From metatracker at psf.upfronthosting.co.za Mon Oct 8 13:15:26 2007 From: metatracker at psf.upfronthosting.co.za (reto) Date: Mon, 08 Oct 2007 11:15:26 -0000 Subject: [Tracker-discuss] =?utf-8?q?=5Bissue162=5D_IDLE_does_not_start_if?= =?utf-8?q?_windows_environment_variable_containing_=27German_Umlau?= =?utf-8?b?dGU6IMOkw7bDvCcgZXhpc3Rz?= Message-ID: <1191842125.98.0.196668271938.issue162@psf.upfronthosting.co.za> New submission from reto: C:\devtools\Python30\Lib\idlelib>set ... eRoomOfflineFiles=Z:\Documents and Settings\tgdwere3\My Documents\eRoom Dateien f?r die Offline-Bearbeitung\ ... C:\devtools\Python30\Lib\idlelib>..\..\python.exe idle.pyw object : UnicodeDecodeError('utf8', b'Z:\\Documents and Settings\\tgdwere3\\My Documents\\eRoom Dateien f\xfcr die Offline-Bearbeitung\\', 63, 69, 'u nsupported Unicode code range') type : UnicodeDecodeError refcount: 4 address : 00B825B0 lost sys.stderr ---------- messages: 817 nosy: reto priority: bug status: unread title: IDLE does not start if windows environment variable containing 'German Umlaute: ???' exists _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Mon Oct 8 13:18:10 2007 From: metatracker at psf.upfronthosting.co.za (Reto Wehrli) Date: Mon, 08 Oct 2007 11:18:10 -0000 Subject: [Tracker-discuss] =?utf-8?q?=5Bissue162=5D_IDLE_does_not_start_if?= =?utf-8?q?_windows_environment_variable_containing_=27German_Umlau?= =?utf-8?b?dGU6IMOkw7bDvCcgZXhpc3Rz?= Message-ID: <1191842290.56.0.452243460798.issue162@psf.upfronthosting.co.za> Reto Wehrli added the comment: after removing the env variable idle starts ---------- status: unread -> chatting _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Mon Oct 8 17:09:22 2007 From: metatracker at psf.upfronthosting.co.za (Paul Dubois) Date: Mon, 08 Oct 2007 15:09:22 -0000 Subject: [Tracker-discuss] =?utf-8?q?=5Bissue162=5D_IDLE_does_not_start_if?= =?utf-8?q?_windows_environment_variable_containing_=27German_Umlau?= =?utf-8?b?dGU6IMOkw7bDvCcgZXhpc3Rz?= Message-ID: <1191856162.31.0.711177655237.issue162@psf.upfronthosting.co.za> Paul Dubois added the comment: Thank you for submitting a Python bug report. Unfortunately you have submitted it to the tracker used to report bugs in the Python bug tracker itself, not Python. Please resubmit it to bugs.python.org. ---------- status: chatting -> resolved _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Mon Oct 8 21:19:58 2007 From: metatracker at psf.upfronthosting.co.za (Brett C.) Date: Mon, 08 Oct 2007 19:19:58 -0000 Subject: [Tracker-discuss] [issue147] Current email headers corrupt mail reader display Message-ID: <1191871198.4.0.166128823056.issue147@psf.upfronthosting.co.za> Brett C. added the comment: `Christian Heimes (tiran) is also affected; noticed that the email notification to python-bugs-list had his name tabbed in on a newline. I see Martin uploaded a script to fix the DB, but I don't know enough SQL to know if it is correct. Plus, should we add something to Roundup to detect this during registration? _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Mon Oct 8 21:32:05 2007 From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=) Date: Mon, 08 Oct 2007 19:32:05 -0000 Subject: [Tracker-discuss] [issue147] Current email headers corrupt mail reader display In-Reply-To: <1191871198.4.0.166128823056.issue147@psf.upfronthosting.co.za> Message-ID: <470A85B3.9070806@v.loewis.de> Martin v. L?wis added the comment: > Plus, should we add something to Roundup to detect this during registration? That's not necessary. I don't think it's possible to fill the form to break it. Regards, Martin _______________________________________________________ Meta Tracker _______________________________________________________ From martin at v.loewis.de Mon Oct 8 21:32:03 2007 From: martin at v.loewis.de (=?ISO-8859-1?Q?=22Martin_v=2E_L=F6wis=22?