Re: [fossil-users] Fossil version 1.32
On Sat, Mar 14, 2015 at 10:24:05PM -0400, Richard Hipp wrote: On 3/14/15, James Turner ja...@calminferno.net wrote: It appears the actual tarballs were also removed. While I understand the reasoning behind this, this will break automated build systems like ports in OpenBSD for stable releases that may reference older versions of fossil. Is there a reason why the tarballs had to be removed? They are still accessible in the old_builds directory. I could move them back. But I decided to make them hard to get to encourage people to upgrade to a version that doesn't have the Ryerson-student-project-eating bug. *If* you can make a compelling argument to move them back, I might. It's not such a big deal for us, OpenBSD, but I wasn't sure for others if having archives go missing would break builds. Then again I guess it's a good way to force people to upgrade :) -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil version 1.32
It appears the actual tarballs were also removed. While I understand the reasoning behind this, this will break automated build systems like ports in OpenBSD for stable releases that may reference older versions of fossil. Is there a reason why the tarballs had to be removed? -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Fossil wiki parsing code
I was curious if anyone has separated the fossil wiki parsing/formatting code from fossil into a standalone library? I was looking to use it in my own project as an alternative to markdown and figured I'd check before doing it myself. Thanks. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil wiki parsing code
On Mon, Dec 01, 2014 at 02:00:45PM -0500, Richard Hipp wrote: On Mon, Dec 1, 2014 at 1:30 PM, James Turner ja...@calminferno.net wrote: I was curious if anyone has separated the fossil wiki parsing/formatting code from fossil into a standalone library? Never been separated out, as far as I know. But you'll find all the code in the single file named wikiformat.c. It shouldn't be that hard to separate from the rest of the system. Yup, was just browsing wikiformat.c. I'll share what I come up with. I also plan on creating a lua wrapper for it as my app is written in lua. Thanks. I was looking to use it in my own project as an alternative to markdown and figured I'd check before doing it myself. Thanks. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil wiki parsing code
On Mon, Dec 01, 2014 at 08:22:49PM +0100, Stephan Beal wrote: On Mon, Dec 1, 2014 at 8:11 PM, James Turner ja...@calminferno.net wrote: Yup, was just browsing wikiformat.c. I'll share what I come up with. I also plan on creating a lua wrapper for it as my app is written in lua. fwiw (oh, come on, most of you knew this was coming ;)... Depending on the features you need, you might want to consider libfossil, which was designed with scripting in mind and contains two proof-of-concept script bindings for my own pet languages: http://fossil.wanderinghorse.net/r/libfossil/ Thanks Stephan, I actually looked at libfossil first. In this particular case I'm just looking for a markdown alternative for a static site generator I'm using to power my personal site. (new URL (no more .cgi on the end), just set up today) The script bindings: https://docs.google.com/document/d/13gRSl6-bj3LV-OKgE-BsqvqF33UFYW3oa3A2OJC5QSY/view and some scripts: http://fossil.wanderinghorse.net/r/libfossil/finfo?name=s2/timeline.s2 http://fossil.wanderinghorse.net/r/libfossil/finfo?name=s2/require.d/fsl/extendFossil.s2 The point being: adding a lua binding should be a snap (and i'd be happy to help - i've got no personal experience with lua but have worked with many scripting engines and written several). -- - stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal Freedom is sloppy. But since tyranny's the only guaranteed byproduct of those who insist on a perfect world, freedom will have to do. -- Bigby Wolf ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil wiki parsing code
On Mon, Dec 01, 2014 at 08:18:06PM +0100, Stephan Beal wrote: On Mon, Dec 1, 2014 at 8:00 PM, Richard Hipp d...@sqlite.org wrote: Never been separated out, as far as I know. But you'll find all the code in the single file named wikiformat.c. It shouldn't be that hard to separate from the rest of the system. Unless i'm mistaken, the block in question is: https://www.fossil-scm.org/index.html/info/fc6a1e15a1733e49c7fc188931cec748d3ead074?ln=1434-1442 that parses the contents of a [...] block. Optionally, you might want to look at: https://www.fossil-scm.org/index.html/info/fc6a1e15a1733e49c7fc188931cec748d3ead074?ln=1193 as the handling might better be added there. Good point, the key parts i want, are the hyperlinks and paragraphs so I may only need 1 or 2 functions. Thanks! -- - stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal Freedom is sloppy. But since tyranny's the only guaranteed byproduct of those who insist on a perfect world, freedom will have to do. -- Bigby Wolf ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] SQLite pointer misuse in head?
Did you remember to run fossil rebuild? On Thu, Oct 30, 2014 at 12:53:54PM -0600, Warren Young wrote: I just updated my Fossil server from version [e061a675e6] 2014-09-26 21:02:03 to the tip of trunk, as of a few minutes ago. On restarting “fossil server”, I got these complaints: SQLITE_MISUSE: API call with NULL database connection pointer SQLITE_MISUSE: misuse at line 103842 of [83afe23e55] SQLITE_MISUSE: API call with NULL database connection pointer SQLITE_MISUSE: misuse at line 103842 of [83afe23e55] I reverted my tree, restarted, and the error doesn’t recur. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Is it possible to add empty folders to the repo?
