrix.com> >> We are still missing a step however. If I have it right, you can tell >> finder that you want a *.py or *.pyw file opened with (*and run by) >> the >> python interpreter. What would be great is if that file go opened be a >> samll utility that looked at the #! line and opened it with the >> correct >> interpreter (I suppose that utility could be a simple shell script, >> come >> to think of it). That way, one could have programs written for various >> versions of Python all on one machine, and be able to run them easily >> from the GUI. I suppose BuildApplet (if it exists for OS-X) can do the >> same thing, but I kind of like the cross-platform elegance of the #! >> line. The Windows folks could do the same thing as well... > > It's not quite the same but if you set the properties such that are > using the PythonLauncher from 2.3 then if you hold down the Alt key > while double clicking a .py or .pyw file in Finder then you will get > an opportunity to specify which python is executed. I missed the original post (or was it to another mailing list?), but I think that a PythonLauncher preference "Honour #! lines in scripts" is a good idea. If you file a feature request on sourceforge I won't forget about it. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From ryanwilcox@mac.com Tue Jan 28 21:29:36 2003 From: ryanwilcox@mac.com (Ryan Wilcox) Date: Tue, 28 Jan 2003 16:29:36 -0500 Subject: [Pythonmac-SIG] Suite Module For CW8? In-Reply-To: <091FD884-32BB-11D7-8301-000A27B19B96@oratrix.com> Message-ID: Greetings, List, Is there a current CodeWarrior (ie: CW8) script-package in Python 2.3? If not, is there any possibility of adding one? (If it's a matter of testing scripts, point me to the correct ones and I'll see what happens.) Thanks!, -Ryan Waiting for 2.3 to get out of alpha before trying it... --------------------------------- Wilcox Design: Understanding Data From Jack.Jansen@oratrix.com Tue Jan 28 23:19:06 2003 From: Jack.Jansen@oratrix.com (Jack Jansen) Date: Wed, 29 Jan 2003 00:19:06 +0100 Subject: [Pythonmac-SIG] Suite Module For CW8? In-Reply-To: Message-ID: On dinsdag, jan 28, 2003, at 22:29 Europe/Amsterdam, Ryan Wilcox wrote: > Greetings, List, > > Is there a current CodeWarrior (ie: CW8) script-package in Python 2.3? > If not, is there any possibility of adding one? You can make it yourself (if you have CW8, but I assume you do). Run gensuitemodule.py (it's in Mac/scripts or :Mac:scripts) on your CW8 IDE. -- - Jack Jansen http://www.cwi.nl/~jack - - If I can't dance I don't want to be part of your revolution -- Emma Goldman - From robin@alldunn.com Wed Jan 29 00:26:00 2003 From: robin@alldunn.com (Robin Dunn) Date: Tue, 28 Jan 2003 16:26:00 -0800 Subject: [Pythonmac-SIG] unix link to framework python In-Reply-To: <091FD884-32BB-11D7-8301-000A27B19B96@oratrix.com> References: <091FD884-32BB-11D7-8301-000A27B19B96@oratrix.com> Message-ID: <3E371F98.2010308@alldunn.com> Jack Jansen wrote: > >>> We are still missing a step however. If I have it right, you can tell >>> finder that you want a *.py or *.pyw file opened with (*and run by) the >>> python interpreter. What would be great is if that file go opened be a >>> samll utility that looked at the #! line and opened it with the correct >>> interpreter (I suppose that utility could be a simple shell script, come >>> to think of it). That way, one could have programs written for various >>> versions of Python all on one machine, and be able to run them easily >>> from the GUI. I suppose BuildApplet (if it exists for OS-X) can do the >>> same thing, but I kind of like the cross-platform elegance of the #! >>> line. The Windows folks could do the same thing as well... >> >> >> It's not quite the same but if you set the properties such that are >> using the PythonLauncher from 2.3 then if you hold down the Alt key >> while double clicking a .py or .pyw file in Finder then you will get >> an opportunity to specify which python is executed. > > > I missed the original post (or was it to another mailing list?), but I > think that a PythonLauncher preference "Honour #! lines in scripts" is a > good idea. If you file a feature request on sourceforge I won't forget > about it. Done. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Wed Jan 29 05:15:51 2003 From: robin@alldunn.com (Robin Dunn) Date: Tue, 28 Jan 2003 21:15:51 -0800 Subject: [Pythonmac-SIG] wxPythonOSX for Python 2.3a1 Message-ID: <3E376387.5050104@alldunn.com> I've uploaded an evaluation release of wxPythonOSX 2.4.02 for MacPython-OSX 2.3a1. You can get it from the bottom of the SF wxPython files page: http://sf.net/project/showfiles.php?group_id=10718 In addition to building for Python 2.3a1 I've greatly changed how the distribution itself is made and now the demo, PyShell, (and others in the future) are now installed as their own App bundles using the new bundlebuilder.py tool. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Wed Jan 29 05:39:47 2003 From: robin@alldunn.com (Robin Dunn) Date: Tue, 28 Jan 2003 21:39:47 -0800 Subject: [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Python 2.3a1 In-Reply-To: <3E376387.5050104@alldunn.com> References: <3E376387.5050104@alldunn.com> Message-ID: <3E376923.2050400@alldunn.com> Robin Dunn wrote: > I've uploaded an evaluation release of wxPythonOSX 2.4.02 for > MacPython-OSX 2.3a1. You can get it from the bottom of the SF wxPython > files page: http://sf.net/project/showfiles.php?group_id=10718 > > In addition to building for Python 2.3a1 I've greatly changed how the > distribution itself is made and now the demo, PyShell, (and others in > the future) are now installed as their own App bundles using the new > bundlebuilder.py tool. > Oh, yeah, one more thing. If you havn't seen it already I have made an installer for MacPython-OSX 2.3a1 available too. It's on the same page in the group just below the wxPythonOSX link. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From ryanwilcox@mac.com Wed Jan 29 06:02:07 2003 From: ryanwilcox@mac.com (Ryan Wilcox) Date: Wed, 29 Jan 2003 01:02:07 -0500 Subject: [Pythonmac-SIG] Suite Module For CW8? In-Reply-To: Message-ID: On Wednesday, January 29, 2003, at 12:19 AM, Jack Jansen wrote: >> Greetings, List, >> >> Is there a current CodeWarrior (ie: CW8) script-package in Python >> 2.3? If not, is there any possibility of adding one? > >You can make it yourself (if you have CW8, but I assume you do). Run >gensuitemodule.py >(it's in Mac/scripts or :Mac:scripts) on your CW8 IDE. Already did that (forgot to mention that, somehow), but was curious about its status in the official distribution. Later, -Ryan --------------------------------- Wilcox Design: Understanding Data From Jack.Jansen@cwi.nl Wed Jan 29 09:26:44 2003 From: Jack.Jansen@cwi.nl (Jack Jansen) Date: Wed, 29 Jan 2003 10:26:44 +0100 Subject: [Pythonmac-SIG] Suite Module For CW8? In-Reply-To: Message-ID: On Wednesday, Jan 29, 2003, at 07:02 Europe/Amsterdam, Ryan Wilcox wrote: >>> Is there a current CodeWarrior (ie: CW8) script-package in Python >>> 2.3? If not, is there any possibility of adding one? >> >> You can make it yourself (if you have CW8, but I assume you do). Run >> gensuitemodule.py >> (it's in Mac/scripts or :Mac:scripts) on your CW8 IDE. > > Already did that (forgot to mention that, somehow), but was curious > about its status in the official distribution. I don't have CW8, I decided that the upgrade from 7 wasn't worth it, so I don't think it'll be in the distribution. Unless someone donates it, of course. (And, even so, the added functionality has to be worth the extra effort, plus the hassle that I won't be able to regenerate the module when something else changes) -- Jack Jansen, , http://www.cwi.nl/~jack If I can't dance I don't want to be part of your revolution -- Emma Goldman From boyle5@llnl.gov Wed Jan 29 16:36:40 2003 From: boyle5@llnl.gov (boyle5@llnl.gov) Date: Wed, 29 Jan 2003 08:36:40 -0800 Subject: [Pythonmac-SIG] MachoPython2.3 and wxpython2.4.0.2 Message-ID: I have a framework build of python2.3.1a on my OS X 10.2.3 box and would like to install wxPython2.4.0.2. From my efforts so far it appears that the installer wxPythonOSX-2.4.0.2 at sourceforge wants to install with python2.2(which I also have). Can I coerce the installer to use python2.3? (my knowledge of frameworks is quite meager). Or should I wait for a new installer? Thanks, Jim From francois.granger@free.fr Wed Jan 29 17:56:57 2003 From: francois.granger@free.fr (Francois Granger) Date: Wed, 29 Jan 2003 18:56:57 +0100 Subject: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1 In-Reply-To: <3E376923.2050400@alldunn.com> References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> Message-ID: At 21:39 -0800 28/01/2003, in message [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Py, Robin Dunn wrote: >Robin Dunn wrote: >>I've uploaded an evaluation release of wxPythonOSX 2.4.02 for >>MacPython-OSX 2.3a1. You can get it from the bottom of the SF >>wxPython files page: >>http://sf.net/project/showfiles.php?group_id=10718 >> >>In addition to building for Python 2.3a1 I've greatly changed how >>the distribution itself is made and now the demo, PyShell, (and >>others in the future) are now installed as their own App bundles >>using the new bundlebuilder.py tool. >> > >Oh, yeah, one more thing. If you havn't seen it already I have made >an installer for MacPython-OSX 2.3a1 available too. It's on the >same page in the group just below the wxPythonOSX link. Thanks a lot for these packages. One remark, the package MacPython-OSX-2.3a1-2.dmg does not work if it is installed on another disk than the one with the system. The links in /usr/local/bin are not created, neither pythonw. -- Recently using MacOSX....... From ranch1@earthlink.net Thu Jan 30 07:38:29 2003 From: ranch1@earthlink.net (Tom) Date: Thu, 30 Jan 2003 02:38:29 -0500 Subject: [Pythonmac-SIG] getting 2.3 working with Tkinter Message-ID: I'm trying to get 2.3 and tkinter working (now that a binary showed up on sourceforge). Installed TclTkAquaBI8.4.1 Trying to work through all the unix stuff to get this working. get this: [Toms-Computer:/usr/local/bin] tom% pythonw /Volumes/80\ Gig\ Stuff/Download_IE/leo-3.10/leo.py Traceback (most recent call last): File "/Volumes/80 Gig Stuff/Download_IE/leo-3.10/leo.py", line 35, in ? from leoGlobals import * File "/Volumes/80 Gig Stuff/Download_IE/leo-3.10/leoGlobals.py", line 8, in ? import exceptions,os,re,string,sys,time,types,Tkinter File "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/lib- tk/Tkinter.py", line 38, in ? import _tkinter # If this fails your Python may not be configured for Tk ImportError: No module named _tkinter Any Ideas? Tom From francois.granger@free.fr Thu Jan 30 09:04:43 2003 From: francois.granger@free.fr (Francois Granger) Date: Thu, 30 Jan 2003 10:04:43 +0100 Subject: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1 In-Reply-To: References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> Message-ID: At 18:56 +0100 29/01/2003, in message [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1, Francois Granger wrote: >At 21:39 -0800 28/01/2003, in message [Pythonmac-SIG] Re: >[wxPython-users] wxPythonOSX for Py, Robin Dunn wrote: >>Robin Dunn wrote: >>>I've uploaded an evaluation release of wxPythonOSX 2.4.02 for >>>MacPython-OSX 2.3a1. You can get it from the bottom of the SF >>>wxPython files page: >>>http://sf.net/project/showfiles.php?group_id=10718 >> >>Oh, yeah, one more thing. If you havn't seen it already I have >>made an installer for MacPython-OSX 2.3a1 available too. It's on >>the same page in the group just below the wxPythonOSX link. > >Thanks a lot for these packages. > >One remark, the package MacPython-OSX-2.3a1-2.dmg does not work if >it is installed on another disk than the one with the system. The >links in /usr/local/bin are not created, neither pythonw. Another remark: On an existing script, working with the 2.2 (pop3proxy.py from Spambayes), I get the following error: File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 16, in open_dbhash import bsddb File "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py", line 