From alex at verk.info Sun May 1 23:04:26 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sun May 1 22:57:55 2005 Subject: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option Message-ID: With much less fanfare than the last time, here is a minor release of the Instiki 0.10 version. Although there are no new features to speak of, if you are playing it safe and still holding on to version 0.9 or earlier, now is the time to consider an upgrade. Instiki 0.10.1 is certainly better (less bugs, more usability) than what you have. Thanks to all the adventurous souls who upgraded to 0.10.0 and provided bug reports and patches included in this new version. Changes: * Upgraded Rails to 0.12.0 * Upgraded rubyzip to version 0.5.8 * BlueCloth is back (RedCloth didn't do pure Markdown well enough) * Handling of line breaks in Textile is as in 0.9 (inserts
tag) * Fixed HTML export (to enclose the output in tags, include the stylesheet etc) * Corrected some compatibility issues with storages from earlier Instiki versions * Some other bug fixes Upgrade instructions: 1. Shut down Instiki. 2. Backup your storage directory. 3. If you are running instiki as gem, "gem update instiki". 4. Otherwise, download the new version and unpack to a new directory. 5. If necessary, copy your storage directory to a new location and update your startup scripts accordingly. Instiki links: * Home page: http://instiki.org * Download: http://rubyforge.org/frs/?group_id=186 * Bug tracker and source browser: http://dev.instiki.org * Subversion repository: http://svn.instiki.org/instiki * Instiki-users maillist: instiki-users@rubyforge.org -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From huw.nichols at gmail.com Mon May 2 08:36:23 2005 From: huw.nichols at gmail.com (Huw) Date: Mon May 2 08:30:14 2005 Subject: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option In-Reply-To: References: Message-ID: First of all thanks for the bug fixes and a wonderful product in general. > * Fixed HTML export (to enclose the output in tags, include the > stylesheet etc) Couple of things: 1) Does anyone else have a problem with the html export showing the wrong pages. i.e. the exported HomePage is showing the correct title but contents from a different page. So are all other exported pages. Maybe something date related as the homepage export is showing a new page I created last night just before upgrading. The running instiki shows the correct page contents however. 2) Is it possible to have an option to export just the page contents as with 0.10 version. This was very useful when the exported pages were run through a script to change the style sheet and other formatting to make it more usable as a stand alone document for my users. Regards Huw From oliver at sanbi.ac.za Mon May 2 11:09:31 2005 From: oliver at sanbi.ac.za (Oliver Hofmann) Date: Mon May 2 11:03:37 2005 Subject: [Instiki] Integration of SmartyPants Message-ID: <427642AB.1020100@sanbi.ac.za> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Greetings! With the new release once again working fine with Markdown (thanks a lot!) I was wondering whether someone successfully integrated [SmartyPants][1] into Instiki? Would be nice to have em dashes and the like. Best wishes, Oliver [1]: http://daringfireball.net/projects/smartypants/ - -- Oliver Hofmann South African National Bioinformatics Institute University of the Western Cape -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFCdkKr9ihIjY7POJMRAhalAKCd0tFmfJW5kLyvZPI7i7FUeXxMfACfShno t47aQ5KNSVrmVPJo1diqOgU= =Ii6d -----END PGP SIGNATURE----- From oliver at sanbi.ac.za Mon May 2 14:27:25 2005 From: oliver at sanbi.ac.za (Oliver Hofmann) Date: Mon May 2 14:21:21 2005 Subject: [Instiki] Potential Markdown bugs Message-ID: <4276710D.4000509@sanbi.ac.za> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Greetings, just two details I noticed: * Rulers (---) need to be separated from lists by at least one blank line, otherwise the last list item tends to get converted into a header * For links in the [Text](url) format the URL needs to be closed with a space followed by a bracket for a number of variants (URLs ending with a slash, .htm, ...) Cheers, Oliver - -- Oliver Hofmann South African National Bioinformatics Institute University of the Western Cape -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFCdnEN9ihIjY7POJMRAmb/AKCR6ZF4v9L8PwK7kJRzYTHDs/8RuwCeNPBY s4L6lJ5yKU3IEPmk55x92c4= =RKUW -----END PGP SIGNATURE----- From alex at verk.info Mon May 2 19:58:30 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 2 19:52:00 2005 Subject: [Instiki] Integration of SmartyPants In-Reply-To: <427642AB.1020100@sanbi.ac.za> References: <427642AB.1020100@sanbi.ac.za> Message-ID: Oliver Hofmann wrote: >-----BEGIN PGP SIGNED MESSAGE----- >With the new release once again working fine with Markdown (thanks a >lot!) I was wondering whether someone successfully integrated >[SmartyPants][1] into Instiki? > > Not that I know of. Some of these things are available in RedCloth, Textile or Mixed modes. Instiki is using an altered version of BlueCloth, anyway - so a nice, well-tested patch could certainly be accepted :) -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Mon May 2 20:52:53 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 2 20:46:22 2005 Subject: [Instiki] Importance of tags Message-ID: With HTML filtering switched off (which is a recommended setting when Instiki is used as a PIM or a project wiki within an intranet), you can put any HTML markup you want inside tags. It will be passed to the browser as is. For example, here is how you can make a working link to a local file (in this example Instiki is running under Windows): QTE File: File Folder: Shared folder Picture:

UNQTE -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From assaph at gmail.com Mon May 2 22:14:26 2005 From: assaph at gmail.com (Assaph Mehr) Date: Mon May 2 22:08:16 2005 Subject: [Instiki] File Uploads for 1.0 Message-ID: <3ff47be905050219143952f9d0@mail.gmail.com> Hi, * Will file uploads make it to Instiki 1.0? Specifically the 'import' option? * How did you keep the snapshot binary files compatible? Reason I ask is that for Pimki the snapshot binary file format will change. So while no upgrading content is needed from Instiki 0.9.2 to 0.10, it doesn't hold for Pimki (i.e. I cannot just reuse the snapshots). Alexey, do you have any insight on keeping the snapshots compatible? Or on performing the import? I'm OK (and I hope my users are as well :) with mandating an export/import upgrade for Pimki, but I don't want to redevelop what is going to be there anyway. Thnaks for your work on Instiki, Assaph From alex at verk.info Mon May 2 22:15:17 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 2 22:08:47 2005 Subject: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option In-Reply-To: References: Message-ID: Huw wrote: >First of all thanks for the bug fixes and a wonderful product in general. > > > >>* Fixed HTML export (to enclose the output in tags, include the >>stylesheet etc) >> >> > > >Couple of things: > >1) Does anyone else have a problem with the html export showing the >wrong pages. i.e. the exported HomePage is showing the correct title >but contents from a different page. So are all other exported pages. >Maybe something date related as the homepage export is showing a new >page I created last night just before upgrading. > > Yeah, I have the same problem. So much for fixing things in the last minute before release, and bypassing unit tests because it is too hard. Sigh... lesson learned. I'll fix it. >2) Is it possible to have an option to export just the page contents >as with 0.10 version. > Ah, so it was a feature! :) OK, check out SVN revision 254, and use it like http://localhost:2500/wiki/export_html?layout=no -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Mon May 2 23:32:47 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 2 23:26:16 2005 Subject: [Instiki] File Uploads for 1.0 In-Reply-To: <3ff47be905050219143952f9d0@mail.gmail.com> References: <3ff47be905050219143952f9d0@mail.gmail.com> Message-ID: Assaph Mehr wrote: >* Will file uploads make it to Instiki 1.0? Specifically the 'import' option? > > Most certainly. The main question is "When Instiki 1.0 is going to happen". And the answer is "fairly soon, on the open source timescale". :) I'll enable them fFileController on the trunk right after fixing the export_html bug (unless there are more embarrassments like this to come in the next few days, of course). Generally, >* How did you keep the snapshot binary files compatible? > > In a way, persisted Ruby object is not much more than a attribute name => attribute value hash (which is the object state), plus the name of its class (which is the object behaviors). So, you have to retain the same class names for all entities in the storage, and if you add new attributes, even if you always initialize them in a constructor, you may not assume that they are initialized, so you must write a getter that will set these attributes to some defaults if they are not defined yet. This is why in the models/web.rb we have snippets like def wiki() @wiki ||= WikiService.instance; end. >Or on performing the import? > Have a look at what the FileController#import does, and do the same. It is easy to get inside the storage and do things there. Just remember to take a snapshot when you are done (or use only public WikiService methods like revise_page, so that they everything is persisted automatically). -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From assaph at gmail.com Mon May 2 23:41:10 2005 From: assaph at gmail.com (Assaph Mehr) Date: Mon May 2 23:35:00 2005 Subject: [Instiki] File Uploads for 1.0 In-Reply-To: References: <3ff47be905050219143952f9d0@mail.gmail.com> Message-ID: <3ff47be905050220416cd1da71@mail.gmail.com> On 5/3/05, Alexey Verkhovsky wrote: > >* Will file uploads make it to Instiki 1.0? Specifically the 'import' option? > > > Most certainly. The main question is "When Instiki 1.0 is going to > happen". And the answer is "fairly soon, on the open source timescale". > :) I'll enable them fFileController on the trunk right after fixing the > export_html bug (unless there are more embarrassments like this to come > in the next few days, of course). Generally, Good. At this point I'm sure we can hold a contest on who can code slowest past midnight :-) > >* How did you keep the snapshot binary files compatible? > > > In a way, persisted Ruby object is not much more than a attribute name > => attribute value hash (which is the object state), plus the name of > its class (which is the object behaviors). I did, yet I still got errors where madeleine will not load the snapshot. No magic solution, need to investigate further. > So, you have to retain the same class names for all entities in the > storage, and if you add new attributes, even if you always initialize > them in a constructor, you may not assume that they are initialized, so > you must write a getter that will set these attributes to some defaults > if they are not defined yet. This is why in the models/web.rb we have > snippets like def wiki() @wiki ||= WikiService.instance; end. That has been my standard practice for pimki 1.x, but I saw you got rid of all those getters in Instiki 0.10 so I thought you found a way around it. Oh well, if I force users to export/import the markup I can leave those out and only introduce them for later versions. This keeps the code somewhat cleaner for the initial release. > >Or on performing the import? > > > Have a look at what the FileController#import does, and do the same. It > is easy to get inside the storage and do things there. Just remember to > take a snapshot when you are done (or use only public WikiService > methods like revise_page, so that they everything is persisted > automatically). I'll take a look, thanks. Cheers, Assaph From huw.nichols at gmail.com Tue May 3 08:51:45 2005 From: huw.nichols at gmail.com (Huw) Date: Tue May 3 08:45:35 2005 Subject: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option In-Reply-To: References:

Message-ID: Blast, gmail sent direct and not to the list, sorry about that ---------- Forwarded message ---------- From: Huw Date: May 3, 2005 1:48 PM Subject: Re: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option To: Alexey Verkhovsky > >Couple of things: > > > >1) Does anyone else have a problem with the html export showing the > >wrong pages. i.e. the exported HomePage is showing the correct title > >but contents from a different page. So are all other exported pages. > >Maybe something date related as the homepage export is showing a new > >page I created last night just before upgrading. > > > > > Yeah, I have the same problem. So much for fixing things in the last > minute before release, and bypassing unit tests because it is too hard. > Sigh... lesson learned. I'll fix it. I know what you mean :) > >2) Is it possible to have an option to export just the page contents > >as with 0.10 version. > > > Ah, so it was a feature! :) Well I found it very useful anyway, strange at first but after I started to use it that way it actually makes life easier for me > OK, check out SVN revision 254, and use it like > http://localhost:2500/wiki/export_html?layout=no Thanks, I will do. Is the "layout=no" going to stay an option from now on then ? Regards Huw From huw.nichols at gmail.com Tue May 3 19:43:32 2005 From: huw.nichols at gmail.com (Huw) Date: Tue May 3 19:37:21 2005 Subject: [Instiki] Rename a WEB Message-ID: Evening all, Just a quick question, is it possible to rename a web ? And if not can a web be exported in markup form and then imported back into a new [correctly named] web ? Thanks -- TTFN H From assaph at gmail.com Tue May 3 19:46:12 2005 From: assaph at gmail.com (Assaph Mehr) Date: Tue May 3 19:40:03 2005 Subject: [Instiki] Rename a WEB In-Reply-To: References: Message-ID: <3ff47be9050503164677038178@mail.gmail.com> On 5/4/05, Huw wrote: > Just a quick question, is it possible to rename a web ? And if not can > a web be exported in markup form and then imported back into a new > [correctly named] web ? In the 'Edit Web' page you can set the name and address of the web. I think the import feature is still disabled in 0.10.1, but will be active soon. HTH, Assaph From huw.nichols at gmail.com Tue May 3 20:05:27 2005 From: huw.nichols at gmail.com (Huw) Date: Tue May 3 19:59:16 2005 Subject: [Instiki] Rename a WEB In-Reply-To: <3ff47be9050503164677038178@mail.gmail.com> References: <3ff47be9050503164677038178@mail.gmail.com> Message-ID: > > Just a quick question, is it possible to rename a web ? And if not can > > a web be exported in markup form and then imported back into a new > > [correctly named] web ? > > In the 'Edit Web' page you can set the name and address of the web. Ah, Silly me, I should have thought of that, somehow I just missed it Thanks for the pointer I will try it > I think the import feature is still disabled in 0.10.1, but will be active soon. Well if I can rename a web I will not need it just yet, be nice to have though if I ever convince the client to install instiki to view their docs :) -- TTFN H From alex at verk.info Wed May 4 00:22:29 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Wed May 4 00:15:57 2005 Subject: [Instiki] [ANN] Instiki 0.10.1 - The Safe Option In-Reply-To: References:

Message-ID: Huw wrote: >Thanks, I will do. Is the "layout=no" going to stay an option from now on then ? > > Yes. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From inevo at netcabo.pt Wed May 4 07:14:42 2005 From: inevo at netcabo.pt (inEvo) Date: Wed May 4 06:08:29 2005 Subject: [Instiki] FileUploadPatch Message-ID: <4278AEA2.2070803@netcabo.pt> Hi, I'm looking for this patch but have been unable to download it from the wiki page. Where can i get it? I know is not for production use but i really need file uploading for my intranet. Where can i get the patch command? I have a dedicated server with Fedora Core 2 on it but there is no patch command. Regards, Nelson From inevo at netcabo.pt Thu May 5 08:21:36 2005 From: inevo at netcabo.pt (inEvo) Date: Thu May 5 07:15:20 2005 Subject: [Instiki] Nowiki tag ?! Message-ID: <427A0FD0.20601@netcabo.pt> Hi, How can i get the nowiki tag to work ?! Regards, Nelson Silva From alex at verk.info Thu May 5 21:04:57 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 5 20:58:21 2005 Subject: [Instiki] Nowiki tag ?! In-Reply-To: <427A0FD0.20601@netcabo.pt> References: <427A0FD0.20601@netcabo.pt> Message-ID: inEvo wrote: > How can i get the nowiki tag to work ?! > > Until 5 minutes ago - by putting both opening and closing nowiki tags on the same line. As of now (SVN revision 261, and a future release 0.10.2), it will work as you expected. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From oliver at sanbi.ac.za Mon May 9 12:15:33 2005 From: oliver at sanbi.ac.za (Oliver Hofmann) Date: Mon May 9 12:09:23 2005 Subject: [Instiki] How to recover from a crash? Message-ID: <427F8CA5.8080207@sanbi.ac.za> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Greetings, been struggling with Instiki all week, getting plenty of errors and crashes when mixing lists with other Markdown constructs (URLs, code fragments etc). Looks like I've finally done it, one page probably got partially saved and prevents Instiki from starting up again (see error log below). Any way I can rummage around the storage folder and identify the offending page? I'd really hate having to reconvert the MoinMoin wiki manually all over again... Cheers, Oliver - --- MiKTeX-pdfetex 2.4.1793 (1.20a) (MiKTeX 2.4) Copyright (C) 1982 D. E. Knuth, (C) 1996-2002 Han The Thanh TeX is a trademark of the American Mathematical Society. => Starting Instiki on http://0.0.0.0:2500 => Data files are stored in C:/Program Files/instiki-0.10.1/storage/2500 Failed rendering page Stack overflow in regexp matcher: / ^ # Start of line <(p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script) # Start tag: \2 \b # word break (.*\n)*? # Any number of lines, m inimal match <\/\1> # Matching end tag [ ]* # trailing spaces (?=\n+|\Z) # End of line or documen t /ix (RegexpError) C:/Program Files/instiki-0.10.1/lib/bluecloth_tweaked.rb:342:in `gsub!' C:/Program Files/instiki-0.10.1/lib/bluecloth_tweaked.rb:342:in `hide_html_block s' C:/Program Files/instiki-0.10.1/lib/bluecloth_tweaked.rb:248:in `apply_block_tra nsforms' C:/Program Files/instiki-0.10.1/lib/bluecloth_tweaked.rb:202:in `to_html' C:/Program Files/instiki-0.10.1/app/models/chunks/engines.rb:40:in `mask' C:/Program Files/instiki-0.10.1/app/models/chunks/engines.rb:17:in `apply_to' C:/Program Files/instiki-0.10.1/app/models/wiki_content.rb:169:in `build_chunks' C:/Program Files/instiki-0.10.1/app/models/wiki_content.rb:147:in `initialize' C:/Program Files/instiki-0.10.1/app/models/revision.rb:80:in `new' C:/Program Files/instiki-0.10.1/app/models/revision.rb:80:in `display_content' C:/Program Files/instiki-0.10.1/app/models/revision.rb:109:in `force_rendering' C:/Program Files/instiki-0.10.1/app/models/page.rb:37:in `revise' C:/Program Files/instiki-0.10.1/app/models/wiki_service.rb:60:in `revise_page' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine/automatic.r b:143:in `send' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine/automatic.r b:143:in `execute' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:183:in ` execute' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:234:in ` recover_log' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:223:in ` recover_logs' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:222:in ` open_uri_original_open' c:/ruby/lib/ruby/1.8/open-uri.rb:87:in `open' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:222:in ` recover_logs' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:221:in ` each' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:221:in ` recover_logs' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:220:in ` recovery' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:220:in ` recover_logs' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine.rb:56:in `n ew' C:/Program Files/instiki-0.10.1/vendor/madeleine-0.7.1/lib/madeleine/automatic.r b:272:in `initialize' C:/Program Files/instiki-0.10.1/app/models/wiki_service.rb:188:in `new' C:/Program Files/instiki-0.10.1/app/models/wiki_service.rb:188:in `initialize' C:/Program Files/instiki-0.10.1/app/models/wiki_service.rb:151:in `new' C:/Program Files/instiki-0.10.1/app/models/wiki_service.rb:151:in `instance' C:/Program Files/instiki-0.10.1/script/server:91 C:/Program Files/instiki-0.10.1/instiki.rb:3:in `load' C:/Program Files/instiki-0.10.1/instiki.rb:3 - ---- - -- Oliver Hofmann South African National Bioinformatics Institute University of the Western Cape -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFCf4yl9ihIjY7POJMRAlq4AKCYXiIt/fOBIKq1yXCo36K8ZW7DdQCfTtTd t/3gMduZ7cRGHIgPsKL4P0k= =AGSI -----END PGP SIGNATURE----- From alex at verk.info Mon May 9 19:58:12 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 9 19:51:26 2005 Subject: [Instiki] How to recover from a crash? In-Reply-To: <427F8CA5.8080207@sanbi.ac.za> References: <427F8CA5.8080207@sanbi.ac.za> Message-ID: Oliver Hofmann wrote: >I'd really hate having to reconvert the MoinMoin wiki manually all >over again... > > No, we shouldn't let this happen! Oliver, I've emailed some questions to you off-list. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From rdlugosz.10120601 at bloglines.com Mon May 9 23:21:59 2005 From: rdlugosz.10120601 at bloglines.com (rdlugosz.10120601@bloglines.com) Date: Mon May 9 23:15:37 2005 Subject: [Instiki] updated import patch? Message-ID: <1115695319.928525947.14223.sendItem@bloglines.com> Anyone have a lead on an updated import patch available that will recreate a wiki from an exported zip of textile? The one on the Instiki page only handles version 0.8.9, which I suppose is one route to take (install 0.8.9 and import). Just wondering if there's anything more current. Thanks! From benschumacher at gmail.com Tue May 10 02:55:55 2005 From: benschumacher at gmail.com (Ben Schumacher) Date: Tue May 10 02:49:35 2005 Subject: [Instiki] Instiki v0.10.1 for OS X Message-ID: <9859143f050509235527a47472@mail.gmail.com> All- I have completed an OS X build of v0.10.1 of Instiki. I haven't been able to test it much (travelling for work and don't have all my resources available), but it should work fine for folks running Tiger. If you'd like to download and try it out, you can grab it from: http://hiro.blahr.com/dev/Instiki_v0_10_1.dmg Unfortunately, I won't be home for a couple weeks and won't be able to try running it with my external drive with Panther installed. I'm fairly certain that I don't have all the required dependancy in the package for pre-OS X 10.4 users. Sorry about the delay, but hopefully I'll be able to get some good tools ready so that building newer OS X package won't take so long in the future. Cheers, Ben P.S. I have read the comments on the Instiki OS X page and will be happy to implement them as I can, but I don't have a lot of time to give to this project right now. Sorry. From alex at verk.info Tue May 10 09:07:12 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Tue May 10 09:00:26 2005 Subject: [Instiki] Instiki v0.10.1 for OS X In-Reply-To: <9859143f050509235527a47472@mail.gmail.com> References: <9859143f050509235527a47472@mail.gmail.com> Message-ID: Ben Schumacher wrote: >I have completed an OS X build of v0.10.1 of Instiki. I haven't been >able to test it much (travelling for work and don't have all my >resources available), but it should work fine for folks running Tiger. >If you'd like to download and try it out, you can grab it from: > > http://hiro.blahr.com/dev/Instiki_v0_10_1.dmg > >Unfortunately, I won't be home for a couple weeks and won't be able to >try running it with my external drive with Panther installed. I'm >fairly certain that I don't have all the required dependancy in the >package for pre-OS X 10.4 users. Sorry about the delay, but hopefully >I'll be able to get some good tools ready so that building newer OS X >package won't take so long in the future. > > Thanks a lot, Ben! Can you send me something (the build files) to commit to SVN? If somebody decides to give this a try on either Panther or Tiger, please let me know the outcome. If I hear positive reports, I'd like to publish this build on the Instiki downloads page. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From rdlugosz.10120601 at bloglines.com Tue May 10 13:00:33 2005 From: rdlugosz.10120601 at bloglines.com (rdlugosz.10120601@bloglines.com) Date: Tue May 10 12:54:10 2005 Subject: [Instiki] updated import patch? Message-ID: <1115744433.2244259241.4607.sendItem@bloglines.com> --- Ryan References: <9859143f050509235527a47472@mail.gmail.com> Message-ID: Instiki crashes upon start for me. Here is the log. ********** Host Name: iMac-g5 Date/Time: 2005-05-10 16:58:59.662 +0300 OS Version: 10.4 Report Version: 3 Command: Instiki Path: /Users/john/Desktop/Instiki.app/Contents/MacOS/Instiki Parent: WindowServer [64] Version: ??? (0.10.1) PID: 318 Thread: 0 Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000000 Thread 0 Crashed: 0 <<00000000>> 0x00000000 0 + 0 1 com.nextangle.instiki 0x000025e0 start + 60 Thread 0 crashed with PPC Thread State: srr0: 0x00000000 srr1: 0x4200d030 vrsave: 0x00000000 cr: 0x42000422 xer: 0x00000000 lr: 0x00002740 ctr: 0x00000000 r0: 0x00002740 r1: 0xbffffcc0 r2: 0xa0001fdc r3: 0x00000002 r4: 0xbffffd84 r5: 0xbffffd90 r6: 0x00000000 r7: 0x0000036a r8: 0x000088ef r9: 0x00000000 r10: 0x90001d20 r11: 0xa0006428 r12: 0x00000000 r13: 0x00000000 r14: 0x00000000 r15: 0x00000000 r16: 0x00000000 r17: 0x00000000 r18: 0x00000000 r19: 0x00000000 r20: 0x00000000 r21: 0x00000000 r22: 0x00000000 r23: 0x00000000 r24: 0x00000000 r25: 0x00000000 r26: 0xbffffd80 r27: 0x0000000c r28: 0x00000002 r29: 0xbffffd84 r30: 0xbffffd90 r31: 0x000025f8 Binary Images Description: 0x1000 - 0x8fff com.nextangle.instiki ??? (0.10.1) /Users/ john/Desktop/Instiki.app/Contents/MacOS/Instiki 0x8fe00000 - 0x8fe50fff dyld 43 /usr/lib/dyld 0x90000000 - 0x901a6fff libSystem.B.dylib /usr/lib/libSystem.B.dylib 0x901fe000 - 0x90202fff libmathCommon.A.dylib /usr/lib/system/ libmathCommon.A.dylib 0x90204000 - 0x90257fff com.apple.CoreText 1.0.0 (???) /System/ Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/CoreText.framework/Versions/A/CoreText 0x90284000 - 0x90335fff ATS /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ATS.framework/ Versions/A/ATS 0x90364000 - 0x9069cfff com.apple.CoreGraphics 1.256.0 (???) / System/Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/CoreGraphics.framework/Versions/A/CoreGraphics 0x90727000 - 0x90800fff com.apple.CoreFoundation 6.4 (368) /System/ Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation 0x90849000 - 0x90849fff com.apple.CoreServices 10.4 (???) /System/ Library/Frameworks/CoreServices.framework/Versions/A/CoreServices 0x9084b000 - 0x9094dfff libicucore.A.dylib /usr/lib/ libicucore.A.dylib 0x909a7000 - 0x90a2bfff libobjc.A.dylib /usr/lib/libobjc.A.dylib 0x90a55000 - 0x90ac9fff com.apple.framework.IOKit 1.4 (???) / System/Library/Frameworks/IOKit.framework/Versions/A/IOKit 0x90ae3000 - 0x90af5fff libauto.dylib /usr/lib/libauto.dylib 0x90afc000 - 0x90dc1fff com.apple.CoreServices.CarbonCore 10.4 (611.1) /System/Library/Frameworks/CoreServices.framework/Versions/ A/Frameworks/CarbonCore.framework/Versions/A/CarbonCore 0x90e24000 - 0x90ea4fff com.apple.CoreServices.OSServices 4.0 (4.0.0) /System/Library/Frameworks/CoreServices.framework/Versions/ A/Frameworks/OSServices.framework/Versions/A/OSServices 0x90eee000 - 0x90f2efff com.apple.CFNetwork 4.0 (80) /System/ Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/ CFNetwork.framework/Versions/A/CFNetwork 0x90f43000 - 0x90f5bfff com.apple.WebServices 1.1.2 (1.1.0) / System/Library/Frameworks/CoreServices.framework/Versions/A/ Frameworks/WebServicesCore.framework/Versions/A/WebServicesCore 0x90f6b000 - 0x90fe9fff com.apple.SearchKit 1.0.3 /System/Library/ Frameworks/CoreServices.framework/Versions/A/Frameworks/ SearchKit.framework/Versions/A/SearchKit 0x9102e000 - 0x91055fff com.apple.Metadata 0.1 (121) /System/ Library/Frameworks/CoreServices.framework/Versions/A/Frameworks/ Metadata.framework/Versions/A/Metadata 0x91066000 - 0x91073fff libz.1.dylib /usr/lib/libz.1.dylib 0x91076000 - 0x91238fff com.apple.security 4.0 (221) /System/ Library/Frameworks/Security.framework/Versions/A/Security 0x9133a000 - 0x91343fff com.apple.DiskArbitration 2.1 /System/ Library/Frameworks/DiskArbitration.framework/Versions/A/DiskArbitration 0x9134a000 - 0x91371fff com.apple.SystemConfiguration 1.8.0 / System/Library/Frameworks/SystemConfiguration.framework/Versions/A/ SystemConfiguration 0x91384000 - 0x9138cfff libbsm.dylib /usr/lib/libbsm.dylib 0x91390000 - 0x9140efff com.apple.audio.CoreAudio 3.0.0 (3.0) / System/Library/Frameworks/CoreAudio.framework/Versions/A/CoreAudio 0x9144c000 - 0x9144cfff com.apple.ApplicationServices 10.4 (???) / System/Library/Frameworks/ApplicationServices.framework/Versions/A/ ApplicationServices 0x9144e000 - 0x91486fff com.apple.AE 1.5 (297) /System/Library/ Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ AE.framework/Versions/A/AE 0x914a1000 - 0x9156cfff com.apple.ColorSync 4.4 /System/Library/ Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ ColorSync.framework/Versions/A/ColorSync 0x915c1000 - 0x91654fff com.apple.print.framework.PrintCore 4.0 (172) /System/Library/Frameworks/ApplicationServices.framework/ Versions/A/Frameworks/PrintCore.framework/Versions/A/PrintCore 0x9169a000 - 0x91757fff com.apple.QD 3.8.5 (???) /System/Library/ Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ QD.framework/Versions/A/QD 0x91795000 - 0x917f3fff com.apple.HIServices 1.5.0 (???) /System/ Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/HIServices.framework/Versions/A/HIServices 0x91821000 - 0x91844fff com.apple.LangAnalysis 1.6 /System/Library/ Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ LangAnalysis.framework/Versions/A/LangAnalysis 0x91858000 - 0x9187dfff com.apple.FindByContent 1.5 /System/ Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/FindByContent.framework/Versions/A/FindByContent 0x91890000 - 0x918d0fff com.apple.LaunchServices 10.4 (118) / System/Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/LaunchServices.framework/Versions/A/LaunchServices 0x918eb000 - 0x918fffff com.apple.speech.synthesis.framework 3.3 / System/Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/SpeechSynthesis.framework/Versions/A/SpeechSynthesis 0x9190d000 - 0x91943fff com.apple.ImageIO.framework 1.0 /System/ Library/Frameworks/ApplicationServices.framework/Versions/A/ Frameworks/ImageIO.framework/Versions/A/ImageIO 0x91957000 - 0x91a19fff libcrypto.0.9.7.dylib /usr/lib/libcrypto. 0.9.7.dylib 0x91a65000 - 0x91a7afff libcups.2.dylib /usr/lib/libcups.2.dylib 0x91a7f000 - 0x91a9bfff libJPEG.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libJPEG.dylib 0x91aa0000 - 0x91b0ffff libJP2.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libJP2.dylib 0x91b26000 - 0x91b2afff libGIF.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libGIF.dylib 0x91b2c000 - 0x91b44fff libRaw.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libRaw.dylib 0x91b47000 - 0x91b8afff libTIFF.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libTIFF.dylib 0x91b91000 - 0x91baafff libPng.dylib /System/Library/Frameworks/ ApplicationServices.framework/Versions/A/Frameworks/ImageIO.framework/ Versions/A/Resources/libPng.dylib 0x91baf000 - 0x91bb2fff libRadiance.dylib /System/Library/ Frameworks/ApplicationServices.framework/Versions/A/Frameworks/ ImageIO.framework/Versions/A/Resources/libRadiance.dylib 0x91bb4000 - 0x91bb4fff com.apple.Accelerate 1.1 (Accelerate 1.1) / System/Library/Frameworks/Accelerate.framework/Versions/A/Accelerate 0x91bb6000 - 0x91ca0fff com.apple.vImage 2.0 /System/Library/ Frameworks/Accelerate.framework/Versions/A/Frameworks/ vImage.framework/Versions/A/vImage 0x91ca8000 - 0x91cc7fff com.apple.Accelerate.vecLib 3.1 (vecLib 3.1) /System/Library/Frameworks/Accelerate.framework/Versions/A/ Frameworks/vecLib.framework/Versions/A/vecLib 0x91d33000 - 0x91d53fff libmx.A.dylib /usr/lib/libmx.A.dylib 0x91d59000 - 0x91dbefff libvMisc.dylib /System/Library/Frameworks/ Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/ A/libvMisc.dylib 0x91dc8000 - 0x91e5afff libvDSP.dylib /System/Library/Frameworks/ Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/ A/libvDSP.dylib 0x91e74000 - 0x92404fff libBLAS.dylib /System/Library/Frameworks/ Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/ A/libBLAS.dylib 0x9244c000 - 0x9275cfff libLAPACK.dylib /System/Library/ Frameworks/Accelerate.framework/Versions/A/Frameworks/ vecLib.framework/Versions/A/libLAPACK.dylib 0x92789000 - 0x92814fff com.apple.DesktopServices 1.3 /System/ Library/PrivateFrameworks/DesktopServicesPriv.framework/Versions/A/ DesktopServicesPriv 0x92856000 - 0x92a7ffff com.apple.Foundation 6.4 (567) /System/ Library/Frameworks/Foundation.framework/Versions/C/Foundation 0x92b9d000 - 0x92c7bfff libxml2.2.dylib /usr/lib/libxml2.2.dylib 0x92c9b000 - 0x92d89fff libiconv.2.dylib /usr/lib/libiconv.2.dylib 0x92d9b000 - 0x92db9fff libGL.dylib /System/Library/Frameworks/ OpenGL.framework/Versions/A/Libraries/libGL.dylib 0x92dc4000 - 0x92e1efff libGLU.dylib /System/Library/Frameworks/ OpenGL.framework/Versions/A/Libraries/libGLU.dylib 0x92e6a000 - 0x92e7afff com.apple.speech.recognition.framework 3.4 /System/Library/Frameworks/Carbon.framework/Versions/A/ Frameworks/SpeechRecognition.framework/Versions/A/SpeechRecognition 0x930d5000 - 0x930f2fff com.apple.audio.SoundManager 3.9 /System/ Library/Frameworks/Carbon.framework/Versions/A/Frameworks/ CarbonSound.framework/Versions/A/CarbonSound 0x9311a000 - 0x93429fff com.apple.HIToolbox 1.4.0 (???) /System/ Library/Frameworks/Carbon.framework/Versions/A/Frameworks/ HIToolbox.framework/Versions/A/HIToolbox 0x93574000 - 0x93580fff com.apple.opengl 1.4.0 /System/Library/ Frameworks/OpenGL.framework/Versions/A/OpenGL 0x93612000 - 0x93612fff com.apple.Cocoa 6.4 (???) /System/Library/ Frameworks/Cocoa.framework/Versions/A/Cocoa 0x93614000 - 0x93c45fff com.apple.AppKit 6.4 (824) /System/Library/ Frameworks/AppKit.framework/Versions/C/AppKit 0x93fd1000 - 0x9403bfff com.apple.CoreData 1.0 (46) /System/ Library/Frameworks/CoreData.framework/Versions/A/CoreData 0x94073000 - 0x9413dfff com.apple.audio.toolbox.AudioToolbox 1.4 / System/Library/Frameworks/AudioToolbox.framework/Versions/A/AudioToolbox 0x94191000 - 0x94191fff com.apple.audio.units.AudioUnit 1.4 / System/Library/Frameworks/AudioUnit.framework/Versions/A/AudioUnit 0x94193000 - 0x942f2fff com.apple.QuartzCore 1.4 /System/Library/ Frameworks/QuartzCore.framework/Versions/A/QuartzCore 0x9433a000 - 0x94377fff libsqlite3.0.dylib /usr/lib/ libsqlite3.0.dylib 0x9437f000 - 0x943cafff libGLImage.dylib /System/Library/ Frameworks/OpenGL.framework/Versions/A/Libraries/libGLImage.dylib 0x95482000 - 0x95505fff