On Mon, Oct 06, 2014 at 04:44:30PM +0200, Stephan Beal wrote: Nope - fossil tracks files only. But you can always add a .keep file or something in a directory you want to keep around and track that. - stephan Sent from a mobile device, possibly from bed. Please excuse brevity and typos. On Oct 6, 2014 4:38 PM, Baruch Burstein bmburst...@gmail.com wrote: I tried doing: mkdir a fossil add a but that didn't work. -- ˙uʍop-ǝpısdn sı ɹoʇıuoɯ ɹnoʎ 'sıɥʇ pɐǝɹ uɐɔ noʎ ɟı -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] importing/forking from Git
On Tue, Mar 04, 2014 at 03:47:31PM -0500, Richard Hipp wrote: On Tue, Mar 4, 2014 at 3:43 PM, Chad Perrin c...@apotheon.net wrote: Trying to fork/import from Git is kinda problematic. I started by creating a new project on a server: $ fossil init projectname.fossil I then cloned locally: $ fossil clone u...@uri.for/projectname/index.cgi projectname.fossil Doesn't the git import create the fossil repo for you? Maybe you should try doing the import into a fossil repo that doesn't already exist. Then you can scp that repo to your server and clone it down. I imported from Git: $ git fast-export --all | fossil import --git --incremental \ /path/to/projectname.fossil I opened the repository: $ mkdir projectname; cd projectname fossil open /path/to/projectname.fossil The above should be all you need to do. It should just work. And it does just work for every git repository that I've tried, though that is, admittedly, not that many repos. If you are willing to share with us the specific repo you are trying to import, prehaps the developers can try it and discover what's going wrong. No files appear. I looked at the timeline: $ fossil timeline === 2014-03-04 === 20:24:06 [178338650a] *CURRENT* initial empty check-in (user: apotheon tags: trunk) === 2013-06-19 === 21:12:33 [34ca661224] last commit message from Git === 2011-10-16 === . . . et cetera The obvious choice seems to be to use revert: $ fossil help revert Usage: fossil revert ?-r REVISION? ?FILE ...? Revert to the current repository version of FILE, or to the version associated with baseline REVISION if the -r flag appears. [. . .] Revert all files if no file name is provided. [. . .] I tried that: $ fossil revert -r 34ca661224 the --revision option does not work for the entire tree How do I actually get a repository import to a usable state so I can push it to the server's repository and get on with my life? -- Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] importing/forking from Git
On Tue, Mar 04, 2014 at 01:55:36PM -0700, Chad Perrin wrote: On Tue, Mar 04, 2014 at 03:47:31PM -0500, Richard Hipp wrote: On Tue, Mar 4, 2014 at 3:43 PM, Chad Perrin c...@apotheon.net wrote: Trying to fork/import from Git is kinda problematic. I started by creating a new project on a server: $ fossil init projectname.fossil I then cloned locally: $ fossil clone u...@uri.for/projectname/index.cgi projectname.fossil I imported from Git: $ git fast-export --all | fossil import --git --incremental \ /path/to/projectname.fossil Bah incremental requires the repo to exist. Without incremental the import is successful and trunk is set correctly. I opened the repository: $ mkdir projectname; cd projectname fossil open /path/to/projectname.fossil The above should be all you need to do. It should just work. And it does just work for every git repository that I've tried, though that is, admittedly, not that many repos. If you are willing to share with us the specific repo you are trying to import, prehaps the developers can try it and discover what's going wrong. git://repo.or.cz/nvi.git For reference, I'm using . . . $ fossil version This is fossil version 1.28 [3d49f04587] 2014-01-27 17:33:44 UTC . . . on FreeBSD, installed from ports, with the JSON and STATIC options both selected at build time. -- Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ] ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] http_ssl paren issue
I think maybe http_ssl needs an extra set of parentheses to deal with
this:
./src/http_ssl.c: In function 'ssl_open':
./src/http_ssl.c:288: warning: cast to pointer from integer of different size
--
James Turner
Index: src/http_ssl.c
==
--- src/http_ssl.c
+++ src/http_ssl.c
@@ -283,11 +283,11 @@
return 1;
}
BIO_get_ssl(iBio, ssl);
#if (SSLEAY_VERSION_NUMBER = 0x00908070) !defined(OPENSSL_NO_TLSEXT)
- if( !SSL_set_tlsext_host_name(ssl,
pUrlData-useProxy?pUrlData-hostname:pUrlData-name) ){
+ if( !SSL_set_tlsext_host_name(ssl,
(pUrlData-useProxy?pUrlData-hostname:pUrlData-name)) ){
fossil_warning(WARNING: failed to set server name indication (SNI),
continuing without it.\n);
}
#endif
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil ui/serv
I should note, this happens on new repos and cloned repos like fossil-scm.org. On Mon, Jan 27, 2014 at 05:10:27PM -0500, James Turner wrote: Am I doing something wrong? I'm suddenly getting Not Found for all URLs. Download lastest fossil release or run trunk. Build with: ./configure make make install Run: fossil ui http://localhost:8080 returns Not Found, all urls return Not Found. This is on OpenBSD and Firefox 26.0. -- James Turner -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil ui/serv
On Mon, Jan 27, 2014 at 05:44:31PM -0500, Richard Hipp wrote: On Mon, Jan 27, 2014 at 5:10 PM, James Turner ja...@calminferno.net wrote: Am I doing something wrong? I'm suddenly getting Not Found for all URLs. Download lastest fossil release or run trunk. Build with: ./configure make make install Run: fossil ui http://localhost:8080 returns Not Found, all urls return Not Found. This is on OpenBSD and Firefox 26.0. Dunno what might be the problem. What you are doing works on Linux, Mac, and Windows. My only OpenBSD access is a shared-host account via ssh, so I cannot run fossil ui there to verify it. But if it works on Linux and Mac, why shouldn't it also work on OpenBSD? -- D. Richard Hipp d...@sqlite.org I dunno either. Everything is fine on my main development machine. I'll chalk it up to a messed up virtual machine I guess. Sorry for the noise. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil 1.28 and SQLite 3.8.0.2
On Mon, Jan 27, 2014 at 06:01:42PM -0500, Richard Hipp wrote: On Mon, Jan 27, 2014 at 5:54 PM, Jan Nijtmans jan.nijtm...@gmail.comwrote: 2014-01-27 Richard Hipp d...@sqlite.org: I think URL like:.../tree?ci=trunk will fail without 3.8.2. This is the url which uses WITHOUT ROWID in trunk, but thanks to the SQLite version check it doesn't depend on it (yet) in Fossil 1.28. Yes, you are correct. I looked only on trunk... FWIW, Fossil 1.29 will very definitely require SQLite 3.8.3 or later in as much as it uses common table expressions to help generate the timeline now. -- D. Richard Hipp d...@sqlite.org Thanks for the clarification guys. I'm thinking it's probably safe then to use 1.28 with our 3.8.0.2 version of SQLite. I'll get our in-tree version of SQLite upgraded to the latest after the unlock. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Tree view and browser back button
On Mon, Jan 13, 2014 at 03:54:34PM -0500, Richard Hipp wrote: On Mon, Jan 13, 2014 at 3:51 PM, Martin Gagnon eme...@gmail.com wrote: I like very much the new tree view but there's one thing that I found a bit annoying. After clicking on a file that is deep in the directory structure, the tree become all collapsed again after pressing the back button. Chrome does that. Firefox, otoh, keeps the tree expanded. I don't know why and don't know if there is anything we can do about it in Javascript/CSS. -- D. Richard Hipp d...@sqlite.org The easiest way would probably be to manipulate the url has as you expand folders. So as an example if you expand ajax the url would become: /index.html/tree?ci=trunk#ajax then another level in it would be /index.html/tree?ci=trunk#ajax/cgi-bin. Then based on the hash in js you could expand the folder structure on load. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Tree view and browser back button