=) Date: Mon, 08 Oct 2007 21:32:03 +0200 Subject: [Tracker-discuss] [issue147] Current email headers corrupt mail reader display In-Reply-To: <1191871198.4.0.166128823056.issue147@psf.upfronthosting.co.za> References: <1191871198.4.0.166128823056.issue147@psf.upfronthosting.co.za> Message-ID: <470A85B3.9070806@v.loewis.de> > Plus, should we add something to Roundup to detect this during registration? That's not necessary. I don't think it's possible to fill the form to break it. Regards, Martin From metatracker at psf.upfronthosting.co.za Fri Oct 12 08:01:13 2007 From: metatracker at psf.upfronthosting.co.za (Paul Dubois) Date: Fri, 12 Oct 2007 06:01:13 -0000 Subject: [Tracker-discuss] [issue163] new traceback, cause unknown Message-ID: <1192168873.7.0.207585629137.issue163@psf.upfronthosting.co.za> New submission from Paul Dubois: Did anybody just do something? We all of a sudden got 38 instances of: Traceback (most recent call last): File "/home/roundup/roundup-production//lib/python2.4/site-packages/roundup/cgi/client.py", line 290, in inner_main self.write_html(self.renderContext()) File "/home/roundup/roundup-production//lib/python2.4/site-packages/roundup/cgi/client.py", line 885, in write_html self.header() File "/home/roundup/roundup-production//lib/python2.4/site-packages/roundup/cgi/client.py", line 926, in header self._socket_op(self.request.start_response, headers, response) File "/home/roundup/roundup-production//lib/python2.4/site-packages/roundup/cgi/client.py", line 871, in _socket_op if err.errno not in self.IGNORE_NET_ERRORS: AttributeError: error instance has no attribute 'errno' ---------- messages: 822 nosy: dubois priority: bug status: unread title: new traceback, cause unknown _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Fri Oct 12 12:40:40 2007 From: metatracker at psf.upfronthosting.co.za (A.M. Kuchling) Date: Fri, 12 Oct 2007 10:40:40 -0000 Subject: [Tracker-discuss] [issue163] new traceback, cause unknown In-Reply-To: <1192168873.7.0.207585629137.issue163@psf.upfronthosting.co.za> Message-ID: <20071012103439.GA15586@amk.local> A.M. Kuchling added the comment: On Fri, Oct 12, 2007 at 06:01:13AM -0000, Paul Dubois wrote: > Did anybody just do something? We all of a sudden got 38 instances of: > Traceback (most recent call last): > > AttributeError: error instance has no attribute 'errno' Neal Norwitz just posted a problem report to the pydotorg list: Date: Thu, 11 Oct 2007 22:07:36 -0700 From: Neal Norwitz To: pydotorg at python.org Subject: [Pydotorg] bugs.python.org painfully slow It's taking minutes to load pages. Which machine is running bugs? Does that server also need a kick from time to time? Maybe some of those errors were causing Neal's problem. Some transient routing error, maybe? --amk ---------- status: unread -> chatting _______________________________________________________ Meta Tracker _______________________________________________________ From nnorwitz at gmail.com Fri Oct 12 20:16:36 2007 From: nnorwitz at gmail.com (Neal Norwitz) Date: Fri, 12 Oct 2007 11:16:36 -0700 Subject: [Tracker-discuss] [issue163] new traceback, cause unknown In-Reply-To: <20071012103439.GA15586@amk.local> References: <1192168873.7.0.207585629137.issue163@psf.upfronthosting.co.za> <20071012103439.GA15586@amk.local> Message-ID: Yeah, the problem cleared itself up after about 15 minutes or so. Perhaps there were too many connections and I was effectively DoSed? n -- On 10/12/07, A.M. Kuchling wrote: > > A.M. Kuchling added the comment: > > On Fri, Oct 12, 2007 at 06:01:13AM -0000, Paul Dubois wrote: > > Did anybody just do something? We all of a sudden got 38 instances of: > > Traceback (most recent call last): > > > > AttributeError: error instance has no attribute 'errno' > > Neal Norwitz just posted a problem report to the pydotorg list: > > Date: Thu, 11 Oct 2007 22:07:36 -0700 > From: Neal Norwitz > To: pydotorg at python.org > Subject: [Pydotorg] bugs.python.org painfully slow > > It's taking minutes to load pages. Which machine is running bugs? > Does that server also need a kick from time to time? > > Maybe some of those errors were causing Neal's problem. Some transient > routing error, maybe? > > --amk > > ---------- > status: unread -> chatting > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________ > Tracker-discuss mailing list > Tracker-discuss at python.org > http://mail.python.org/mailman/listinfo/tracker-discuss > From metatracker at psf.upfronthosting.co.za Fri Oct 12 20:16:40 2007 From: metatracker at psf.upfronthosting.co.za (nnorwitz) Date: Fri, 12 Oct 2007 18:16:40 -0000 Subject: [Tracker-discuss] [issue163] new traceback, cause unknown In-Reply-To: <20071012103439.GA15586@amk.local> Message-ID: nnorwitz added the comment: Yeah, the problem cleared itself up after about 15 minutes or so. Perhaps there were too many connections and I was effectively DoSed? n -- On 10/12/07, A.M. Kuchling wrote: > > A.M. Kuchling added the comment: > > On Fri, Oct 12, 2007 at 06:01:13AM -0000, Paul Dubois wrote: > > Did anybody