libstdc++.6.dylib /usr/lib/libstdc++.6.dylib 0x95583000 - 0x9558bfff libgcc_s.1.dylib /usr/lib/libgcc_s.1.dylib 0x968e1000 - 0x968e2fff com.apple.zerolink 1.2 (3) /System/Library/ PrivateFrameworks/ZeroLink.framework/Versions/A/ZeroLink On May 10, 2005, at 9:55 AM, Ben Schumacher wrote: > All- > > I have completed an OS X build of v0.10.1 of Instiki. I haven't been > able to test it much (travelling for work and don't have all my > resources available), but it should work fine for folks running Tiger. > If you'd like to download and try it out, you can grab it from: > > http://hiro.blahr.com/dev/Instiki_v0_10_1.dmg > > Unfortunately, I won't be home for a couple weeks and won't be able to > try running it with my external drive with Panther installed. I'm > fairly certain that I don't have all the required dependancy in the > package for pre-OS X 10.4 users. Sorry about the delay, but hopefully > I'll be able to get some good tools ready so that building newer OS X > package won't take so long in the future. > > Cheers, > > Ben > > P.S. I have read the comments on the Instiki OS X page and will be > happy to implement them as I can, but I don't have a lot of time to > give to this project right now. Sorry. > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > =================================== I am modifying all my mail accounts so that they route via john@3eyes.co.uk. Please update your address books. However mail to my UIAH address will still get to me for the time being. This is more a long term strategy, but please take notice -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instiki-users/attachments/20050510/5826d190/attachment-0001.htm From list at tylerhunt.com Wed May 11 15:48:52 2005 From: list at tylerhunt.com (Tyler Hunt) Date: Wed May 11 15:42:29 2005 Subject: [Instiki] Modifying Data Message-ID: <24C4B441-2731-4253-A0AE-615F06D9DB44@tylerhunt.com> I just set up Instiki last night, and so far everything is working pretty well. I'm frustrated by the lack of administration controls, however, in particular the lack of page renaming. I'm also currently trying to figure out how to resolve a situation in which I need to remove authors, or more specifically revisions from certain authors. I have tried going into the command_log and editing out a series of revision to remove the references to that particular author, but I get "dump format error(0x74)" on restart when doing that. I have also looked into the export/import patches discussed on instiki.org, but it doesn't seem like any of these work with the current version. Is there a way to export all of the data, modify certain bits, and then import it all back into Instiki again, ideally retaining revisions? Are there working patches that I haven't found? From tyler at tylerhunt.com Wed May 11 15:46:35 2005 From: tyler at tylerhunt.com (Tyler Hunt) Date: Wed May 11 19:53:36 2005 Subject: [Instiki] Modifying Data Message-ID: <515DCEDA-ED64-4EEC-BCA0-2BAD1325953B@tylerhunt.com> I just set up Instiki last night, and so far everything is working pretty well. I'm frustrated by the lack of administration controls, however, in particular the lack of page renaming. I'm also currently trying to figure out how to resolve a situation in which I need to remove authors, or more specifically revisions from certain authors. I have tried going into the command_log and editing out a series of revision to remove the references to that particular author, but I get "dump format error(0x74)" on restart when doing that. I have also looked into the export/import patches discussed on instiki.org, but it doesn't seem like any of these work with the current version. Is there a way to export all of the data, modify certain bits, and then import it all back into Instiki again, ideally retaining revisions? Are there working patches that I haven't found? From tamassmith at mac.com Thu May 12 12:42:48 2005 From: tamassmith at mac.com (=?ISO-8859-1?Q?Tam=E1s_Smith?=) Date: Thu May 12 12:36:26 2005 Subject: [Instiki] Modifying Data In-Reply-To: <515DCEDA-ED64-4EEC-BCA0-2BAD1325953B@tylerhunt.com> References: <515DCEDA-ED64-4EEC-BCA0-2BAD1325953B@tylerhunt.com> Message-ID: <82d676a5186b238a5c19dfabdbe624e1@mac.com> Tyler As I understand it, Instiki does not have any real concept of authors or a user database. 'Authors' just consist of a page and a cookie tracks your 'author page' to put it in the little text field at bottom after you edit the first page during a session. For secured sites, a password allows you to have binary access control, but has nothing to do with authorship. I think if you just get forget the concept of an author beyond a linked page, it will help. Also, regarding page renaming, you can just: 1. Create another link next to the old link name on the existing page that references it. 2. Edit the page you want to rename, copy the content from text field, and cancel the editing. 3. Go to the existing page the references it, edit it, and change the name for the page there. 4. Click on the new page name question mark on this existing page to create the new page and copy the content into the text field You might also want to go in and remove all the content of the old page, or replace its contents with 'DELETE' so it does not interfere with search results. A facility exists to delete all orphaned pages that do have links from any of the existing pages connected to the HomePage 'origin', so this page will get cleared up eventually if nothing else links to it. I agree it would be nice to have a way to rename without going through these steps, but it adds a lot of overhead, especially considering that you have n possible pages with links to the page that would either: 1. Need to have their content programmatically modified to point to the new link 2. Have links to a page with no content since the content moved with the rename I would rather not have the content prone to automated search and replace in this fashion as it means a rename of a page has the *potential* to corrupt all the pages that link to it by a bad search and replace. Also, to what author to you ascribe the change on all these linking pages? What if others are currently editing some of those pages? You can see the complications become legion. Basically, Instiki is simpler than it seems, but in some ways, that's the beauty and power of it. It actually needs to get *simpler* then it is now. The alternative is Twiki, which has tons of administration controls, but is basically unusable because of them. If you stop and think, do I really *need* X feature or administrative control you will find, in most cases, you do not. Sorry for the long message, but I felt the same way you did when I first started using it, but over time, became more and more comfortable. Tam?s On May 11, 2005, at 3:46 PM, Tyler Hunt wrote: I just set up Instiki last night, and so far everything is working pretty well. I'm frustrated by the lack of administration controls, however, in particular the lack of page renaming. I'm also currently trying to figure out how to resolve a situation in which I need to remove authors, or more specifically revisions from certain authors. I have tried going into the command_log and editing out a series of revision to remove the references to that particular author, but I get "dump format error(0x74)" on restart when doing that. I have also looked into the export/import patches discussed on instiki.org, but it doesn't seem like any of these work with the current version. Is there a way to export all of the data, modify certain bits, and then import it all back into Instiki again, ideally retaining revisions? Are there working patches that I haven't found? _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From assaph at gmail.com Thu May 12 21:20:19 2005 From: assaph at gmail.com (Assaph Mehr) Date: Thu May 12 21:13:52 2005 Subject: [Instiki] Modifying Data In-Reply-To: <515DCEDA-ED64-4EEC-BCA0-2BAD1325953B@tylerhunt.com> References: <515DCEDA-ED64-4EEC-BCA0-2BAD1325953B@tylerhunt.com> Message-ID: <3ff47be9050512182044017bc7@mail.gmail.com> Hi Tyler, On 5/12/05, Tyler Hunt wrote: > I just set up Instiki last night, and so far everything is working > pretty well. I'm frustrated by the lack of administration controls, > however, in particular the lack of page renaming. In Instiki you have to follow the steps Tom?s mentioned to rename pages, specifically the delete orphaned pages. (I didn't like it either so I forked Instiki). > I'm also currently trying to figure out how to resolve a situation in > which I need to remove authors, or more specifically revisions from > certain authors. I have tried going into the command_log and editing > out a series of revision to remove the references to that particular > author, but I get "dump format error(0x74)" on restart when doing > that. I have also looked into the export/import patches discussed on > instiki.org, but it doesn't seem like any of these work with the > current version. Is there a way to export all of the data, modify > certain bits, and then import it all back into Instiki again, ideally > retaining revisions? Are there working patches that I haven't found? Do you mean delete discreet revisions of pages? You can always rollback to a particular revision of a page. Do you mean delete all revisions by a certain author across all pages? There is no way to currently do that. As you've discovered, editing the binary madeleine dump files is not a good idea :-). As for export/import: the functionality is there - but disabled - in Instiki 0.10.1 as it's not production ready. Someone (Ryan I think) posted a couple of days ago that he did make it work with minimal changes. Ryan, perhaps you can post a patch? Alexey, I know it'll break things, but it's easy to make madeleine use YAML as the dump file format (zipped up). I make a patch on Instiki core to enable this as a command line option. This will of course mean that you cannot interchange snapshots files between the two formats (unless I come up with an auto-detection way), but given enough upfront warnings some people may choose to always use this format. ps. As Tom?s pointed out re user security, explicit simplicity and lack of security is actually better than attempted incomplete security. Do you have a particular need that you're trying to solve? Reading between the lines it sounds like you have a public wiki that has been spammed. To my understanding, there is no way to combat this effectively on public wikis, except making then password protected (i.e. only if you're one of a privileged few who know the edit password, you can only read). This is a global issue with wikis and blog, not just for Instiki. I actually prefer the Instiki way - by not attempting to solve it (which you can't) you significantly decrease the application complexity. HTH, Assaph From rdlugosz.10120601 at bloglines.com Thu May 12 23:24:58 2005 From: rdlugosz.10120601 at bloglines.com (rdlugosz.10120601@bloglines.com) Date: Thu May 12 23:18:33 2005 Subject: [Instiki] Modifying Data Message-ID: <1115954698.397983974.30145.sendItem@bloglines.com> --- Assaph Mehr As for export/import: the functionality is there - but disabled - in > Instiki 0.10.1 as it's not production ready. Someone (Ryan I think) > posted a couple of days ago that he did make it work with minimal > changes. Ryan, perhaps you can post a patch? I can't do a diff from where I'm at, but basically you just need to make two simple edits to the app/controllers/file_controller.rb file. First, in the method "import", comment out the "return if file_uploads_disabled?" line. Second, go down to the next method "check_allow_uploads" and change the first line from "unless RAILS_ENV == 'test'" to "unless RAILS_ENV == 'production'". Now, go to yourWiki/import and you'll get an interface which will let you upload a zip full of textile. It will import each page in the zip into your wiki. I haven't seen any glaring problems yet, but stay tuned for my next email which may be related to this. (but may not). From rdlugosz.10120601 at bloglines.com Thu May 12 23:30:13 2005 From: rdlugosz.10120601 at bloglines.com (rdlugosz.10120601@bloglines.com) Date: Thu May 12 23:23:53 2005 Subject: [Instiki] Instiki not writing snapshots? Message-ID: <1115955013.3717863143.5874.sendItem@bloglines.com> Lately my instiki hasn't been outputting any snapshots to the storage directory. I'm running 0.10.1 and the only "strange" thing about my setup is that I've imported the content via the (production disabled) /import action that I've discussed in other threads here. I've got an ever-growing 00...01.command_log file, but there are *no* snapshot files! I guess this probably isn't the end of the world, presuming Instiki will restart correctly should I bounce it, but something doesn't seem right. Any thoughts? From alex at verk.info Thu May 12 23:48:16 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 12 23:41:23 2005 Subject: [Instiki] Instiki not writing snapshots? In-Reply-To: <1115955013.3717863143.5874.sendItem@bloglines.com> References: <1115955013.3717863143.5874.sendItem@bloglines.com> Message-ID: rdlugosz.10120601@bloglines.com wrote: >I've got an ever-growing 00...01.command_log >file, but there are *no* snapshot files! I guess this probably isn't the >end of the world, presuming Instiki will restart correctly should I bounce >it, but something doesn't seem right. Any thoughts? > > The thread responsible for taking snapshots is in wiki_service.rb, start_snapshot_thread method. It will always try to write a snapshot on startup, if there is a command.log; if it cannot (for whatever reason), it should log an error (including full stack trace) in stderr. Please try to run instiki from command line, without --daemon option, and see what it has to say. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From chris at seagul.co.uk Fri May 13 17:16:44 2005 From: chris at seagul.co.uk (Chris Roos) Date: Fri May 13 17:10:54 2005 Subject: [Instiki] --binding param not working on freebsd Message-ID: <4285193B.4010100@seagul.co.uk> Hi, I'm running instiki 0.10.1 on FreeBsd 4.9 with Apache 1.3 utilising mod_proxy. Because of this, I wanted to ensure that instiki was only listening to the local machine. As far as I was aware, I should use the binding parameter to achieve this. However, after changing this to 127.0.0.1 (both on the command line and manually in the server ruby file) I can still connect from remote machines direct to :2500. I also tried this with the svn version of instiki and the same thing happens. I then tried this on two windows boxes with both the 0.10.1 and svn versions. In both cases the --binding flag worked and limited connections to the local machine. I also tried it on FreeBsd 5.3R with 0.10.1 installed via Gems and the --binding flag also worked as expected. I'm guessing that I've done something stupid on the 4.9 box but can't think of what it may be or of anything else to test at the moment... Any thoughts? Chris From rdlugosz.10120601 at bloglines.com Sat May 14 13:51:40 2005 From: rdlugosz.10120601 at bloglines.com (rdlugosz.10120601@bloglines.com) Date: Sat May 14 13:45:11 2005 Subject: [Instiki] Instiki not writing snapshots? Message-ID: <1116093100.1764243869.2675.sendItem@bloglines.com> --- Instiki-users Please try to run instiki from command line, without --daemon option, > and see what it has to say. Alexey - thanks for the reply. I did this just now, and it successfully completed writing the snapshot. Must have been some sort of transient error! -Ryan From alex at verk.info Sat May 14 21:32:18 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sat May 14 21:25:21 2005 Subject: [Instiki] --binding param not working on freebsd In-Reply-To: <4285193B.4010100@seagul.co.uk> References: <4285193B.4010100@seagul.co.uk> Message-ID: Chris Roos wrote: > I should use the binding parameter to achieve this. However, after > changing this to 127.0.0.1 (both on the command line and manually in > the server ruby file) I can still connect from remote machines direct > to :2500. The argument of --binding option is eventually passed to the underlying Socket object. If it doesn't do what it is supposed to do, this must be between your Ruby build and your OS. It does work on FreeBSD 5.3 (running instiki.org and rubyonrails.com). -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From list at tylerhunt.com Sun May 15 20:13:33 2005 From: list at tylerhunt.com (Tyler Hunt) Date: Sun May 15 20:07:00 2005 Subject: [Instiki] Modifying Data In-Reply-To: <1115954698.397983974.30145.sendItem@bloglines.com> References: <1115954698.397983974.30145.sendItem@bloglines.com> Message-ID: On May 12, 2005, at 23:24, rdlugosz.10120601@bloglines.com wrote: > I can't do a diff from where I'm at, but basically you just need to > make two simple > edits to the app/controllers/file_controller.rb file. First, in > the method > "import", comment out the "return if file_uploads_disabled?" line. > > Second, > go down to the next method "check_allow_uploads" and change the > first line > from "unless RAILS_ENV == 'test'" to "unless RAILS_ENV == > 'production'". > > Now, go to yourWiki/import and you'll get an interface which will > let you > upload a zip full of textile. It will import each page in the zip > into your > wiki. I followed these instructions, and with a little further tweaking to change the max_upload_size was able to get import working. I was looking for something that would maintain the revisions and authorship, however, so I've decided to suck up my perfectionism and try to let the wiki