On Mon, Jan 13, 2014 at 04:03:14PM -0500, James Turner wrote: On Mon, Jan 13, 2014 at 03:54:34PM -0500, Richard Hipp wrote: On Mon, Jan 13, 2014 at 3:51 PM, Martin Gagnon eme...@gmail.com wrote: I like very much the new tree view but there's one thing that I found a bit annoying. After clicking on a file that is deep in the directory structure, the tree become all collapsed again after pressing the back button. Chrome does that. Firefox, otoh, keeps the tree expanded. I don't know why and don't know if there is anything we can do about it in Javascript/CSS. -- D. Richard Hipp d...@sqlite.org The easiest way would probably be to manipulate the url has as you expand folders. Should read: The easiest way would probably be to manipulate the url hash as you expand folders. So as an example if you expand ajax the url would become: /index.html/tree?ci=trunk#ajax then another level in it would be /index.html/tree?ci=trunk#ajax/cgi-bin. Then based on the hash in js you could expand the folder structure on load. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Problem with compilation under MINGW
On Fri, Jan 03, 2014 at 10:17:47PM -0500, James Turner wrote: [snip] I'll check with others but I'm not sure reliability is really the concern. We imported SQLite into our base tree. Because of this we try, when possible, to limit duplicating libraries in ports to reduce having to patch multiple versions of the same libraries. A good example of this is Firefox and newer software that uses WebKit. These projects like to ship their own copies of libraries to make their building process easier. For us, it just increases our headaches when it comes to maintaining required patches. SQLite is clearly a different case. I believe the only major patch we have in our tree relates to adding arc4random(3) support. If fossil removes --disable-internal-sqlite I highly doubt I'll be asked to maintain this feature in our ports tree (again I'll check with others and get back with you). -- James Turner I didn't get a strong outcry from other OpenBSD porters. We understand embedding SQLite is the typical way it's used and therefor somewhat of a different case. We would prefer --disable-internal-sqlite to remain but again you are the maintainers and of course must make the final decision. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Problem with compilation under MINGW
On Fri, Jan 03, 2014 at 09:28:52PM -0500, Richard Hipp wrote: On Fri, Jan 3, 2014 at 9:14 PM, Richie Adler richiead...@gmail.com wrote: After checkin [bd1151126a], compilation under MINGW produces the following error: Yes, it fails for me too OK, so I propose the following fix: (1) Move [bd1151126a] into a branch. (2) Remove the --disable-internal-sqlite option on trunk. Require the use of the built-in SQLite only, since SQLite needs to be built with non-standard compile-time options to fully meet the needs of Fossil. (3) Cherry-pick check-ins after [bd1151126a] onto the trunk. (4) When distributions complain, we simply explain that we tried using an external SQLite but it introduced too many complications and bugs and that we now require statically linking SQLite for reasons of security and reliability. Comments? -- D. Richard Hipp d...@sqlite.org I won't begin to tell you how to develop fossil but I do want to throw out there that OpenBSD has been providing fossil with --disable-internal-sqlite via it's ports tree and packages fairly successfully (and with 'fossil sqlite3' support) for the last year+. I am the maintainer if you have any questions. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Problem with compilation under MINGW
On Fri, Jan 03, 2014 at 10:04:25PM -0500, Richard Hipp wrote: On Fri, Jan 3, 2014 at 9:33 PM, James Turner ja...@calminferno.net wrote: I won't begin to tell you how to develop fossil but I do want to throw out there that OpenBSD has been providing fossil with --disable-internal-sqlite via it's ports tree and packages fairly successfully (and with 'fossil sqlite3' support) for the last year+. I am the maintainer if you have any questions. How much push-back are we going to get from the OpenBSD community if --disable-internal-sqlite goes away due to reliability concerns? I appreciate the desire to keep libraries like libz and libjpeg as shared libraries so that if security problems are encountered they can be fixed in one place. But those libraries are dealing with unvetted input from untrusted sources. SQLite does not work that way. Any application that allows unvetted SQL text through from untrusted sources has way worse problems than any bugs in SQLite. For this reason, while there have been many bugs reported against SQLite from the field in its 14 year history, no security vulnerabilities have ever been reported. Not one. This in spite of SQLite being using in over 1 million different applications with over 2 billion deployments. Security vulnerabilities are just not an issue like they are with libraries that process raw user input. Fossil also uses libz, but we have no issue with using a shared library for that. A version of libz is included in the Fossil source tree, but that is merely to simplify compilation on windows systems which do not commonly have libz handy. -- D. Richard Hipp d...@sqlite.org I'll check with others but I'm not sure reliability is really the concern. We imported SQLite into our base tree. Because of this we try, when possible, to limit duplicating libraries in ports to reduce having to patch multiple versions of the same libraries. A good example of this is Firefox and newer software that uses WebKit. These projects like to ship their own copies of libraries to make their building process easier. For us, it just increases our headaches when it comes to maintaining required patches. SQLite is clearly a different case. I believe the only major patch we have in our tree relates to adding arc4random(3) support. If fossil removes --disable-internal-sqlite I highly doubt I'll be asked to maintain this feature in our ports tree (again I'll check with others and get back with you). -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Build is broke on OpenBSD
On Thu, Dec 12, 2013 at 07:50:54PM -0500, Richard Hipp wrote: Looks like it was busted on linux too. Should be fixed in the latest check-in, though. Yup everything is building correctly again. Thanks for the quick fix. On Thu, Dec 12, 2013 at 7:26 PM, James Turner ja...@calminferno.net wrote: The fossil build is currently broke on OpenBSD. SQLite now relies on pthread functions but does not link against -lpthread. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Version 1.27
On Wed, Sep 11, 2013 at 10:52:44AM -0500, djg...@gmail.com wrote: I didn't know there was an official testing document, I just dropped the exe in place and ran: fossil server file.fossil I mostly just use fossil for the wiki which it looked fine so just to clarify I just glanced over the Wiki portion of Fossil and nothing else. Jonathan Otsuka Fossil has tons of regressions tests. Download the source, tclsh /path/to/fossil/source/test/tester.tcl /path/to/fossil On Sep 11, 2013, at 10:00 AM, djg...@gmail.com wrote: This worked on Windows 2008 R2 64-bit, Windows 7 32-bit, Mountain Lion 10.8.4 64-bit. Jonathan Otuska On Sep 11, 2013, at 9:03 AM, Richard Hipp d...@sqlite.org wrote: Fossil version 1.27 has been tagged and precompiled binaries have been uploaded to http://www.fossil-scm.org/download.html I do this infrequently enough that I keep forgetting exactly how things were compiled the previous time. So please try out the precompiled binaries and let me know if you encounter any problems. I started a wiki page (http://www.fossil-scm.org/fossil/wiki?name=Release+Build+How-To) describing how the precompiled binaries are produced. If adjustments need to be made for compatibility purposes, please let me know. For example, will the Windows8 build using MSVC-x86 work for everybody, or do I need to redo that build on (say) Windows7 or perhaps using MinGW? Does the Mac binary work on all intel-based Mac systems? Etc. I'll send out an official release announcement once a few people have verified the current binaries. -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] 1.26 release archive