just do something? We all of a sudden got 38 instances of: > > Traceback (most recent call last): > > > > AttributeError: error instance has no attribute 'errno' > > Neal Norwitz just posted a problem report to the pydotorg list: > > Date: Thu, 11 Oct 2007 22:07:36 -0700 > From: Neal Norwitz > To: pydotorg at python.org > Subject: [Pydotorg] bugs.python.org painfully slow > > It's taking minutes to load pages. Which machine is running bugs? > Does that server also need a kick from time to time? > > Maybe some of those errors were causing Neal's problem. Some transient > routing error, maybe? > > --amk > > ---------- > status: unread -> chatting > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________ > Tracker-discuss mailing list > Tracker-discuss at python.org > http://mail.python.org/mailman/listinfo/tracker-discuss > _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Sun Oct 14 13:24:10 2007 From: metatracker at psf.upfronthosting.co.za (Erik Forsberg) Date: Sun, 14 Oct 2007 11:24:10 -0000 Subject: [Tracker-discuss] [issue147] Current email headers corrupt mail reader display Message-ID: <1192361050.72.0.804065096623.issue147@psf.upfronthosting.co.za> Erik Forsberg added the comment: fixuser.py has been run. ---------- status: chatting -> resolved _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Sun Oct 14 16:22:32 2007 From: metatracker at psf.upfronthosting.co.za (Erik Forsberg) Date: Sun, 14 Oct 2007 14:22:32 -0000 Subject: [Tracker-discuss] [issue141] Should send email on assignment Message-ID: <1192371752.12.0.450841435865.issue141@psf.upfronthosting.co.za> Erik Forsberg added the comment: The mail detectors have now been rewritten. I rearranged the code into one file - functionality previously in tellteam.py, busybody.py and nosyreaction.py are now in one (new) file - sendmail.py, which hopefully makes it clearer when and how mail is sent. Mail is now sent to tellteam for new messages only, and to busybody and nosy list for all other changes (property-only changes as well as additions to the list of messages). Also made nosyreaction add the creator of new issues without any messages to the nosy list (a corner case previously not covered). One known misfeature: Property-only modifications sent to nosy members currently are presented as "System Message", not "Changes by " - this will be fixed when the next upgrade of roundup is made. Sorry for the delay, there's been little time for leisure computing lately. ---------- status: chatting -> resolved _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Mon Oct 15 22:29:48 2007 From: metatracker at psf.upfronthosting.co.za (Brett C.) Date: Mon, 15 Oct 2007 20:29:48 -0000 Subject: [Tracker-discuss] [issue141] Should send email on assignment In-Reply-To: <1192371752.12.0.450841435865.issue141@psf.upfronthosting.co.za> Message-ID: Brett C. added the comment: No need to apologize. We all have a limited amount of time to work on these pet projects of ours. On 10/14/07, Erik Forsberg wrote: > > Erik Forsberg added the comment: > > The mail detectors have now been rewritten. I rearranged the code into one file > - functionality previously in tellteam.py, busybody.py and nosyreaction.py are > now in one (new) file - sendmail.py, which hopefully makes it clearer when and > how mail is sent. > > Mail is now sent to tellteam for new messages only, and to busybody and nosy > list for all other changes (property-only changes as well as additions to the > list of messages). > > Also made nosyreaction add the creator of new issues without any messages to the > nosy list (a corner case previously not covered). > > One known misfeature: Property-only modifications sent to nosy members currently > are presented as "System Message", not "Changes by " - this > will be fixed when the next upgrade of roundup is made. > > Sorry for the delay, there's been little time for leisure computing lately. > > ---------- > status: chatting -> resolved > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________ > Tracker-discuss mailing list > Tracker-discuss at python.org > http://mail.python.org/mailman/listinfo/tracker-discuss > ---------- status: resolved -> chatting _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Tue Oct 16 17:25:35 2007 From: metatracker at psf.upfronthosting.co.za (Guido van Rossum) Date: Tue, 16 Oct 2007 15:25:35 -0000 Subject: [Tracker-discuss] [issue164] System message mail Message-ID: <1192548335.38.0.960756685273.issue164@psf.upfronthosting.co.za> New submission from Guido van Rossum: Since a few days (a week?) I keep getting a lot of emails related to tracker issues (in whose nosy list I am) where the body of the message is simply "System message:". The sender is "admin". What's up with these? I'm attaching the latest example. ---------- messages: 828 nosy: guido priority: urgent status: unread title: System message mail _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Tue Oct 16 19:08:11 2007 From: metatracker at psf.upfronthosting.co.za (Paul Dubois) Date: Tue, 16 Oct 2007 17:08:11 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: <1192548335.38.0.960756685273.issue164@psf.upfronthosting.co.za> Message-ID: Paul Dubois added the comment: Guido, By popular demand a change was made to send email on EVERY change to an issue, including ones for which no human typed a message. Roundup's default behavior in that case is to make up a message that says 'System message'. For example if someone just goes in and changes the priority, but doesn't type a message as to why, you'll get such a message. It is possible to do better and replace this message with something that has at least the name of the user who made the change, such as, 'Action taken by Dubois.' but to see what the change is about you have to look at the stuff at the bottom like priority->urgent. We will make this improvement when someone gets a chance but it isn't going to be all that much of an improvement. It would therefore be best if people making a change type a message. My experience with my own tracker made me feel these 'no message' changes should not trigger an email, but I seem to be the only one who thinks so and that is why we made the change to the current behavior. On 10/16/07, Guido van Rossum wrote: > > > New submission from Guido van Rossum: > > Since a few days (a week?) I keep getting a lot of emails related to > tracker > issues (in whose nosy list I am) where the body of the message is simply > "System > message:". The sender is "admin". What's up with these? I'm attaching > the > latest example. > > ---------- > messages: 828 > nosy: guido > priority: urgent > status: unread > title: System message mail > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________ > Tracker-discuss mailing list > Tracker-discuss at python.org > http://mail.python.org/mailman/listinfo/tracker-discuss > ---------- status: unread -> chatting _______________________________________________________ Meta Tracker _______________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20071016/aaf97fcc/attachment.htm From metatracker at psf.upfronthosting.co.za Tue Oct 16 20:05:02 2007 From: metatracker at psf.upfronthosting.co.za (Guido van Rossum) Date: Tue, 16 Oct 2007 18:05:02 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: Message-ID: Guido van Rossum added the comment: Then the message sent should contain a hint as to what was changed. In the example I attached, there was *nothing*. On 10/16/07, Paul Dubois wrote: > > Paul Dubois added the comment: > > Guido, > > By popular demand a change was made to send email on EVERY change to an > issue, including ones for which no human typed a message. Roundup's default > behavior in that case is to make up a message that says 'System message'. > For example if someone just goes in and changes the priority, but doesn't > type a message as to why, you'll get such a message. > > It is possible to do better and replace this message with something that has > at least the name of the user who made the change, such as, 'Action taken by > Dubois.' but to see what the change is about you have to look at the stuff > at the bottom like priority->urgent. We will make this improvement when > someone gets a chance but it isn't going to be all that much of an > improvement. It would therefore be best if people making a change type a > message. > > My experience with my own tracker made me feel these 'no message' changes > should not trigger an email, but I seem to be the only one who thinks so and > that is why we made the change to the current behavior. > > On 10/16/07, Guido van Rossum wrote: > > > > > > New submission from Guido van Rossum: > > > > Since a few days (a week?) I keep getting a lot of emails related to > > tracker > > issues (in whose nosy list I am) where the body of the message is simply > > "System > > message:". The sender is "admin". What's up