live on it's own for the time being. Thanks for the help and suggestions, Tyler Hunt From chris at seagul.co.uk Wed May 18 03:45:29 2005 From: chris at seagul.co.uk (Chris Roos) Date: Wed May 18 03:39:02 2005 Subject: [Instiki] --binding param not working on freebsd In-Reply-To: References: <4285193B.4010100@seagul.co.uk> Message-ID: <428AF299.4090609@seagul.co.uk> Alexey Verkhovsky wrote: > Chris Roos wrote: > >> I should use the binding parameter to achieve this. However, after >> changing this to 127.0.0.1 (both on the command line and manually in >> the server ruby file) I can still connect from remote machines direct >> to :2500. > > > The argument of --binding option is eventually passed to the underlying > Socket object. If it doesn't do what it is supposed to do, this must be > between your Ruby build and your OS. It does work on FreeBSD 5.3 > (running instiki.org and rubyonrails.com). > Right, I've got a bit more info. The problems I am having are because the FreeBsd that I am running is actually an instance in a Jail. In this case, the Jail only knows about the public IP, it doesn't have access to the Loopback (or any other local) address. As a fix, I have requested that a firewall rule be set-up to block access to port 2500 on this IP. However, it's been suggested that if I could configure the daemon to use Unix Domain Sockets rather than IP sockets that I could get round the current problem. I realise this becomes more of a general Ruby question but just wonder if you know whether it would be at all possible? I'm guessing that the code I'd be interested in lies in the socket.so library? Thanks for your help, Chris From nicholas_wieland at yahoo.it Wed May 18 09:26:35 2005 From: nicholas_wieland at yahoo.it (Nicholas Wieland) Date: Wed May 18 09:19:04 2005 Subject: [Instiki] Change charset in instiki Message-ID: <20050518132635.GA10197@pixie.subbacultcha.home> Ciao *, I'm working on setting up instiki for rubyonrails.it, and want to add a few custom headers to make it work with italian people. What I understood is that the best way to accomplish such task is to edit app/views/layouts/default.rhtml, so I changed to and to . The problem is that I've also added a custom header (the green one you can see at www.rubyonrails.it), so I have a lot of strange characters instead of "?????" ... I really can't see a problem with my approach: as far as the page is served as ISO-8859-1 everything should work fine (and in fact www.rubyonrails.it renders ok). I'm using Webrick, so my guess is that it has a default charset that overwrites the one I set, but to be honest I'm quite clueless now :) TIA, ngw -- checking for life_signs in -lKenny... no Oh my god, make (1) killed Kenny ! You, bastards ! nicholas_wieland-at-yahoo-dot-it From dmertz at online.de Wed May 18 10:26:50 2005 From: dmertz at online.de (Denis Mertz) Date: Wed May 18 10:16:40 2005 Subject: [Instiki] Change charset in instiki In-Reply-To: <20050518132635.GA10197@pixie.subbacultcha.home> References: <20050518132635.GA10197@pixie.subbacultcha.home> Message-ID: <200505181626.50291.dmertz@online.de> Le Mercredi 18 Mai 2005 15:26, Nicholas Wieland a ?crit?: > Ciao *, > I'm working on setting up instiki for rubyonrails.it, and want to > add a few custom headers to make it work with italian people. > What I understood is that the best way to accomplish such task is to > edit app/views/layouts/default.rhtml, so I changed > > > > to > > > > and > > > > to > > . > Hello, What motivates this change ? Is'nt utf8 just fine, even for an italian site ? > The problem is that I've also added a custom header (the green one you > can see at www.rubyonrails.it), so I have a lot of strange characters > instead of "?????" ... > > I really can't see a problem with my approach: as far as the page is > served as ISO-8859-1 everything should work fine (and in fact > www.rubyonrails.it renders ok). Where is the problem exactly then ? Where do you see the strange chars ? > > I'm using Webrick, so my guess is that it has a default charset that > overwrites the one I set, but to be honest I'm quite clueless now :) > > TIA, > ngw From alex at verk.info Thu May 19 00:22:08 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 19 00:15:01 2005 Subject: [Instiki] --binding param not working on freebsd In-Reply-To: <428AF299.4090609@seagul.co.uk> References: <4285193B.4010100@seagul.co.uk> <428AF299.4090609@seagul.co.uk> Message-ID: Chris Roos wrote: > However, it's been suggested that if I could configure the daemon to > use Unix Domain Sockets rather than IP sockets that I could get round > the current problem. Admittedly, I don't even know if it's possible to create Unix domain sockets with Ruby :) Probably it is. Whether it's possible to make WebRick bind to it, is another interesting question... You should try to ask about it on Ruby-Talk. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Thu May 19 00:26:15 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 19 00:19:08 2005 Subject: [Instiki] Change charset in instiki In-Reply-To: <200505181626.50291.dmertz@online.de> References: <20050518132635.GA10197@pixie.subbacultcha.home> <200505181626.50291.dmertz@online.de> Message-ID: Denis Mertz wrote: >What motivates this change ? Is'nt utf8 just fine, even for an italian site ? > > >Where is the problem exactly then ? Where do you see the strange chars ? > > Nicholas, I have the same questions as Denis :) -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From chris at seagul.co.uk Thu May 19 02:22:06 2005 From: chris at seagul.co.uk (Chris Roos) Date: Thu May 19 02:15:40 2005 Subject: [Instiki] --binding param not working on freebsd In-Reply-To: References: <4285193B.4010100@seagul.co.uk> <428AF299.4090609@seagul.co.uk> Message-ID: <428C308E.6030808@seagul.co.uk> Alexey Verkhovsky wrote: > Chris Roos wrote: > >> However, it's been suggested that if I could configure the daemon to >> use Unix Domain Sockets rather than IP sockets that I could get round >> the current problem. > > > Admittedly, I don't even know if it's possible to create Unix domain > sockets with Ruby :) > Probably it is. Whether it's possible to make WebRick bind to it, is > another interesting question... > > You should try to ask about it on Ruby-Talk. > Thanks for your help Alexey. I've just posted a question on ruby-talk now. Chris From nicholas_wieland at yahoo.it Thu May 19 05:59:27 2005 From: nicholas_wieland at yahoo.it (Nicholas Wieland) Date: Thu May 19 05:53:41 2005 Subject: [Instiki] Change charset in instiki In-Reply-To: <428C1567.40407@verk.info> References: <20050518132635.GA10197@pixie.subbacultcha.home> <200505181626.50291.dmertz@online.de> <428C1567.40407@verk.info> Message-ID: <20050519095927.GA31512@pixie.subbacultcha.home> - Alexey Verkhovsky : > I have the same questions as Denis :) Even if I use UTF-8 the problem remains, so the charset doesn't count. Does Webrick use a default charset ? If you or Denis could try to add a pair of ????? to the default layout and see if the page renders correctly I'd know if I have to mess with instiki or just my system ... Thank you both, ngw -- checking for life_signs in -lKenny... no Oh my god, make (1) killed Kenny ! You, bastards ! nicholas_wieland-at-yahoo-dot-it From dmertz at online.de Thu May 19 06:45:30 2005 From: dmertz at online.de (Denis Mertz) Date: Thu May 19 06:35:21 2005 Subject: [Instiki] Change charset in instiki In-Reply-To: <20050519095927.GA31512@pixie.subbacultcha.home> References: <20050518132635.GA10197@pixie.subbacultcha.home> <428C1567.40407@verk.info> <20050519095927.GA31512@pixie.subbacultcha.home> Message-ID: <200505191245.31466.dmertz@online.de> Le Jeudi 19 Mai 2005 11:59, Nicholas Wieland a ?crit?: > - Alexey Verkhovsky : > > I have the same questions as Denis :) > > Even if I use UTF-8 the problem remains, so the charset doesn't count. > Does Webrick use a default charset ? > If you or Denis could try to add a pair of ????? to the default layout > and see if the page renders correctly I'd know if I have to mess with > instiki or just my system ... Just a guess, Ruby scripts are encoded either in ASCII or UTF-8 or EUC-JP or Shift-JIS, the last two are for Japanese. Instiki uses UTF-8, and you probably dont want to change that. But this means that you need to encode the Instiki source files (including the .rhtml templates) in UTF-8. I will make a test myself later, do not have enough time now. Ciao Denis > > Thank you both, > ngw From chris at seagul.co.uk Mon May 23 10:00:45 2005 From: chris at seagul.co.uk (Chris Roos) Date: Mon May 23 09:54:19 2005 Subject: [Instiki] Madeleine corruption Message-ID: <4291E20D.1060500@seagul.co.uk> I have instiki 0.10.1 configured with multiple webs on FreeBsd. It has been working fine for quite a while now. Unfortunately I was just editing a page and on saving received an error. From that point on I was unable to access that particular web - all of the others could be accessed fine. I copied the snapshot files to a local windows machine (with Instiki from svn) and tested on there. I get an error "undefined method `length' for nil:NilClass" when clicking on the corrupt web. In addition, the web is opened in edit mode automatically. No changes can be made. I have copied the log file (when running on windows) at the bottom of the email. I removed the latest snapshot file and the corrupt web can be accessed correctly, so the corruption is only in the latest snapshot. Unfortunately the previous snapshot is four days old and so I would prefer it if somehow I could recover the information from the latest snapshot. The question is, I don't really have a clue as to how I would about doing that..? Any help and advice would be much appreciated. Chris D:\_svn\instiki\trunk>instiki --storage=C:\TEMP\instiki => Starting Instiki on http://0.0.0.0:2500 => Data files are stored in C:\TEMP\instiki/2500 [2005-05-23 14:51:50] INFO WEBrick 1.3.1 [2005-05-23 14:51:50] INFO ruby 1.8.2 (2004-12-25) [i386-mswin32] [2005-05-23 14:51:50] INFO WEBrick::HTTPServer#start: pid=2544 port=2500 Processing WikiController#index (for 127.0.0.1 at Mon May 23 14:51:54 GMT Daylig ht Time 2005) Parameters: {"action"=>"index", "controller"=>"wiki"} Redirected to http://localhost:2500/web_list Completed in 0.01000 (100 reqs/sec) 127.0.0.1 - - [23/May/2005:14:51:53 GMT Daylight Time] "GET / HTTP/1.1" 302 96 - -> / Processing WikiController#web_list (for 127.0.0.1 at Mon May 23 14:51:55 GMT Day light Time 2005) Parameters: {"action"=>"web_list", "controller"=>"wiki"} Rendering wiki/web_list within layouts/default Rendering layouts/default (200 OK) Completed in 0.10000 (9 reqs/sec) | Rendering: 0.09000 (90%) 127.0.0.1 - - [23/May/2005:14:51:55 GMT Daylight Time] "GET /web_list HTTP/1.1" 200 2860 - -> /web_list Processing WikiController#show (for 127.0.0.1 at Mon May 23 14:51:59 GMT Dayligh t Time 2005) Parameters: {"action"=>"show", "id"=>"HomePage", "controller"=>"wiki", "web"=> "versitec"} Rendering wiki/page within layouts/default undefined method `length' for nil:NilClass (ActionView::TemplateError) On line #81 of /wiki/page.rhtml 78: <% end %> 79: 80: 81: <% unless @page.linked_from.empty? %> 82: 83: | Linked from: 84: <%= @page.linked_from.collect { |referring_page| /RedCloth-3.0.3/lib/redcloth.rb:533:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' /RedCloth-3.0.3/lib/redcloth.rb:522:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:518:in `gsub!' /RedCloth-3.0.3/lib/redcloth.rb:518:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:612:in `call' /RedCloth-3.0.3/lib/redcloth.rb:612:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:611:in `each' /RedCloth-3.0.3/lib/redcloth.rb:611:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:586:in `collect' /RedCloth-3.0.3/lib/redcloth.rb:586:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:288:in `to_html' D:/_svn/instiki/trunk/app/models/chunks/engines.rb:34:in `mask' D:/_svn/instiki/trunk/app/models/chunks/engines.rb:17:in `apply_to' D:/_svn/instiki/trunk/app/models/wiki_content.rb:169:in `build_chunks' D:/_svn/instiki/trunk/app/models/wiki_content.rb:147:in `initialize' D:/_svn/instiki/trunk/app/models/revision.rb:80:in `new' D:/_svn/instiki/trunk/app/models/revision.rb:80:in `display_content' D:/_svn/instiki/trunk/app/models/revision.rb:58:in `wiki_words' D:/_svn/instiki/trunk/app/models/page.rb:111:in `send' D:/_svn/instiki/trunk/app/models/page.rb:111:in `method_missing' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `select' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' D:/_svn/instiki/trunk/app/models/page.rb:78:in `linked_from' (erb):81:in `evaluate_locals' /rails/actionpack/lib/action_view/base.rb:265:in `evaluate_locals' /rails/actionpack/lib/action_view/base.rb:276:in `rhtml_render' /rails/actionpack/lib/action_view/base.rb:201:in `send' /rails/actionpack/lib/action_view/base.rb:201:in `render_template' /rails/actionpack/lib/action_view/base.rb:181:in `render_file' /rails/actionpack/lib/action_controller/layout.rb:208:in `render_without_ben chmark' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `measure' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' /rails/actionpack/lib/action_controller/base.rb:447:in `render_action' D:/_svn/instiki/trunk/app/controllers/wiki_controller.rb:203:in `show' /rails/actionpack/lib/action_controller/base.rb:708:in `send' /rails/actionpack/lib/action_controller/base.rb:708:in `perform_action_witho ut_filters' /rails/actionpack/lib/action_controller/filters.rb:294:in `perform_action_wi thout_benchmark' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `perform_actio n_without_rescue' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `measure' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `perform_actio n_without_rescue' /rails/actionpack/lib/action_controller/rescue.rb:79:in `perform_action' /rails/actionpack/lib/action_controller/base.rb:330:in `send' /rails/actionpack/lib/action_controller/base.rb:330:in `process' /rails/actionpack/lib/action_controller/base.rb:273:in `process' /rails/railties/lib/dispatcher.rb:32:in `dispatch' D:/_svn/instiki/trunk/public/dispatch.rb:10 /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' /rails/railties/lib/webrick_server.rb:82:in `handle_dispatch' /rails/railties/lib/webrick_server.rb:35:in `service' c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:104:in `service' c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:65:in `run' c:/ruby/lib/ruby/1.8/webrick/server.rb:155:in `start_thread' c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start_thread' c:/ruby/lib/ruby/1.8/webrick/server.rb:94:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `each' c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' /rails/railties/lib/webrick_server.rb:21:in `dispatch' D:/_svn/instiki/trunk/script/server:92 D:/_svn/instiki/trunk/instiki.rb:3:in `load' D:/_svn/instiki/trunk/instiki.rb:3 Redirected to http://localhost:2500/versitec/edit/HomePage Completed in 0.66000 (1 reqs/sec) 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET /versitec/show/HomeP age HTTP/1.1" 302 110 http://localhost:2500/web_list -> /versitec/show/HomePage Processing WikiController#edit (for 127.0.0.1 at Mon May 23 14:51:59 GMT Dayligh t Time 2005) Parameters: {"action"=>"edit", "id"=>"HomePage", "controller"=>"wiki", "web"=> "versitec"} Rendering wiki/edit within layouts/default Rendering layouts/default (200 OK) Completed in 0.09000 (11 reqs/sec) | Rendering: 0.08000 (88%) 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET /versitec/edit/HomeP age HTTP/1.1" 200 3960 http://localhost:2500/web_list -> /versitec/edit/HomePage From tamassmith at mac.com Mon May 23 15:34:25 2005 From: tamassmith at mac.com (=?ISO-8859-1?Q?Tam=E1s_Smith?