The 1.26 release archive is tagged as 20130618210923 yet the extracted folder is fossil-src-20130618210323. Any chance a new tarball can be rolled? Saves me from having to put a work around in for the OpenBSD port I maintain. Thanks. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] 1.26 release archive
On Tue, Jun 18, 2013 at 06:43:38PM -0400, Richard Hipp wrote: On Tue, Jun 18, 2013 at 6:39 PM, James Turner ja...@calminferno.net wrote: The 1.26 release archive is tagged as 20130618210923 yet the extracted folder is fossil-src-20130618210323. Any chance a new tarball can be rolled? Saves me from having to put a work around in for the OpenBSD port I maintain. I uploaded the new tarball moments before your email arrived. Please try again. -- D. Richard Hipp d...@sqlite.org Looks good. Thanks. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Core dump in th_lang
Every once and awhile fossil core dumps on me while serving up sites using cgi. Here is the key information from the backtrace. If this isn't helpful I can provide the core file. This is with 1.25 but it doesn't look like anything has changes in trunk that would fix this. #0 0x1c0e2036 in proc_command (interp=0x80ce8e00, ctx=0x0, argc=4, argv=0x87cdda00, argl=0x87cdda10) at ./src/th_lang.c:463 463 ./src/th_lang.c: No such file or directory. in ./src/th_lang.c -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil 1.25 Windows XP - open command 'fails'
On Thu, Apr 04, 2013 at 09:40:19PM +0200, Jan wrote: Sorry, I forgot to mention that the repo size is ok and fossil ui tells me that all the content is there - in theory. But not in the check out. fossil test-integrity -R test.fossil 1677 non-phantom blobs (out of 1677 total) checked: 0 errors I also ran fossil rebuild test.fossil Probably it is an XP thing? I'll try it on Win7 as well. Jan After you run fossil open, maybe you just need to checkout the correct branch. Try fossil up trunk. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Chiselapp.com shutting down
On Fri, Mar 29, 2013 at 02:29:54PM +0100, Martijn Coppoolse wrote: Op 29-3-2013 1:52, James Turner schreef: After a couple weeks of debate, I've decided to shut down Chiselapp.com. As the message on the homepage states, new account and repository creation has been disabled. Access to the website and repositories will remain open until May 1st 2013. Sad to see it go, I really liked its simplicity (both in usage and looks). If someone wanted to carry on the initiative, could you tell us approximately what the load would be, both in terms of bandwidth and storage space? A big thanks for everyone who's used the service, sadly I just don't have the time to maintain it and hosting it has started to become a burden. A big thank you for having hosted it all these years! The Chisel codebase, Flint, is available under the AGPLv3 license so if your interested now is probably the time to grab it. I’d already done that. :-) -- Martijn Coppoolse ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users Martijn, I don't really have numbers for bandwidth usage, I host it at my home on a business internet connection so I don't keep track since it's unlimited. However, space wise Chisel has about 11.4G worth of repositories. It's a fairly small operation I just don't have the time or frankly the interest in running it anymore. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Chiselapp.com shutting down
On Fri, Mar 29, 2013 at 05:09:46PM +, John Coulter wrote: James Turner james@... writes: [message clipped] The Chisel codebase, Flint, is available under the AGPLv3 license so if your interested now is probably the time to grab it. James, would you consider changing the license to something more permissive, perhaps BSD-style, or at least the GPL? I understand the intent behind the AGPL, but I think the project is more likely to survive in the community with a less- restrictive license. Thanks for hosting the service for all this time! Done. Flint is now licensed under the more permissive ISC license [0]. [0] http://opensource.org/licenses/ISC -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Chiselapp.com shutting down
After a couple weeks of debate, I've decided to shut down Chiselapp.com. As the message on the homepage states, new account and repository creation has been disabled. Access to the website and repositories will remain open until May 1st 2013. Richard, if you wouldn't mind removing the link to chiselapp.com from the Fossil homepage I would appreciate it. A diff is attached for your convenience. A big thanks for everyone who's used the service, sadly I just don't have the time to maintain it and hosting it has started to become a burden. The Chisel codebase, Flint, is available under the AGPLv3 license so if your interested now is probably the time to grab it. -- James Turner Index: www/index.wiki == --- www/index.wiki +++ www/index.wiki @@ -142,12 +142,10 @@ * How Fossil does [./password.wiki | password management]. * On-line [/help | help]. * Documentation on the [http://www.sqliteconcepts.org/THManual.pdf | TH1 Script Language] used to configure the ticketing subsystem. - * A free hosting server for Fossil repositories is available at - [http://chiselapp.com/]. * How to [./server.wiki | set up a server] for your repository. * Customizing the [./custom_ticket.wiki | ticket system]. * Methods to [./checkin_names.wiki | identify a specific check-in]. * [./inout.wiki | Import and export] from and to Git. * [./fossil-v-git.wiki | Fossil versus Git]. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Side-by-side diff and non-English text
On Sat, Mar 02, 2013 at 08:01:33PM +0300, Sergei Gavrikov wrote: On Thu, 28 Feb 2013, Richard Hipp wrote: On Thu, Feb 28, 2013 at 4:20 AM, Jan Nijtmans jan.nijtm...@gmail.com wrote: 2013/2/28 Sergei Gavrikov sergei.gavri...@gmail.com: On Wed, 27 Feb 2013, Joe Mistachkin wrote: Do you have an example that is viewable online? Find, please, a test http://chiselapp.com/user/sg/repository/pangrams Look at side-by-side diff views from 2nd to 3rd check-in and from 3rd to 4th. It looks like the span class=diffrm, which is inserted, is dropped in the middle of a 2-byte UTF-8 character, splitting it in two invalid characters. That should never be done! Fixed here: http://www.fossil-scm.org/fossil/info/e2b3490676 Thank you. Unfortunately, splitting did not go away. As chiselapp.com has not got this workaround yet, I created two test repositories (Lorem Ipsum in Greek and Russian, UTF-8 encoded): (1) https://chiselapp.com/user/sg/repository/lipsum (2) http://91.208.39.24:8080/lipsum The second service is managed by the latest Fossil (+ e2b3490676) Please, compare results https://chiselapp.com/user/sg/repository/lipsum/vdiff?from=8b0984ab46648f58to=121bf8035b966861 http://91.208.39.24:8080/lipsum/vdiff?from=8b0984ab46648f58to=121bf8035b966861 Sergei Chiselapp is a third party service and uses the latest stable fossil. I don't really have the time to give it the attention it needs and have been debating shutting it down. -- James Turner ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] New regexp OpenBSD build error