40, in ? import _bsddb ImportError: Failure linking new module: /sw/lib/libdb-3.3.dylib: dyld: /usr/local/bin/python can't open library: /sw/lib/libdb-3.3.dylib (No such file or directory, errno = 2) I have fink installed. But I never downloaded the fink version of Python. Is it needed for your distribution to work ? -- Recently using MacOSX....... From francois.granger@free.fr Thu Jan 30 15:34:43 2003 From: francois.granger@free.fr (Francois Granger) Date: Thu, 30 Jan 2003 16:34:43 +0100 Subject: [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Python 2.3a1 In-Reply-To: <3E376923.2050400@alldunn.com> References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> Message-ID: At 21:39 -0800 28/01/2003, in message [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Py, Robin Dunn wrote: >Robin Dunn wrote: >>I've uploaded an evaluation release of wxPythonOSX 2.4.02 for >>MacPython-OSX 2.3a1. You can get it from the bottom of the SF >>wxPython files page: >>http://sf.net/project/showfiles.php?group_id=10718 >> >>In addition to building for Python 2.3a1 I've greatly changed how >>the distribution itself is made and now the demo, PyShell, (and >>others in the future) are now installed as their own App bundles >>using the new bundlebuilder.py tool. >> > >Oh, yeah, one more thing. If you havn't seen it already I have made >an installer for MacPython-OSX 2.3a1 available too. It's on the >same page in the group just below the wxPythonOSX link. MacG4 400 640MB memory. MacOS X 10.2.3 Installed Python 2.3a1 Installed wxPythonOSX 2.4.02 Run wxPythonOSX-2.4.0.2/RunDemo by double clicking on it successfully. Installed PythonCardPrototype-0.7 with the command line: python setup.py install I got an error message at end about permissions: error: /Library/Frameworks/Python.framework/Versions/2.3/bin/install-pythoncard.py: Permission denied I did : sudo python setup.py install No error message. When I do "Openwith" -> "PythonLauncher" for codeeditor.py in the Finder, I get: [fbg:~] fgranger% "/usr/local/bin/python" "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/PythonCardPrototype/components/codeeditor.py" && echo Exit status: $? && exit 1 16:28:35: Debug: wxMac library installation name is '/usr/local/lib/libwx_macd-2.4.0.dylib' 16:28:35: Debug: wxMac resources file name is '/usr/local/lib/libwx_macd-2.4.0.rsrc' Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/PythonCardPrototype/components/codeeditor.py", line 11, in ? from wxPython import stc, wx File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/stc.py", line 7, in ? from stc_ import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/stc_.py", line 6, in ? from misc2 import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/misc2.py", line 4, in ? from windows import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/windows.py", line 6, in ? from gdi import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/gdi.py", line 7, in ? import wx File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/wx.py", line 22, in ? from mdi import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/mdi.py", line 14, in ? from frames import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/frames.py", line 14, in ? from stattool import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/stattool.py", line 14, in ? from controls import * File "//Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/wxPython/controls.py", line 16, in ? class wxControlPtr(wxWindowPtr): NameError: name 'wxWindowPtr' is not defined [fbg:~] fgranger% If you need more info..... -- Recently using MacOSX....... From robin@alldunn.com Thu Jan 30 16:19:05 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 08:19:05 -0800 Subject: [wxPython-mac] [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Python 2.3a1 In-Reply-To: References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> Message-ID: <3E395079.7060306@alldunn.com> Francois Granger wrote: > At 21:39 -0800 28/01/2003, in message [Pythonmac-SIG] Re: > > I got an error message at end about permissions: > > error: > /Library/Frameworks/Python.framework/Versions/2.3/bin/install-pythoncard.py: > Permission denied Hmmm... I made the site-packages writeable by admin but didn't think anywhere else would be necessary. I'll change that. > > No error message. > > When I do "Openwith" -> "PythonLauncher" for codeeditor.py in the > Finder, I get: > > [fbg:~] fgranger% "/usr/local/bin/python" BTW, you probably want to either change that to a .pyw file extension, or change PythonLauncher to use /usr/local/bin/pythonw so that the codeeditor can interact with the GUI properly. > "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/PythonCardPrototype/components/codeeditor.py" > && echo Exit status: $? && exit 1 > 16:28:35: Debug: wxMac library installation name is > '/usr/local/lib/libwx_macd-2.4.0.dylib' > 16:28:35: Debug: wxMac resources file name is > '/usr/local/lib/libwx_macd-2.4.0.rsrc' > Traceback (most recent call last): > File > "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/PythonCardPrototype/components/codeeditor.py", > line 11, in ? > from wxPython import stc, wx Try changing the order of these imports. If this is the first time that wx is imported then importing other wx modules before it could cause a problem like this. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Thu Jan 30 16:23:12 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 08:23:12 -0800 Subject: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1 In-Reply-To: References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> Message-ID: <3E395170.1030209@alldunn.com> Francois Granger wrote: > > Another remark: > On an existing script, working with the 2.2 (pop3proxy.py from > Spambayes), I get the following error: > > File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 16, in > open_dbhash > import bsddb > File > "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py", > line 40, in ? > import _bsddb > ImportError: Failure linking new module: /sw/lib/libdb-3.3.dylib: dyld: > /usr/local/bin/python can't open library: /sw/lib/libdb-3.3.dylib (No > such file or directory, errno = 2) > > I have fink installed. But I never downloaded the fink version of Python. > > Is it needed for your distribution to work ? > No, the problem is that Python's build only looks in places like /usr/include and /usr/local/include for the dh.h it need to build the extension module in the bsddb package. I'll work that out and make a new build. Has anyone noticed any other "standard" extension modules that are missing? -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Thu Jan 30 16:25:35 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 08:25:35 -0800 Subject: [Pythonmac-SIG] MachoPython2.3 and wxpython2.4.0.2 In-Reply-To: References: Message-ID: <3E3951FF.3080901@alldunn.com> boyle5@llnl.gov wrote: > I have a framework build of python2.3.1a on my OS X 10.2.3 box and > would like to install wxPython2.4.0.2. > From my efforts so far it appears that the installer > wxPythonOSX-2.4.0.2 at sourceforge wants to install with python2.2(which I > also have). > Can I coerce the installer to use python2.3? (my knowledge of frameworks > is quite meager). Or should I wait for a new > installer? > There is already a new installer there. Look at the bottom of the files page in the "Builds for Python 2.3a1" section. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Thu Jan 30 16:29:16 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 08:29:16 -0800 Subject: [Pythonmac-SIG] getting 2.3 working with Tkinter In-Reply-To: References: Message-ID: <3E3952DC.9080404@alldunn.com> Tom wrote: > I'm trying to get 2.3 and tkinter working (now that a binary showed up > on sourceforge). > > Installed TclTkAquaBI8.4.1 > > Trying to work through all the unix stuff to get this working. > > get this: > [Toms-Computer:/usr/local/bin] tom% pythonw /Volumes/80\ Gig\ > Stuff/Download_IE/leo-3.10/leo.py > Traceback (most recent call last): > File "/Volumes/80 Gig Stuff/Download_IE/leo-3.10/leo.py", line 35, in ? > from leoGlobals import * > File "/Volumes/80 Gig Stuff/Download_IE/leo-3.10/leoGlobals.py", line > 8, in ? > import exceptions,os,re,string,sys,time,types,Tkinter > File > "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/lib- > tk/Tkinter.py", line 38, in ? > import _tkinter # If this fails your Python may not be configured > for Tk > ImportError: No module named _tkinter > If you're talking about my MacPython binary (in the wxPython files area at SF) then it's simply because I don't have TclTk installed and so when I built Python it skipped the _tkinter module. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From ranch1@earthlink.net Thu Jan 30 17:00:50 2003 From: ranch1@earthlink.net (Tom) Date: Thu, 30 Jan 2003 12:00:50 -0500 Subject: [Pythonmac-SIG] getting 2.3 working with Tkinter Message-ID: <62712062-3474-11D7-AE85-003065DFF1FA@earthlink.net> > From: Tom > Date: Thu Jan 30, 2003 11:46:40 AM US/Eastern > To: Robin Dunn > Subject: Re: [Pythonmac-SIG] getting 2.3 working with Tkinter > > > On Thursday, January 30, 2003, at 11:29 AM, Robin Dunn wrote: > >> >> If you're talking about my MacPython binary (in the wxPython files >> area at SF) then it's simply because I don't have TclTk installed and >> so when I built Python it skipped the _tkinter module. >> > > Darn, I thought that might be the case (given that the binary was on > an alternative GUI sit.e). > > I don't suppose I can just drop in a few modules, I sue would like to > run leo, X11 is too slow on this G4 400, and I don't seem to be smart > enough to get things working if I try to compile things myself (plus > source download over dailup is insane). > > Tom > From francois.granger@free.fr Thu Jan 30 17:30:31 2003 From: francois.granger@free.fr (Francois Granger) Date: Thu, 30 Jan 2003 18:30:31 +0100 Subject: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1 In-Reply-To: <3E395170.1030209@alldunn.com> References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> <3E395170.1030209@alldunn.com> Message-ID: At 08:23 -0800 30/01/2003, in message Re: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1, Robin Dunn wrote: > >No, the problem is that Python's build only looks in places like >/usr/include and /usr/local/include for the dh.h it need to build >the extension module in the bsddb package. I'll work that out and >make a new build. [fbg:~] fgranger% locate dh.h /usr/include/openssl/dh.h Is it this one ? -- Recently using MacOSX....... From francois.granger@free.fr Thu Jan 30 18:05:59 2003 From: francois.granger@free.fr (Francois Granger) Date: Thu, 30 Jan 2003 19:05:59 +0100 Subject: [wxPython-mac] [Pythonmac-SIG] Re: [wxPython-users] wxPythonOSX for Python 2.3a1 In-Reply-To: <3E395079.7060306@alldunn.com> References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> <3E395079.7060306@alldunn.com> Message-ID: At 08:19 -0800 30/01/2003, in message Re: [wxPython-mac] [Pythonmac-SIG] Re: [wxPython-users], Robin Dunn wrote: >Francois Granger wrote: >> >>When I do "Openwith" -> "PythonLauncher" for codeeditor.py in the >>Finder, I get: >> >>[fbg:~] fgranger% "/usr/local/bin/python" > >BTW, you probably want to either change that to a .pyw file extension It seems to launch twice, no terminal window open. >, or change PythonLauncher to use /usr/local/bin/pythonw it seems to launch twice then nothing. The Terminal says: Last login: Thu Jan 30 18:48:21 on ttyp4 "/usr/local/bin/pythonw" -d "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-pWelcome to Darwin! ackages/PythonCardPrototype/components/codeeditor.py" && echo Exit status: $? && exit 1 [fbg:~] fgranger% "/usr/local/bin/pythonw" -d "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/PythonCardPrototype/components/codeeditor.py" && echo Exit status: $? && exit 1 18:52:01: Debug: wxMac library installation name is '/usr/local/lib/libwx_macd-2.4.0.dylib' [Process completed]Mac resources file name is '/usr/local/lib/libwx_macd-2.4.0.rsrc' I guess that I should reset my association prefs, because when I do an OpenWith