=) Date: Mon May 23 15:27:43 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <4291E20D.1060500@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> Message-ID: <31bf4c4369111c66245ceb752bfca4f4@mac.com> Chris I had this same thing happen to me after a user edit. What I did to save from having to revert back to an older snapshot was go into the command_log file in the storage directory and take out the last couple of entries. First make a straight backup of the whole storage directory, then edit the log file. However, when you do this, you need to do it in a binary editing mode so that the file does not get an EOF character attached, which will break the Madeleine processing of the object changes log upon startup of Instiki. Therefore, you should use something like: vim -b Which will edit the log in straight binary format. You can get a feel for where the entries break by looking for the @myidi field or something similar. The command log is essentially the object changes to the database since the last snapshot. Hopefully what will happen is that you will take out the last couple of Wiki entries causing the problem. Note that I don't think this is a Madeline corruption issue, but something to do with bad markup that causes a blow out of the markdown system that goes all the way up the chain. In the bug reports database, I saw that a change was put in to keep such bad markup from blowing up the application all the way to the top, but I do not think that code change is in 0.10.1. Editing the command_log is not a good idea in general and just described here for emergency purposes. Tam?s On May 23, 2005, at 10:00 AM, Chris Roos wrote: I have instiki 0.10.1 configured with multiple webs on FreeBsd. It has been working fine for quite a while now. Unfortunately I was just editing a page and on saving received an error. From that point on I was unable to access that particular web - all of the others could be accessed fine. I copied the snapshot files to a local windows machine (with Instiki from svn) and tested on there. I get an error "undefined method `length' for nil:NilClass" when clicking on the corrupt web. In addition, the web is opened in edit mode automatically. No changes can be made. I have copied the log file (when running on windows) at the bottom of the email. I removed the latest snapshot file and the corrupt web can be accessed correctly, so the corruption is only in the latest snapshot. Unfortunately the previous snapshot is four days old and so I would prefer it if somehow I could recover the information from the latest snapshot. The question is, I don't really have a clue as to how I would about doing that..? Any help and advice would be much appreciated. Chris D:\_svn\instiki\trunk>instiki --storage=C:\TEMP\instiki => Starting Instiki on http://0.0.0.0:2500 => Data files are stored in C:\TEMP\instiki/2500 [2005-05-23 14:51:50] INFO WEBrick 1.3.1 [2005-05-23 14:51:50] INFO ruby 1.8.2 (2004-12-25) [i386-mswin32] [2005-05-23 14:51:50] INFO WEBrick::HTTPServer#start: pid=2544 port=2500 Processing WikiController#index (for 127.0.0.1 at Mon May 23 14:51:54 GMT Daylig ht Time 2005) Parameters: {"action"=>"index", "controller"=>"wiki"} Redirected to http://localhost:2500/web_list Completed in 0.01000 (100 reqs/sec) 127.0.0.1 - - [23/May/2005:14:51:53 GMT Daylight Time] "GET / HTTP/1.1" 302 96 - -> / Processing WikiController#web_list (for 127.0.0.1 at Mon May 23 14:51:55 GMT Day light Time 2005) Parameters: {"action"=>"web_list", "controller"=>"wiki"} Rendering wiki/web_list within layouts/default Rendering layouts/default (200 OK) Completed in 0.10000 (9 reqs/sec) | Rendering: 0.09000 (90%) 127.0.0.1 - - [23/May/2005:14:51:55 GMT Daylight Time] "GET /web_list HTTP/1.1" 200 2860 - -> /web_list Processing WikiController#show (for 127.0.0.1 at Mon May 23 14:51:59 GMT Dayligh t Time 2005) Parameters: {"action"=>"show", "id"=>"HomePage", "controller"=>"wiki", "web"=> "versitec"} Rendering wiki/page within layouts/default undefined method `length' for nil:NilClass (ActionView::TemplateError) On line #81 of /wiki/page.rhtml 78: <% end %> 79: 80: 81: <% unless @page.linked_from.empty? %> 82: 83: | Linked from: 84: <%= @page.linked_from.collect { |referring_page| /RedCloth-3.0.3/lib/redcloth.rb:533:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each' /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' /RedCloth-3.0.3/lib/redcloth.rb:522:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:518:in `gsub!' /RedCloth-3.0.3/lib/redcloth.rb:518:in `block_textile_lists' /RedCloth-3.0.3/lib/redcloth.rb:612:in `call' /RedCloth-3.0.3/lib/redcloth.rb:612:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:611:in `each' /RedCloth-3.0.3/lib/redcloth.rb:611:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:586:in `collect' /RedCloth-3.0.3/lib/redcloth.rb:586:in `blocks' /RedCloth-3.0.3/lib/redcloth.rb:288:in `to_html' D:/_svn/instiki/trunk/app/models/chunks/engines.rb:34:in `mask' D:/_svn/instiki/trunk/app/models/chunks/engines.rb:17:in `apply_to' D:/_svn/instiki/trunk/app/models/wiki_content.rb:169:in `build_chunks' D:/_svn/instiki/trunk/app/models/wiki_content.rb:147:in `initialize' D:/_svn/instiki/trunk/app/models/revision.rb:80:in `new' D:/_svn/instiki/trunk/app/models/revision.rb:80:in `display_content' D:/_svn/instiki/trunk/app/models/revision.rb:58:in `wiki_words' D:/_svn/instiki/trunk/app/models/page.rb:111:in `send' D:/_svn/instiki/trunk/app/models/page.rb:111:in `method_missing' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `select' D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' D:/_svn/instiki/trunk/app/models/page.rb:78:in `linked_from' (erb):81:in `evaluate_locals' /rails/actionpack/lib/action_view/base.rb:265:in `evaluate_locals' /rails/actionpack/lib/action_view/base.rb:276:in `rhtml_render' /rails/actionpack/lib/action_view/base.rb:201:in `send' /rails/actionpack/lib/action_view/base.rb:201:in `render_template' /rails/actionpack/lib/action_view/base.rb:181:in `render_file' /rails/actionpack/lib/action_controller/layout.rb:208:in `render_without_ben chmark' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `measure' /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' /rails/actionpack/lib/action_controller/base.rb:447:in `render_action' D:/_svn/instiki/trunk/app/controllers/wiki_controller.rb:203:in `show' /rails/actionpack/lib/action_controller/base.rb:708:in `send' /rails/actionpack/lib/action_controller/base.rb:708:in `perform_action_witho ut_filters' /rails/actionpack/lib/action_controller/filters.rb:294:in `perform_action_wi thout_benchmark' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `perform_actio n_without_rescue' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `measure' /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `perform_actio n_without_rescue' /rails/actionpack/lib/action_controller/rescue.rb:79:in `perform_action' /rails/actionpack/lib/action_controller/base.rb:330:in `send' /rails/actionpack/lib/action_controller/base.rb:330:in `process' /rails/actionpack/lib/action_controller/base.rb:273:in `process' /rails/railties/lib/dispatcher.rb:32:in `dispatch' D:/_svn/instiki/trunk/public/dispatch.rb:10 /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' /rails/railties/lib/webrick_server.rb:82:in `handle_dispatch' /rails/railties/lib/webrick_server.rb:35:in `service' c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:104:in `service' c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:65:in `run' c:/ruby/lib/ruby/1.8/webrick/server.rb:155:in `start_thread' c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start_thread' c:/ruby/lib/ruby/1.8/webrick/server.rb:94:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `each' c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' /rails/railties/lib/webrick_server.rb:21:in `dispatch' D:/_svn/instiki/trunk/script/server:92 D:/_svn/instiki/trunk/instiki.rb:3:in `load' D:/_svn/instiki/trunk/instiki.rb:3 Redirected to http://localhost:2500/versitec/edit/HomePage Completed in 0.66000 (1 reqs/sec) 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET /versitec/show/HomeP age HTTP/1.1" 302 110 http://localhost:2500/web_list -> /versitec/show/HomePage Processing WikiController#edit (for 127.0.0.1 at Mon May 23 14:51:59 GMT Dayligh t Time 2005) Parameters: {"action"=>"edit", "id"=>"HomePage", "controller"=>"wiki", "web"=> "versitec"} Rendering wiki/edit within layouts/default Rendering layouts/default (200 OK) Completed in 0.09000 (11 reqs/sec) | Rendering: 0.08000 (88%) 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET /versitec/edit/HomeP age HTTP/1.1" 200 3960 http://localhost:2500/web_list -> /versitec/edit/HomePage _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From alex at verk.info Mon May 23 15:37:25 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 23 15:30:10 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <4291E20D.1060500@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> Message-ID: Chris Roos wrote: > I have instiki 0.10.1 configured with multiple webs on FreeBsd. It > has been working fine for quite a while now. Unfortunately I was just > editing a page and on saving received an error. From that point on I > was unable to access that particular web - all of the others could be > accessed fine. > > I copied the snapshot files to a local windows machine (with Instiki > from svn) and tested on there. I get an error "undefined method > `length' for nil:NilClass" when clicking on the corrupt web. In > addition, the web is opened in edit mode automatically. No changes > can be made. I have copied the log file (when running on windows) at > the bottom of the email. > > I removed the latest snapshot file and the corrupt web can be accessed > correctly, so the corruption is only in the latest snapshot. > Unfortunately the previous snapshot is four days old and so I would > prefer it if somehow I could recover the information from the latest > snapshot. The question is, I don't really have a clue as to how I > would about doing that..? > > Any help and advice would be much appreciated. One oif these days I will write some thing to enable users poke in the Madeleine storage and save their results. Until then, please email me your last snapshot and tell me the name of the page that you edited. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From tamassmith at mac.com Mon May 23 18:10:34 2005 From: tamassmith at mac.com (=?ISO-8859-1?Q?Tam=E1s_Smith?=) Date: Mon May 23 18:03:51 2005 Subject: [Instiki] Instiki not writing snapshots? In-Reply-To: References: <1115955013.3717863143.5874.sendItem@bloglines.com> Message-ID: Alexey et al I see this behavior as well on Mac OS X Server 10.3. I get very strange snapshot schedules that seem to stop after the first snapshot: drwxr-xr-x 25 myuser myuser 850 23 May 19:06 . drwxr-xr-x 3 myuser myuser 102 12 May 05:36 .. -rw-rw-rw- 1 myuser myuser 1153232 22 May 18:23 000000000000000000001.command_log -rw-r--r-- 1 myuser myuser 2032 11 May 18:16 000000000000000000001.snapshot -rw-r--r-- 1 myuser myuser 5177 11 May 18:16 000000000000000000002.snapshot -rw-r--r-- 1 myuser myuser 6219 11 May 18:16 000000000000000000003.snapshot -rw-r--r-- 1 myuser myuser 8450 11 May 18:16 000000000000000000004.snapshot -rw-r--r-- 1 myuser myuser 9085 11 May 18:16 000000000000000000005.snapshot -rw-r--r-- 1 myuser myuser 9682 11 May 18:16 000000000000000000006.snapshot -rw-r--r-- 1 myuser myuser 9973 11 May 18:16 000000000000000000007.snapshot -rw-r--r-- 1 myuser myuser 11001 11 May 18:16 000000000000000000008.snapshot -rw-r--r-- 1 myuser myuser 12485 11 May 18:16 000000000000000000009.snapshot -rw-r--r-- 1 myuser myuser 12717 11 May 18:16 000000000000000000010.snapshot -rw-r--r-- 1 myuser myuser 12701 11 May 18:16 000000000000000000011.snapshot -rw-r--r-- 1 myuser myuser 79400 11 May 18:16 000000000000000000012.snapshot -rw-r--r-- 1 myuser myuser 81774 11 May 18:24 000000000000000000013.snapshot -rw-r--r-- 1 myuser myuser 94442 11 May 19:24 000000000000000000014.snapshot -rw-r--r-- 1 myuser myuser 95895 11 May 20:24 000000000000000000015.snapshot -rw-r--r-- 1 myuser myuser 100655 11 May 21:24 000000000000000000016.snapshot -rw-r--r-- 1 myuser myuser 105795 11 May 22:24 000000000000000000017.snapshot -rw-r--r-- 1 myuser myuser 106570 11 May 23:24 000000000000000000018.snapshot -rw-r--r-- 1 myuser myuser 107239 12 May 00:24 000000000000000000019.snapshot -rw-r--r-- 1 myuser myuser 108299 12 May 01:24 000000000000000000020.snapshot That listed ran just now on May 23. Note when the snapshots stop and the size and date of the command_log. You can also see that the snapshots may correspond with the start of the daemon: thomassmith:~/daemons/instipackage/instiki-0.10.1 myuser$ ps -uxM USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME PRI STIME UTIME COMMAND myuser 18444 0.0 0.1 27552 532 ?? S 8May05 0:00.00 31 0:00.00 0:00.00 tim myuser 18745 0.0 5.5 66128 28656 ?? S 12May05 12:34.64 31 1:05.38 11:29.26 ruby/bin/ruby -I /Users/myuser/daemons/instipackage/ruby myuser 27701 0.0 0.1 30740 448 ?? S 9:26PM 0:00.43 31 0:00.25 0:00.18 /usr/sbin/sshd -i myuser 27702 0.0 0.2 18644 816 std S 9:26PM 0:00.37 31 0:00.30 0:00.07 -bash Hard to read in the above, but it shows the Instiki process starting at about the time of the last two snapshots, making it appear that Instiki started, wrote a snapshot via the snapshot thread an hour later, then lost the snapshot thread. Note that running ps with the -M switch should show additional threads for the processes, if any. In this case, we only see one thread - the main thread, for the ruby process. I'll add this up to the bug report system, but wanted to run this by people first. This is a 0.10.1, but the webs running at 0.10.0. Tam?s On May 12, 2005, at 11:48 PM, Alexey Verkhovsky wrote: rdlugosz.10120601@bloglines.com wrote: > I've got an ever-growing 00...01.command_log > file, but there are *no* snapshot files! I guess this probably isn't > the > end of the world, presuming Instiki will restart correctly should I > bounce > it, but something doesn't seem right. Any thoughts? > The thread responsible for taking snapshots is in wiki_service.rb, start_snapshot_thread method. It will always try to write a snapshot on startup, if there is a command.log; if it cannot (for whatever reason), it should log an error (including full stack trace) in stderr. Please try to run instiki from command line, without --daemon option, and see what it has to say. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From alex at verk.info Mon May 23 18:46:18 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 23 18:39:02 2005 Subject: [Instiki] Instiki not writing snapshots? In-Reply-To: References: <1115955013.3717863143.5874.sendItem@bloglines.com> Message-ID: Tam?s Smith wrote: > Note that running ps with the -M switch should show additional threads > for the processes, if any. In this case, we only see one thread - the > main thread, for the ruby process. Ruby threads are green threads - Ruby is not using native OS threads. So, if you do something like ruby -e '10.times { Thread.new { sleep 1000 } }', you won't see multiple threads either. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From ben at incomumdesign.com Mon May 23 23:11:58 2005 From: ben at incomumdesign.com (Ben Jackson) Date: Tue May 24 00:14:21 2005 Subject: [Instiki] Password reset Message-ID: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> I don't remember the password. No luck with googling "instiki password reset". How is it done? Ben From eduardo.fernandez at gmail.com Tue May 24 03:50:17 2005 From: eduardo.fernandez at gmail.com (Eduardo Fernandez Corrales) Date: Tue May 24 03:43:30 2005 Subject: [Instiki] Password reset In-Reply-To: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> Message-ID: <3bf06c66050524005072207934@mail.gmail.com> >From http://instiki.org/show/FAQ : jsierles?The above code should be added now to wiki_controller.rb, andshould look like this simply: def setPasswiki.system[:password] = "newpassword"end Visit the /setPass URL and watch an exception thrown. Your passwordshould be reset. Comment code afterwards, of course. Hope that helps.-- Edu On 5/24/05, Ben Jackson wrote:> I don't remember the password. No luck with googling "instiki password> reset". How is it done?> > Ben> > _______________________________________________> Instiki-users mailing list> Instiki-users@rubyforge.org> http://rubyforge.org/mailman/listinfo/instiki-users> From ben at incomumdesign.com Tue May 24 08:59:18 2005 From: ben at incomumdesign.com (Ben Jackson) Date: Tue May 24 08:55:01 2005 Subject: [Instiki] Password reset In-Reply-To: <3bf06c66050524005072207934@mail.gmail.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> Message-ID: <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> This didn't work for me... I shut down the wiki, added the lines from the FAQ (the lines below are a little chopped up and won't compile), restarted and went to http://localhost:2500/setPass, saw the exception "Unknown web 'setPass'", but my password still isn't being reset. Thanks, - B On 24 May 2005, at 04:50, Eduardo Fernandez Corrales wrote: >> From http://instiki.org/show/FAQ : > jsierles?The above code should be added now to > wiki_controller.rb, andshould look like this simply: > def setPasswiki.system[:password] = "newpassword"end > Visit the /setPass URL and watch an exception thrown. Your > passwordshould be reset. Comment code afterwards, of course. > Hope that helps.-- Edu > On 5/24/05, Ben Jackson wrote:> I don't > remember the password. No luck with googling "instiki password> > reset". How is it done?> > Ben> > > _______________________________________________> Instiki-users mailing > list> Instiki-users@rubyforge.org> > http://rubyforge.org/mailman/listinfo/instiki-users> > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > From assaph at gmail.com Tue May 24 18:47:21 2005 From: assaph at gmail.com (Assaph Mehr) Date: Tue May 24 18:40:36 2005 Subject: [Instiki] Password reset In-Reply-To: <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> Message-ID: <3ff47be90505241547744abdd6@mail.gmail.com> Easier to just retrieve the password :-) If you can isolate the wiki for a while (i.e. where no one else canview it) just edit app/views/wiki/page.rhtml (or any other) and insertthis line somewhere near the first div: Password: <%= @wiki.system['password'] || wiki.system[:password] %> Reload the home page and you'll see the password in all its glory :-)(If you're using Instiki 0.9.2 - vs 0.10.x - just drop the @ sign inthe above line). The reason you should isolate the wiki is that anyone else viewing atthis time will also see the password. Don't forget to delete the lineonce you're done. Cheers,Assaph On 5/24/05, Ben Jackson wrote:> This didn't work for me... I shut down the wiki, added the lines from> the FAQ (the lines below are a little chopped up and won't compile),> restarted and went to http://localhost:2500/setPass, saw the exception> "Unknown web 'setPass'", but my password still isn't being reset.> Thanks,> > - B> > On 24 May 2005, at 04:50, Eduardo Fernandez Corrales wrote:> > >> From http://instiki.org/show/FAQ :> > jsierles?The above code should be added now to> > wiki_controller.rb, andshould look like this simply:> > def setPasswiki.system[:password] = "newpassword"end> > Visit the /setPass URL and watch an exception thrown. Your> > passwordshould be reset. Comment code afterwards, of course.