On Sat, Jan 05, 2013 at 08:03:49PM +0100, Baptiste Daroussin wrote: 2013/1/2 Stephan Beal sgb...@googlemail.com: On Wed, Jan 2, 2013 at 8:56 AM, Baptiste Daroussin baptiste.darous...@gmail.com wrote: 2013/1/2 James Turner ja...@calminferno.net: ... in.mx = nIn=0 ? nIn : strlen(zIn); ... pRe-sIn.mx = strlen(pRe-sIn.z); those are fixed now, too. Thanks again for the report. http://fossil-scm.org/index.html/info/46af6b8260 -- - stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users Thanks it now builds and works without any warnings on FreeBSD Bapt OpenBSD as well. -- James Turner | 0x560B0C49 ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] New regexp OpenBSD build error
On OpenBSD re_exec is already defined in unistd.h causing the latest trunk to fail to build. In file included from ./src/diff.c:22: bld/diff.h:287: error: conflicting types for 're_exec' /usr/include/unistd.h:478: error: previous declaration of 're_exec' was here *** Error 1 in /home/james/code/fossil (./src/main.mk:557 'bld/diff.o') It's obsolete but sadly still present. http://www.openbsd.org/cgi-bin/man.cgi?query=re_execapropos=0sektion=0manpath=OpenBSD+Currentarch=i386format=html -- James Turner | 0x560B0C49 ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] applied DVCS for collaborative work - on the fossil project itself, or otherwise
On Thu, Dec 13, 2012 at 09:27:03AM -0700, Matt Welland wrote: One partial solution available today is to use http://chiselapp.com. Simply use their clone repo feature with regular pull. What is missing is a way to register the fork with the original project. It would be cool if it was possible to submit the URL of the forked repo to the parent. The parent can verify that the URL is a real fork and not some random URL and then make the forks (with description?) browseable on a Forks page. I like this, I should add the ability to easily fork chisel repositories. It would be easy to maintain the association and provide a page to show forks of projects. I'll add this to my chiselapp todo list. Thanks for the idea! -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Add NetSurf to allowed UserAgents
On Thu, Nov 22, 2012 at 10:40:32PM -0500, James Turner wrote: NetSurf [0] is probably a lesser known browser, but I run it on my Lemote Yeeloong, which is mips based and doesn't support larger browsers like Firefox under OpenBSD. It might be nice to have it added. Thanks. [0] http://www.netsurf-browser.org/ -- James Turner ja...@calminferno.net I know there has been more important things brewing recently on the mailing list but anyone with commit rights willing to commit this quick hit for us NetSurf users? Thanks. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] howto `grep' through old revisions
What about just using regex(3)? It's included with BSD and quite possibly Linux distributions, but I have no way of checking. No idea how good the library is, but it might cut down on dependencies? -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Add NetSurf to allowed UserAgents
NetSurf [0] is probably a lesser known browser, but I run it on my Lemote Yeeloong, which is mips based and doesn't support larger browsers like Firefox under OpenBSD. It might be nice to have it added. Thanks. [0] http://www.netsurf-browser.org/ -- James Turner ja...@calminferno.net Index: src/login.c == --- src/login.c +++ src/login.c @@ -404,10 +404,11 @@ return 0; } if( memcmp(zAgent, Opera/, 6)==0 ) return 1; if( memcmp(zAgent, Safari/, 7)==0 ) return 1; if( memcmp(zAgent, Lynx/, 5)==0 ) return 1; + if( memcmp(zAgent, NetSurf/, 8)==0 ) return 1; return 0; } /* ** COMMAND: test-ishuman ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] why does `fossil rm' not do the real thing?
On Tue, Nov 20, 2012 at 03:39:06PM +, David Given wrote: Richard Hipp wrote: [...] CVS did not couple the actions, and I copied CVS in this regard. I agree with you now, that coupling them is the right thing to do. But I fear to change it because that might cause problems for existing scripts. Add a -p for physical option to actually change the files, and leave the default as is? I agree, changing the existing behaviour would be a recipe for disaster. I'd suggest -f like cvs rm uses. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil all rebuild seg faults
With the latest fossil trunk (This is fossil version 1.24 [bdbe6c74b8] 2012-10-30 18:14:27 UTC) fossil all rebuild is seg faulting for me. fossil all rebuild Segmentation fault (core dumped) gdb is showing the below: #0 collect_arguments (zArg=0x7f7f Address 0x7f7f out of bounds) at allrepo.c:61 61 allrepo.c: No such file or directory. in allrepo.c -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil all rebuild seg faults
Looks good. fossil all rebuild is working for me again. If it helps explain anything, I'm running OpenBSD. On Tue, Oct 30, 2012 at 08:11:53PM -0400, Richard Hipp wrote: Please try the latest and let me know whether or not the problem is fixed. Tnx for the report. On Tue, Oct 30, 2012 at 6:31 PM, James Turner ja...@calminferno.net wrote: With the latest fossil trunk (This is fossil version 1.24 [bdbe6c74b8] 2012-10-30 18:14:27 UTC) fossil all rebuild is seg faulting for me. fossil all rebuild Segmentation fault (core dumped) gdb is showing the below: #0 collect_arguments (zArg=0x7f7f Address 0x7f7f out of bounds) at allrepo.c:61 61 allrepo.c: No such file or directory. in allrepo.c -- James Turner ja...@calminferno.net -- D. Richard Hipp d...@sqlite.org -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil all rebuild seg faults
On Tue, Oct 30, 2012 at 08:22:25PM -0400, Richard Hipp wrote: On Tue, Oct 30, 2012 at 8:18 PM, Steve Bennett ste...@workware.net.auwrote: On 31/10/2012, at 10:11 AM, Richard Hipp wrote: Please try the latest and let me know whether or not the problem is fixed. Tnx for the report. Regarding your latest commit, I've run across this on 64 bit too. The problem is the '0' at the end of the variable args. Use NULL instead, otherwise you only get a 32 bit zero value instead of 64 bit. I bet you're right. I've previously made the same mistake using using Tcl_AppendResult() Yeah I'm running amd64, so definitely 64bit over here. Cheers, Steve Embedded Systems Specialists - http://workware.net.au/ WorkWare Systems Pty Ltd W: www.workware.net.au P: +61 434 921 300 E: ste...@workware.net.au F: +61 7 3391 6002 -- D. Richard Hipp d...@sqlite.org -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Wiki naming limitations
On Tue, Oct 23, 2012 at 11:36:34PM +, K wrote:
Is someone able to explain to me why there is a 3-character minimum to wiki
names? This seems to me like a strange arbitrary constraint to impose.
The minimum-3-chararacter rule is subjective. When I wrote Fossil (6 years
ago) it never occurred to me that anybody would want a wiki page name of
less than 3 characters.
The patch to relax that rule should be trivial.
I just tried fossil 1.24 and wiki pages still must be 3 characters or
greater. I would appreciate receiving instruction on how to go about getting
this changed to a lower number of characters. I personally would suggest a
minimum of 1 character as I find anything greater to be arbitrary.
Please advise.
Kind regards,
^K
K,
The attached diff should decrease the limit to 1 character.
--
James Turner
ja...@calminferno.net
Index: src/wiki.c
==
--- src/wiki.c
+++ src/wiki.c
@@ -27,11 +27,11 @@
** Return true if the input string is a well-formed wiki page name.
**
** Well-formed wiki page names do not begin or end with whitespace,
** and do not contain tabs or other control characters and do not
** contain more than a single space character in a row. Well-formed
-** names must be between 3 and 100 chracters in length, inclusive.
+** names must be between 1 and 100 chracters in length, inclusive.