with these? I'm attaching > > the > > latest example. > > > > ---------- > > messages: 828 > > nosy: guido > > priority: urgent > > status: unread > > title: System message mail > > > > _______________________________________________________ > > Meta Tracker > > > > _______________________________________________________ > > _______________________________________________ > > Tracker-discuss mailing list > > Tracker-discuss at python.org > > http://mail.python.org/mailman/listinfo/tracker-discuss > > > > ---------- > status: unread -> chatting > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Tue Oct 16 20:49:11 2007 From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=) Date: Tue, 16 Oct 2007 18:49:11 -0000 Subject: [Tracker-discuss] [issue164] System message mail Message-ID: <1192560551.82.0.0880616094249.issue164@psf.upfronthosting.co.za> Martin v. L?wis added the comment: Guido, somehow, your attachment was not recorded. Can you please retry? Ideally, attach the entire RFC822 message that you received. The messages I receive have bodies like ##### SNIP ############ System message: ---------- resolution: -> duplicate status: open -> closed superseder: -> Text.edit_modified() fails _____________________________________ Tracker _____________________________________ ########## SNAP ################ _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Tue Oct 16 20:54:21 2007 From: metatracker at psf.upfronthosting.co.za (Guido van Rossum) Date: Tue, 16 Oct 2007 18:54:21 -0000 Subject: [Tracker-discuss] [issue164] System message mail Message-ID: <1192560861.94.0.788634727113.issue164@psf.upfronthosting.co.za> Guido van Rossum added the comment: Here it is again. _______________________________________________________ Meta Tracker _______________________________________________________ -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: exampl.txt Url: http://mail.python.org/pipermail/tracker-discuss/attachments/20071016/ef13afd9/attachment.txt From metatracker at psf.upfronthosting.co.za Tue Oct 16 21:12:07 2007 From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=) Date: Tue, 16 Oct 2007 19:12:07 -0000 Subject: [Tracker-discuss] [issue164] System message mail Message-ID: <1192561927.54.0.568789814935.issue164@psf.upfronthosting.co.za> Martin v. L?wis added the comment: Looking at this issue, I see that a lot of changes only changed the files that are attached. Perhaps it generates such empty messages for addition and deletion of files? (atleast the change in exampl.txt, from 06:33:39 UTC seems to correlate to such a change on the bug by gregory.p.smith). Assigning to Erik; as a quick fix, I suggest that messages with changes to just the list of associated files are suppressed. In the longer term, it would be useful to send out such attachments to to people on the nosy list, of course. ---------- assignedto: -> forsberg nosy: +forsberg _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Wed Oct 17 00:16:14 2007 From: metatracker at psf.upfronthosting.co.za (Paul Dubois) Date: Tue, 16 Oct 2007 22:16:14 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: <1192561927.54.0.568789814935.issue164@psf.upfronthosting.co.za> Message-ID: Paul Dubois added the comment: I'm sorry I did not make this clear. Prior behavior: only a change in which a message was also entered generated email about that change. The message was displayed, and below that an indication of what else had been changed, as in ----- assignedto -> forsberg When something appears in that section it is a change of value, not a restatement of a previous truth. New behavior: ANY change (including attaching a file, changing a priority, etc. ) generates email. If there is no message, it says System message. The trailing indicators of other changes are also present. Requested behavior from Martin is somewhere in between, to discriminate about what kind of messageless changes will generate mail. On 10/16/07, Martin v. L?wis wrote: > > > Martin v. L?wis added the comment: > > Looking at this issue, I see that a lot of changes only changed the files > that > are attached. Perhaps it generates such empty messages for addition and > deletion > of files? (atleast the change in exampl.txt, from 06:33:39 UTC seems to > correlate to such a change on the bug by gregory.p.smith). > > Assigning to Erik; as a quick fix, I suggest that messages with changes to > just > the list of associated files are suppressed. In