with now, it give me a full menu of all apps. But I don't know wich file to trash. -- Recently using MacOSX....... From robin@alldunn.com Thu Jan 30 18:57:33 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 10:57:33 -0800 Subject: [Pythonmac-SIG] Re: wxPythonOSX for Python 2.3a1 In-Reply-To: <3E395170.1030209@alldunn.com> References: <3E376387.5050104@alldunn.com> <3E376923.2050400@alldunn.com> <3E395170.1030209@alldunn.com> Message-ID: <3E39759D.5030004@alldunn.com> Robin Dunn wrote: > Francois Granger wrote: > >> >> Another remark: >> On an existing script, working with the 2.2 (pop3proxy.py from >> Spambayes), I get the following error: >> >> File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 16, >> in open_dbhash >> import bsddb >> File >> "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py", >> line 40, in ? >> import _bsddb >> ImportError: Failure linking new module: /sw/lib/libdb-3.3.dylib: >> dyld: /usr/local/bin/python can't open library: >> /sw/lib/libdb-3.3.dylib (No such file or directory, errno = 2) >> >> I have fink installed. But I never downloaded the fink version of Python. >> >> Is it needed for your distribution to work ? >> > > No, the problem is that Python's build only looks in places like > /usr/include and /usr/local/include for the dh.h it need to build the > extension module in the bsddb package. I'll work that out and make a > new build. Oops, I didn't look closly enough at that traceback. I'll change things so it static links with libdb... -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Fri Jan 31 01:01:27 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 17:01:27 -0800 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary Message-ID: <3E39CAE7.2010605@alldunn.com> Ok, I've just uploaded a new MacPython binary to https://sourceforge.net/project/showfiles.php?group_id=10718 This one should fix the link problem with the _bsddb module, and also includes the (ug!) tkinter extension module. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Fri Jan 31 04:18:31 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 20:18:31 -0800 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E39CAE7.2010605@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> Message-ID: <3E39F917.1010903@alldunn.com> Robin Dunn wrote: > Ok, I've just uploaded a new MacPython binary to > https://sourceforge.net/project/showfiles.php?group_id=10718 This one > should fix the link problem with the _bsddb module, and also includes > the (ug!) tkinter extension module. > Kevin Altis reported that he is not able to install this binary because it is preventing him from putting it on his only disk. I added a RootVolumeOnly directive to the package info file since somebody mentioned troubles putting it on other volumes. It works fine for me, my second cvolume is diabled and I am only able to unstall on the root. Does anybody else see this problem? Does anybody know a way to fix it besides removing the RootVolumeOnly? -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Fri Jan 31 04:55:21 2003 From: robin@alldunn.com (Robin Dunn) Date: Thu, 30 Jan 2003 20:55:21 -0800 Subject: [wxPython-mac] Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E39F917.1010903@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> Message-ID: <3E3A01B9.4060200@alldunn.com> Robin Dunn wrote: > Robin Dunn wrote: > >> Ok, I've just uploaded a new MacPython binary to >> https://sourceforge.net/project/showfiles.php?group_id=10718 This one >> should fix the link problem with the _bsddb module, and also includes >> the (ug!) tkinter extension module. >> > > Kevin Altis reported that he is not able to install this binary because > it is preventing him from putting it on his only disk. I added a > RootVolumeOnly directive to the package info file since somebody > mentioned troubles putting it on other volumes. It works fine for me, > my second cvolume is diabled and I am only able to unstall on the root. Sheesh! One of these days you'd think I would lear to type! That should be "second volume" and "install on the root" -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From altis@semi-retired.com Fri Jan 31 05:01:52 2003 From: altis@semi-retired.com (Kevin Altis) Date: Thu, 30 Jan 2003 21:01:52 -0800 Subject: [wxPython-mac] Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E3A01B9.4060200@alldunn.com> Message-ID: > From: Robin Dunn > > > > Kevin Altis reported that he is not able to install this binary because > > it is preventing him from putting it on his only disk. I added a > > RootVolumeOnly directive to the package info file since somebody > > mentioned troubles putting it on other volumes. It works fine for me, > > my second cvolume is diabled and I am only able to unstall on the root. > > Sheesh! One of these days you'd think I would lear to type! > > That should be "second volume" and "install on the root" Well these things happen when you are mentally "diabled". ;-) ka From francois.granger@free.fr Fri Jan 31 12:22:16 2003 From: francois.granger@free.fr (Francois Granger) Date: Fri, 31 Jan 2003 13:22:16 +0100 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E39F917.1010903@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> Message-ID: At 20:18 -0800 30/01/2003, in message Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary, Robin Dunn wrote: > >RootVolumeOnly directive to the package info file since somebody >mentioned troubles putting it on other volumes. The directive works. It forbid install on anything else that system volume. It is me. I said in a previous message: >One remark, the package MacPython-OSX-2.3a1-2.dmg does not work if it >is installed on another disk than the one with the system. The links >in /usr/local/bin are not created, neither pythonw. >Does anybody else see this problem? Does anybody know a way to fix >it besides removing the RootVolumeOnly? Downloaded and installed. It did an update. There is a new IDLE program in the Application/MacPython-2.3 folder. I doubleclicked it and it launch and quit. I tried to run the script wich