*/
int wiki_name_is_wellformed(const unsigned char *z){
int i;
if( z[0]=0x20 ){
return 0;
@@ -39,11 +39,11 @@
for(i=1; z[i]; i++){
if( z[i]0x20 ) return 0;
if( z[i]==0x20 z[i-1]==0x20 ) return 0;
}
if( z[i-1]==' ' ) return 0;
- if( i3 || i100 ) return 0;
+ if( i1 || i100 ) return 0;
return 1;
}
/*
** Output rules for well-formed wiki pages
@@ -52,11 +52,11 @@
@ ul
@ li Must not begin or end with a space./li
@ li Must not contain any control characters, including tab or
@ newline./li
@ li Must not have two or more spaces in a row internally./li
- @ li Must be between 3 and 100 characters in length./li
+ @ li Must be between 1 and 100 characters in length./li
@ /ul
}
/*
** Check a wiki name. If it is not well-formed, then issue an error
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Warning fix
Fix for: warning: passing argument 2 of 'db_get' discards qualifiers from
pointer target type from latest diffcmd change.
--
James Turner
ja...@calminferno.net
Index: src/diffcmd.c
==
--- src/diffcmd.c
+++ src/diffcmd.c
@@ -471,11 +471,11 @@
zName = gdiff-command;
}else{
zDefault = 0;
zName = diff-command;
}
- return db_get(zName, zDefault);
+ return db_get(zName, (char*)zDefault);
}
/*
** COMMAND: diff
** COMMAND: gdiff
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Corrupt cloning of private repo
On Thu, Aug 23, 2012 at 01:35:31PM -0400, Tomek Kott wrote: Bill, Nothing is shown in the timeline besides the initial commit entry, and no files are shown in the browser. Manually going to artifact pages also results in nothing. I also want to clarify that I am in no way trying to rush this: I have access to (two) full repos. I just want to diagnose the cloning process / private repo process for future use. Thanks, Tomek Is the branch private? Don't you need to pass --private while cloning? -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil version 1.23
On Wed, Aug 08, 2012 at 08:10:58AM -0400, Richard Hipp wrote: Fossil version 1.23 is now available on the website. See http://www.fossil-scm.org/download.html for additional information. Please report any problems to this mailing list. [snip] D. Richard Hipp d...@sqlite.org I just download the tar.gz provided on the website and ran ./configure make on OpenBSD 5.0 I get the following make failure. make: don't know how to make ./src/../manifest.uuid. Stop in /home/james/fossil-src-20120808112557. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil version 1.23
On Wed, Aug 08, 2012 at 10:47:44AM -0400, Richard Hipp wrote: On Wed, Aug 8, 2012 at 10:42 AM, James Turner ja...@calminferno.net wrote: I just download the tar.gz provided on the website and ran ./configure make on OpenBSD 5.0 I get the following make failure. make: don't know how to make ./src/../manifest.uuid. Stop in /home/james/fossil-src-20120808112557. Please try again with the updated tarball and let me know either way. Tnx. -- D. Richard Hipp d...@sqlite.org New tarball is good, everything compiles fine. Thanks. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] althttpd router patch
On Thu, Jun 21, 2012 at 04:25:15PM -0400, James Turner wrote:
Attached is a patch that adds a new -router option to alhttpd. If the
requested file is not present on disk the filename passed to -router
will be ran instead. This allows you to then use REQUEST_URI to handle
URL routing within your application. By default -router will look in
docroot so there is no need to pass a forward slash.
Ex. althttpd -router router.cgi
If you navigate to http://localhost/ or http://localhost/blog/1
http://localhost/router.cgi is loaded but the URL stays the same and
REQUEST_URI will be / or /blog/1
-router is checked in 2 places once before checking for not-found.html
and once after determining index.html and index.cgi don't exist, since
the first check wasn't enough.
With this change I'm no longer dependent on Apache with mod_rewrite.
Not sure if anyone is interested in this feature, but attached is an
updated patch that applies cleanly after the latest althttpd changes.
--
James Turner
ja...@calminferno.net
--- althttpd.c.orig Sun Jul 8 19:10:17 2012
+++ althttpd.c Sun Jul 8 19:48:04 2012
@@ -150,6 +150,7 @@ static char *zHttp = http; /* http or https */
static int useTimeout = 1; /* True to use times */
static int ipv6Only = 0; /* Use IPv6 only */
static int ipv4Only = 0; /* Use IPv4 only */
+static char *zRouterFile = 0;/* Use this file as a router */
/*
** Change every space or unprintable character in the zAgent[] string
@@ -1161,6 +1162,15 @@ void ProcessOneRequest(int forceClose){
}
zLine[i+1] = 0;
if( stat(zLine,statbuf)!=0 ){
+ if( zRouterFile ){
+strcpy(zLine[1],/);
+strcpy(zLine[2],zRouterFile);
+if( stat(zLine,statbuf)==0 S_ISREG(statbuf.st_mode)
+ access(zLine,R_OK)==0 ){
+ zRealScript = StrDup(zLine[1]);
+ break;
+}
+ }
int stillSearching = 1;
while( stillSearching i0 ){
while( i0 zLine[i]!='/' ){ i--; }
@@ -1191,6 +1201,15 @@ void ProcessOneRequest(int forceClose){
strcpy(zLine[i+1],/index.cgi);
if( stat(zLine,statbuf)!=0 || !S_ISREG(statbuf.st_mode)
|| access(zLine,R_OK) ){
+ if( zRouterFile ){
+strcpy(zLine[1],/);
+strcpy(zLine[2],zRouterFile);
+if( stat(zLine,statbuf)==0 S_ISREG(statbuf.st_mode)
+ access(zLine,R_OK)==0 ){
+ zRealScript = StrDup(zLine[1]);
+ break;
+}
+ }
NotFound(__LINE__);
}
}
@@ -1655,6 +1674,8 @@ int main(int argc, char **argv){
if( atoi(zArg) ){
useTimeout = 0;
}
+}else if( strcmp(z, -router)==0 ){
+ zRouterFile = zArg;
}else{
Malfunction(__LINE__, unknown argument: [%s], z);
}
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] althttpd router patch
Attached is a patch that adds a new -router option to alhttpd. If the
requested file is not present on disk the filename passed to -router
will be ran instead. This allows you to then use REQUEST_URI to handle
URL routing within your application. By default -router will look in
docroot so there is no need to pass a forward slash.
Ex. althttpd -router router.cgi
If you navigate to http://localhost/ or http://localhost/blog/1
http://localhost/router.cgi is loaded but the URL stays the same and
REQUEST_URI will be / or /blog/1
-router is checked in 2 places once before checking for not-found.html
and once after determining index.html and index.cgi don't exist, since
the first check wasn't enough.
With this change I'm no longer dependent on Apache with mod_rewrite.