> > Hope that helps.-- Edu> > On 5/24/05, Ben Jackson wrote:> I don't> > remember the password. No luck with googling "instiki password>> > reset". How is it done?> > Ben> >> > _______________________________________________> Instiki-users mailing> > list> Instiki-users@rubyforge.org >> > http://rubyforge.org/mailman/listinfo/instiki-users>> > _______________________________________________> > Instiki-users mailing list> > Instiki-users@rubyforge.org > > http://rubyforge.org/mailman/listinfo/instiki-users> >> > > _______________________________________________> Instiki-users mailing list> Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users> From alex at verk.info Tue May 24 20:06:17 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Tue May 24 19:58:59 2005 Subject: [Instiki] Change charset in instiki In-Reply-To: <20050519095927.GA31512@pixie.subbacultcha.home> References: <20050518132635.GA10197@pixie.subbacultcha.home> <200505181626.50291.dmertz@online.de> <428C1567.40407@verk.info> <20050519095927.GA31512@pixie.subbacultcha.home> Message-ID: Nicholas Wieland wrote: > - Alexey Verkhovsky : > > >>I have the same questions as Denis :) >> >> > >Even if I use UTF-8 the problem remains, so the charset doesn't count. >Does Webrick use a default charset ? >If you or Denis could try to add a pair of ????? to the default layout >and see if the page renders correctly I'd know if I have to mess with >instiki or just my system ... > > If I put this line into any .rhtml file, it comes out as a ???? Whereas if I put the same line into the page content, it is rendered fine and dandy. Hmm... has to be something in Rails. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From chris at seagul.co.uk Thu May 26 07:33:26 2005 From: chris at seagul.co.uk (Chris Roos) Date: Thu May 26 07:26:42 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <31bf4c4369111c66245ceb752bfca4f4@mac.com> References: <4291E20D.1060500@seagul.co.uk> <31bf4c4369111c66245ceb752bfca4f4@mac.com> Message-ID: <4295B406.9090902@seagul.co.uk> Thanks for the advice Tamas but unfortunately my storage directory didn't have any .command_log files (I just extracted the backup, that I did as soon as I saw the problem, to double check). I'm going to send my .snapshot file to Alexey, as per his reply to this same thread, and see if he can work some magic. Chris Tam?s Smith wrote: > Chris > > I had this same thing happen to me after a user edit. What I did to > save from having to revert back to an older snapshot was go into the > command_log file in the storage directory and take out the last couple > of entries. First make a straight backup of the whole storage > directory, then edit the log file. However, when you do this, you need > to do it in a binary editing mode so that the file does not get an EOF > character attached, which will break the Madeleine processing of the > object changes log upon startup of Instiki. Therefore, you should use > something like: > > vim -b > > Which will edit the log in straight binary format. You can get a feel > for where the entries break by looking for the @myidi field or something > similar. The command log is essentially the object changes to the > database since the last snapshot. Hopefully what will happen is that you > will take out the last couple of Wiki entries causing the problem. > > Note that I don't think this is a Madeline corruption issue, but > something to do with bad markup that causes a blow out of the markdown > system that goes all the way up the chain. In the bug reports database, > I saw that a change was put in to keep such bad markup from blowing up > the application all the way to the top, but I do not think that code > change is in 0.10.1. > > Editing the command_log is not a good idea in general and just > described here for emergency purposes. > > Tam?s > > > On May 23, 2005, at 10:00 AM, Chris Roos wrote: > > I have instiki 0.10.1 configured with multiple webs on FreeBsd. It has > been working fine for quite a while now. Unfortunately I was just > editing a page and on saving received an error. From that point on I > was unable to access that particular web - all of the others could be > accessed fine. > > I copied the snapshot files to a local windows machine (with Instiki > from svn) and tested on there. I get an error "undefined method > `length' for nil:NilClass" when clicking on the corrupt web. In > addition, the web is opened in edit mode automatically. No changes can > be made. I have copied the log file (when running on windows) at the > bottom of the email. > > I removed the latest snapshot file and the corrupt web can be accessed > correctly, so the corruption is only in the latest snapshot. > Unfortunately the previous snapshot is four days old and so I would > prefer it if somehow I could recover the information from the latest > snapshot. The question is, I don't really have a clue as to how I would > about doing that..? > > Any help and advice would be much appreciated. > > Chris > > > > D:\_svn\instiki\trunk>instiki --storage=C:\TEMP\instiki > => Starting Instiki on http://0.0.0.0:2500 > => Data files are stored in C:\TEMP\instiki/2500 > [2005-05-23 14:51:50] INFO WEBrick 1.3.1 > [2005-05-23 14:51:50] INFO ruby 1.8.2 (2004-12-25) [i386-mswin32] > [2005-05-23 14:51:50] INFO WEBrick::HTTPServer#start: pid=2544 port=2500 > > > Processing WikiController#index (for 127.0.0.1 at Mon May 23 14:51:54 > GMT Daylig > ht Time 2005) > Parameters: {"action"=>"index", "controller"=>"wiki"} > Redirected to http://localhost:2500/web_list > Completed in 0.01000 (100 reqs/sec) > 127.0.0.1 - - [23/May/2005:14:51:53 GMT Daylight Time] "GET / HTTP/1.1" > 302 96 > - -> / > > > Processing WikiController#web_list (for 127.0.0.1 at Mon May 23 14:51:55 > GMT Day > light Time 2005) > Parameters: {"action"=>"web_list", "controller"=>"wiki"} > Rendering wiki/web_list within layouts/default > Rendering layouts/default (200 OK) > Completed in 0.10000 (9 reqs/sec) | Rendering: 0.09000 (90%) > 127.0.0.1 - - [23/May/2005:14:51:55 GMT Daylight Time] "GET /web_list > HTTP/1.1" > 200 2860 > - -> /web_list > > > Processing WikiController#show (for 127.0.0.1 at Mon May 23 14:51:59 GMT > Dayligh > t Time 2005) > Parameters: {"action"=>"show", "id"=>"HomePage", "controller"=>"wiki", > "web"=> > "versitec"} > Rendering wiki/page within layouts/default > undefined method `length' for nil:NilClass (ActionView::TemplateError) > On line #81 of /wiki/page.rhtml > > 78: <% end %> > 79: > 80: > 81: <% unless @page.linked_from.empty? %> > 82: > 83: | Linked from: > 84: <%= @page.linked_from.collect { |referring_page| > > /RedCloth-3.0.3/lib/redcloth.rb:533:in `block_textile_lists' > /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' > /RedCloth-3.0.3/lib/redcloth.rb:522:in `each' > /RedCloth-3.0.3/lib/redcloth.rb:522:in `each_with_index' > /RedCloth-3.0.3/lib/redcloth.rb:522:in `block_textile_lists' > /RedCloth-3.0.3/lib/redcloth.rb:518:in `gsub!' > /RedCloth-3.0.3/lib/redcloth.rb:518:in `block_textile_lists' > /RedCloth-3.0.3/lib/redcloth.rb:612:in `call' > /RedCloth-3.0.3/lib/redcloth.rb:612:in `blocks' > /RedCloth-3.0.3/lib/redcloth.rb:611:in `each' > /RedCloth-3.0.3/lib/redcloth.rb:611:in `blocks' > /RedCloth-3.0.3/lib/redcloth.rb:586:in `collect' > /RedCloth-3.0.3/lib/redcloth.rb:586:in `blocks' > /RedCloth-3.0.3/lib/redcloth.rb:288:in `to_html' > D:/_svn/instiki/trunk/app/models/chunks/engines.rb:34:in `mask' > D:/_svn/instiki/trunk/app/models/chunks/engines.rb:17:in `apply_to' > D:/_svn/instiki/trunk/app/models/wiki_content.rb:169:in `build_chunks' > D:/_svn/instiki/trunk/app/models/wiki_content.rb:147:in `initialize' > D:/_svn/instiki/trunk/app/models/revision.rb:80:in `new' > D:/_svn/instiki/trunk/app/models/revision.rb:80:in `display_content' > D:/_svn/instiki/trunk/app/models/revision.rb:58:in `wiki_words' > D:/_svn/instiki/trunk/app/models/page.rb:111:in `send' > D:/_svn/instiki/trunk/app/models/page.rb:111:in `method_missing' > D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' > D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `select' > D:/_svn/instiki/trunk/app/models/page_set.rb:39:in `pages_that_link_to' > D:/_svn/instiki/trunk/app/models/page.rb:78:in `linked_from' > (erb):81:in `evaluate_locals' > /rails/actionpack/lib/action_view/base.rb:265:in `evaluate_locals' > /rails/actionpack/lib/action_view/base.rb:276:in `rhtml_render' > /rails/actionpack/lib/action_view/base.rb:201:in `send' > /rails/actionpack/lib/action_view/base.rb:201:in `render_template' > /rails/actionpack/lib/action_view/base.rb:181:in `render_file' > /rails/actionpack/lib/action_controller/layout.rb:208:in > `render_without_ben > chmark' > /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' > /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `measure' > /rails/actionpack/lib/action_controller/benchmarking.rb:23:in `render' > /rails/actionpack/lib/action_controller/base.rb:447:in `render_action' > D:/_svn/instiki/trunk/app/controllers/wiki_controller.rb:203:in `show' > /rails/actionpack/lib/action_controller/base.rb:708:in `send' > /rails/actionpack/lib/action_controller/base.rb:708:in > `perform_action_witho > ut_filters' > /rails/actionpack/lib/action_controller/filters.rb:294:in > `perform_action_wi > thout_benchmark' > /rails/actionpack/lib/action_controller/benchmarking.rb:34:in > `perform_actio > n_without_rescue' > /rails/actionpack/lib/action_controller/benchmarking.rb:34:in `measure' > /rails/actionpack/lib/action_controller/benchmarking.rb:34:in > `perform_actio > n_without_rescue' > /rails/actionpack/lib/action_controller/rescue.rb:79:in > `perform_action' > /rails/actionpack/lib/action_controller/base.rb:330:in `send' > /rails/actionpack/lib/action_controller/base.rb:330:in `process' > /rails/actionpack/lib/action_controller/base.rb:273:in `process' > /rails/railties/lib/dispatcher.rb:32:in `dispatch' > D:/_svn/instiki/trunk/public/dispatch.rb:10 > /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' > /rails/activesupport/lib/active_support/dependencies.rb:189:in `load' > /rails/railties/lib/webrick_server.rb:82:in `handle_dispatch' > /rails/railties/lib/webrick_server.rb:35:in `service' > c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:104:in `service' > c:/ruby/lib/ruby/1.8/webrick/httpserver.rb:65:in `run' > c:/ruby/lib/ruby/1.8/webrick/server.rb:155:in `start_thread' > c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start' > c:/ruby/lib/ruby/1.8/webrick/server.rb:144:in `start_thread' > c:/ruby/lib/ruby/1.8/webrick/server.rb:94:in `start' > c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `each' > c:/ruby/lib/ruby/1.8/webrick/server.rb:89:in `start' > c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' > c:/ruby/lib/ruby/1.8/webrick/server.rb:79:in `start' > /rails/railties/lib/webrick_server.rb:21:in `dispatch' > D:/_svn/instiki/trunk/script/server:92 > D:/_svn/instiki/trunk/instiki.rb:3:in `load' > D:/_svn/instiki/trunk/instiki.rb:3 > Redirected to http://localhost:2500/versitec/edit/HomePage > Completed in 0.66000 (1 reqs/sec) > 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET > /versitec/show/HomeP > age HTTP/1.1" 302 110 > http://localhost:2500/web_list -> /versitec/show/HomePage > > > Processing WikiController#edit (for 127.0.0.1 at Mon May 23 14:51:59 GMT > Dayligh > t Time 2005) > Parameters: {"action"=>"edit", "id"=>"HomePage", "controller"=>"wiki", > "web"=> > "versitec"} > Rendering wiki/edit within layouts/default > Rendering layouts/default (200 OK) > Completed in 0.09000 (11 reqs/sec) | Rendering: 0.08000 (88%) > 127.0.0.1 - - [23/May/2005:14:51:59 GMT Daylight Time] "GET > /versitec/edit/HomeP > age HTTP/1.1" 200 3960 > http://localhost:2500/web_list -> /versitec/edit/HomePage > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > > From chris at seagul.co.uk Thu May 26 07:39:58 2005 From: chris at seagul.co.uk (Chris Roos) Date: Thu May 26 07:33:12 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> Message-ID: <4295B58E.40400@seagul.co.uk> Alexey Verkhovsky wrote: > Chris Roos wrote: > >> I have instiki 0.10.1 configured with multiple webs on FreeBsd. It >> has been working fine for quite a while now. Unfortunately I was just >> editing a page and on saving received an error. From that point on I >> was unable to access that particular web - all of the others could be >> accessed fine. >> >> I copied the snapshot files to a local windows machine (with Instiki >> from svn) and tested on there. I get an error "undefined method >> `length' for nil:NilClass" when clicking on the corrupt web. In >> addition, the web is opened in edit mode automatically. No changes >> can be made. I have copied the log file (when running on windows) at >> the bottom of the email. >> >> I removed the latest snapshot file and the corrupt web can be accessed >> correctly, so the corruption is only in the latest snapshot. >> Unfortunately the previous snapshot is four days old and so I would >> prefer it if somehow I could recover the information from the latest >> snapshot. The question is, I don't really have a clue as to how I >> would about doing that..? >> >> Any help and advice would be much appreciated. > > > One oif these days I will write some thing to enable users poke in the > Madeleine storage and save their results. Until then, please email me > your last snapshot and tell me the name of the page that you edited. > Before I send you the .snapshot file I wanted to say that I can't remember which page it would have been that I was editing. Is it (relatively) easy for you to spot corruption in a page or would it be a case of manually looking into each page to check for malformed html (which I'm led to believe is the cause of the problem from Tamas)? In addition do you have any pointers as to how one would go about replicating the procedure that you will use to attempt to retrieve a corrupt snapshot file? Maybe this is something I could take a stab at myself. I realise that I could just find this out myself by digging into the code but if you have any pointers then it makes that process a lot easier ;-) Thanks in advance, Chris From huw.nichols at gmail.com Thu May 26 08:29:16 2005 From: huw.nichols at gmail.com (Huw) Date: Thu May 26 08:22:24 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> Message-ID: > > I copied the snapshot files to a local windows machine (with Instiki > > from svn) and tested on there. I get an error "undefined method > > `length' for nil:NilClass" when clicking on the corrupt web. In > > addition, I too have seen this exact same thing, yesterday I was doing a lot of work with nested lists and indented paragraphs under a list item etc and got a rails application error followed by the above length thing several times. To fix it all I had to do was hit back in the browser to get back to the edit screen for the last page I touched and remove the edit [usually either a div or blank line between items] and re-save. All was then fine. Not much help for you at the moment I know but maybe worth trying if it happens to you again. My guess would be that it is a bug in the Textile parsing -- annoying for me as it means my formatting does not work on indent paras under a list item -- but only sometimes Ah, not sure if it is relevant or not but I am running version 0.10, ruby 1.8.2 and latest gems under freebsd Huw From alex at verk.info Thu May 26 11:31:10 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 26 11:23:46 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <4295B58E.40400@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> <4295B58E.40400@seagul.co.uk> Message-ID: Chris Roos wrote: > Before I send you the .snapshot file I wanted to say that I can't > remember which page it would have been that I was editing. Is it > (relatively) easy for you to spot corruption in a page or would it be > a case of manually looking into each page to check for malformed html > (which I'm led to believe is the cause of the problem from Tamas)? No, it is easy. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Thu May 26 11:32:09 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Thu May 26 11:24:47 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> Message-ID: Huw wrote: >To fix it all I had to do was hit back in the browser to get back to >the edit screen for the last page > Or you can type in the edit URL manually. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From chris at seagul.co.uk Thu May 26 11:36:51 2005 From: chris at seagul.co.uk (Chris Roos) Date: Thu May 26 11:30:05 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> <4295B58E.40400@seagul.co.uk> Message-ID: <4295ED13.8080700@seagul.co.uk> Alexey Verkhovsky wrote: > Chris Roos wrote: > >> Before I send you the .snapshot file I wanted to say that I can't >> remember which page it would have been that I was editing. Is it >> (relatively) easy for you to spot corruption in a page or would it be >> a case of manually looking into each page to check for malformed html >> (which I'm led to believe is the cause of the problem from Tamas)? > > > No, it is easy. > Cool. I've replied off list with the snapshot file. Cheers, Chris From tamassmith at mac.com Thu May 26 17:43:22 2005 From: tamassmith at mac.com (=?ISO-8859-1?Q?Tam=E1s_Smith?