gave me this problem, pop3proxy.py from the spambayes project. Error below. I retested with a smaller script. the line: dbm = bsddb.hashopen('/Volumes/OS99/spambayesdev/hammie.db', 'r') give the same error. I don't know who is concerned with this. [fbg:/Volumes/OS99/spambayesdev] fgranger% which python /usr/local/bin/python [fbg:/Volumes/OS99/spambayesdev] fgranger% python -v pop3proxy.py [...] import bsddb # directory /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb # /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.pyc matches /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py import bsddb # precompiled from /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.pyc import _bsddb # dynamically loaded from /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/lib-dynload/_bsddb.so Traceback (most recent call last): File "pop3proxy.py", line 1557, in ? run() File "pop3proxy.py", line 1531, in run state.createWorkers() File "pop3proxy.py", line 1145, in createWorkers self.bayes = storage.DBDictClassifier( \ File "/Volumes/OS99/spambayesdev/spambayes/storage.py", line 140, in __init__ self.load() File "/Volumes/OS99/spambayesdev/spambayes/storage.py", line 148, in load self.dbm = dbmstorage.open(self.db_name, self.mode) File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 54, in open return f(*args) File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 17, in open_dbhash return bsddb.hashopen(*args) File "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py", line 162, in hashopen d.open(file, _db.DB_HASH, flags, mode) bsddb._db.DBInvalidArgError: (22, 'Invalid argument -- memp_fopen: page sizes must be a power-of-2') [...] -- Recently using MacOSX....... From marcus.h.mendenhall@vanderbilt.edu Fri Jan 31 15:26:50 2003 From: marcus.h.mendenhall@vanderbilt.edu (Marcus Mendenhall) Date: Fri, 31 Jan 2003 09:26:50 -0600 Subject: [Pythonmac-SIG] threads & sockets crash MacPython 2.2.1 In-Reply-To: <20030130170001.21573.80517.Mailman@mail.python.org> Message-ID: <6B45C3A6-3530-11D7-8671-003065A81A70@vanderbilt.edu> I am experimenting with multithreaded code and sockets to consider building a distributed control system for our Free Electron Laser using python. The following fairly straightforward snippet causes a crash and traceback, which it doesn't do with the Unix build of python. from threading import * import time import socket import SocketServer import traceback socketno=8000 def request_handler(request, client, server_self): print client, request class myserver(SocketServer.UDPServer): allow_reuse_address=1 def serve(self): try: print "threaded server running." while(1): self.handle_request() except: traceback.print_exc() server=myserver(('',socketno), request_handler) try: server_thread=Thread(target=server.serve, name="Server") server_thread.start() time.sleep(0.1) print "server up!" s=socket.socket(socket.AF_INET, socket.SOCK_DGRAM) s.connect(('127.0.0.1',socketno)) for i in range(20): s.send('packet %d\n' % i) time.sleep(0.001) finally: time.sleep(0.1) #allow threads to rationalize server.socket.close() #force server down del server the crash happens when I try to force the server down with the server.socket.close(). Here is the traceback: ********** Date/Time: 2003-01-31 09:24:11 -0600 OS Version: 10.2.3 (Build 6G30) Host: mendenhall.fel.Vanderbilt.Edu Command: Python IDE PID: 6018 Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000058 Thread 0: #0 0x90073c48 in mach_msg_trap #1 0x90005f90 in mach_msg #2 0x9022a75c in SwitchContexts #3 0x90225008 in YieldToThread #4 0x9022abbc in YieldToAnyThread #5 0x001cdf44 in 0x1cdf44 #6 0x0020c168 in 0x20c168 #7 0x0020c070 in 0x20c070 #8 0x0020bfe8 in 0x20bfe8 #9 0x0020cb64 in 0x20cb64 #10 0x002111f4 in 0x2111f4 #11 0x002381c0 in 0x2381c0 #12 0x00238134 in mac_1str #13 0x002380a8 in mac_chdir #14 0x001fc004 in PyCFunction_Call #15 0x00259ea8 in 0x259ea8 #16 0x00256278 in PyEval_EvalCodeEx #17 0x0025d3dc in fast_function #18 0x00259f94 in 0x259f94 #19 0x00256278 in PyEval_EvalCodeEx #20 0x0025d3dc in fast_function #21 0x00259f94 in 0x259f94 #22 0x00256278 in PyEval_EvalCodeEx #23 0x0025597c in function_call #24 0x001da51c in PyObject_Call #25 0x00255238 in instancemethod_call #26 0x001da51c in PyObject_Call #27 0x001d59e4 in PyEval_CallObjectWithKeywords #28 0x002b8024 in builtin_apply #29 0x001fc004 in PyCFunction_Call #30 0x00259ea8 in 0x259ea8 #31 0x00256278 in PyEval_EvalCodeEx #32 0x0025d3dc in fast_function #33 0x00259f94 in 0x259f94 #34 0x00256278 in PyEval_EvalCodeEx #35 0x0025d3dc in fast_function #36 0x00259f94 in 0x259f94 #37 0x00256278 in PyEval_EvalCodeEx #38 0x0025d3dc in fast_function #39 0x00259f94 in 0x259f94 #40 0x00256278 in PyEval_EvalCodeEx #41 0x0025d3dc in fast_function #42 0x00259f94 in 0x259f94 #43 0x00256278 in PyEval_EvalCodeEx #44 0x0025d3dc in fast_function #45 0x00259f94 in 0x259f94 #46 0x00256278 in PyEval_EvalCodeEx #47 0x0025d3dc in fast_function #48 0x00259f94 in 0x259f94 #49 0x00256278 in PyEval_EvalCodeEx #50 0x0025d3dc in fast_function #51 0x00259f94 in 0x259f94 #52 0x00256278 in PyEval_EvalCodeEx #53 0x0025d3dc in fast_function #54 0x00259f94 in 0x259f94 #55 0x00256278 in PyEval_EvalCodeEx #56 0x0025597c in function_call #57 0x001da51c in PyObject_Call #58 0x00255238 in instancemethod_call #59 0x001da51c in PyObject_Call #60 0x001d59e4 in PyEval_CallObjectWithKeywords #61 0x00296078 in PyInstance_New #62 0x001da51c in PyObject_Call #63 0x0025d000 in do_call #64 0x00259fac in 0x259fac #65 0x00256278 in PyEval_EvalCodeEx #66 0x0025eb70 in PyEval_EvalCode #67 0x002a7638 in PyImport_ExecCodeModuleEx #68 0x002a6060 in load_source_module #69 0x00217d24 in load_module #70 0x00217760 in import_submodule #71 0x00217198 in load_next #72 0x00216a38 in import_module_ex #73 0x001d6704 in PyImport_ImportModuleEx #74 0x002b7e64 in builtin___import__ #75 0x001fc004 in PyCFunction_Call #76 0x001da51c in PyObject_Call #77 0x001d59e4 in PyEval_CallObjectWithKeywords #78 0x002599dc in 0x2599dc #79 0x00256278 in PyEval_EvalCodeEx #80 0x0025eb70 in PyEval_EvalCode #81 