From schwarz.labecedaire at st.com Mon Jan 3 06:34:18 2005 From: schwarz.labecedaire at st.com (Christian SCHWARZ) Date: Mon Jan 3 06:34:18 2005 Subject: [Moin-devel] TWikiDraw plugin Message-ID: <41D957BD.4030703@st.com> Hi, I am running into troubles similar to those already solved when creating images with multi-line code in it. The original problem was the accidental creation of a "\r\n" string instead of a simple "\n". My problem is the creation of a "^M\n" string instead of the simple "\n". I guess, this is because some people use windows to edit the file and then the ^M gets inserted for the Linux environment (with apache). A first solution would be to write single line and use multiple of them, but it is a lot of work and not very artistic. Is there another solution? Thanks, Christian PS: How do I do a simple scale of a resulting image without manually editing the image in a separate tool? From mailinglist at alexanderweb.de Mon Jan 3 08:06:08 2005 From: mailinglist at alexanderweb.de (Alexander Schremmer) Date: Mon Jan 3 08:06:08 2005 Subject: [Moin-devel] TWikiDraw plugin In-Reply-To: <41D957BD.4030703@st.com> Message-ID: Christian SCHWARZ wrote on Monday, January 03, 2005 3:34 PM CEST: > The original problem was the accidental creation of a "\r\n" string > instead of a simple "\n". My problem is the creation of a "^M\n" > string instead of the simple "\n". MoinMoin 1.2 or 1.3? > PS: How do I do a simple scale of a resulting image without manually > editing the image in a separate tool? See MoinMoin:MacroMarket, there is a macro for your needs. Kind regards, Alexander From schwarz.labecedaire at st.com Mon Jan 3 08:57:18 2005 From: schwarz.labecedaire at st.com (Christian SCHWARZ) Date: Mon Jan 3 08:57:18 2005 Subject: [Moin-devel] TWikiDraw plugin In-Reply-To: References: Message-ID: <41D9792E.7090209@st.com> > MoinMoin 1.2 or 1.3? the version is 1.2.4 (at least the tar file has this name) does updating help? > See MoinMoin:MacroMarket, there is a macro for your needs. yep, thanks, the ImageLink looks quite promising. I basically wondered whether or not there is already a formatting character or not. thanks, Christian From 2004b at usenet.alexanderweb.de Mon Jan 3 09:21:08 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Mon Jan 3 09:21:08 2005 Subject: [Moin-devel] Re: TWikiDraw plugin References: <41D9792E.7090209@st.com> Message-ID: <2bo0jgjnpdlf.dlg@usenet.alexanderweb.de> On Mon, 03 Jan 2005 17:56:14 +0100, Christian SCHWARZ wrote: >> MoinMoin 1.2 or 1.3? > the version is 1.2.4 (at least the tar file has this name) > does updating help? Yes, upgrading would help. In 1.3, the writing of line endings was refactored completly. It should solve all problems. But be aware that upgrading to 1.3 is a rather complex process. Kind regards, Alexander From noreply at sourceforge.net Tue Jan 4 07:01:26 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 4 07:01:26 2005 Subject: [Moin-devel] [ moin-Patches-1092353 ] Better reStructuredText support Message-ID: Patches item #1092353, was opened at 2004-12-28 14:17 Message generated for change (Comment added) made by orutherfurd You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Matthew Gilbert (mmgilbe) Assigned to: Nobody/Anonymous (nobody) Summary: Better reStructuredText support Initial Comment: Updated version of my previous patch. This adds much better reStructuredText support to MoinMoin. It is a drop in replacement for the existing rst.py. This version is significantly refactored and has numerous bug fixes. It supports MoinMoin 1.2.X and 1.3.X and requires docutils version 0.3.3 or later. ---------------------------------------------------------------------- Comment By: Ollie Rutherfurd (orutherfurd) Date: 2005-01-04 10:00 Message: Logged In: YES user_id=265596 Please consider accepting this patch -- it greatly improves the existing reST support. I've been using in without problems on 1.3.1. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 From alan.ezust at gmail.com Tue Jan 4 07:27:24 2005 From: alan.ezust at gmail.com (Alan Ezust) Date: Tue Jan 4 07:27:24 2005 Subject: [Moin-devel] [[Navigation(children,2)]] really slow now in 1.3 Message-ID: Hi everybody. I have a bookmarks page that is supposed to list all of its children pages. [[Navigation(children,2)]] In order to perform Navigation(children,2), the operation is now significantly slower than before. And it makes my hard drive go all NUTS for like 20 seconds, when it is performed for the first time in a while. It could be that the macro can be improved in performance, but it could also be due to the new file/directory format of data/text, I am not exactly sure. From alan.ezust at gmail.com Tue Jan 4 07:33:30 2005 From: alan.ezust at gmail.com (Alan Ezust) Date: Tue Jan 4 07:33:30 2005 Subject: [Moin-devel] how do you debug? Message-ID: I was just wondering, if you want to step through moin code as it is running, what is the preferred way to debug it? Do you capture user interactions to a log file and then re-run moin.cgi with some special option? I'd like to step through moin code in my debugger (eric3). From alan.ezust at gmail.com Tue Jan 4 07:42:32 2005 From: alan.ezust at gmail.com (Alan Ezust) Date: Tue Jan 4 07:42:32 2005 Subject: [Moin-devel] Re: how do you debug? In-Reply-To: References: Message-ID: Ah, dunno why I keep forgetting that cgi scripts can be run from the command line... But actually I was wondering more specifically about how to debug twisted mod-python apps. I suppose that would be more appropriately posted on a twisted or mod-python forum. So you can ignore this post. On Tue, 4 Jan 2005 10:32:27 -0500, Alan Ezust wrote: > I was just wondering, if you want to step through moin code as it is > running, what is the preferred way to debug it? Do you capture user > interactions to a log file and then re-run moin.cgi with some special > option? I'd like to step through moin code in my debugger (eric3). > From pneustrom at yahoo.com Tue Jan 4 14:58:32 2005 From: pneustrom at yahoo.com (Philip Neustrom) Date: Tue Jan 4 14:58:32 2005 Subject: [Moin-devel] Re: how do you debug? In-Reply-To: Message-ID: <20050104225730.1310.qmail@web41405.mail.yahoo.com> They can be run from the command line -- but how do you pass the normal CGI GET/POST to them? --- Alan Ezust wrote: > Ah, dunno why I keep forgetting that cgi scripts can > be run from the > command line... > > But actually I was wondering more specifically about > how to debug > twisted mod-python apps. I suppose that would be > more appropriately > posted on a twisted or mod-python forum. So you can > ignore this post. > > > On Tue, 4 Jan 2005 10:32:27 -0500, Alan Ezust > wrote: > > I was just wondering, if you want to step through > moin code as it is > > running, what is the preferred way to debug it? Do > you capture user > > interactions to a log file and then re-run > moin.cgi with some special > > option? I'd like to step through moin code in my > debugger (eric3). > > > > > ------------------------------------------------------- > The SF.Net email is sponsored by: Beat the > post-holiday blues > Get a FREE limited edition SourceForge.net t-shirt > from ThinkGeek. > It's fun and FREE -- well, > almost....http://www.thinkgeek.com/sfshirt > _______________________________________________ > Moin-devel mailing list > Moin-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/moin-devel > __________________________________ Do you Yahoo!? Yahoo! Mail - You care about security. So do we. http://promotions.yahoo.com/new_mail From alan.ezust at gmail.com Tue Jan 4 18:14:10 2005 From: alan.ezust at gmail.com (Alan Ezust) Date: Tue Jan 4 18:14:10 2005 Subject: [Moin-devel] Re: how do you debug? In-Reply-To: <20050104225730.1310.qmail@web41405.mail.yahoo.com> References: <20050104225730.1310.qmail@web41405.mail.yahoo.com> Message-ID: Actually, I am no longer sure of the answer. I thought that all CGI apps could be debugged the same way: by pickling the "request" object and then unpickling it again later, to simulate the user requests. however, the Moin request object is not picklable. I was examining it and noticed it has a lot of references to other objects which also might not be easily picklable. So actually, I should have phrased my question in a similar way to yours. What is the recommended way of capturing the GET/POST request data from the user so that we can re-feed it to the moin program for debugging? Alternately, how does one do remote debugging? On Tue, 4 Jan 2005 14:57:30 -0800 (PST), Philip Neustrom wrote: > They can be run from the command line -- but how do > you pass the normal CGI GET/POST to them? From gilbert at voxmea.net Tue Jan 4 23:02:08 2005 From: gilbert at voxmea.net (Matthew Gilbert) Date: Tue Jan 4 23:02:08 2005 Subject: [Moin-devel] reStructuredText support Message-ID: <41DB90C4.1080806@voxmea.net> Hi All, I recently submitted a new rst.py to sourceforge. Its a drop in replacement for the existing rst.py. It adds much better support for reStructuredText than what currently ships with MoinMoin. I also submitted a similar patch in September, and frankly this one is much better. I was hoping at least to get some feedback as to what I can improve to get this considered to be accepted. I have tested this quite a bit and have a few users who have tested it without issue. I would greatly appreciate anyone who would be willing to try it or who would be willing to provide feedback. The patch can be found here: http://sourceforge.net/tracker/index.php?func=detail&aid=1092353&group_id=8482&atid=308482 Highlights of the new patch include: * rest image support * inline: attachment support * docinfo elements * much better MoinMoin like formatting * many, many bug fixes * MoinMoin 1.3.x support Any help is appreciated. Thanks _matt From nirs at actcom.net.il Wed Jan 5 05:05:30 2005 From: nirs at actcom.net.il (Nir Soffer) Date: Wed Jan 5 05:05:30 2005 Subject: [Moin-devel] Re: Moin-devel digest, Vol 1 #517 - 6 msgs In-Reply-To: <20050105042730.0603F886C2@sc8-sf-spam1.sourceforge.net> References: <20050105042730.0603F886C2@sc8-sf-spam1.sourceforge.net> Message-ID: <5B9F6E0E-5F1A-11D9-A5D8-000A95B45AA0@actcom.net.il> On 5 Jan, 2005, at 6:25, moin-devel-request at lists.sourceforge.net wrote: > I have a bookmarks page that is supposed to list all of its children > pages. > [[Navigation(children,2)]] > > In order to perform Navigation(children,2), the operation is now > significantly slower than before. And it makes my hard drive go all > NUTS for like 20 seconds, when it is performed for the first time in a > while. > > It could be that the macro can be improved in performance, but it > could also be due to the new file/directory format of data/text, I am > not exactly sure. In 1.3, getting a page list is very expensive, about 5 times more expensive than in 1.2. Number of system page also grown to 556 from 324. The result is that getting a page list is 5-10 times slower on 1.3. Every operation that need a page list is effected. On my machine, getting page list takes about 0.06s in 1.2.4 (350 pages) and 0.6s in 1.3.2 dev (650 pages). Navigation macro get a page list once, but this does not explain 20s delay, unless your machine is very old or sick (or both), or you try to serve the wiki from the network instead of local disk. We will optimize this in 1.3.3 and remove the slowdown in 1.4. Best Regards, Nir Soffer From Nigel.Metheringham at dev.intechnology.co.uk Wed Jan 5 05:25:26 2005 From: Nigel.Metheringham at dev.intechnology.co.uk (Nigel Metheringham) Date: Wed Jan 5 05:25:26 2005 Subject: [Moin-devel] Re: Moin-devel digest, Vol 1 #517 - 6 msgs In-Reply-To: <5B9F6E0E-5F1A-11D9-A5D8-000A95B45AA0@actcom.net.il> References: <20050105042730.0603F886C2@sc8-sf-spam1.sourceforge.net> <5B9F6E0E-5F1A-11D9-A5D8-000A95B45AA0@actcom.net.il> Message-ID: <1104931467.14712.2.camel@angua.localnet> On Wed, 2005-01-05 at 15:04 +0200, Nir Soffer wrote: > In 1.3, getting a page list is very expensive, about 5 times more > expensive than in 1.2. Number of system page also grown to 556 from > 324. The result is that getting a page list is 5-10 times slower on > 1.3. Every operation that need a page list is effected. On my machine, > getting page list takes about 0.06s in 1.2.4 (350 pages) and 0.6s in > 1.3.2 dev (650 pages). Is this list cached? I have some pages in a FAQ hierarchy on one wiki (still at 1.2.4) which has a total of 4 Navigation macros in it (sibs and children in both top and bottom head/trailers). I can see this being more than a bit of a pain :-) Nigel. -- [ Nigel Metheringham Nigel.Metheringham at InTechnology.co.uk ] [ - Comments in this message are my own and not ITO opinion/policy - ] From 2004b at usenet.alexanderweb.de Wed Jan 5 08:21:59 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Wed Jan 5 08:21:59 2005 Subject: [Moin-devel] Re: reStructuredText support References: <41DB90C4.1080806@voxmea.net> Message-ID: <1hvl4ix37zsdt.dlg@usenet.alexanderweb.de> On Wed, 05 Jan 2005 00:01:24 -0700, Matthew Gilbert wrote: > I recently submitted a new rst.py to sourceforge. Its a drop in > replacement for the existing rst.py. It adds much better support for > reStructuredText than what currently ships with MoinMoin. I also > submitted a similar patch in September, and frankly this one is much > better. I was hoping at least to get some feedback as to what I can > improve to get this considered to be accepted. I have tested this quite > a bit and have a few users who have tested it without issue. I would > greatly appreciate anyone who would be willing to try it or who would be > willing to provide feedback. I think that we should include it in the main distribution. The only drawback is that it does not use the formatter (at least I guess so). Kind regards, Alexander From 2004b at usenet.alexanderweb.de Wed Jan 5 08:33:00 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Wed Jan 5 08:33:00 2005 Subject: [Moin-devel] Re: Re: Moin-devel digest, Vol 1 #517 - 6 msgs References: <20050105042730.0603F886C2@sc8-sf-spam1.sourceforge.net> <5B9F6E0E-5F1A-11D9-A5D8-000A95B45AA0@actcom.net.il> <1104931467.14712.2.camel@angua.localnet> Message-ID: <76rfhsq16r4w$.dlg@usenet.alexanderweb.de> On Wed, 05 Jan 2005 13:24:26 +0000, Nigel Metheringham wrote: > On Wed, 2005-01-05 at 15:04 +0200, Nir Soffer wrote: >> In 1.3, getting a page list is very expensive, about 5 times more >> expensive than in 1.2. Number of system page also grown to 556 from >> 324. The result is that getting a page list is 5-10 times slower on >> 1.3. Every operation that need a page list is effected. On my machine, >> getting page list takes about 0.06s in 1.2.4 (350 pages) and 0.6s in >> 1.3.2 dev (650 pages). > > Is this list cached? No, it is not. > I have some pages in a FAQ hierarchy on one wiki > (still at 1.2.4) which has a total of 4 Navigation macros in it (sibs > and children in both top and bottom head/trailers). I can see this > being more than a bit of a pain :-) I think that we should cache it per request. It might need some memory but the garbage collector should easily free that ... Kind regards, Alexander From gilbert at voxmea.net Wed Jan 5 09:03:23 2005 From: gilbert at voxmea.net (Matthew Gilbert) Date: Wed Jan 5 09:03:23 2005 Subject: [Moin-devel] Re: reStructuredText support In-Reply-To: <1hvl4ix37zsdt.dlg@usenet.alexanderweb.de> References: <41DB90C4.1080806@voxmea.net> <1hvl4ix37zsdt.dlg@usenet.alexanderweb.de> Message-ID: <41DC1D38.5060800@voxmea.net> Alexander Schremmer wrote: > On Wed, 05 Jan 2005 00:01:24 -0700, Matthew Gilbert wrote: > > >>I recently submitted a new rst.py to sourceforge. Its a drop in >>replacement for the existing rst.py. It adds much better support for >>reStructuredText than what currently ships with MoinMoin. I also >>submitted a similar patch in September, and frankly this one is much >>better. I was hoping at least to get some feedback as to what I can >>improve to get this considered to be accepted. I have tested this quite >>a bit and have a few users who have tested it without issue. I would >>greatly appreciate anyone who would be willing to try it or who would be >>willing to provide feedback. > > > I think that we should include it in the main distribution. The only > drawback is that it does not use the formatter (at least I guess so). You are correct, for many things it does not use the formatter. However, I've tried to use it whenever possible. Right now those cases are limited to things such as emphasis, code, strong, bullet_list, etc. Thanks! _matt From noreply at sourceforge.net Wed Jan 5 09:55:03 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 09:55:03 2005 Subject: [Moin-devel] [ moin-Bugs-1074301 ] Actions on the Get Infos page Message-ID: Bugs item #1074301, was opened at 2004-11-27 20:43 Message generated for change (Comment added) made by thomaswaldmann You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Actions on the Get Infos page Initial Comment: The list of actions in each line on the Get Infos page is separated by non-breaking spaces (nbsp entity). If the list get too long, it will collide with the menu or page border, as the browser is not authorized to split it in two lines. Replacing the nbsp between each action name by a regular space would solve this issue nicely. ---------------------------------------------------------------------- >Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 18:48 Message: Logged In: YES user_id=100649 iirc, the nbsp was put there on purpose. if replaced by a space, some browsers break the actions into 2 lines too early (instead of breaking the comment field) So I am not sure if that would be really an improvement. It is known that esp. the rightsidebar theme has multiple problems with non-breaking stuff to the left, those actions are only a very minor part of that problem. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:14:57 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:14:57 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None >Status: Closed Priority: 5 Submitted By: Markus Elfring (elfring) Assigned to: Nobody/Anonymous (nobody) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:15:50 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:15:50 2005 Subject: [Moin-devel] [ moin-Feature Requests-1063485 ] Multi-level numbered lists Message-ID: Feature Requests item #1063485, was opened at 2004-11-10 00:29 Message generated for change (Settings changed) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1063485&group_id=8482 Category: None Group: None Status: Open >Priority: 3 Submitted By: Allan Mertner (mertner) Assigned to: Nobody/Anonymous (nobody) Summary: Multi-level numbered lists Initial Comment: I need to produce a page that is a TOC of other pages - something like this: 1. IntroPage 2. StuffPage 2.1. DetailedStuffPage 2.2. MoreDetailsStuffPage 2.2.1. VeryDetailedStuffPage 2.3. ... etc. However, MoinMoin does not appear to like this - at the 2nd level, it adds too much line spacing and at the 3rd level it does not turn the items into
  • items in an
      at all. Any suggestions, please? ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2004-11-11 22:21 Message: Logged In: YES user_id=100649 Moin currently only support this style: 1. first item first level 2. second item first level 1. first item second level 2. second item second level 3. third item third level As this is no bug, I move it to feature request. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1063485&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:19:27 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:19:27 2005 Subject: [Moin-devel] [ moin-Feature Requests-748254 ] ISSN: Message-ID: Feature Requests item #748254, was opened at 2003-06-03 17:53 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=748254&group_id=8482 Category: None Group: None >Status: Closed Priority: 5 Submitted By: Jonathan Smith (jcwinnie) Assigned to: Nobody/Anonymous (nobody) Summary: ISSN: Initial Comment: Herr Hermann: I like the MoinMoin feature ISBN: Would it be possible to add ISSN: It might appeal to the academic community. Best regards, JonathanSmith ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:17 Message: Logged In: YES user_id=254738 Please add a working entry to the MoinMaster interwiki map if you still want to see this feature. Otherwise modify your local interwiki map. ---------------------------------------------------------------------- Comment By: Jonathan Smith (jcwinnie) Date: 2003-06-03 18:19 Message: Logged In: YES user_id=645989 The International Standard Serial Number (ISSN) is an internationally recognized identification number for serial publications. An ISSN is an eight-digit standard number that, once assigned, becomes a permanent attribute of the individual serial for as long as it is issued under a given title. Each time the title of a serial is changed, a new ISSN must be assigned. ISSN must never be reused for new or changed serial titles. ISSN are assigned by national and regional centres of the international ISSN Network. This system of international coordination ensures that each ISSN is unique to one serial publication. ISSN can be used wherever information on serials needs to be recorded or communicated. ---- I believe that U.S. Library of Congress Online Catalog could be used for this purpose. Example: http://catalog.loc.gov/cgi-bin/Pwebrecon.cgi? Search_Arg=1502- 8003&Search_Code=KNUM&PID=8907&SEQ=200306031 21246&CNT=25&HIST=1 ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=748254&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:24:22 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:24:22 2005 Subject: [Moin-devel] [ moin-Bugs-1074301 ] Actions on the Get Infos page Message-ID: Bugs item #1074301, was opened at 2004-11-27 20:43 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Actions on the Get Infos page Initial Comment: The list of actions in each line on the Get Infos page is separated by non-breaking spaces (nbsp entity). If the list get too long, it will collide with the menu or page border, as the browser is not authorized to split it in two lines. Replacing the nbsp between each action name by a regular space would solve this issue nicely. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:23 Message: Logged In: YES user_id=254738 Is this problem relevant for the modern theme? Or did you just see it with the rightsidebar theme? The last one does not look nice if your screen resolution is too low. That is a known problem. Please reopen this bug if it is relevant for the modern theme. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 18:48 Message: Logged In: YES user_id=100649 iirc, the nbsp was put there on purpose. if replaced by a space, some browsers break the actions into 2 lines too early (instead of breaking the comment field) So I am not sure if that would be really an improvement. It is known that esp. the rightsidebar theme has multiple problems with non-breaking stuff to the left, those actions are only a very minor part of that problem. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:25:02 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:25:02 2005 Subject: [Moin-devel] [ moin-Bugs-1074301 ] Actions on the Get Infos page Message-ID: Bugs item #1074301, was opened at 2004-11-27 20:43 Message generated for change (Settings changed) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 Category: None Group: None >Status: Closed Resolution: None Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Actions on the Get Infos page Initial Comment: The list of actions in each line on the Get Infos page is separated by non-breaking spaces (nbsp entity). If the list get too long, it will collide with the menu or page border, as the browser is not authorized to split it in two lines. Replacing the nbsp between each action name by a regular space would solve this issue nicely. ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:23 Message: Logged In: YES user_id=254738 Is this problem relevant for the modern theme? Or did you just see it with the rightsidebar theme? The last one does not look nice if your screen resolution is too low. That is a known problem. Please reopen this bug if it is relevant for the modern theme. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 18:48 Message: Logged In: YES user_id=100649 iirc, the nbsp was put there on purpose. if replaced by a space, some browsers break the actions into 2 lines too early (instead of breaking the comment field) So I am not sure if that would be really an improvement. It is known that esp. the rightsidebar theme has multiple problems with non-breaking stuff to the left, those actions are only a very minor part of that problem. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:32:00 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:32:00 2005 Subject: [Moin-devel] [ moin-Feature Requests-708690 ] moin-dump should optionally include attachments Message-ID: Feature Requests item #708690, was opened at 2003-03-24 09:19 Message generated for change (Comment added) made by thomaswaldmann You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=708690&group_id=8482 Category: None Group: None >Status: Deleted Priority: 5 Submitted By: Robert Pollak (jondo) Assigned to: Nobody/Anonymous (nobody) Summary: moin-dump should optionally include attachments Initial Comment: .. e.g. pictures inlined with "attachment:filename". ---------------------------------------------------------------------- >Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 19:31 Message: Logged In: YES user_id=100649 one of the next major releases will have no "attachments" any more. files, pages, etc. will all be stored as first class items (including acls and revisions). So that issue will then be gone, too, I hope. At least we won't fix it before. :) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=708690&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:36:27 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:36:27 2005 Subject: [Moin-devel] [ moin-Bugs-1074304 ] Attaching a file and giving mime type text/plain changes the Message-ID: Bugs item #1074304, was opened at 2004-11-27 20:46 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074304&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Duplicate Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Attaching a file and giving mime type text/plain changes the Initial Comment: If you attach a .po file to a page, and give it MIME type text/plain, an .conf extension is added to the file. Example: File name: fr.po MIME type: text/plain -> The file is attached without problem but its name is: fr.po.conf ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:35 Message: Logged In: YES user_id=254738 I can confirm this bug. Please switch to this wiki page for further bug tracking: http://moinmoin.wikiwikiweb.de:8000/MoinMoinBugs/AttachFileAddsStupidExtensions I will close it because it will be a duplicate of the bug I filed in our wiki. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074304&group_id=8482 From noreply at sourceforge.net Wed Jan 5 10:45:01 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 10:45:01 2005 Subject: [Moin-devel] [ moin-Bugs-1096637 ] PLEASE REPORT BUGS ON http://tinyurl.com/6nku2 Message-ID: Bugs item #1096637, was opened at 2005-01-05 19:44 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 7 Submitted By: Alexander Schremmer (alexanderweb) Assigned to: Alexander Schremmer (alexanderweb) Summary: PLEASE REPORT BUGS ON http://tinyurl.com/6nku2 Initial Comment: Please report new bugs on our wiki: http://moinmoin.wikiwikiweb.de/MoinMoinBugs Thanks. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:10:03 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:10:03 2005 Subject: [Moin-devel] [ moin-Patches-1051035 ] A Better ?test Message-ID: Patches item #1051035, was opened at 2004-10-20 22:49 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1051035&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Rejected Priority: 5 Submitted By: Brad Clarke (yourgod) Assigned to: Nobody/Anonymous (nobody) Summary: A Better ?test Initial Comment: I've made a patch that keeps the moin.cgi?test output from failing before it can start outputting. It also outputs the server's python version before doing anything else. These were both done in attempting to install on sf.net's web server only to find their python version is too old to support your software: http://yourgod.sourceforge.net/cgi-bin/moin.cgi?test ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:09 Message: Logged In: YES user_id=254738 Unfortunately, test code was moved out of the .cgi file in 1.3. Please update the docs if you think that the advises about the required Python version are not clear enough. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1051035&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:11:27 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:11:27 2005 Subject: [Moin-devel] [ moin-Patches-1030782 ] rst parser using docutils html4css1 writer Message-ID: Patches item #1030782, was opened at 2004-09-19 20:22 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1030782&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Duplicate Priority: 5 Submitted By: Matthew Gilbert (mmgilbe) Assigned to: Nobody/Anonymous (nobody) Summary: rst parser using docutils html4css1 writer Initial Comment: Alternative rst parser using the docutils code rather than the moinmoin code for formatting. In my test cases handles more of the docutils structure than the existing code. Supports all of the moinmoin's linking syntax but using restructuredtext directives. Tested with docutils 0.3.5. So for now, requires docutils 0.3.5. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:10 Message: Logged In: YES user_id=254738 Thanks for submitting another patch in the patch tracker. This is a duplicate therefore. ---------------------------------------------------------------------- Comment By: Ollie Rutherfurd (orutherfurd) Date: 2004-12-02 20:04 Message: Logged In: YES user_id=265596 I'd love to see this in a future release of MoinMoin. In case providing feedback increases the chance of this being merged, I've been using it with 1.3rc1, and it's working nicely. I wasn't able to apply the patch to rst.py -- maybe a line-ending issue, so I just overwrote the existing file with everything from the "+" lines. ---------------------------------------------------------------------- Comment By: Matthew Gilbert (mmgilbe) Date: 2004-09-27 00:01 Message: Logged In: YES user_id=1002483 Tested with docutils 0.3.3 which I see there is a package for in debian unstable? Thomas, will you install something from unstable? The url is: http://packages.debian.org/unstable/python/python-docutils It seems to work fine with 0.3.3. Will you try the patch the now? Thanks ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2004-09-25 04:16 Message: Logged In: YES user_id=100649 as long as the efforts on rst require more than what's in debian, I can't even test it. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1030782&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:13:23 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:13:23 2005 Subject: [Moin-devel] [ moin-Patches-969170 ] Patch to add hyperlinks to reStructured text parser Message-ID: Patches item #969170, was opened at 2004-06-08 21:56 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=969170&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Fixed Priority: 5 Submitted By: Ralf Schlatterbeck (schlatterbeck) Assigned to: Nobody/Anonymous (nobody) Summary: Patch to add hyperlinks to reStructured text parser Initial Comment: Currently (release 1.1, but apparently also in 1.2.2) the reStructured Text parser in parser/rst.py does not render links correctly. This patch fixes the links by - applying some transformations to the read-in rst - then rendering the links accordingly by calling appropriate api functions of the formatter. Patch is against 1.1 but not much has changed in 1.2.2, will provide patch against 1.2.2 if needed. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:12 Message: Logged In: YES user_id=254738 Please check if the issue is fixed in the new rst parser of MoinMoin 1.3.2. If not, submit a patch against the new code. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=969170&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:16:21 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:16:21 2005 Subject: [Moin-devel] [ moin-Patches-801787 ] Added strike-through wiki markup Message-ID: Patches item #801787, was opened at 2003-09-06 23:43 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=801787&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Out of Date Priority: 5 Submitted By: Jason Dillon (user57) Assigned to: Nobody/Anonymous (nobody) Summary: Added strike-through wiki markup Initial Comment: This patch adds strike-through wiki markup. For example: ---Strike this stuff--- Not striked --- Strike all of this --- I used '---' to avoid problems when people sign names like: --jason And also because --- is not a horizontal rule. Basically I need a way to mark bits in the wiki as crap, but not delete them right off the bat... ya know? ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:15 Message: Logged In: YES user_id=254738 Please supply a patch against the latest release if you want to see this in the official code base. And discuss on the page MoinMoin:MarkupProposals about the markup which should be used. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2004-09-25 03:59 Message: Logged In: YES user_id=100649 See comments there: http://moinmoin.wikiwikiweb.de/MarkupProposals?action=diff&date2=0&date1=1094038974 ---------------------------------------------------------------------- Comment By: Jason Dillon (user57) Date: 2003-09-07 12:03 Message: Logged In: YES user_id=15045 The regex used before was not valid, as it caught rules as well. The latest patch resolves this. ---------------------------------------------------------------------- Comment By: Jason Dillon (user57) Date: 2003-09-07 02:14 Message: Logged In: YES user_id=15045 Looks like there is a problem with this... some pages get everything striked... :-| ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=801787&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:17:24 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:17:24 2005 Subject: [Moin-devel] [ moin-Patches-1092353 ] Better reStructuredText support Message-ID: Patches item #1092353, was opened at 2004-12-28 20:17 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 Category: None Group: None Status: Open >Resolution: Accepted Priority: 5 Submitted By: Matthew Gilbert (mmgilbe) >Assigned to: Alexander Schremmer (alexanderweb) Summary: Better reStructuredText support Initial Comment: Updated version of my previous patch. This adds much better reStructuredText support to MoinMoin. It is a drop in replacement for the existing rst.py. This version is significantly refactored and has numerous bug fixes. It supports MoinMoin 1.2.X and 1.3.X and requires docutils version 0.3.3 or later. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:16 Message: Logged In: YES user_id=254738 This will be committed into the main code base and should be available in MoinMoin 1.3.2 ---------------------------------------------------------------------- Comment By: Ollie Rutherfurd (orutherfurd) Date: 2005-01-04 16:00 Message: Logged In: YES user_id=265596 Please consider accepting this patch -- it greatly improves the existing reST support. I've been using in without problems on 1.3.1. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 From noreply at sourceforge.net Wed Jan 5 11:59:03 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 11:59:03 2005 Subject: [Moin-devel] [ moin-Patches-736885 ] Identify users using client certificates Message-ID: Patches item #736885, was opened at 2003-05-13 09:45 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=736885&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Martin v. L?wis (loewis) Assigned to: Thomas Waldmann (thomaswaldmann) Summary: Identify users using client certificates Initial Comment: This patch allows to identifiy Moin users using SSL client certificates. Specifically, it uses the common name and the email address from the cerificate's subject's distinguished name. Cookies and Moin user ids are still used, and finding users works like this 1. If there is a cookie, use that 2. If there is no cookie, iterate over all users, and try to find one with the same email address or where the X.509 common name is the same as the Moin user name. 3. If no user is found, but either the email address or the common name is set, create a new user. This patch works only with Apache mod_ssl, as it relies on the environment variables SSL_CLIENT_S_DN* being set. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:58 Message: Logged In: YES user_id=254738 So, will someone adapt the patch to 1.3? ---------------------------------------------------------------------- Comment By: Martin v. L?wis (loewis) Date: 2004-09-25 09:07 Message: Logged In: YES user_id=21627 I can certainly test it, yes. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2004-09-25 04:13 Message: Logged In: YES user_id=100649 Martin, I don't think any of the devels uses client certificates, so we could not test it. Would you test it if we include your patch into 1.2.x (and later, into 1.3)? I would make the necessary changes to adapt it to the current version (like using request object). ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=736885&group_id=8482 From noreply at sourceforge.net Wed Jan 5 13:08:00 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 13:08:00 2005 Subject: [Moin-devel] [ moin-Bugs-1074301 ] Actions on the Get Infos page Message-ID: Bugs item #1074301, was opened at 2004-11-27 20:43 Message generated for change (Comment added) made by fevrier_o_pekoe You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 Category: None Group: None >Status: Open Resolution: None Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Actions on the Get Infos page Initial Comment: The list of actions in each line on the Get Infos page is separated by non-breaking spaces (nbsp entity). If the list get too long, it will collide with the menu or page border, as the browser is not authorized to split it in two lines. Replacing the nbsp between each action name by a regular space would solve this issue nicely. ---------------------------------------------------------------------- >Comment By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Date: 2005-01-05 22:07 Message: Logged In: YES user_id=1167328 I have seen this bug with all available themes shipped with MoinMoin. I am therefore re-opening it. As a result of this bug, the "Get Info" page tends to be too long to be seen without using the horizontal scroll bar. I cannot comment on the broken browsers issues, but with a regular browser, it seems to worsen the situation rather than improving it. ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:23 Message: Logged In: YES user_id=254738 Is this problem relevant for the modern theme? Or did you just see it with the rightsidebar theme? The last one does not look nice if your screen resolution is too low. That is a known problem. Please reopen this bug if it is relevant for the modern theme. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 18:48 Message: Logged In: YES user_id=100649 iirc, the nbsp was put there on purpose. if replaced by a space, some browsers break the actions into 2 lines too early (instead of breaking the comment field) So I am not sure if that would be really an improvement. It is known that esp. the rightsidebar theme has multiple problems with non-breaking stuff to the left, those actions are only a very minor part of that problem. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 From noreply at sourceforge.net Wed Jan 5 13:28:04 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 13:28:04 2005 Subject: [Moin-devel] [ moin-Bugs-1074301 ] Actions on the Get Infos page Message-ID: Bugs item #1074301, was opened at 2004-11-27 21:43 Message generated for change (Comment added) made by nirs You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Assigned to: Nobody/Anonymous (nobody) Summary: Actions on the Get Infos page Initial Comment: The list of actions in each line on the Get Infos page is separated by non-breaking spaces (nbsp entity). If the list get too long, it will collide with the menu or page border, as the browser is not authorized to split it in two lines. Replacing the nbsp between each action name by a regular space would solve this issue nicely. ---------------------------------------------------------------------- Comment By: Nir Soffer (nirs) Date: 2005-01-05 23:27 Message: Logged In: YES user_id=832344 This is not a bug but a design that needs wide window. If your window is narrow and you use big fonts, you will have to use the horizontal scroll bar. MoinMoin uses the font you setup in your browser. This problem is typical with rightsidebar theme, which does not work well with narrow windows and big fonts. I think we can replace the nbsp with regular space as the design - whitespace in this case should be set by the css. ---------------------------------------------------------------------- Comment By: Jean-Philippe Gu?rard (fevrier_o_pekoe) Date: 2005-01-05 23:07 Message: Logged In: YES user_id=1167328 I have seen this bug with all available themes shipped with MoinMoin. I am therefore re-opening it. As a result of this bug, the "Get Info" page tends to be too long to be seen without using the horizontal scroll bar. I cannot comment on the broken browsers issues, but with a regular browser, it seems to worsen the situation rather than improving it. ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:23 Message: Logged In: YES user_id=254738 Is this problem relevant for the modern theme? Or did you just see it with the rightsidebar theme? The last one does not look nice if your screen resolution is too low. That is a known problem. Please reopen this bug if it is relevant for the modern theme. ---------------------------------------------------------------------- Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-05 19:48 Message: Logged In: YES user_id=100649 iirc, the nbsp was put there on purpose. if replaced by a space, some browsers break the actions into 2 lines too early (instead of breaking the comment field) So I am not sure if that would be really an improvement. It is known that esp. the rightsidebar theme has multiple problems with non-breaking stuff to the left, those actions are only a very minor part of that problem. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1074301&group_id=8482 From noreply at sourceforge.net Wed Jan 5 13:48:33 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 13:48:33 2005 Subject: [Moin-devel] [ moin-Bugs-1096637 ] PLEASE REPORT BUGS ON http://tinyurl.com/6nku2 Message-ID: Bugs item #1096637, was opened at 2005-01-05 20:44 Message generated for change (Comment added) made by nirs You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 7 Submitted By: Alexander Schremmer (alexanderweb) Assigned to: Alexander Schremmer (alexanderweb) Summary: PLEASE REPORT BUGS ON http://tinyurl.com/6nku2 Initial Comment: Please report new bugs on our wiki: http://moinmoin.wikiwikiweb.de/MoinMoinBugs Thanks. ---------------------------------------------------------------------- Comment By: Nir Soffer (nirs) Date: 2005-01-05 23:47 Message: Logged In: YES user_id=832344 This is not a bug :-) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 From noreply at sourceforge.net Wed Jan 5 14:06:05 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Wed Jan 5 14:06:05 2005 Subject: [Moin-devel] [ moin-Patches-1092353 ] Better reStructuredText support Message-ID: Patches item #1092353, was opened at 2004-12-28 20:17 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 Category: None Group: None Status: Open Resolution: Accepted Priority: 5 Submitted By: Matthew Gilbert (mmgilbe) Assigned to: Alexander Schremmer (alexanderweb) Summary: Better reStructuredText support Initial Comment: Updated version of my previous patch. This adds much better reStructuredText support to MoinMoin. It is a drop in replacement for the existing rst.py. This version is significantly refactored and has numerous bug fixes. It supports MoinMoin 1.2.X and 1.3.X and requires docutils version 0.3.3 or later. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 23:04 Message: Logged In: YES user_id=254738 I am waiting for documentation update/validation of the MoinMaster documentation ... Then I will close this patch tracker item. ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 20:16 Message: Logged In: YES user_id=254738 This will be committed into the main code base and should be available in MoinMoin 1.3.2 ---------------------------------------------------------------------- Comment By: Ollie Rutherfurd (orutherfurd) Date: 2005-01-04 16:00 Message: Logged In: YES user_id=265596 Please consider accepting this patch -- it greatly improves the existing reST support. I've been using in without problems on 1.3.1. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1092353&group_id=8482 From noreply at sourceforge.net Thu Jan 6 10:37:33 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Thu Jan 6 10:37:33 2005 Subject: [Moin-devel] [ moin-Bugs-1097377 ] Umlaut-in-Link problem 1.3.1 Message-ID: Bugs item #1097377, was opened at 2005-01-06 19:36 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1097377&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Martin Jahr (martinjahr) Assigned to: Nobody/Anonymous (nobody) Summary: Umlaut-in-Link problem 1.3.1 Initial Comment: When installing Wiki 1.3.1, Links with Umlauts dont work properly in my installation. While the pages show the Links correctly, the URL seems to break when referring to a page containing escaped characters. So will the page link displaying as "Aktuelle?nderungen" lead to a URL "...mywiki/Aktuelle%c3%84nderungen" which itself points to the page "Aktuelle?derungen". I tested this with these environments: Browser -------- IE 6.0SP1 Mozilla/Firefox 1.? Web Server ----------- IIS 6.0 Apache 5.0.22 OS -- Win2k Server English Win2003 Server German Python ------ Python23 Python24 all with the same results. charset is unchanged everywhere in the config files (iso) Other observation: action=test shows 2 failed assertions - escape html markup in getText formatted call with link (expected "<escape;>" but got "

      <a class="nonexistent" href="...mywiki/%3cescape% 3e">&lt;escape&gt;</a>

      ") - escape html markup in getText formatted call (expected "<escape;>" but got "

      &lt;escape&gt;

      ") Any idea? Looks like a messed-up escape mechanism somewhere... ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1097377&group_id=8482 From noreply at sourceforge.net Thu Jan 6 11:12:08 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Thu Jan 6 11:12:08 2005 Subject: [Moin-devel] [ moin-Bugs-1096637 ] PLEASE REPORT BUGS ONLY ON http://tinyurl.com/6nku2 Message-ID: Bugs item #1096637, was opened at 2005-01-05 19:44 Message generated for change (Settings changed) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 7 Submitted By: Alexander Schremmer (alexanderweb) Assigned to: Alexander Schremmer (alexanderweb) >Summary: PLEASE REPORT BUGS ONLY ON http://tinyurl.com/6nku2 Initial Comment: Please report new bugs on our wiki: http://moinmoin.wikiwikiweb.de/MoinMoinBugs Thanks. ---------------------------------------------------------------------- Comment By: Nir Soffer (nirs) Date: 2005-01-05 22:47 Message: Logged In: YES user_id=832344 This is not a bug :-) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1096637&group_id=8482 From noreply at sourceforge.net Thu Jan 6 11:12:26 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Thu Jan 6 11:12:26 2005 Subject: [Moin-devel] [ moin-Bugs-1097377 ] Umlaut-in-Link problem 1.3.1 Message-ID: Bugs item #1097377, was opened at 2005-01-06 19:36 Message generated for change (Settings changed) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1097377&group_id=8482 Category: None Group: None >Status: Closed >Resolution: Duplicate Priority: 5 Submitted By: Martin Jahr (martinjahr) Assigned to: Nobody/Anonymous (nobody) Summary: Umlaut-in-Link problem 1.3.1 Initial Comment: When installing Wiki 1.3.1, Links with Umlauts dont work properly in my installation. While the pages show the Links correctly, the URL seems to break when referring to a page containing escaped characters. So will the page link displaying as "Aktuelle?nderungen" lead to a URL "...mywiki/Aktuelle%c3%84nderungen" which itself points to the page "Aktuelle?derungen". I tested this with these environments: Browser -------- IE 6.0SP1 Mozilla/Firefox 1.? Web Server ----------- IIS 6.0 Apache 5.0.22 OS -- Win2k Server English Win2003 Server German Python ------ Python23 Python24 all with the same results. charset is unchanged everywhere in the config files (iso) Other observation: action=test shows 2 failed assertions - escape html markup in getText formatted call with link (expected "<escape;>" but got "

      <a class="nonexistent" href="...mywiki/%3cescape% 3e">&lt;escape&gt;</a>

      ") - escape html markup in getText formatted call (expected "<escape;>" but got "

      &lt;escape&gt;

      ") Any idea? Looks like a messed-up escape mechanism somewhere... ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=108482&aid=1097377&group_id=8482 From noreply at sourceforge.net Fri Jan 7 02:29:08 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Fri Jan 7 02:29:08 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by elfring You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None >Status: Open Priority: 5 Submitted By: Markus Elfring (elfring) >Assigned to: Alexander Schremmer (alexanderweb) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Markus Elfring (elfring) Date: 2005-01-07 11:27 Message: Logged In: YES user_id=572001 Can you imagine a similar API for Python? Would you like to reuse anything from other software? http://www.magnolia.info/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Fri Jan 7 07:49:23 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Fri Jan 7 07:49:23 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Markus Elfring (elfring) Assigned to: Alexander Schremmer (alexanderweb) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-07 16:48 Message: Logged In: YES user_id=254738 I do not know any. You might want to look for one. But first, you should sketch the advantages of being integrated into such a framework (optionally). I think that we will never require such a beast :-) ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-07 11:27 Message: Logged In: YES user_id=572001 Can you imagine a similar API for Python? Would you like to reuse anything from other software? http://www.magnolia.info/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Mon Jan 10 09:17:05 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Mon Jan 10 09:17:05 2005 Subject: [Moin-devel] [ moin-Patches-1099576 ] Automatic account creation Message-ID: Patches item #1099576, was opened at 2005-01-10 17:16 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1099576&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Brian E. Gallew (geekatcmu) Assigned to: Nobody/Anonymous (nobody) Summary: Automatic account creation Initial Comment: Premise: http_auth_enabled isn't very interesting (as is) when all of your users are authenticated via a trusted mechanism as they all must go through a second sign-on process (admittedly, only once). My patch includes a couple bits: 1) adds an auto_account variable to multiconfig.py (defaults to 0 of course). 2) Modifies User.__init__ in user.py such that, if auto_account and http_auth_enabled are both set, default some user information and save it. 3) Modifies normalizeName in user.py such that "." and "@" are permissible characters in usernames. This is necessary for my environment because we use mod_pubcookie for authentication. mod_pubcookie returns userids like "geek at ANDREW.CMU.EDU" (e.g. the Kerberos domain is part of the id). If this patch is accepted I'll update the appropriate documentation. If the documentation has to come first, let me know and I'll deal with it. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1099576&group_id=8482 From noreply at sourceforge.net Mon Jan 10 15:17:00 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Mon Jan 10 15:17:00 2005 Subject: [Moin-devel] [ moin-Feature Requests-1099811 ] Author information in RSS feed Message-ID: Feature Requests item #1099811, was opened at 2005-01-11 12:16 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Simon Lieschke (slieschke) Assigned to: Nobody/Anonymous (nobody) Summary: Author information in RSS feed Initial Comment: It would be nice to include author information in the recent changes RSS feed. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 From noreply at sourceforge.net Tue Jan 11 05:40:11 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 11 05:40:11 2005 Subject: [Moin-devel] [ moin-Feature Requests-1099811 ] Author information in RSS feed Message-ID: Feature Requests item #1099811, was opened at 2005-01-11 00:16 Message generated for change (Comment added) made by twerschlein You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Simon Lieschke (slieschke) Assigned to: Nobody/Anonymous (nobody) Summary: Author information in RSS feed Initial Comment: It would be nice to include author information in the recent changes RSS feed. ---------------------------------------------------------------------- Comment By: Thomas Werschlein (twerschlein) Date: 2005-01-11 14:39 Message: Logged In: YES user_id=679755 adding the line: handler.simpleNode(('dc', 'creator'), edname) after handler.endNode(('dc', 'contributor')) in $SITE_PACKAGES/MoinMoin/action/rss_rc.py does the job. However, Nick Bradbury, the author of FeedDemon states here (http://www.bradsoft.com/forums/shwmessage.aspx?forumid=22&messageid=5650#bm5658), "Although dc:contributor is a valid Dublin Core element, my understanding is that it denotes someone other than the author who contributed to the item. The author should be specified using the dc:creator element." Do we have the information about the initial author of a page in MoinMoin? But I agree with you, that having the last editor as "author" is still better than having no author at all. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 From noreply at sourceforge.net Tue Jan 11 06:13:10 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 11 06:13:10 2005 Subject: [Moin-devel] [ moin-Feature Requests-1099811 ] Author information in RSS feed Message-ID: Feature Requests item #1099811, was opened at 2005-01-11 12:16 Message generated for change (Comment added) made by slieschke You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Simon Lieschke (slieschke) Assigned to: Nobody/Anonymous (nobody) Summary: Author information in RSS feed Initial Comment: It would be nice to include author information in the recent changes RSS feed. ---------------------------------------------------------------------- >Comment By: Simon Lieschke (slieschke) Date: 2005-01-12 03:12 Message: Logged In: YES user_id=578756 Hmm, I would think the creator line you gave should always be called if the recent changes feed has the ddiffs=1 query parameter. In this case each item in the feed represents a wiki page revision. The author/creator of the item clearly is the person who made edits to the page If ddiffs=1 is not present then yes, it doesn't make sense to include the creator line (unless perhaps the initial creator is available) as each item in the feed represents a wiki page. Cool. Thanks anyway. ---------------------------------------------------------------------- Comment By: Thomas Werschlein (twerschlein) Date: 2005-01-12 02:39 Message: Logged In: YES user_id=679755 adding the line: handler.simpleNode(('dc', 'creator'), edname) after handler.endNode(('dc', 'contributor')) in $SITE_PACKAGES/MoinMoin/action/rss_rc.py does the job. However, Nick Bradbury, the author of FeedDemon states here (http://www.bradsoft.com/forums/shwmessage.aspx?forumid=22&messageid=5650#bm5658), "Although dc:contributor is a valid Dublin Core element, my understanding is that it denotes someone other than the author who contributed to the item. The author should be specified using the dc:creator element." Do we have the information about the initial author of a page in MoinMoin? But I agree with you, that having the last editor as "author" is still better than having no author at all. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 From noreply at sourceforge.net Tue Jan 11 06:43:07 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 11 06:43:07 2005 Subject: [Moin-devel] [ moin-Feature Requests-1099811 ] Author information in RSS feed Message-ID: Feature Requests item #1099811, was opened at 2005-01-11 00:16 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Simon Lieschke (slieschke) Assigned to: Nobody/Anonymous (nobody) Summary: Author information in RSS feed Initial Comment: It would be nice to include author information in the recent changes RSS feed. ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-11 15:42 Message: Logged In: YES user_id=254738 I do not think that the original creator of a wiki page is interesting for viewers in the RecentChanges view, especially if the page is very old. IMHO we should add editor information to the RSS output for 1.3.2 ---------------------------------------------------------------------- Comment By: Simon Lieschke (slieschke) Date: 2005-01-11 15:12 Message: Logged In: YES user_id=578756 Hmm, I would think the creator line you gave should always be called if the recent changes feed has the ddiffs=1 query parameter. In this case each item in the feed represents a wiki page revision. The author/creator of the item clearly is the person who made edits to the page If ddiffs=1 is not present then yes, it doesn't make sense to include the creator line (unless perhaps the initial creator is available) as each item in the feed represents a wiki page. Cool. Thanks anyway. ---------------------------------------------------------------------- Comment By: Thomas Werschlein (twerschlein) Date: 2005-01-11 14:39 Message: Logged In: YES user_id=679755 adding the line: handler.simpleNode(('dc', 'creator'), edname) after handler.endNode(('dc', 'contributor')) in $SITE_PACKAGES/MoinMoin/action/rss_rc.py does the job. However, Nick Bradbury, the author of FeedDemon states here (http://www.bradsoft.com/forums/shwmessage.aspx?forumid=22&messageid=5650#bm5658), "Although dc:contributor is a valid Dublin Core element, my understanding is that it denotes someone other than the author who contributed to the item. The author should be specified using the dc:creator element." Do we have the information about the initial author of a page in MoinMoin? But I agree with you, that having the last editor as "author" is still better than having no author at all. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1099811&group_id=8482 From noreply at sourceforge.net Tue Jan 11 07:11:23 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 11 07:11:23 2005 Subject: [Moin-devel] [ moin-Feature Requests-1100206 ] twisted reverse proxy support Message-ID: Feature Requests item #1100206, was opened at 2005-01-11 16:10 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1100206&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Thomas Werschlein (twerschlein) Assigned to: Nobody/Anonymous (nobody) Summary: twisted reverse proxy support Initial Comment: I run twistedmoin behind an Apache reverse proxy using mod_proxy. Everything works fine, but the RSS feed shows the internal host/port as source instead of the published URL (i.e. the reverse proxy host/port). adding these two lines to MoinRequest(server.Request): if getattr(config, 'proxyhost', None) and getattr(config, 'proxyport', None): self.setHost(config.proxyhost, config.proxyport, ssl=getattr(config, 'proxyssl', 0)) after self.host = self.channel.transport.getHost() in MoinMoin/server/twistedmoin.py seems to solve the problem. Obviously, this is a quick and dirty hack and I am not sure, if this has some bad side effects I am not aware of. I am also not sure, if it is the right place to do it. And, of course it would be better to include some default settings in the Config class for the new config attributes proxyhost, proxyport and proxyssl instead of the getattr quirk. BTW, this is the docstring of setHost in twisted/protocols/http.py: """Change the host and port the request thinks it's using. This method is useful for working with reverse HTTP proxies (e.g. both Squid and Apache's mod_proxy can do this), then the address the HTTP client is using is different than the one we're listening on. For example, Apache may be listening on https://www.example.com, and then forwarding requests to http://localhost:8080, but we don't want HTML produced by Twisted to say 'http://localhost:8080', they should say 'https://www.example.com', so we do:: request.setHost('www.example.com', 443, ssl=1) This method is experimental. """ ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1100206&group_id=8482 From noreply at sourceforge.net Fri Jan 21 01:19:24 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Fri Jan 21 01:19:24 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by elfring You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Markus Elfring (elfring) Assigned to: Alexander Schremmer (alexanderweb) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Markus Elfring (elfring) Date: 2005-01-21 10:18 Message: Logged In: YES user_id=572001 The interest seems to be growing... I guess that the more examples are available the easier it will become to map or port an approach to the Python environment. http://phpCR.org/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-07 16:48 Message: Logged In: YES user_id=254738 I do not know any. You might want to look for one. But first, you should sketch the advantages of being integrated into such a framework (optionally). I think that we will never require such a beast :-) ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-07 11:27 Message: Logged In: YES user_id=572001 Can you imagine a similar API for Python? Would you like to reuse anything from other software? http://www.magnolia.info/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Fri Jan 21 05:53:24 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Fri Jan 21 05:53:24 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Markus Elfring (elfring) Assigned to: Alexander Schremmer (alexanderweb) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-21 14:52 Message: Logged In: YES user_id=254738 Maybe you should look for examples written in Python. And "sketch the advantages of being integrated into such a framework". ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-21 10:18 Message: Logged In: YES user_id=572001 The interest seems to be growing... I guess that the more examples are available the easier it will become to map or port an approach to the Python environment. http://phpCR.org/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-07 16:48 Message: Logged In: YES user_id=254738 I do not know any. You might want to look for one. But first, you should sketch the advantages of being integrated into such a framework (optionally). I think that we will never require such a beast :-) ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-07 11:27 Message: Logged In: YES user_id=572001 Can you imagine a similar API for Python? Would you like to reuse anything from other software? http://www.magnolia.info/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Fri Jan 21 07:14:13 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Fri Jan 21 07:14:13 2005 Subject: [Moin-devel] [ moin-Feature Requests-863361 ] Standard Content Storage API Message-ID: Feature Requests item #863361, was opened at 2003-12-20 05:07 Message generated for change (Comment added) made by elfring You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Markus Elfring (elfring) Assigned to: Alexander Schremmer (alexanderweb) Summary: Standard Content Storage API Initial Comment: Would you like to support content storage independence? Content Repository http://jcp.org/en/jsr/detail?id=170 ---------------------------------------------------------------------- >Comment By: Markus Elfring (elfring) Date: 2005-01-21 16:13 Message: Logged In: YES user_id=572001 I don't know any Python examples so far. Can you see the advantages from this approach? http://incubator.apache.org/jackrabbit/arch/overview.html ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-21 14:52 Message: Logged In: YES user_id=254738 Maybe you should look for examples written in Python. And "sketch the advantages of being integrated into such a framework". ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-21 10:18 Message: Logged In: YES user_id=572001 The interest seems to be growing... I guess that the more examples are available the easier it will become to map or port an approach to the Python environment. http://phpCR.org/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-07 16:48 Message: Logged In: YES user_id=254738 I do not know any. You might want to look for one. But first, you should sketch the advantages of being integrated into such a framework (optionally). I think that we will never require such a beast :-) ---------------------------------------------------------------------- Comment By: Markus Elfring (elfring) Date: 2005-01-07 11:27 Message: Logged In: YES user_id=572001 Can you imagine a similar API for Python? Would you like to reuse anything from other software? http://www.magnolia.info/ ---------------------------------------------------------------------- Comment By: Alexander Schremmer (alexanderweb) Date: 2005-01-05 19:12 Message: Logged In: YES user_id=254738 It looks quite Java specific. In 1.4, there will be an abstracted storage interface which lets you plug in your own storage system. Maybe that will help. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=863361&group_id=8482 From noreply at sourceforge.net Sat Jan 22 05:33:17 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Sat Jan 22 05:33:17 2005 Subject: [Moin-devel] [ moin-Patches-1100206 ] Twisted reverse proxy support Message-ID: Patches item #1100206, was opened at 2005-01-11 16:10 Message generated for change (Settings changed) made by alexanderweb You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1100206&group_id=8482 Category: None Group: None Status: Open Resolution: None >Priority: 6 Submitted By: Thomas Werschlein (twerschlein) Assigned to: Nobody/Anonymous (nobody) >Summary: Twisted reverse proxy support Initial Comment: I run twistedmoin behind an Apache reverse proxy using mod_proxy. Everything works fine, but the RSS feed shows the internal host/port as source instead of the published URL (i.e. the reverse proxy host/port). adding these two lines to MoinRequest(server.Request): if getattr(config, 'proxyhost', None) and getattr(config, 'proxyport', None): self.setHost(config.proxyhost, config.proxyport, ssl=getattr(config, 'proxyssl', 0)) after self.host = self.channel.transport.getHost() in MoinMoin/server/twistedmoin.py seems to solve the problem. Obviously, this is a quick and dirty hack and I am not sure, if this has some bad side effects I am not aware of. I am also not sure, if it is the right place to do it. And, of course it would be better to include some default settings in the Config class for the new config attributes proxyhost, proxyport and proxyssl instead of the getattr quirk. BTW, this is the docstring of setHost in twisted/protocols/http.py: """Change the host and port the request thinks it's using. This method is useful for working with reverse HTTP proxies (e.g. both Squid and Apache's mod_proxy can do this), then the address the HTTP client is using is different than the one we're listening on. For example, Apache may be listening on https://www.example.com, and then forwarding requests to http://localhost:8080, but we don't want HTML produced by Twisted to say 'http://localhost:8080', they should say 'https://www.example.com', so we do:: request.setHost('www.example.com', 443, ssl=1) This method is experimental. """ ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1100206&group_id=8482 From noreply at sourceforge.net Sun Jan 23 05:42:10 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Sun Jan 23 05:42:10 2005 Subject: [Moin-devel] [ moin-Patches-1100206 ] Twisted reverse proxy support Message-ID: Patches item #1100206, was opened at 2005-01-11 16:10 Message generated for change (Comment added) made by thomaswaldmann You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1100206&group_id=8482 Category: None Group: None Status: Open Resolution: None Priority: 6 Submitted By: Thomas Werschlein (twerschlein) Assigned to: Nobody/Anonymous (nobody) Summary: Twisted reverse proxy support Initial Comment: I run twistedmoin behind an Apache reverse proxy using mod_proxy. Everything works fine, but the RSS feed shows the internal host/port as source instead of the published URL (i.e. the reverse proxy host/port). adding these two lines to MoinRequest(server.Request): if getattr(config, 'proxyhost', None) and getattr(config, 'proxyport', None): self.setHost(config.proxyhost, config.proxyport, ssl=getattr(config, 'proxyssl', 0)) after self.host = self.channel.transport.getHost() in MoinMoin/server/twistedmoin.py seems to solve the problem. Obviously, this is a quick and dirty hack and I am not sure, if this has some bad side effects I am not aware of. I am also not sure, if it is the right place to do it. And, of course it would be better to include some default settings in the Config class for the new config attributes proxyhost, proxyport and proxyssl instead of the getattr quirk. BTW, this is the docstring of setHost in twisted/protocols/http.py: """Change the host and port the request thinks it's using. This method is useful for working with reverse HTTP proxies (e.g. both Squid and Apache's mod_proxy can do this), then the address the HTTP client is using is different than the one we're listening on. For example, Apache may be listening on https://www.example.com, and then forwarding requests to http://localhost:8080, but we don't want HTML produced by Twisted to say 'http://localhost:8080', they should say 'https://www.example.com', so we do:: request.setHost('www.example.com', 443, ssl=1) This method is experimental. """ ---------------------------------------------------------------------- >Comment By: Thomas Waldmann (thomaswaldmann) Date: 2005-01-23 14:41 Message: Logged In: YES user_id=100649 the current 1.3 development version (soon to be released as 1.3.2) uses such code only for POST. so i think your patch can't be applied there. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=308482&aid=1100206&group_id=8482 From msoulier at digitaltorque.ca Sun Jan 23 19:39:04 2005 From: msoulier at digitaltorque.ca (Michael P. Soulier) Date: Sun Jan 23 19:39:04 2005 Subject: [Moin-devel] Handling a form submission Message-ID: <20050124033802.GU6975@digitaltorque.ca> People, I'm plugging in an app of mine using a processor plugin that I'm writing. What I'm not sure how to do with Moin is present a form, and handle the submission. I do this in CGI all the time, but I'm not sure where to hook this in to Moin's architecture. Currently, the output of my plugin is done with a simple processor, and that's easy enough to do. But, I'm not sure how to do a form submission. Can someone point me in the right direction for how to do this with Moin? Also, are there processor API docs anywhere? I didn't see anything in HelpOnProcessors. I'm using Moin 1.2.4. Thanks, Mike -- Michael P. Soulier http://www.digitaltorque.ca http://opag.ca python -c 'import this' Jabber: msoulier at digitaltorque.ca -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From tw-public at gmx.de Mon Jan 24 04:08:11 2005 From: tw-public at gmx.de (Thomas Waldmann) Date: Mon Jan 24 04:08:11 2005 Subject: [Moin-devel] Handling a form submission In-Reply-To: <20050124033802.GU6975@digitaltorque.ca> References: <20050124033802.GU6975@digitaltorque.ca> Message-ID: <41F4E194.2090100@gmx.de> > I'm plugging in an app of mine using a processor plugin that I'm writing. What > I'm not sure how to do with Moin is present a form, and handle the submission. > I do this in CGI all the time, but I'm not sure where to hook this in to > Moin's architecture. > > Currently, the output of my plugin is done with a simple processor, and that's > easy enough to do. But, I'm not sure how to do a form submission. > > Can someone point me in the right direction for how to do this with Moin? Stuff using forms in moin is e.g. in userform.py and in AttachFile.py. Often a combination of an action with a macro (processor, parser) is used for processing (or generating) forms. Form values can be retrieved from request.form, maybe grep for that. > Also, are there processor API docs anywhere? I didn't see anything in > HelpOnProcessors. There isn't much except the source. And except some ancient beginning of a Form handling (no current developer ever used or touched it), there isn't much of an API for that stuff. If you want to work on it, feel free (but development should be based at minimum on 1.3.current, or 1.4.current - but 1.4 for sure isn't intended for production). > I'm using Moin 1.2.4. You maybe want to upgrade to 1.3.2 now. :) From robert.seeger at orsoft.de Mon Jan 24 08:17:19 2005 From: robert.seeger at orsoft.de (Robert Seeger) Date: Mon Jan 24 08:17:19 2005 Subject: [Moin-devel] 1.3.2 does not run on IIS due to error in request.py In-Reply-To: <20050123042541.70BD488844@sc8-sf-spam1.sourceforge.net> Message-ID: <001201c50230$c3c1a220$ed02a8c0@orsoft.de> In request.py, function RequestBase::_setup_vars_from_std_env there is some special code to handle some strange stuff on windows. The first branch is supposed to be checking for apache and if this fails, the second branch is checking for IIS. However the second branch is never executed due to an obvious error in the first check: # First check if we can use Apache's request_uri variable in order to # gather the correct user input if (os.name != 'posix' and server_software.find('Apache/1.3') == -1 and self.request_uri != ''): import urllib path_info = urllib.unquote(self.request_uri.replace(self.script_name, '', 1).replace('?' + query_string, '', 1)) Obviously if we are using IID, then server_software.find('Apache/1.3') == -1 is always true. The correct check should be if (os.name != 'posix' and server_software.find('Apache/1.3') != -1 The symptoms of this error were not easy to understand, because most of the unit tests (I re-enabled them) worked OK and I could perfectly work with FrontPage - but only with FrontPage! Whichever page was requested I always got just FrontPage. Kind Regards, Robert Seeger From 2004b at usenet.alexanderweb.de Mon Jan 24 08:25:21 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Mon Jan 24 08:25:21 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py References: <20050123042541.70BD488844@sc8-sf-spam1.sourceforge.net> <001201c50230$c3c1a220$ed02a8c0@orsoft.de> Message-ID: <1w0c33jan9yol$.dlg@usenet.alexanderweb.de> On Mon, 24 Jan 2005 17:21:30 +0100, Robert Seeger wrote: > In request.py, function RequestBase::_setup_vars_from_std_env there is some > special code to handle some strange stuff on windows. > The first branch is supposed to be checking for apache and if this fails, > the second branch is checking for IIS. > However the second branch is never executed due to an obvious error in the > first check: > > # First check if we can use Apache's request_uri variable in order > to > # gather the correct user input > if (os.name != 'posix' and server_software.find('Apache/1.3') == -1 > and self.request_uri != ''): > import urllib > path_info = > urllib.unquote(self.request_uri.replace(self.script_name, '', 1).replace('?' > + query_string, '', 1)) > > > Obviously if we are using IID, then > server_software.find('Apache/1.3') == -1 > is always true. > > The correct check should be > if (os.name != 'posix' and server_software.find('Apache/1.3') != -1 > > The symptoms of this error were not easy to understand, because most of the > unit tests (I re-enabled them) worked OK and I could perfectly work with > FrontPage - but only with FrontPage! Whichever page was requested I always > got just FrontPage. Can you give me the URL to your site? Did you change the if line to if (os.name != 'posix' and server_software.find('Apache/1.3') != -1:? Can I have the output of your CGI environment (is printed if you request the tests from localhost)? I would like to know if IIS gives you the request_uri. If it does not, the above patch does not make any sense because it sets path_info to an empty string. Kind regards, Alexander From 2004b at usenet.alexanderweb.de Mon Jan 24 09:06:09 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Mon Jan 24 09:06:09 2005 Subject: [Moin-devel] ANN: Security hole in 1.3.2 - Patch attached Message-ID: <11yg0gc5hy0n3.dlg@usenet.alexanderweb.de> Hi, there was a report about a security issue today. We were able to confirm it and created a patch that will fix this issue. Either apply the following patch or copy the linked file into the directory which contains the MoinMoin package. This is mostly the directory lib/site-packages. The bug causes search results from pages the user can not read to be displayed. When clicking on the link to the page, the user is not able to read the full page, though. For more information see Fixed search.py: http://moinmoin.wikiwikiweb.de/MoinMoinBugs/FullTextSearchIgnoresAcl?action=AttachFile&do=get&target=search.py I hope that there will be a new release today that contains this patch. Kind regards, Alexander --- orig/MoinMoin/search.py +++ mod/MoinMoin/search.py @@ -1014,7 +1014,7 @@ if result: if not filter: # Filter deleted pages or pages the user can't read. - if not page.exists() and request.user.may.read(name): + if not (page.exists() and request.user.may.read(name)): continue hits.append(FoundPage(name, result)) From 2004b at usenet.alexanderweb.de Mon Jan 24 12:45:17 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Mon Jan 24 12:45:17 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py References: <20050123042541.70BD488844@sc8-sf-spam1.sourceforge.net> <001201c50230$c3c1a220$ed02a8c0@orsoft.de> <1w0c33jan9yol$.dlg@usenet.alexanderweb.de> Message-ID: <741yxt3nvfd6$.dlg@usenet.alexanderweb.de> On Mon, 24 Jan 2005 17:24:42 +0100, Alexander Schremmer wrote: >> In request.py, function RequestBase::_setup_vars_from_std_env there is some >> special code to handle some strange stuff on windows. Yeah, it is even new in 1.3.2. >> The first branch is supposed to be checking for apache and if this fails, >> the second branch is checking for IIS. The first branch is just used if the server supplies request_uri. IIS does not, so it will be relevant if the CGI is just executed on IIS. If the first branch was not used, the second checks if IIS called the CGI script. If this is true, it tries to decode the path_info. >> However the second branch is never executed due to an obvious error in the >> first check: I still do not think that this is an error. >> The symptoms of this error were not easy to understand, because most of the >> unit tests (I re-enabled them) worked OK and I could perfectly work with >> FrontPage - but only with FrontPage! Whichever page was requested I always >> got just FrontPage. Funnily, this should be the consequence after applying your hints :-) I have installed IIS just because I wanted to make sure that my addition in request.py was correct. Besides a minor error that disallowes the page view of pages containing characters that cannot be displayed in the system code page (e.g. Asian ones), it was perfectly valid and worked without any further problems on IIS 5.1. A little fix for that glitch will be in the hotfix release that ThomasWaldmann will finish in the next hours. Maybe you should check in your configuration if you enabled "Check that file exists". This may not be enabled because IIS would check if the (virtual) page name exists in the directory script. If this still does not fix your problem, you should try to contact my personally. Kind regards, Alexander From robert.seeger at orsoft.de Tue Jan 25 03:37:01 2005 From: robert.seeger at orsoft.de (Robert Seeger) Date: Tue Jan 25 03:37:01 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py In-Reply-To: <20050125041702.1FAB913367@sc8-sf-spam2.sourceforge.net> Message-ID: <000a01c502d2$9cf5f680$ed02a8c0@orsoft.de> > Message: 5 > To: moin-devel at lists.sourceforge.net > From: Alexander Schremmer <2004b at usenet.alexanderweb.de> > Date: Mon, 24 Jan 2005 21:44:44 +0100 > Organization: XOR AX,AX > Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to > error in request.py > > On Mon, 24 Jan 2005 17:24:42 +0100, Alexander Schremmer wrote: > > >> In request.py, function > RequestBase::_setup_vars_from_std_env there is some > >> special code to handle some strange stuff on windows. > > Yeah, it is even new in 1.3.2. > > >> The first branch is supposed to be checking for apache and > if this fails, > >> the second branch is checking for IIS. > > The first branch is just used if the server supplies > request_uri. IIS does > not, so it will be relevant if the CGI is just executed on IIS. > > If the first branch was not used, the second checks if IIS > called the CGI > script. If this is true, it tries to decode the path_info. > > >> However the second branch is never executed due to an > obvious error in the > >> first check: > > I still do not think that this is an error. Sorry, I forgot to mention that I had fixed the request_uri issue as well. IIS does indeed not supply 'REQUEST_URI', so my workaround for an issue with this (I think I fixed this in 1.3.0 already and did not think about it) was self.request_uri = env.get('REQUEST_URI', '%s'%self.script_name) (this is probably not quite correct, I just needed something useful in that variable for the request.url variable that is used to find a wiki in a farm config). Thus, the first check was always true in my case and the problem was self-made. Nevertheless I think that server_software.find('Apache/1.3') == -1 was meant to be server_software.find('Apache/1.3') != -1 if the first check was intended for apache only. In any case, changing this fixed my problem just fine. The second check was executed and did its magic. However, as there are some places that use self.request_uri there should be a safe setting of this variable other than '' even in the case that the server does not deliver 'REQUEST_URI'. My fix is probably not the best way to do so. Unfortunately my wiki farm (with a total of 4 wikis, a lot of patches and special development including forms-based database access, currently migrating from 1.1 on IIS 5 to 1.3 on IIS 6) runs on our intranet and I can't supply an external url (which means I have to solve all the riddles by myself ;-( Kind Regards, Robert Seeger > > >> The symptoms of this error were not easy to understand, > because most of the > >> unit tests (I re-enabled them) worked OK and I could > perfectly work with > >> FrontPage - but only with FrontPage! Whichever page was > requested I always > >> got just FrontPage. > > Funnily, this should be the consequence after applying your hints :-) > > I have installed IIS just because I wanted to make sure that > my addition in > request.py was correct. Besides a minor error that disallowes > the page view > of pages containing characters that cannot be displayed in > the system code > page (e.g. Asian ones), it was perfectly valid and worked without any > further problems on IIS 5.1. > > A little fix for that glitch will be in the hotfix release that > ThomasWaldmann will finish in the next hours. > > Maybe you should check in your configuration if you enabled > "Check that > file exists". This may not be enabled because IIS would check if the > (virtual) page name exists in the directory script. > > If this still does not fix your problem, you should try to contact my > personally. > > Kind regards, > Alexander > > > > > --__--__-- > > _______________________________________________ > Moin-devel mailing list > Moin-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/moin-devel > > > End of Moin-devel Digest > From R.Bauer at fz-juelich.de Tue Jan 25 10:11:02 2005 From: R.Bauer at fz-juelich.de (Reimar Bauer) Date: Tue Jan 25 10:11:02 2005 Subject: [Moin-devel] > MoinMoin1.3.1 Message-ID: <41F68B8F.3030504@fz-juelich.