the longer term, it would > be > useful to send out such attachments to to people on the nosy list, of > course. > > ---------- > assignedto: -> forsberg > nosy: +forsberg > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________ > Tracker-discuss mailing list > Tracker-discuss at python.org > http://mail.python.org/mailman/listinfo/tracker-discuss > _______________________________________________________ Meta Tracker _______________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.python.org/pipermail/tracker-discuss/attachments/20071016/cde6bf68/attachment-0001.htm From metatracker at psf.upfronthosting.co.za Wed Oct 17 08:07:04 2007 From: metatracker at psf.upfronthosting.co.za (=?utf-8?q?Martin_v._L=C3=B6wis?=) Date: Wed, 17 Oct 2007 06:07:04 -0000 Subject: [Tracker-discuss] [issue164] System message mail Message-ID: <1192601224.35.0.27640371346.issue164@psf.upfronthosting.co.za> Martin v. L?wis added the comment: Apparently, I did not make myself clear, either. I do fully understand what the old and the new behavior is. The issue here seems to be that the new behavior is slightly different from what Paul describes: in the new behavior, the trailing indicators may be empty in certain cases, which is apparently what Guido complains about (not that they say "System message", not that they are generated in the first place, but that that they *simply* say "System message" and nothing else). Such messages should indeed not be generated, IMO; it's a plain bug that they are. Either there is a change, then the nature of the change should be documented in the system message, or there is no change, then no system message should be generated. I also speculated about what the cause for these empty system messages is, namely, that the only change was to the list of attached files, and the system message failed to document that this was the change. I proposed to suppress them for the near future. In the longer term, the system message should be generated, even if only the attached files change. It then should include added files as an attachment; removed files should be documented with a text such as "removed file", or "-fileXXX". _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Wed Oct 17 11:34:15 2007 From: metatracker at psf.upfronthosting.co.za (Erik Forsberg) Date: Wed, 17 Oct 2007 09:34:15 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: <1192548335.38.0.960756685273.issue164@psf.upfronthosting.co.za> Message-ID: <20071017113407.32aca250@impertigo.lkpg.cendio.se> Erik Forsberg added the comment: On Tue, 16 Oct 2007 15:25:35 -0000 Guido van Rossum wrote: > > New submission from Guido van Rossum: > > Since a few days (a week?) I keep getting a lot of emails related to > tracker issues (in whose nosy list I am) where the body of the > message is simply "System message:". The sender is "admin". What's > up with these? Ah. That's an effect me rewriting the e-mail detectors to fix http://psf.upfronthosting.co.za/roundup/meta/issue141. I've seen these messages in the tracker I run at work - I think they occur when files are added or removed. I'll have a look. \EF -- Erik Forsberg http://efod.se GPG/PGP Key: 1024D/0BAC89D9 _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Wed Oct 17 18:30:04 2007 From: metatracker at psf.upfronthosting.co.za (Guido van Rossum) Date: Wed, 17 Oct 2007 16:30:04 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: <1192601224.35.0.27640371346.issue164@psf.upfronthosting.co.za> Message-ID: Guido van Rossum added the comment: On 10/16/07, Martin v. L?wis wrote: > > Martin v. L?wis added the comment: > > Apparently, I did not make myself clear, either. I do fully understand what the > old and the new behavior is. The issue here seems to be that the new behavior is > slightly different from what Paul describes: in the new behavior, the trailing > indicators may be empty in certain cases, which is apparently what Guido > complains about (not that they say "System message", not that they are generated > in the first place, but that that they *simply* say "System message" and nothing > else). Right. Though it would also be helpful if the user who made the change (if there is one) is named as the sender. And instead of "System message" it could say something like "Attribute change". But these are minor -- I wouldn't have brought this up if there weren't system messages without change indicators. > Such messages should indeed not be