=) Date: Thu May 26 17:36:39 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> Message-ID: <9e94eceaf106fbec6c93182592280acf@mac.com> Huw et al I will note that the time I got this similar bug, which is reported in the bug tracking system as bug 161, a user was also using complex list and paragraph indentations. Btw, Alexey, I emailed you the snapshot for that bug and made a comment on the bug to that affect, though I may have labeled it incorrectly as pertaining to bug 163, sorry. Tam?s On May 26, 2005, at 8:29 AM, Huw wrote: >> I copied the snapshot files to a local windows machine (with Instiki >> from svn) and tested on there. I get an error "undefined method >> `length' for nil:NilClass" when clicking on the corrupt web. In >> addition, I too have seen this exact same thing, yesterday I was doing a lot of work with nested lists and indented paragraphs under a list item etc and got a rails application error followed by the above length thing several times. To fix it all I had to do was hit back in the browser to get back to the edit screen for the last page I touched and remove the edit [usually either a div or blank line between items] and re-save. All was then fine. Not much help for you at the moment I know but maybe worth trying if it happens to you again. My guess would be that it is a bug in the Textile parsing -- annoying for me as it means my formatting does not work on indent paras under a list item -- but only sometimes Ah, not sure if it is relevant or not but I am running version 0.10, ruby 1.8.2 and latest gems under freebsd Huw _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From huw.nichols at gmail.com Thu May 26 19:05:29 2005 From: huw.nichols at gmail.com (Huw) Date: Thu May 26 18:58:38 2005 Subject: Fwd: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> Message-ID: > I will note that the time I got this similar bug, which is reported in > the bug tracking system as bug 161, a user was also using complex list > and paragraph indentations. Yes it does seem that the more complex but at the same time perfectly valid formatting is prone to throw a fit from time to time. I have also had numerous problems when editing pages with html markup, as for a side bar etc. It seems really east to mess up. All that said I have yet to find anything that comes close to the speed and ease of use that we have here :) Huw -- TTFN H From chris at seagul.co.uk Fri May 27 05:14:20 2005 From: chris at seagul.co.uk (Chris Roos) Date: Fri May 27 05:07:32 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <9e94eceaf106fbec6c93182592280acf@mac.com> References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> Message-ID: <4296E4EC.7050800@seagul.co.uk> Does anyone currently have the definite cause for this problem? Is the cause definitely pointing to redcloth or could it still be something in instiki? All I'm thinking is that if we can come up with an example of a page that causes the error we can add it to the tests that are bundled with redcloth. Any thoughts? Chris Tam?s Smith wrote: > Huw et al > > I will note that the time I got this similar bug, which is reported in > the bug tracking system as bug 161, a user was also using complex list > and paragraph indentations. Btw, Alexey, I emailed you the snapshot for > that bug and made a comment on the bug to that affect, though I may have > labeled it incorrectly as pertaining to bug 163, sorry. > > Tam?s > > > On May 26, 2005, at 8:29 AM, Huw wrote: > >>> I copied the snapshot files to a local windows machine (with Instiki >>> from svn) and tested on there. I get an error "undefined method >>> `length' for nil:NilClass" when clicking on the corrupt web. In >>> addition, > > > I too have seen this exact same thing, yesterday I was doing a lot of > work with nested lists and indented paragraphs under a list item etc > and got a rails application error followed by the above length thing > several times. > > To fix it all I had to do was hit back in the browser to get back to > the edit screen for the last page I touched and remove the edit > [usually either a div or blank line between items] and re-save. All > was then fine. > > Not much help for you at the moment I know but maybe worth trying if > it happens to you again. > > My guess would be that it is a bug in the Textile parsing -- annoying > for me as it means my formatting does not work on indent paras under a > list item -- but only sometimes > > Ah, not sure if it is relevant or not but I am running version 0.10, > ruby 1.8.2 and latest gems under freebsd > > Huw > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > > From chris at seagul.co.uk Fri May 27 13:42:02 2005 From: chris at seagul.co.uk (Chris Roos) Date: Fri May 27 13:35:18 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <4296E4EC.7050800@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> <4296E4EC.7050800@seagul.co.uk> Message-ID: <42975BEA.2060402@seagul.co.uk> I've just been doing some investigation. I can reliably reproduce the following error in Instiki 0.10.1 on Windows XP with Rails 0.12.1 and Firefox 1.0.4. "NoMethodError in Wiki#save" "undefined method `length' for nil:NilClass" If we don't go back and edit the change that caused this problem and then try to re-visit the wiki we get the following error. "Redirection limit for this URL exceeded. Unable to load the requested page. This may be caused by cookies that are blocked." All we need to do is create a list where there are more indentations at the top than the bottom. For example, creating any of these lists will cause the error. ### three deep ## two deep ##### five deep # one deep ** two deep * one deep **** four deep *** three deep However, if we duplicate the bottom line in each case at the top of the list then all works fine, e.g. *** three deep **** four deep *** three deep We can trace the problem to RedCloth by emulating what instiki will do in irb. require 'redcloth' r = RedCloth.new("** two deep\r\n* one deep") r.to_html => returns the undefined method length for nil:NilClass error. This doesn't appear as a problem in the actual Textile online implementation at http://www.textism.com/tools/textile/, i.e. putting ** two deep * one deep doesn't error but returns the following (albeit not well formed) html.

two deep

one deep

I'd appreciate if someone could confirm this to be true and then I guess we contact _why (I'm pretty sure RedCloth is one of his). Chris Chris Roos wrote: > Does anyone currently have the definite cause for this problem? Is the > cause definitely pointing to redcloth or could it still be something in > instiki? All I'm thinking is that if we can come up with an example of > a page that causes the error we can add it to the tests that are bundled > with redcloth. > > Any thoughts? > > Chris > > Tam?s Smith wrote: > >> Huw et al >> >> I will note that the time I got this similar bug, which is reported in >> the bug tracking system as bug 161, a user was also using complex list >> and paragraph indentations. Btw, Alexey, I emailed you the snapshot >> for that bug and made a comment on the bug to that affect, though I >> may have labeled it incorrectly as pertaining to bug 163, sorry. >> >> Tam?s >> >> >> On May 26, 2005, at 8:29 AM, Huw wrote: >> >>>> I copied the snapshot files to a local windows machine (with Instiki >>>> from svn) and tested on there. I get an error "undefined method >>>> `length' for nil:NilClass" when clicking on the corrupt web. In >>>> addition, >> >> >> >> I too have seen this exact same thing, yesterday I was doing a lot of >> work with nested lists and indented paragraphs under a list item etc >> and got a rails application error followed by the above length thing >> several times. >> >> To fix it all I had to do was hit back in the browser to get back to >> the edit screen for the last page I touched and remove the edit >> [usually either a div or blank line between items] and re-save. All >> was then fine. >> >> Not much help for you at the moment I know but maybe worth trying if >> it happens to you again. >> >> My guess would be that it is a bug in the Textile parsing -- annoying >> for me as it means my formatting does not work on indent paras under a >> list item -- but only sometimes >> >> Ah, not sure if it is relevant or not but I am running version 0.10, >> ruby 1.8.2 and latest gems under freebsd >> >> Huw >> >> _______________________________________________ >> Instiki-users mailing list >> Instiki-users@rubyforge.org >> http://rubyforge.org/mailman/listinfo/instiki-users >> >> >> _______________________________________________ >> Instiki-users mailing list >> Instiki-users@rubyforge.org >> http://rubyforge.org/mailman/listinfo/instiki-users >> >> > > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > > From huw.nichols at gmail.com Fri May 27 13:59:01 2005 From: huw.nichols at gmail.com (Huw) Date: Fri May 27 13:52:07 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <42975BEA.2060402@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> <4296E4EC.7050800@seagul.co.uk> <42975BEA.2060402@seagul.co.uk> Message-ID: That sounds like you may have hit it, As I said before, I often get this sort of error but have not had the time to look into it properly, I will go check some of my pages that give problems and report back BTW: Thanks for putting in the effort to get us started:) Regards H > I've just been doing some investigation. I can reliably reproduce the > following error in Instiki 0.10.1 on Windows XP with Rails 0.12.1 and > Firefox 1.0.4. > > "NoMethodError in Wiki#save" > "undefined method `length' for nil:NilClass" > > If we don't go back and edit the change that caused this problem and > then try to re-visit the wiki we get the following error. > > "Redirection limit for this URL exceeded. Unable to load the requested > page. This may be caused by cookies that are blocked." > > All we need to do is create a list where there are more indentations at > the top than the bottom. For example, creating any of these lists will > cause the error. > > ### three deep > ## two deep From alex at verk.info Fri May 27 21:37:47 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Fri May 27 21:30:22 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <42975BEA.2060402@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> <4296E4EC.7050800@seagul.co.uk> <42975BEA.2060402@seagul.co.uk> Message-ID: Chris Roos wrote: > I've just been doing some investigation. I can reliably reproduce the > following error in Instiki 0.10.1 on Windows XP with Rails 0.12.1 and > Firefox 1.0.4. > > "Redirection limit for this URL exceeded. Unable to load the > requested page. This may be caused by cookies that are blocked." There are two factors at play here. One is a RedCloth bug that causes RedCloth to raise an error while trying to parse this markup. That's something worth reporting in RedCloth tracker. The other is how Instiki 0.10.1 handles the situation ("Redirection limit..." etc). This is "fixed in Subversion trunk" (TM) -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From ben at incomumdesign.com Sat May 28 12:15:56 2005 From: ben at incomumdesign.com (Ben Jackson) Date: Sat May 28 12:11:35 2005 Subject: [Instiki] Password reset In-Reply-To: <3ff47be90505241547744abdd6@mail.gmail.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> <3ff47be90505241547744abdd6@mail.gmail.com> Message-ID: <741d466dda2939f051b0054eb2484d16@incomumdesign.com> Putting in Password: <%= @wiki.system[:password] %> Spit out my expected password, however when tried to use it to log in, I still got access denied. What could be going wrong? On 24 May 2005, at 19:47, Assaph Mehr wrote: > Password: <%= @wiki.system['password'] || wiki.system[:password] %> From alex at verk.info Sat May 28 18:37:46 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sat May 28 18:30:18 2005 Subject: [Instiki] Password reset In-Reply-To: <741d466dda2939f051b0054eb2484d16@incomumdesign.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> <3ff47be90505241547744abdd6@mail.gmail.com> <741d466dda2939f051b0054eb2484d16@incomumdesign.com> Message-ID: Ben Jackson wrote: > Putting in > Password: <%= @wiki.system[:password] %> > Spit out my expected password, however when tried to use it to log in, > I still got access denied. What could be going wrong? @wiki.system[:password] is an admin password (which you can use to reset a password of a web). Web's password is @wiki.webs[].password -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From ben at incomumdesign.com Sat May 28 19:02:10 2005 From: ben at incomumdesign.com (Ben Jackson) Date: Sat May 28 18:57:52 2005 Subject: [Instiki] Password reset In-Reply-To: References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> <3ff47be90505241547744abdd6@mail.gmail.com> <741d466dda2939f051b0054eb2484d16@incomumdesign.com> Message-ID: <4943e5e4b2830feb0e2c774b870aac0e@incomumdesign.com> Ah ok. Out of curiosity, how do I get to the admin screen to change the password? On 28 May 2005, at 19:37, Alexey Verkhovsky wrote: > Ben Jackson wrote: > >> Putting in >> Password: <%= @wiki.system[:password] %> >> Spit out my expected password, however when tried to use it to log >> in, I still got access denied. What could be going wrong? > > @wiki.system[:password] is an admin password (which you can use to > reset a password of a web). > Web's password is @wiki.webs[].password > > -- > Best regards, > > Alexey Verkhovsky > > Ruby Forum: http://ruby-forum.org (moderator) > RForum: http://rforum.andreas-s.net (co-author) > Instiki: http://instiki.org (maintainer) > > _______________________________________________ > Instiki-users mailing list > Instiki-users@rubyforge.org > http://rubyforge.org/mailman/listinfo/instiki-users > From alex at verk.info Sat May 28 19:26:09 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sat May 28 19:18:41 2005 Subject: [Instiki] Password reset In-Reply-To: <4943e5e4b2830feb0e2c774b870aac0e@incomumdesign.com> References: <10ff776f825c9b3c8390852042413bc8@incomumdesign.com> <3bf06c66050524005072207934@mail.gmail.com> <776e91843fa7a3234a6d67e3f1378574@incomumdesign.com> <3ff47be90505241547744abdd6@mail.gmail.com> <741d466dda2939f051b0054eb2484d16@incomumdesign.com> <4943e5e4b2830feb0e2c774b870aac0e@incomumdesign.com> Message-ID: Ben Jackson wrote: > Ah ok. Out of curiosity, how do I get to the admin screen to change > the password? Something like http://localhost:2500/mywiki/edit_web -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Sun May 29 10:56:44 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sun May 29 10:49:16 2005 Subject: [Instiki] Instiki v0.10.1 for OS X In-Reply-To: References: <9859143f050509235527a47472@mail.gmail.com> Message-ID: Ben, based on this report I didn't publish your OS X build yet. What do you think about it? John Evans wrote: > Instiki crashes upon start for me. Here is the log. > > > ********** > > Host Name: iMac-g5 > Date/Time: 2005-05-10 16:58:59.662 +0300 > OS Version: 10.4 > Report Version: 3 > > Command: Instiki > Path: /Users/john/Desktop/Instiki.app/Contents/MacOS/Instiki > Parent: WindowServer [64] > > Version: ??? (0.10.1) > > PID: 318 > Thread: 0 > > Exception: EXC_BAD_ACCESS (0x0001) > Codes: KERN_PROTECTION_FAILURE (0x0002) at 0x00000000 -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From alex at verk.info Sun May 29 13:08:59 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Sun May 29 13:01:29 2005 Subject: [Instiki] Poking in storage Message-ID: Hi all, I've written some script to help people poke around in the Madeleine storage. It will be included in the 0.10.2 release, but it works with any earlier 0.10 version (including the current Instiki trunk, of course). You can get it at http://svn.instiki.org/instiki/trunk/script/debug_storage and copy to your instiki/script directory. Two caveats: 1. You MUST be a reasonably good Ruby programmer to use it successfully for anything non-trivial. 2. It's very easy to screw up something by poking in the storage internals. If you do, please undo your changes by deleting the most recent snapshot(s) and don't ask for help. Usage example: E:\eclipse\workspace\instiki\script>irb irb(main):001:0> load 'debug_storage' Enter path to storage [E:/eclipse/workspace/instiki/storage/2500]: Loading storage from the default storage path (E:/eclipse/workspace/instiki/storage/2500) Instiki storage from E:/eclipse/workspace/instiki/storage/2500 is loaded. Access it a global variable $wiki. Happy poking! => true irb(main):003:0> $wiki.system => {"password"=>"foo"} irb(main):005:0> $wiki.system['password'] = 'bar' => "bar" irb(main):006:0> $wiki.webs.keys => ["wiki1", "wiki2"] irb(main):007:0> $wiki.webs['wiki1'].password = 'the_password' => "the_password" irb(main):008:0> WikiService::snapshot => [] Things that are possible: # cleaning old revisions $wiki.webs['wiki'].pages['HomePage'].revisions = $wiki.webs['wiki'].pages['HomePage'].revisions[-1..-1] # Changing contents of a revision $wiki.webs['wiki'].pages['HomePage'].revisions[-1] = 'new content' # Checking that all pages can be rendered by the markup engine $wiki.webs['wiki'].pages.each_pair do |name, page| page.revisions.each_with_index do |revision, i| begin revision.display_content rescue => puts "Error when rendering revision ##{i} of page #{name.inspect}:" puts e.message puts e.backtrace.join("\n") end end -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From semt at psitrans.de Mon May 30 05:23:14 2005 From: semt at psitrans.de (Semt, Andreas) Date: Mon May 30 05:16:17 2005 Subject: [Instiki] How to include images in Instiki? Message-ID: <81DBB1DB4CC3B348AA5E2B892428AACC0B5DF6@exchangeptr.psitrans.de> Hello list! I use Instiki 0.10.0 and try to show some images in my wiki. With the HTML tag a image

there's no effect ... the image is not shown. The Textile Reference has a solution: ---

Bunny.