--
James Turner
ja...@calminferno.net
--- althttpd.c.orig 2012-06-21 12:31:12.0 -0400
+++ althttpd.c 2012-06-21 15:07:23.0 -0400
@@ -147,6 +147,7 @@ static int omitLog = 0; /* Do n
static int useHttps = 0; /* True to use HTTPS: instead of HTTP: */
static char *zHttp = http; /* http or https */
static int useTimeout = 1; /* True to use times */
+static char *zRouterFile = 0;/* Use this file as a router */
/*
** Change every space or unprintable character in the zAgent[] string
@@ -1153,6 +1154,15 @@ void ProcessOneRequest(int forceClose){
}
zLine[i+1] = 0;
if( stat(zLine,statbuf)!=0 ){
+ if( zRouterFile ){
+strcpy(zLine[1],/);
+strcpy(zLine[2],zRouterFile);
+if( stat(zLine,statbuf)==0 S_ISREG(statbuf.st_mode)
+ access(zLine,R_OK)==0 ){
+ zRealScript = StrDup(zLine[1]);
+ break;
+}
+ }
int stillSearching = 1;
while( stillSearching i0 ){
while( i0 zLine[i]!='/' ){ i--; }
@@ -1183,6 +1193,15 @@ void ProcessOneRequest(int forceClose){
strcpy(zLine[i+1],/index.cgi);
if( stat(zLine,statbuf)!=0 || !S_ISREG(statbuf.st_mode)
|| access(zLine,R_OK) ){
+ if( zRouterFile ){
+strcpy(zLine[1],/);
+strcpy(zLine[2],zRouterFile);
+if( stat(zLine,statbuf)==0 S_ISREG(statbuf.st_mode)
+ access(zLine,R_OK)==0 ){
+ zRealScript = StrDup(zLine[1]);
+ break;
+}
+ }
NotFound(__LINE__);
}
}
@@ -1590,6 +1609,8 @@ int main(int argc, char **argv){
if( atoi(zArg) ){
useTimeout = 0;
}
+}else if( strcmp(z, -router)==0 ){
+ zRouterFile = zArg;
}else{
Malfunction(__LINE__, unknown argument: [%s], z);
}
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] althttpd router patch
On Thu, Jun 21, 2012 at 04:25:15PM -0400, James Turner wrote: Attached is a patch that adds a new -router option to alhttpd. If the requested file is not present on disk the filename passed to -router will be ran instead. This allows you to then use REQUEST_URI to handle URL routing within your application. By default -router will look in docroot so there is no need to pass a forward slash. Ex. althttpd -router router.cgi If you navigate to http://localhost/ or http://localhost/blog/1 http://localhost/router.cgi is loaded but the URL stays the same and REQUEST_URI will be / or /blog/1 -router is checked in 2 places once before checking for not-found.html and once after determining index.html and index.cgi don't exist, since the first check wasn't enough. With this change I'm no longer dependent on Apache with mod_rewrite. Just to clarify you can already achieve something similar by having a front loading index.cgi like fossil-scm.org does, this method just makes the URLs a bit cleaner by removing the router file from the URL. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] althttpd.c apache log analyzers
On Wed, May 30, 2012 at 08:09:54PM -0400, Richard Hipp wrote:
On Wed, May 30, 2012 at 6:52 PM, Thomas Stover c...@thomasstover.com wrote:
On Wed, 30 May 2012 18:14:54 -0400
Richard Hipp wrote:
Can you send the output of diff -u please? I can't quite figure
out how to apply the diff below.
See if that works. Admittedly I never use diff.
The new diff is much better. Thanks.
But now I see that the patch is not quite right:
(1) You always use GET instead of the value in the zMethod variable
(2) You always use HTTP/1.0 instead of the value in zProtocol
(3) The date format is wrong
(4) You append two extra fields not mentioned in the wikipedia
documentation on the Common Log Format
The diff is actually for what apache calls the combined log format which
includes the referer and user-agent at the end.
That's all I see at the moment. Maybe fix those thing and send me a new
diff and we'll try again?
--- src/althttpd.c 2011-12-28 15:42:28.0 -0500
+++ althttpd.c 2012-05-30 17:49:59.0 -0400
@@ -205,6 +205,11 @@
rScale = 1.0/(double)sysconf(_SC_CLK_TCK);
chdir((zRoot zRoot[0]) ? zRoot : /);
if( (log = fopen(zLogFile,a))!=0 ){
+#ifdef COMMON_LOG_FORMAT
+ fprintf(log, %s - - [%s +] \GET %s HTTP/1.0\ %d %d \%s\
\%s\,
+ zRemoteAddr, zDate, zScript, zReplyStatus, nOut,
zReferer, zAgent); +
+#else
fprintf(log, %s %s %s://%s%s %s %s %d %d %g %g %g %g %d %d %s
%s\n, zDate, zRemoteAddr, zHttp, zHttpHost, zScript, zReferer,
zReplyStatus, nIn, nOut,
@@ -213,8 +218,8 @@
rScale*sTms.tms_cutime,
rScale*sTms.tms_cstime,
(int)(now - beginTime),
- nRequest, zAgent, zRM
- );
+ nRequest, zAgent, zRM);
+#endif
fclose(log);
nIn = nOut = 0;
}
--
www.thomasstover.com
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
--
D. Richard Hipp
d...@sqlite.org
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
--
James Turner
ja...@calminferno.net
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil push doesn't show up on the server
On Mon, May 07, 2012 at 06:23:53PM +0200, Pascal J. Bourguignon wrote: What's wrong? What should I do to push my commits? Your answer is in the output of your failed fossil push: Error: not authorized to write -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] How to set use gmt from command line?
On Wed, Apr 18, 2012 at 05:04:15PM -0400, Richard Hipp wrote:
On Wed, Apr 18, 2012 at 4:56 PM, Matt Welland estifo...@gmail.com wrote:
How can I set a project to use GMT for timeline times from the command
line? We have around 120 fossils and opening the ui to each by hand to
correct this setting will be a little tedious
sqlite3 repo.fossil REPLACE INTO config(name,value)
VALUES('timeline-utc','1')
I could be wrong here, but isn't UTC/GMT the default?
--
James Turner
ja...@calminferno.net
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] turning off gpg
On Sat, Apr 14, 2012 at 05:28:41PM +0200, Marek wrote:
Yes it works, thanks!
It seems that fossil help branch doesn't list -nosign option. Also, when
clearsign is globally off shouldn't nosign be automatically turned on when
creating new branches?
just a thought,
Marek
The following patch make branch check the clearsign setting.