0x002a7638 in PyImport_ExecCodeModuleEx #82 0x002bf824 in PyImport_ExecCodeModule #83 0x002c9578 in run_main_resource #84 0x002c9440 in PyMac_InitApplet #85 0x0019c1cc in main Thread 1: #0 0x9003eaa8 in semaphore_wait_signal_trap #1 0x9003e8c4 in _pthread_cond_wait #2 0x90233368 in TSWaitOnSemaphoreCommon #3 0x90248a10 in AsyncFileThread(void*) #4 0x90020d48 in _pthread_body Thread 2: #0 0x90042688 in semaphore_timedwait_signal_trap #1 0x9003e8b4 in _pthread_cond_wait #2 0x9022ffd4 in MPWaitOnQueue #3 0x90736fc4 in TNodeSyncTask::SyncTaskProc(void*) #4 0x9025e3dc in PrivateMPEntryPoint #5 0x90020d48 in _pthread_body Thread 3: #0 0x9000514c in syscall #1 0x90515d6c in BSD_waitevent #2 0x9051573c in CarbonSelectThreadFunc #3 0x90020d48 in _pthread_body Thread 4: #0 0x9003eaa8 in semaphore_wait_signal_trap #1 0x9003e8c4 in _pthread_cond_wait #2 0x9051dc30 in CarbonOperationThreadFunc #3 0x90020d48 in _pthread_body Thread 5 Crashed: #0 0x00000058 in 0x58 #1 0x00288b24 in 0x288b24 #2 0x00288960 in PySocketSock_recvfrom #3 0x001fc004 in PyCFunction_Call #4 0x00259ea8 in 0x259ea8 #5 0x00256278 in PyEval_EvalCodeEx #6 0x0025d3dc in fast_function #7 0x00259f94 in 0x259f94 #8 0x00256278 in PyEval_EvalCodeEx #9 0x0025d3dc in fast_function #10 0x00259f94 in 0x259f94 #11 0x00256278 in PyEval_EvalCodeEx #12 0x0025597c in function_call #13 0x001da51c in PyObject_Call #14 0x00255238 in instancemethod_call #15 0x001da51c in PyObject_Call #16 0x001d59e4 in PyEval_CallObjectWithKeywords #17 0x002b8024 in builtin_apply #18 0x001fc004 in PyCFunction_Call #19 0x00259ea8 in 0x259ea8 #20 0x00256278 in PyEval_EvalCodeEx #21 0x0025d3dc in fast_function #22 0x00259f94 in 0x259f94 #23 0x00256278 in PyEval_EvalCodeEx #24 0x0025597c in function_call #25 0x001da51c in PyObject_Call #26 0x00255238 in instancemethod_call #27 0x001da51c in PyObject_Call #28 0x001d59e4 in PyEval_CallObjectWithKeywords #29 0x0028bf1c in t_bootstrap #30 0x0028c84c in 0x28c84c #31 0x9023bdac in CooperativeThread #32 0x90020d48 in _pthread_body Thread 6: #0 0x9003eaa8 in semaphore_wait_signal_trap #1 0x9003e8c4 in _pthread_cond_wait #2 0x90233368 in TSWaitOnSemaphoreCommon #3 0x9023c170 in TimerThread #4 0x90020d48 in _pthread_body PPC Thread State: srr0: 0x00000058 srr1: 0x4000f030 vrsave: 0x00000000 xer: 0x20000000 lr: 0x00285ca4 ctr: 0x0000005a mq: 0x00000000 r0: 0x0000005a r1: 0xf0283ff0 r2: 0x003f3344 r3: 0x01fa2860 r4: 0x004ca860 r5: 0x00000000 r6: 0x1a47d000 r7: 0x00000000 r8: 0x87c26960 r9: 0x00000000 r10: 0x0000a832 r11: 0x00000000 r12: 0x004e40b0 r13: 0x00000001 r14: 0x024995f0 r15: 0x02416420 r16: 0x02417200 r17: 0x02416578 r18: 0x00000000 r19: 0x00000001 r20: 0x024ff8c0 r21: 0x00000000 r22: 0x00000000 r23: 0x02438640 r24: 0x02417200 r25: 0x01fa2860 r26: 0xf02840b8 r27: 0xf0284148 r28: 0xf0284138 r29: 0x00000000 r30: 0x0006074c r31: 0x0043ae48 Is anyone out there enough of an expert in MacPython's threading code to see why this should melt down? From the "Thread 6 crashed" dump, it looks as if the socket being closed is not returning an error correctly. Thanks in advance. Marcus Mendenhall From marcus.h.mendenhall@vanderbilt.edu Fri Jan 31 15:29:39 2003 From: marcus.h.mendenhall@vanderbilt.edu (Marcus Mendenhall) Date: Fri, 31 Jan 2003 09:29:39 -0600 Subject: [Pythonmac-SIG] threads & sockets crash MacPython 2.2.2 (was ...2.2.1) In-Reply-To: <20030130170001.21573.80517.Mailman@mail.python.org> Message-ID: My previous message about a crash in MacPython 2.2.1 was really in 2.2.2, if it makes any difference. Sorry. Marcus Mendenhall From robin@alldunn.com Fri Jan 31 17:25:59 2003 From: robin@alldunn.com (Robin Dunn) Date: Fri, 31 Jan 2003 09:25:59 -0800 Subject: bsddb.hashopen problem [was: Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary] In-Reply-To: References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> Message-ID: <3E3AB1A7.2040209@alldunn.com> Francois Granger wrote: [...] > > I tried to run the script wich gave me this problem, pop3proxy.py from > the spambayes project. Error below. I retested with a smaller script. > the line: > > dbm = bsddb.hashopen('/Volumes/OS99/spambayesdev/hammie.db', 'r') > > give the same error. I don't know who is concerned with this. > > [fbg:/Volumes/OS99/spambayesdev] fgranger% which python > /usr/local/bin/python > [fbg:/Volumes/OS99/spambayesdev] fgranger% python -v pop3proxy.py > > [...] > import bsddb # directory > /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb > # > /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.pyc > matches > /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py > > import bsddb # precompiled from > /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.pyc > > import _bsddb # dynamically loaded from > /Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/lib-dynload/_bsddb.so > > Traceback (most recent call last): > File "pop3proxy.py", line 1557, in ? > run() > File "pop3proxy.py", line 1531, in run > state.createWorkers() > File "pop3proxy.py", line 1145, in createWorkers > self.bayes = storage.DBDictClassifier( \ > File "/Volumes/OS99/spambayesdev/spambayes/storage.py", line 140, in > __init__ > self.load() > File "/Volumes/OS99/spambayesdev/spambayes/storage.py", line 148, in load > self.dbm = dbmstorage.open(self.db_name, self.mode) > File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 54, in > open > return f(*args) > File "/Volumes/OS99/spambayesdev/spambayes/dbmstorage.py", line 17, in > open_dbhash > return bsddb.hashopen(*args) > File > "/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/bsddb/__init__.py", > line 162, in hashopen > d.open(file, _db.DB_HASH, flags, mode) > bsddb._db.DBInvalidArgError: (22, 'Invalid argument -- memp_fopen: page > sizes must be a power-of-2') > [...] > Sounds like either spambayes is passing a bad value for the pgsize parameter, or bsddb.hashopen should have an explicit default value for pgsize... -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From robin@alldunn.com Fri Jan 31 17:37:26 2003 From: robin@alldunn.com (Robin Dunn) Date: Fri, 31 Jan 2003 09:37:26 -0800 