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dear all there should be a hint about the __init__.py needed since Version 1.3.2 in the plugin dir of a current instance. Because in previous versions the wiki doesn't crash if this file was missing. mywiki/data/plugin | more __init__.py # *** Do not remove this! *** # Although being empty, the presence of this file is important for plugins # working correctly. The hint comes by the marvellous error handler after it occurs. But there are too much probabilities for other errors too. If you have followed the instructions in UPDATE.html for 1.2 it should not occure. I have missed this point. I don't know why all our wikis have worked before 1.3.2. But now this file is important. Probably one of the migration tools should be enhanced to fix this too. cheers Reimar - -- Reimar Bauer Institut fuer Stratosphaerische Chemie (ICG-I) Forschungszentrum Juelich email: R.Bauer at fz-juelich.de - ------------------------------------------------------------------- ~ a IDL library at ForschungsZentrum Juelich ~ http://www.fz-juelich.de/icg/icg-i/idl_icglib/idl_lib_intro.html =================================================================== -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (GNU/Linux) iD8DBQFB9ouO5aOc3Q9hk/kRApo/AJ9P+wbPt1f5as2Y+GvYl/NiA9v7mgCeNPQM eJgJch0tLrgagiwVrr1Fbsk= =4owW -----END PGP SIGNATURE----- From noreply at sourceforge.net Tue Jan 25 15:30:13 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 25 15:30:13 2005 Subject: [Moin-devel] [ moin-Feature Requests-1109474 ] filter recent changes Message-ID: Feature Requests item #1109474, was opened at 2005-01-25 23:28 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1109474&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Chris Akre (netmanakre) Assigned to: Nobody/Anonymous (nobody) Summary: filter recent changes Initial Comment: I'd like the option to filter recent changes based on an arbitrary full text search. The use case is the need to see all recently changed pages in a category. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1109474&group_id=8482 From noreply at sourceforge.net Tue Jan 25 15:33:35 2005 From: noreply at sourceforge.net (SourceForge.net) Date: Tue Jan 25 15:33:35 2005 Subject: [Moin-devel] [ moin-Feature Requests-1109476 ] email notification based on text search Message-ID: Feature Requests item #1109476, was opened at 2005-01-25 23:30 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1109476&group_id=8482 Category: None Group: None Status: Open Priority: 5 Submitted By: Chris Akre (netmanakre) Assigned to: Nobody/Anonymous (nobody) Summary: email notification based on text search Initial Comment: I'd like to have the ability to set an email notification for the modification of any page containing an arbitrary full text search. The use case is receiving notification when a page belonging to a category changes. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=358482&aid=1109476&group_id=8482 From 2004b at usenet.alexanderweb.de Tue Jan 25 15:53:05 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Tue Jan 25 15:53:05 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py References: <20050125041702.1FAB913367@sc8-sf-spam2.sourceforge.net> <000a01c502d2$9cf5f680$ed02a8c0@orsoft.de> Message-ID: <1egdy4hwxtubm$.dlg@usenet.alexanderweb.de> On Tue, 25 Jan 2005 12:40:01 +0100, Robert Seeger wrote: > Sorry, I forgot to mention that I had fixed the request_uri issue as well. > IIS does indeed not supply 'REQUEST_URI', so my workaround for an issue with > this (I think I fixed this in 1.3.0 already and did not think about it) was > self.request_uri = env.get('REQUEST_URI', '%s'%self.script_name) > (this is probably not quite correct, I just needed something useful in that > variable for the request.url variable that is used to find a wiki in a farm > config). > Thus, the first check was always true in my case and the problem was > self-made. Nevertheless I think that > server_software.find('Apache/1.3') == -1 > was meant to be > server_software.find('Apache/1.3') != -1 > if the first check was intended for apache only. No, it was not. The first check should just run if a) there is an request_uri (a valid one) and b) the server is not Apache 1.3. I guess you screwed up request_uri (your code does not seem to assemble it correctly) and glitched into this problem. I am not sure if I should have expected such a broken request_uri :-) Maybe you give us some reasons why MoinMoin needs that request_uri ... Normally, it should not. Please try to run MoinMoin 1.3.3 with no local modifications (at least not in that function :-)) and the following patch applied: (see first link on that page) http://moinmoin.wikiwikiweb.de/KnownIssues Kind regards, Alexander From robert.seeger at orsoft.de Wed Jan 26 03:46:02 2005 From: robert.seeger at orsoft.de (Robert Seeger) Date: Wed Jan 26 03:46:02 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py In-Reply-To: <20050126042702.0F1BEF5A1@sc8-sf-spam2.sourceforge.net> Message-ID: <000801c5039d$23a06470$ed02a8c0@orsoft.de> > Message: 5 > To: moin-devel at lists.sourceforge.net > From: Alexander Schremmer <2004b at usenet.alexanderweb.de> > Date: Wed, 26 Jan 2005 00:51:15 +0100 > Organization: XOR AX,AX > Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to > error in request.py > > On Tue, 25 Jan 2005 12:40:01 +0100, Robert Seeger wrote: > > > Sorry, I forgot to mention that I had fixed the request_uri > issue as well. > > IIS does indeed not supply 'REQUEST_URI', so my workaround > for an issue with > > this (I think I fixed this in 1.3.0 already and did not > think about it) was > > self.request_uri = env.get('REQUEST_URI', > '%s'%self.script_name) > > (this is probably not quite correct, I just needed > something useful in that > > variable for the request.url variable that is used to find > a wiki in a farm > > config). > > Thus, the first check was always true in my case and the problem was > > self-made. Nevertheless I think that > > server_software.find('Apache/1.3') == -1 > > was meant to be > > server_software.find('Apache/1.3') != -1 > > if the first check was intended for apache only. > > No, it was not. The first check should just run if a) there is an > request_uri (a valid one) and b) the server is not Apache 1.3. > I guess you screwed up request_uri (your code does not seem > to assemble it > correctly) and glitched into this problem. I am not sure if I > should have > expected such a broken request_uri :-) > > Maybe you give us some reasons why MoinMoin needs that request_uri ... > Normally, it should not. Please try to run MoinMoin 1.3.3 > with no local > modifications (at least not in that function :-)) and the > following patch > applied: (see first link on that page) > http://moinmoin.wikiwikiweb.de/KnownIssues > > Kind regards, > Alexander Sorry, I messed it all up with wrong assumptions about Alexander's intentions. Next time I'd better ask first and only then try to fix it by myself ;-). So, I revoked my changes in request.py and applied the patch from http://moinmoin.wikiwikiweb.de/MoinMoinBugs/request%2epy_broken_on_IIS (as this was the only relevant line changed from 1.3.2 to 1.3.3, I ignored the other 1.3.3 modules for now). As a result, however, I get a configuration error: MoinMoin Configuration Error Could not find a match for url: "ors-x-know". Check your URL regular expressions in the "wikis" list in "farmconfig.py". The reason for this is that the checked url (in multiconfig.getConfig(url)) only contains the server name, but not the part that identifies a wiki instance. This again is due to request.py, line 253: self.url = self.server_name + self.request_uri which does not work if there is no significant self.request_uri Here is some details about my installation: I'm running a farm on a single server {{{ors-x-know}}} with a number of wiki instances. The following table lists their url in IIS, the regex and wikiname used in farmconfig.py ||URL ||wikiname||regex|| ||{{{http://ors-x-know/master11cgi/moin.cgi}}}||master11||{{{r".*master11.*$ "}}}|| ||{{{http://ors-x-know/master13cgi/moin.cgi}}}||master13||{{{r".*master13.*$ "}}}|| ||{{{http://ors-x-know/devinfocgi/moin.cgi}}}||master13||{{{r".*devinfo.*$"} }}|| With the above setting, the checked url is only {{{ors-x-know}}} This is why I did the mentioned hack: self.request_uri = env.get('REQUEST_URI', '%s'%self.script_name) and glitched into the wrong check. I have now modified this hack: * In the IIS specific patch starting on line 233, add: self.request_uri = '%s'%self.script_name This makes the checked url e.g. {{{ors-x-know/master13cgi/moin.cgi}}} which makes my setting happy, though it might not be the correct value for request_uri (should probably include the full path and parameters?) Maybe the proper solution would be to set self.url differently in the RequestCGI class (some other derived classes do this)? Kind Regards, Robert From 2004b at usenet.alexanderweb.de Wed Jan 26 07:50:03 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Wed Jan 26 07:50:03 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py References: <20050126042702.0F1BEF5A1@sc8-sf-spam2.sourceforge.net> <000801c5039d$23a06470$ed02a8c0@orsoft.de> Message-ID: On Wed, 26 Jan 2005 12:49:48 +0100, Robert Seeger wrote: >> No, it was not. The first check should just run if a) there is an >> request_uri (a valid one) and b) the server is not Apache 1.3. >> I guess you screwed up request_uri (your code does not seem >> to assemble it >> correctly) and glitched into this problem. I am not sure if I >> should have >> expected such a broken request_uri :-) >> [...] > As a result, however, I get a configuration error: > > MoinMoin Configuration Error > Could not find a match for url: "ors-x-know". > > Check your URL regular expressions in the "wikis" list in "farmconfig.py". > > The reason for this is that the checked url (in multiconfig.getConfig(url)) > only contains the server name, but not the part that identifies a wiki > instance. > This again is due to request.py, line 253: > self.url = self.server_name + self.request_uri > > which does not work if there is no significant self.request_uri > > Here is some details about my installation: > > I'm running a farm on a single server {{{ors-x-know}}} with a number of wiki > instances. The following table lists their url in IIS, the regex and > wikiname used in farmconfig.py > > ||URL ||wikiname||regex|| > >||{{{http://ors-x-know/master11cgi/moin.cgi}}}||master11||{{{r".*master11.*$ > "}}}|| > >||{{{http://ors-x-know/master13cgi/moin.cgi}}}||master13||{{{r".*master13.*$ > "}}}|| > >||{{{http://ors-x-know/devinfocgi/moin.cgi}}}||master13||{{{r".*devinfo.*$"} > }}|| > > With the above setting, the checked url is only {{{ors-x-know}}} > > This is why I did the mentioned hack: > self.request_uri = env.get('REQUEST_URI', '%s'%self.script_name) > and glitched into the wrong check. > > I have now modified this hack: > * In the IIS specific patch starting on line 233, add: > self.request_uri = '%s'%self.script_name > > This makes the checked url e.g. {{{ors-x-know/master13cgi/moin.cgi}}} which > makes my setting happy, though it might not be the correct value for > request_uri (should probably include the full path and parameters?) > > Maybe the proper solution would be to set self.url differently in the > RequestCGI class (some other derived classes do this)? Yeah, it should be. Assuming that request_uri contains something useful should be considered evil. :-) I will look into this ... Kind regards, Alexander From tw-public at gmx.de Wed Jan 26 08:22:00 2005 From: tw-public at gmx.de (Thomas Waldmann) Date: Wed Jan 26 08:22:00 2005 Subject: [Moin-devel] > MoinMoin1.3.1 In-Reply-To: <41F68B8F.3030504@fz-juelich.de> References: <41F68B8F.3030504@fz-juelich.de> Message-ID: <41F7BFE9.5070408@gmx.de> > there should be a hint about the __init__.py needed since Version 1.3.2 I don't think this requirement is new. It should be there for quite a while (and being documented in CHANGES hopefully). > I don't know why all our wikis have worked before 1.3.2. But now this > file is important. The early versions of the plugin loader were severely broken (don't remember until when, but maybe around 1.1 or early 1.2). Then it got replaced by a working plugin loader. And even later, I think it was modified again. > Probably one of the migration tools should be enhanced to fix this too. Yes, we could just copy those __init__.py in there. But it is documented that plugins aren't fixed by the mig scripts, so this is a missing feature, not a bug. :) From robert.seeger at orsoft.de Thu Jan 27 03:50:21 2005 From: robert.seeger at orsoft.de (Robert Seeger) Date: Thu Jan 27 03:50:21 2005 Subject: [Moin-devel] > MoinMoin1.3.1 In-Reply-To: <20050127042606.29DB913E3E@sc8-sf-spam2.sourceforge.net> Message-ID: <000c01c50465$c37ffc80$ed02a8c0@orsoft.de> > Message: 3 > Date: Wed, 26 Jan 2005 17:06:01 +0100 > From: Thomas Waldmann > To: moin-devel at lists.sourceforge.net > Subject: Re: [Moin-devel] > MoinMoin1.3.1 > > > there should be a hint about the __init__.py needed since > Version 1.3.2 > > I don't think this requirement is new. It should be there for quite a > while (and being documented in CHANGES hopefully). > > > I don't know why all our wikis have worked before 1.3.2. > But now this > > file is important. > > The early versions of the plugin loader were severely broken (don't > remember until when, but maybe around 1.1 or early 1.2). > > Then it got replaced by a working plugin loader. > > And even later, I think it was modified again. > > > Probably one of the migration tools should be enhanced to > fix this too. > > Yes, we could just copy those __init__.py in there. > > But it is documented that plugins aren't fixed by the mig scripts, so > this is a missing feature, not a bug. :) > Sorry to object, but this is a serious bug. Just remove __init__.py from the data/plugin directory of a wiki which does n o t have any plugins under data/plugin and see what happens! In my CGI (IIS) version I just get: CGI Error The specified CGI application misbehaved by not returning a complete set of HTTP headers. which isn't very helpful anyway. Only after forcing a http header (ASCII) on sys.stdout you see the proper error message: [Thu Jan 27 12:31:50 2005] MoinMoin Configuration Error: Could not import plugin package from "D:\WEBServer\Homepage\moin\master11\data\" because of ImportError: No module named plugin. Make sure your data directory path is correct, check permissions, and that the data/plugin directory has an __init__.py file. Thus, not having this file breaks any migrated wiki instance, not just a specific plugin. This should definitely be avoided by a fixed migration script. Kind Regards, Robert From clan at ziplip.com Fri Jan 28 15:46:11 2005 From: clan at ziplip.com (=?Windows-1251?Q?