--- However, i don't know the correct location of the image in the wiki directory ... any ideas? Perhaps another ways to include a image in the wiki? THANKS for any help! Greetings, Andreas Semt From alex at verk.info Mon May 30 09:00:23 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 30 08:52:53 2005 Subject: [Instiki] How to include images in Instiki? In-Reply-To: <81DBB1DB4CC3B348AA5E2B892428AACC0B5DF6@exchangeptr.psitrans.de> References: <81DBB1DB4CC3B348AA5E2B892428AACC0B5DF6@exchangeptr.psitrans.de> Message-ID: Semt, Andreas wrote: >I use Instiki 0.10.0 and try to show some images in my wiki. > > 1. To be able to use raw HTML in the pages, make sure that it is allowed in the web settings. Yours most probably is set to filter it out. 2. The right place to put images is instiki/public/images directory. instiki/public is Inastiki's app root for static resources, therefore the URL to instiki/public/images/a.jpg would be something like http://localhost:2500/images/a.jpg By the way, if anybody has figured out the right Apache proxy configuration for 0.10, would you be so kind to post it here or write it up on the wiki? -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From semt at psitrans.de Mon May 30 09:27:04 2005 From: semt at psitrans.de (Semt, Andreas) Date: Mon May 30 09:20:07 2005 Subject: AW: [Instiki] How to include images in Instiki? Message-ID: <81DBB1DB4CC3B348AA5E2B892428AACC083049@exchangeptr.psitrans.de> Thanks Alexey! Your helpful answer leads (me) to a new question: Most of my wiki content is still under Instiki 0.9.2. So: Can I use the Export feature to transfer all 0.9.2 content to Instiki version 0.10.0/0.10.1 ? If not - another possible way? Thanks again! Greetings, Andreas Semt -----Urspr?ngliche Nachricht----- Von: instiki-users-bounces@rubyforge.org [mailto:instiki-users-bounces@rubyforge.org]Im Auftrag von Alexey Verkhovsky Gesendet: Montag, 30. Mai 2005 15:00 An: Instiki-users Betreff: Re: [Instiki] How to include images in Instiki? Semt, Andreas wrote: >I use Instiki 0.10.0 and try to show some images in my wiki. > > 1. To be able to use raw HTML in the pages, make sure that it is allowed in the web settings. Yours most probably is set to filter it out. 2. The right place to put images is instiki/public/images directory. instiki/public is Inastiki's app root for static resources, therefore the URL to instiki/public/images/a.jpg would be something like http://localhost:2500/images/a.jpg By the way, if anybody has figured out the right Apache proxy configuration for 0.10, would you be so kind to post it here or write it up on the wiki? -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From tamassmith at mac.com Mon May 30 12:36:15 2005 From: tamassmith at mac.com (=?ISO-8859-1?Q?Tam=E1s_Smith?=) Date: Mon May 30 12:29:20 2005 Subject: AW: [Instiki] How to include images in Instiki? In-Reply-To: <81DBB1DB4CC3B348AA5E2B892428AACC083049@exchangeptr.psitrans.de> References: <81DBB1DB4CC3B348AA5E2B892428AACC083049@exchangeptr.psitrans.de> Message-ID: Andreas You should make sure to create a backup of your storage directory before doing this, but you should be able to just run Insitki 0.10.1 with your existing storage directory without doing export/import (which is not currently available with direct editing of the Instiki code). Note that you should not try this with 0.10.0 as a few problems cropped up when upgrading from a 0.9x storage directory. However, as far as I can tell, 0.10.1 ironed them out. Tam?s On May 30, 2005, at 9:27 AM, Semt, Andreas wrote: Thanks Alexey! Your helpful answer leads (me) to a new question: Most of my wiki content is still under Instiki 0.9.2. So: Can I use the Export feature to transfer all 0.9.2 content to Instiki version 0.10.0/0.10.1 ? If not - another possible way? Thanks again! Greetings, Andreas Semt -----Urspr?ngliche Nachricht----- Von: instiki-users-bounces@rubyforge.org [mailto:instiki-users-bounces@rubyforge.org]Im Auftrag von Alexey Verkhovsky Gesendet: Montag, 30. Mai 2005 15:00 An: Instiki-users Betreff: Re: [Instiki] How to include images in Instiki? Semt, Andreas wrote: > I use Instiki 0.10.0 and try to show some images in my wiki. > > 1. To be able to use raw HTML in the pages, make sure that it is allowed in the web settings. Yours most probably is set to filter it out. 2. The right place to put images is instiki/public/images directory. instiki/public is Inastiki's app root for static resources, therefore the URL to instiki/public/images/a.jpg would be something like http://localhost:2500/images/a.jpg By the way, if anybody has figured out the right Apache proxy configuration for 0.10, would you be so kind to post it here or write it up on the wiki? -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users _______________________________________________ Instiki-users mailing list Instiki-users@rubyforge.org http://rubyforge.org/mailman/listinfo/instiki-users From alex at verk.info Mon May 30 20:06:26 2005 From: alex at verk.info (Alexey Verkhovsky) Date: Mon May 30 19:58:58 2005 Subject: AW: [Instiki] How to include images in Instiki? In-Reply-To: References: <81DBB1DB4CC3B348AA5E2B892428AACC083049@exchangeptr.psitrans.de> Message-ID: Tam?s Smith wrote: > However, as far as I can tell, 0.10.1 ironed them out. Most of them :) I should release 0.10.2 soon, this will take care of the "endless redirection loop" story and provide better handling of RedCloth / BlueCloth errors. -- Best regards, Alexey Verkhovsky Ruby Forum: http://ruby-forum.org (moderator) RForum: http://rforum.andreas-s.net (co-author) Instiki: http://instiki.org (maintainer) From chris at seagul.co.uk Tue May 31 05:02:47 2005 From: chris at seagul.co.uk (Chris Roos) Date: Tue May 31 04:56:01 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: References: <4291E20D.1060500@seagul.co.uk> <9e94eceaf106fbec6c93182592280acf@mac.com> <4296E4EC.7050800@seagul.co.uk> <42975BEA.2060402@seagul.co.uk> Message-ID: <429C2837.1050608@seagul.co.uk> Alexey Verkhovsky wrote: > > There are two factors at play here. One is a RedCloth bug that causes > RedCloth to raise an error while trying to parse this markup. That's > something worth reporting in RedCloth tracker. > Just so that this is closed properly if you like, having checked the RedCloth issue tracker, this bug has already been reported in #577. From chris at seagul.co.uk Tue May 31 05:04:24 2005 From: chris at seagul.co.uk (Chris Roos) Date: Tue May 31 04:57:39 2005 Subject: [Instiki] Madeleine corruption In-Reply-To: <4295ED13.8080700@seagul.co.uk> References: <4291E20D.1060500@seagul.co.uk> <4295B58E.40400@seagul.co.uk> <4295ED13.8080700@seagul.co.uk> Message-ID: <429C2898.9060203@seagul.co.uk> Chris Roos wrote: > Alexey Verkhovsky wrote: > >> Chris Roos wrote: >> >>> Before I send you the .snapshot file I wanted to say that I can't >>> remember which page it would have been that I was editing. Is it >>> (relatively) easy for you to spot corruption in a page or would it be >>> a case of manually looking into each page to check for malformed html >>> (which I'm led to believe is the cause of the problem from Tamas)? >> >> >> >> No, it is easy. >> > Cool. I've replied off list with the snapshot file. > Big thanks to Alexey for fixing the problem for me. Cheers, Chris From cb.lists+instiki at gmail.com Tue May 31 20:33:49 2005 From: cb.lists+instiki at gmail.com (CB) Date: Tue May 31 20:27:27 2005 Subject: [Instiki] Madeleine error after class name change Message-ID: <429D026D.1090700@gmail.com> Firstly, apologies if this is not too bright a question. I'm at the baby-steps stage with Ruby. I've made a Deplate (http://deplate.sourceforge.net) formatter available in my installation of Instiki. My aims were (1) to have deplate available as my instiki markup, as it's what I write my docs in, and (2) to start learning some Ruby. I've hardly done the latter, as it all proved too easy: - add a new AbstractEngine subclass in engines.rb, calling on the deplate html formatting class - add a *_help.rhtml view for the new formatter - hook this into edit_web.rhtml So far so good. The problem came when I renamed my new class from 'Test' to 'Deplate' (also editing edit_web.rhtml accordingly, and creating deplate_help.rhtml). Now when I start Instwiki I get the following error: -- 8< -- %programfiles%/ruby/lib/ruby/gems/1.8/gems/madeleine-0.7.1/lib/madeleine/automatic.rb:230:in `load': undefined class/odule Engines::Test (ArgumentError) -- 8< -- Madeleine is presumably persisting some of the app state here, and still has a reference to the old Test class. So my question is: if I want to rename the class, is there some way I can prevent this error without having to set up an entirely new wiki? (I do want to retain my data). From assaph at gmail.com Tue May 31 20:59:30 2005 From: assaph at gmail.com (Assaph Mehr) Date: Tue May 31 20:52:30 2005 Subject: [Instiki] Madeleine error after class name change In-Reply-To: <429D026D.1090700@gmail.com> References: <429D026D.1090700@gmail.com> Message-ID: <3ff47be90505311759638ae13d@mail.gmail.com> > The problem came when I renamed my new class from 'Test' to 'Deplate' > (also editing edit_web.rhtml accordingly, and creating > deplate_help.rhtml). Now when I start Instwiki I get the following error: > > -- 8< -- > %programfiles%/ruby/lib/ruby/gems/1.8/gems/madeleine-0.7.1/lib/madeleine/automatic.rb:230:in > `load': undefined class/odule Engines::Test (ArgumentError) > -- 8< -- > > Madeleine is presumably persisting some of the app state here, and still > has a reference to the old Test class. Madeleine is a (binary) storage, it persists all objects in the object graph. When it comes to load those objects next time, it tries to match the objects to their classes. If you've renamed the class than madeleine can't find it when it loads the snapshot and stops. > So my question is: if I want to rename the class, is there some way I > can prevent this error without having to set up an entirely new wiki? (I > do want to retain my data). First backup your data (always a good idea :). You can try Alexey's script for poking in the snapshot files (http://svn.instiki.org/instiki/trunk/script/debug_storage - look at the list archives), or you can export the web contents and recreate it later. Importing is currently disabled in Instiki, but should be easy to uncomment the relevant sections to enable it back. I have not tried any of these methods yet, but am sure i'll need them soon ;-) HTH, Assaph From cb.lists+instiki at gmail.com Tue May 31 21:36:12 2005 From: cb.lists+instiki at gmail.com (CB) Date: Tue May 31 21:29:23 2005 Subject: [Instiki] Madeleine error after class name change In-Reply-To: <3ff47be90505311759638ae13d@mail.gmail.com> References: <429D026D.1090700@gmail.com> <3ff47be90505311759638ae13d@mail.gmail.com> Message-ID: <429D110C.70609@gmail.com> Assaph Mehr wrote: > > >First backup your data (always a good idea :). > > Oh yes. Backups of backups. >You can try Alexey's script for poking in the snapshot files >(http://svn.instiki.org/instiki/trunk/script/debug_storage - look at >the list archives), or you can export the web contents and recreate it >later. Importing is currently disabled in Instiki, but should be easy >to uncomment the relevant sections to enable it back. >I have not tried any of these methods yet, but am sure i'll need them soon ;-) > > OK, thanks. All this must make *using* instiki to keep & hold information a bit of a pain for anyone who's simultaneously hacking it? I guess if the import works OK then it should solve the problem. From assaph at gmail.com Tue May 31 21:54:35 2005 From: assaph at gmail.com (Assaph Mehr) Date: Tue May 31 21:47:34 2005 Subject: [Instiki] Madeleine error after class name change In-Reply-To: <429D110C.70609@gmail.com> References: <429D026D.1090700@gmail.com> <3ff47be90505311759638ae13d@mail.gmail.com> <429D110C.70609@gmail.com> Message-ID: <3ff47be9050531185415a85e4@mail.gmail.com> > All this must make *using* instiki to keep & hold information a bit of a > pain for anyone who's simultaneously hacking it? I guess if the import > works OK then it should solve the problem. I have been bitten before :-) One other thing to watch out for is if you add persistent attributes to the web (or similar classes). You should manually provide the getter if nil is not a valid option, as the objects loaded from the previous snapshot will not have the new instance variable. e.g. attr_reader :funk_level def funk_level @funk_level ||= 'groovy' end Of course, this is not a problem if nil is a valid value. Of course I once had a bunch of such things, so I mechanically added getters for all new stuff, including a bollean property that was true by default: def bool_stuff @bool_stuff ||= true end :-) Cheers, Assaph Cheers, Assaph