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> Message-ID: <3E3AB456.8090106@alldunn.com> Francois Granger wrote: > > There is a new IDLE program in the Application/MacPython-2.3 folder. I > doubleclicked it and it launch and quit. > Do you have TclTkAqua installed? -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From francois.granger@free.fr Fri Jan 31 18:19:22 2003 From: francois.granger@free.fr (Francois Granger) Date: Fri, 31 Jan 2003 19:19:22 +0100 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E3AB456.8090106@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> <3E3AB456.8090106@alldunn.com> Message-ID: At 09:37 -0800 31/01/2003, in message Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary, Robin Dunn wrote: >Francois Granger wrote: >> >>There is a new IDLE program in the Application/MacPython-2.3 >>folder. I doubleclicked it and it launch and quit. >> > >Do you have TclTkAqua installed? Sorry, no. I am going stp by step so that as not having too much "beta" software at the same time. ;-) But that was just a try. I think that wxPython and Tk are for the same purpose. So i'll prefer not add another library. I'll stick to PythonIDE wich I am used to and maybe later to PythonCard codeeditor when it will work. -- Recently using MacOSX....... From robin@alldunn.com Fri Jan 31 18:45:18 2003 From: robin@alldunn.com (Robin Dunn) Date: Fri, 31 Jan 2003 10:45:18 -0800 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> <3E3AB456.8090106@alldunn.com> Message-ID: <3E3AC43E.1040705@alldunn.com> Francois Granger wrote: > At 09:37 -0800 31/01/2003, in message Re: [Pythonmac-SIG] New > MacPython-OSX 2.3a1 binary, Robin Dunn wrote: > >> Francois Granger wrote: >> >>> >>> There is a new IDLE program in the Application/MacPython-2.3 folder. >>> I doubleclicked it and it launch and quit. >>> >> >> Do you have TclTkAqua installed? > > > Sorry, no. I am going stp by step so that as not having too much "beta" > software at the same time. ;-) > > But that was just a try. I think that wxPython and Tk are for the same > purpose. So i'll prefer not add another library. I'll stick to PythonIDE > wich I am used to and maybe later to PythonCard codeeditor when it will > work. > That's okay, I wouldn't have it installed either except people were asking me to include _tkinter in my Python binary... I just wanted to be sure that I didn't leave anything out of the Installer. -- Robin Dunn Software Craftsman http://wxPython.org Java give you jitters? Relax with wxPython! From tony@lownds.com Fri Jan 31 20:42:34 2003 From: tony@lownds.com (Tony Lownds) Date: Fri, 31 Jan 2003 12:42:34 -0800 Subject: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: <3E39F917.1010903@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> Message-ID: >Kevin Altis reported that he is not able to install this binary >because it is preventing him from putting it on his only disk. I >added a RootVolumeOnly directive to the package info file since >somebody mentioned troubles putting it on other volumes. It works >fine for me, my second cvolume is diabled and I am only able to >unstall on the root. > >Does anybody else see this problem? Does anybody know a way to fix >it besides removing the RootVolumeOnly? Yes, I am seeing this as well. I only have one volume (its a laptop with one disk). I tried with two versions of the installer tool (1.3.1 and 1.3.2), it did not make a difference. The error occurs on the select destination step; if I go back and then continue again, the error is gone and I can continue installing. By the way, thank you for including Tkinter!! -Tony From francois.granger@free.fr Fri Jan 31 21:26:12 2003 From: francois.granger@free.fr (Francois Granger) Date: Fri, 31 Jan 2003 22:26:12 +0100 Subject: bsddb.hashopen problem [was: Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary] In-Reply-To: <3E3AB1A7.2040209@alldunn.com> References: <3E39CAE7.2010605@alldunn.com> <3E39F917.1010903@alldunn.com> <3E3AB1A7.2040209@alldunn.com> Message-ID: At 09:25 -0800 31/01/2003, in message bsddb.hashopen problem [was: Re: [Pythonmac-SIG] New Ma, Robin Dunn wrote: >Francois Granger wrote: >[...] >> >>I tried to run the script wich gave me this problem, pop3proxy.py >>from the spambayes project. Error below. I retested with a smaller >>script. the line: >> >>dbm = bsddb.hashopen('/Volumes/OS99/spambayesdev/hammie.db', 'r') >> >>give the same error. I don't know who is concerned with this. > >Sounds like either spambayes is passing a bad value for the pgsize >parameter, or bsddb.hashopen should have an explicit default value >for pgsize... My little script did not passed any parameter. So, I guess that it is the later, not the former ? I can report it to the spambayes list where there are some really well known python programmers.... -- Recently using MacOSX....... From altis@semi-retired.com Fri Jan 31 21:57:54 2003 From: altis@semi-retired.com (Kevin Altis) Date: Fri, 31 Jan 2003 13:57:54 -0800 Subject: [wxPython-mac] Re: [Pythonmac-SIG] New MacPython-OSX 2.3a1 binary In-Reply-To: Message-ID: > From: Tony Lownds > > >Kevin Altis reported that he is not able to install this binary > >because it is preventing him from putting it on his only disk. I > >added a RootVolumeOnly directive to the package info file since > >somebody mentioned troubles putting it on other volumes. It works > >fine for me, my second cvolume is diabled and I am only able to > >unstall on the root. > > > >Does anybody else see this problem? Does anybody know a way to fix > >it besides removing the RootVolumeOnly? > > Yes, I am seeing this as well. I only have one volume (its a laptop > with one disk). I tried with two versions of the installer tool > (1.3.1 and 1.3.2), it did not make a difference. > > The error occurs on the select destination step; if I go back and > then continue again, the error is gone and I can continue installing. Indeed, clicking Go Back and then continuing solves the problem. Previously I had only tried quitting the installer and then trying again. This is still a weird problem. ka