=C1=E8=E7=ED=E5=F1-=D6=E5=ED=F2p_H=E0=F6=E8o=EDa=EB=FC=ED=FB=E9?=) Date: Fri Jan 28 15:46:11 2005 Subject: [Moin-devel] =?Windows-1251?Q?Re[2]:_C=F2=F0a=F2=E5=E3=E8=FF_win-win?= Message-ID: <6437833505.20040923613035@201-1-145-215.dsl.telesp.net.br> ??????????? ??????????: ????????? WIN-WIN ??????? ? 3 ??????? 2005 ???? ?????? ??????????????????? ???? ??????? ?????? ?? ????????? ???????????? ??? ????-??, ?? ???? ????????? ? ?????? ?????????. ???-??, ????????? ????? ???????? ???????? ?????? ?????? ? ?????? ? ?????????????. ? ???-?? ?????? ??? ????????, ??? ???, ??? ?? ????? ?? ??? ??? ? ?? ????? ????????? ? ???????????????, ? ???, ???????????, ???? ?????????. ?? ???? ?? ????????????, ??? ????? ????????????? ? ?????????? ????????: ??????? ???????? ??????????? ? ??????????? ???????? ???????????? ?? ??? ?? ?????? ???????! ??????? ?????? ??? ??? ????? ? ???????????? ? ??????? ??????-??????, ?????? ????, ?????????, ??????? ? ????? ?????????? ??????????? ??????? ?????????, ??????????????? ?? ?? ????????. ???????????, ????????????? ?????? ?. ???????? ?????????? ????????????? ?????????, ????????????? ????? ???????? ??????? ? ?????????, ?????? ??????? ??????????? ???????????. ???????????? ??????? ??????????: ?? ?????? 10-??????? ????? ?????? ?????????? ?????????? ??????? ? ?????? ???????? ? ??????????? ??????? ????? PR, ?? ?????? ?????? ?? ????? ??????????????? ??????????? ? ?????? ????????, ????? ?????? ? ?????????-?????????, ??????????? ? ???????, ??????, ????????, ????????. ??????? ????????? ? ???????????? ? ????????? ??????? ? ???????? ????? ????????, ????????? ?? ???????, ????????? ??????? ?????, ?????????????? ? ???????????? ??????, ??? ??, ??? ???????? ???????????? ???????? ???????? ????????? ??????-??????????. ??????? ???????? ? ??????? ??????? ?????????? ? 10-?????? ???? ?????????? ??????????????? ????????? ? ?????? ?????? ?? ????????? ?????????? ???????, ??????????????, ???????? ??????????? ??????? ??????, ??????? ???????????. ???????? PR-???????? ?PR??????; ??????? ??????? ????????? ??? ???????; ???????? - ???????? ? ????? ????????????? ?????? ????????. ????????? ???????? 1. ??????????? ????? ????????, ?? ????????????? ? ???????? ??????????????????? ??????????? ? ???? ???????????. ? ???? ? ??????? ???????????. ? ????? ?????????? ???????????. ? Small talk ? ???????? ?????????? ???????????. ? ???????? ??????????? ????????? ?????-??????? ? ???????? ?????????? ??????. ? ?????????: ??? ??, ??? ??????? ?? ?????? ???????????? ???????????????? ???????????? ????? ??????? ???????????. ????????-??????????? ???????? ????????. ???????????? ??????? ????????? ????????? ?? ?????? ???????????. ?????? ? ?????? ?? ?????????????. ?????? ?????????? ????-?????. 2. ?????????????? ?????? ? ???? ??????? ???????????. ? ????? ??? ?????????. ? ????????????? ??????? ??? ? ??????????? ?? ???????????. ? ?????????????, ????????????? ? ?????????? ???????????? ?????????? ? ????????. ??????? ????????????? ??????????? ?????? ? ?????? ??????? ?? ?? ???????????. ????? ????????. ?????????? ??????. ???????? ?????????. ??????? ???????. ?????????? ??????: ?????????? ? ??????????. ? ????? ???????. ? ??????? ????????????. ? ????????? ?? ????????????? ?????????. ? ??????????? ?????? ? ?????. 3. ??????? ??????? ??????? ???????????. ? ??????? ???????????, ???????????? ?? ???????? ???? ?? ?????. ??? ???????? ??????. ? ????????, ?????????? ?????????? ???? ?? ????? ? ?????????? ?? ???????????. ? ??????? ??????? ? ????????, ?? ??????????? ???? ?????????????. ? ??? ??????????? ?? ???????: ???????????? ? ??????????????? ? ??????? ??????????? ? ??????????? ????????. ???? ? ????? ??????? ?????: ????????? ?????, ?? ??? ???, ???? ? ??? ?? ?????? ??????????? ????? ??? ??? ??????? ??????????? ? ?????????????? ????? ???????? ??? ????. ? ???? ??????? ??????? ? ?????? ??? ??????? ?????? ? ??????????????? ???????? (?????????). ? ??????? ?????? ??? ??????? ????????????. ??????? ?????????? ??????? ????????????. ? ?????? ??????? ?? ?????? ?????????? ??? ??? ?? ???????? ??????, ????? ???? ??????????. ? ???????? ??????? ??????? ??? ??????? ??????? ??????. ????????? ?????????? ???????? ? ???????? ????????. ? ??????? ???????? ? ???????? ???????????. ??????? ????????????? ? ????? ??????. ? ????????? ??????? ? ???????? ???????????. ????????? ?? ???????. ????????? ? ??????. ????? ?? ? ???????? ?????????, ?? ???? ? ?????? ???????????? ? ???????? 100 000 ?????? ??? ??? ???????? ??????. ? ??????? ?????? ? ????????? ???????? ?? ???????? ??? ??????????? ??????????. ? ?????????? ?????? ? ???????? ???????????. ???? ?????? ? ?????? ??? ??????????. ? ????? ??????????? ?????? ???????? ? ??????????? ??????? ?? ?????? ???????????. ???? ????????? ?? ??????????? ????? ??????. ? ?????? ?????????? ???????? ??? ????? ???????????? ?????????????. ? ?????????? ? ????. ??????? ??????????? ????. ?????? 7 ?????? ? 2 ???????. ? ?????????? ???????? ????????? ??????: ? ???????? ? ?????????????? ???????? ????????? ??????? ??????????. ? ???? ???????????? ? ???????? ???????????. ? ???????? ????????????? ?????? ??????? ? ????? ?????. ? ??????????? ???????? ? ????????????. ? ?????????????? ???????? ??????????????? ???????? ???????. ? ????????????? ????????? ? ??????? ??????????? ?? ???????? ???????. ? ????? ??????????????, ????????????? ????????? ? ???????????????? ???????????? ???????, (????????). ? ???????? ??????????? ?????????? ??????? ??????????? ? ?????????????????. ????????? ? 570.00 ???. ? ?? ?????? ?????????. ? ??? ??????? ? ???????? ????????? ?????? ? 10% ? 15% ??????????????. ? ? ????????? ??????: ?????????????-???????????????? ???????????? ?? ????????, ??????? ??????????, ????-?????, ???? ? ?????????, ?????????? ???????? ? ????? ???????? ? ????????. ????????? ? 9.30 ? 17.00 ? ??????? 13.00 ?14.00 ? ??????????? ? 9.00 ? ?????????-???? ????? ??????????, ?. ????, ??. ???????????? 1/3. ? ?????????? ????? ? ????????? ?????????????? ????????? ??? ???????????. ??????????? ? ???.: (044) 233-4669, 237-9005 ? ????: (044) 290-1044 -------------- next part -------------- An HTML attachment was scrubbed... URL: From a.hanson at f5.com Fri Jan 28 17:00:16 2005 From: a.hanson at f5.com (Aaron Hanson) Date: Fri Jan 28 17:00:16 2005 Subject: [Moin-devel] links from print layout pages Message-ID: <1106960140.24351.117.camel@maxwell.dev.net> Hi All - I have a need to make a printed copy of all the pages in one of my Moins. It would be great if the links from a 'print layout' page requested a print layout of the linked page, but they don't. I tried to look through the source for a single point to patch in this feature, but I'm getting lost. Anybody have a suggestion of the right place to put such a patch? Or have I overlooked some similar techinique in the documentation? If this could be done, then I could just use wget on the site to get a complete, print-formatted doc... Thanks. -Aaron From 2004b at usenet.alexanderweb.de Fri Jan 28 17:37:45 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Fri Jan 28 17:37:45 2005 Subject: [Moin-devel] Re: links from print layout pages References: <1106960140.24351.117.camel@maxwell.dev.net> Message-ID: <1rblkrpm20m1i.dlg@usenet.alexanderweb.de> On Fri, 28 Jan 2005 16:55:40 -0800, Aaron Hanson wrote: > I have a need to make a printed copy of all the pages in one of my > Moins. It would be great if the links from a 'print layout' page > requested a print layout of the linked page, but they don't. You can capture the normal view of every page as well because it looks exactly like the print view when printed with a modern browser. If your browser does not support CSS good enough for that, you might want to look at the script in the MoinMoin/scripts/ directory that saves slim HTML pages for every wiki page. Kind regards, alexander From 2004b at usenet.alexanderweb.de Fri Jan 28 17:49:40 2005 From: 2004b at usenet.alexanderweb.de (Alexander Schremmer) Date: Fri Jan 28 17:49:40 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py References: <20050126042702.0F1BEF5A1@sc8-sf-spam2.sourceforge.net> <000801c5039d$23a06470$ed02a8c0@orsoft.de> Message-ID: <14fs7impm6z2y.dlg@usenet.alexanderweb.de> On Wed, 26 Jan 2005 16:49:28 +0100, Alexander Schremmer wrote: >> As a result, however, I get a configuration error: >> >> MoinMoin Configuration Error >> Could not find a match for url: "ors-x-know". >> >> Check your URL regular expressions in the "wikis" list in "farmconfig.py". >> >> With the above setting, the checked url is only {{{ors-x-know}}} >> [...] >> This makes the checked url e.g. {{{ors-x-know/master13cgi/moin.cgi}}} which >> makes my setting happy, though it might not be the correct value for >> request_uri (should probably include the full path and parameters?) >> >> Maybe the proper solution would be to set self.url differently in the >> RequestCGI class (some other derived classes do this)? > > Yeah, it should be. Assuming that request_uri contains something useful > should be considered evil. :-) I will look into this ... I fixed the bug in patch-600. You may want to download the code snapshot for that release: http://arch.thinkmo.de/cgi-bin/archzoom.cgi/arch at arch.thinkmo.de--2003-archives/moin--main--1.3--patch-600.tar.gz It should be as stable as 1.3.3 ... Just replace the MoinMoin dir. Please let us know if it works. Kind regards, Alexander From msoulier at digitaltorque.ca Sat Jan 29 21:00:25 2005 From: msoulier at digitaltorque.ca (Michael P. Soulier) Date: Sat Jan 29 21:00:25 2005 Subject: [Moin-devel] Handling a form submission In-Reply-To: <41F4E194.2090100@gmx.de> References: <20050124033802.GU6975@digitaltorque.ca> <41F4E194.2090100@gmx.de> Message-ID: <20050130045835.GO7124@digitaltorque.ca> On 24/01/05 Thomas Waldmann said: > Stuff using forms in moin is e.g. in userform.py and in AttachFile.py. > > Often a combination of an action with a macro (processor, parser) is > used for processing (or generating) forms. Form values can be retrieved > from request.form, maybe grep for that. AttachFile is very educational. > There isn't much except the source. And except some ancient beginning of > a Form handling (no current developer ever used or touched it), there > isn't much of an API for that stuff. > > If you want to work on it, feel free (but development should be based at > minimum on 1.3.current, or 1.4.current - but 1.4 for sure isn't intended > for production). Ok. > You maybe want to upgrade to 1.3.2 now. :) Yeah, I probably should. I am working on a simple CMS plugin for Moin. I wanted a simple news item submission tool, with news archives and excerpts, and it should be possible now that I see how to work with Moin. Thanks, Mike -- Michael P. Soulier http://www.digitaltorque.ca http://opag.ca python -c 'import this' Jabber: msoulier at digitaltorque.ca -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: From robert.seeger at orsoft.de Mon Jan 31 03:51:09 2005 From: robert.seeger at orsoft.de (Robert Seeger) Date: Mon Jan 31 03:51:09 2005 Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to error in request.py In-Reply-To: <20050129042711.3E04B881A3@sc8-sf-spam1.sourceforge.net> Message-ID: <001701c5078b$003db5e0$ed02a8c0@orsoft.de> > Message: 4 > To: moin-devel at lists.sourceforge.net > From: Alexander Schremmer <2004b at usenet.alexanderweb.de> > Date: Sat, 29 Jan 2005 02:49:22 +0100 > Organization: XOR AX,AX > Subject: [Moin-devel] Re: 1.3.2 does not run on IIS due to > error in request.py > [...] > I fixed the bug in patch-600. You may want to download the > code snapshot > for that release: > http://arch.thinkmo.de/cgi-bin/archzoom.cgi/arch at arch.thinkmo.de--2003-archi ves/moin--main--1.3--patch-600.tar.gz > >It should be as stable as 1.3.3 ... Just replace the MoinMoin dir. >Please let us know if it works. I can't use archzoom tarballs on my windows system (probably due to too long directory names or outdated WinZip tool), however I applied the changes from request.py and wikiutil.py. It now looks good, I haven't found any more problems. The only thing that frequently annoys me is that the CGI version does not display proper errors if anything unexpected happens, I always have to add a line like sys.stdout.write("Content-type: text/plain;charset=ASCII\r\n") somwhere near the possible code point for debugging purposes. It would be nice if somebody who knows more about this could modify the CGI version so that it always displays at least the call stack if something crashes. Thanks for your assistance. Kind Regards, Robert