generated, IMO; it's a plain bug that they > are. Either there is a change, then the nature of the change should be > documented in the system message, or there is no change, then no system message > should be generated. > > I also speculated about what the cause for these empty system messages is, > namely, that the only change was to the list of attached files, and the system > message failed to document that this was the change. I proposed to suppress them > for the near future. In the longer term, the system message should be generated, > even if only the attached files change. It then should include added files as an > attachment; removed files should be documented with a text such as "removed > file", or "-fileXXX". > > _______________________________________________________ > Meta Tracker > > _______________________________________________________ > _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Wed Oct 17 21:15:34 2007 From: metatracker at psf.upfronthosting.co.za (Erik Forsberg) Date: Wed, 17 Oct 2007 19:15:34 -0000 Subject: [Tracker-discuss] [issue164] System message mail In-Reply-To: Message-ID: <20071017211527.53ecfedf@uterus> Erik Forsberg added the comment: > > Apparently, I did not make myself clear, either. I do fully > > understand what the old and the new behavior is. The issue here > > seems to be that the new behavior is slightly different from what > > Paul describes: in the new behavior, the trailing indicators may be > > empty in certain cases, which is apparently what Guido complains > > about (not that they say "System message", not that they are > > generated in the first place, but that that they *simply* say > > "System message" and nothing else). > > Right. Though it would also be helpful if the user who made the change > (if there is one) is named as the sender. And instead of "System > message" it could say something like "Attribute change". But these are > minor -- I wouldn't have brought this up if there weren't system > messages without change indicators. I've modified the detector that sends out mail to nosy list, python-bugs-list and new-bugs-announce. The mails will now contain a section that lists the URLs of added and removed files. The behaviour is unfortunately not completely consistent for all cases - roundup does need some improvement in the code that sends messages. Some oddities: *) The message sent out when a file is added or removed (now) always contains a link to the file. *) If there is a change note in the change _and_ a file is added _and_ the recipient gets the mail because he/she is on the nosy list, the file is also attached to the mail. *) Files are never attached to mail destined for python-bugs-list/new-bugs-announce - they only gets links. A bit confusing, but hopefully not too confusing. There's been some talk about improving the mail sending code in roundup lately, so hopefully, it will be possible to make this less confusing in the future. The "System message" message and admin as sender will be replaced with real user names when we upgrade to the next (yet to be released) version of roundup. Cheers, \EF -- Erik Forsberg http://efod.se GPG/PGP Key: 1024D/0BAC89D9 ---------- status: chatting -> resolved _______________________________________________________ Meta Tracker _______________________________________________________ From draghuram at gmail.com Thu Oct 18 18:22:22 2007 From: draghuram at gmail.com (Raghuram Devarakonda) Date: Thu, 18 Oct 2007 12:22:22 -0400 Subject: [Tracker-discuss] replies appearing as attachments Message-ID: <2c51ecee0710180922s7fd9cc90t182db64347ad8eb7@mail.gmail.com> Hi, Check http://bugs.python.org/issue1290. In addition to the actual attachment "testdata.txt", two other attachments "unnamed" appear and they seem to be the email replies to the bug. This is not the expected behaviour. Right? Thanks, Raghu From forsberg at efod.se Thu Oct 18 21:07:35 2007 From: forsberg at efod.se (Erik Forsberg) Date: Thu, 18 Oct 2007 21:07:35 +0200 Subject: [Tracker-discuss] replies appearing as attachments In-Reply-To: <2c51ecee0710180922s7fd9cc90t182db64347ad8eb7@mail.gmail.com> References: <2c51ecee0710180922s7fd9cc90t182db64347ad8eb7@mail.gmail.com> Message-ID: <20071018210735.3b3e496a@uterus> On Thu, 18 Oct 2007 12:22:22 -0400 "Raghuram Devarakonda" wrote: > Hi, > > Check http://bugs.python.org/issue1290. > > In addition to the actual attachment "testdata.txt", two other > attachments "unnamed" appear and they seem to be the email replies to > the bug. This is not the expected behaviour. Right? I think that is the result of the replies being written using a mail program that sends mail as both text/plain and text/html. Roundup will parse the mail and add the text/plain part as a message on the issue, and add the text/html part as an attachment. Inspecting the unnamed attachments on the issue you're linking to, they are indeed duplicate copies of the messages added at the same time. Slightly odd, but how is roundup supposed to know that the text/html is just a duplicate of the text/plain part? It could be a real text/html attachment added on purpose. I guess this became more visible now with the changes I made yesterday to fix http://psf.upfronthosting.co.za/roundup/meta/issue164, as each attachment is now listed in all mail sent to nosy and python-bugs-list. We could filter out attachments with the name "unnamed". Hmm.. \EF -- Erik Forsberg http://efod.se GPG/PGP Key: 1024D/0BAC89D9 -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://mail.python.org/pipermail/tracker-discuss/attachments/20071018/9988d5c3/attachment.pgp From metatracker at psf.upfronthosting.co.za Mon Oct 22 14:32:40 2007 From: metatracker at psf.upfronthosting.co.za (Stefan Seefeld) Date: Mon, 22 Oct 2007 12:32:40 -0000 Subject: [Tracker-discuss] [issue165] Preventing 'Broken Pipe' and similar exceptions Message-ID: <1193056360.43.0.794459675532.issue165@psf.upfronthosting.co.za> New submission from Stefan Seefeld: The python- as well as the meta-tracker generate an incredible amount of benign exceptions, such as 'Broken Pipe', when a client connection is shut down prior to roundup getting a chance to honor a request. I remember seing a proposed patch for this, probably on the roundup ML. As this (and similar) issue causes a lot of noise in my inbox (from 'roundup-admin'), I classified this as 'critical'. Thanks, ---------- assignedto: forsberg messages: 839 nosy: forsberg, stefan priority: critical status: unread title: Preventing 'Broken Pipe' and similar exceptions _______________________________________________________ Meta Tracker _______________________________________________________ From metatracker at psf.upfronthosting.co.za Tue Oct 23 21:00:33 2007 From: metatracker at psf.upfronthosting.co.za (Erik Forsberg) Date: Tue, 23 Oct 2007 19:00:33 -0000 Subject: [Tracker-discuss] [issue165] Preventing 'Broken Pipe' and similar exceptions Message-ID: <1193166033.6.0.749395284979.issue165@psf.upfronthosting.co.za> Erik Forsberg added the comment: This is hardly critical, affecting five people, none of the python developers. This will be fixed when we upgrade our roundup installation, after release of the next roundup. If you feel very annoyed, I suggest that you fix the problem in our vendor branch of roundup and upgrade on psf. ---------- assignedto: forsberg -> priority: critical -> bug status: unread -> chatting _______________________________________________________ Meta Tracker _______________________________________________________ From facundobatista at gmail.com Tue Oct 23 22:44:27 2007 From: facundobatista at gmail.com (Facundo Batista) Date: Tue, 23 Oct 2007 17:44:27 -0300 Subject: [Tracker-discuss] Python tickets summary In-Reply-To: References: Message-ID: 2007/9/13, Facundo Batista : > All the listings are accesible from the same pages, start here: > > http://www.taniquetil.com.ar/facundo/py_tickets.html > > (remember to refresh) > > Any idea to improve these pages is welcomed. Following an idea of John Lenton, now every page shows a graph of the activity per day of all open tickets. Look at the bottom: http://www.taniquetil.com.ar/facundo/py_tickets.html Regards, -- . Facundo Blog: http://www.taniquetil.com.ar/plog/ PyAr: http://www.python.org/ar/ From metatracker at psf.upfronthosting.co.za Wed Oct 24 21:25:26 2007 From: metatracker at psf.upfronthosting.co.za (Raghuram Devarakonda) Date: Wed, 24 Oct 2007 19:25:26 -0000 Subject: [Tracker-discuss] [issue154] auto-link to svn.python.org for revision IDs Message-ID: <1193253926.23.0.999020030202.issue154@psf.upfronthosting.co.za> Raghuram Devarakonda added the comment: Can the same auto-linking be done for "rev NNNNN" please? Issue 1290 has such usage. Thanks. ---------- status: done-cbb -> chatting _______________________________________________________ Meta Tracker _______________________________________________________