--
James Turner
ja...@calminferno.net
Index: src/branch.c
==
--- src/branch.c
+++ src/branch.c
@@ -52,11 +52,12 @@
if( g.argc5 ){
usage(new BRANCH-NAME CHECK-IN ?-bgcolor COLOR?);
}
db_find_and_open_repository(0, 0);
noSign = db_get_int(omitsign, 0)|noSign;
-
+ if( db_get_boolean(clearsign, 0)==0 ){ noSign = 1; }
+
/* fossil branch new name */
zBranch = g.argv[3];
if( zBranch==0 || zBranch[0]==0 ){
fossil_panic(branch name cannot be empty);
}
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] turning off gpg
On Fri, Apr 13, 2012 at 12:09:22AM +0200, Marek wrote: Hi, I'd like to ask whether it's possible to turn off gpg signing entirely, for example I've set clearsign to off and running: fossil settings clearsign gives me: clearsign (global) off Running commit however tells me that gpg command has not been found and asks me whether I want to proceed anyway. Is there something else I'm missing? thanks, Marek . What version of fossil are you running? clearsign has been off by default for quite awhile now. If you set clearsign off --global after you already created or cloned a repository where clearsign was enabled you need to go and set it off for those repositories as well or remove the setting so the global option gets used. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] turning off gpg
On Sat, Apr 14, 2012 at 01:19:11AM +0200, Marek wrote: I'm sorry should have been before running fossil init so fossil is set to clearsign off before I init a new repository but it still keeps looking for gpg. Am I doing something wrong or could this be a bug? thanks, Marek . You shouldn't need to set clearsign to off with fossil 1.22 it's off by default. I'm not a fossil developer so I really don't know what could be going on here. But if you start with a fresh fossil init and then fossil open then fossil add and finally fossil commit you should never see anything about gpg. If you have cloned another repo clearsign settings could have been pulled in via .fossil-settings or other means that may be taking precedence over the global setting? Those are my only thoughts, maybe someone else has other clues. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] An idea to make command line life easier
On Tue, Jan 31, 2012 at 06:45:11PM +, Jacek Ca??a wrote: Hi all, I've just got an idea about how to improve the use of fossil command line a bit. The main thing is to add an additional index column when running 'fossil changes', so instead of: fossil cha EDITEDsomefile.x EDITEDdir\anotherone.xx EDITEDdir\somefile.x ... it would produce: fossil cha 1 EDITED somefile.x 2 EDITED dir\anotherone.xx 3 EDITED dir\somefile.x Sounds like a perl script I recently came across for git. https://github.com/holygeek/git-number -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Add LDFLAGS support to ./configure
Minor tweak to get LDFLAGS to work with ./configure. Previously you
would need to set CFLAGS to get ./configure to find sqlite in
/usr/local/lib. This now works:
LDFLAGS=-L/usr/local/lib ./configure --disable-internal-sqlite
--
James Turner
ja...@calminferno.net
Index: autosetup/cc.tcl
==
--- autosetup/cc.tcl
+++ autosetup/cc.tcl
@@ -454,11 +454,11 @@
switch -exact -- $opts(-lang) {
c++ {
lappend cmdline {*}[get-define CXX] {*}[get-define
CXXFLAGS]
}
c {
- lappend cmdline {*}[get-define CC] {*}[get-define
CFLAGS]
+ lappend cmdline {*}[get-define CC] {*}[get-define
CFLAGS] {*}[get-define LDFLAGS]
}
default {
autosetup-error cctest called with unknown language:
$opts(-lang)
}
}
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil on Earthlink Web Hosting?
On Tue, Jan 03, 2012 at 03:34:51PM -0800, Ross Berteig wrote: I have a request to share a repository with one of my customers, largely to provide them access to the ticket system for their product development (and future maintenance) efforts. We use Earthlink now as a web and email host so an ideal solution would seem to be to install fossil there. Hosting a fossil repository at Earthlink would have a zero marginal cost to our project if it is possible to do at all. An alternative would be to poke a pinhole in our firewall and host it on a system in our office, but since we expose no public services through that firewall now, there is some nervousness about that approach. Are there any fossil users out there with success stories or cautionary tales they'd like to share? Especially with respect to Earthlink? Or with self-hosting on relatively low-bandwidth DSL? Ross Berteig r...@cheshireeng.com Cheshire Engineering Corp. http://www.CheshireEng.com/ If you don't mind hosting your project with a third party, I run a free hosting website at [0]. Repositories are served by a chrooted apache server running on openbsd 5.0. I also do nightly backups to tarsnap [1]. [0] http://chiselapp.com/ [1] http://tarsnap.com/ -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Using LDFLAGS with configure
On Wed, Dec 28, 2011 at 09:34:12PM +1100, Steve Bennett wrote: Put the LDFLAGS= after ./configure WorkWare Systems P:0434 921 300 www.workware.net.au F:07 3102 9221 I have tried this as well. LDFLAGS is still not used when testing for sqlite during the ./configure stage. If I set the LDFLAGS and don't pass --disable-internal-sqlite then LDFLAGS is used during the make process only. My problem is getting the sqlite test to use LDFLAGS. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Using LDFLAGS with configure
On Thu, Dec 29, 2011 at 08:17:12AM +1100, Steve Bennett wrote: Try CFLAGS instead WorkWare Systems P:0434 921 300 www.workware.net.au F:07 3102 9221 CFLAGS did the trick, thanks for your help. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Using LDFLAGS with configure
How do I specify additional lib directories during the configure stage? Using: LDFLAGS=-L/usr/local/lib ./configure --disable-internal-sqlite fails with Error: system sqlite3 not found. It doesn't look like LDFLAGS is being used when testing for sqlite. Thanks. -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Error: wrong project
Just a quick reminder, please submit bug reports for chisel at [0]. I don't always watch this list. Thanks. [0] https://chiselapp.com/user/james/repository/chisel On Tue, Dec 20, 2011 at 11:05:26AM +0100, bobef...@free.fr wrote: Hello, I have a fossil repository that I would like to put on chiselapp.com. I cannot upload it because it is larger than 8M. Chiselapp website suggests to create a new project then push to it: Limit 8M in size. If your repository is larger than this, create a new empty project and push to it instead. When I try to push from my current local repository to new on chiselapp, I get this error: Error: wrong project Is there a way to push/pull/sync between unrelated repositories? Thanks. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- James Turner ja...@calminferno.net ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] http_socket.c warning patch
Attached is a diff that fixes a (probably) harmless compiler warning on OpenBSD. -- James Turner ja...@calminferno.net --- src/http_socket.c.orig Tue Dec 13 09:10:57 2011 +++ src/http_socket.c Fri Dec 16 09:36:11 2011 @@ -35,10 +35,10 @@ #include ws2tcpip.h # endif #else +# include netinet/in.h # include arpa/inet.h # include sys/socket.h # include netdb.h -# include netinet/in.h #endif #include assert.h #include sys/types.h ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Presentation slides for Fossil?
On Mon, Sep 12, 2011 at 09:48:01AM -0400, Martin S. Weber wrote: On 09/10/11 15:54, Ron Wilson wrote: On Thu, Sep 8, 2011 at 4:08 PM, Ingo Kochingo.k...@gmx.de wrote: Take a look at http://arnebachmann.pytalhost.de/pdf/integrated-version-control-with-fossil-scm.pdf Good presentation. I noted it said that configuration is not shared. Actually, it is possible to share configuration between Fossil instances. It wrongly claims that fossil is GPL. -Martin Fossil was originally licensed under the GPL. When that presentation was given in 2009 sharing configurations was not possible and the codebase was GPL. -- James Turner ja...@calminferno.net signature.asc Description: Digital signature ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] cannot push or commit
On Fri, Sep 02, 2011 at 01:44:51PM -0400, ron georgia wrote: At a complete loss. This is what i get every time, no matter what i do. Perhaps there is some fundamental thing i am not understanding. rgeorgia@rgeorgia-c1:~/workspace/FOSSIL/ntracker$ fossil push Server:http://carepo01.us.alcatel-lucent.com/cgi-bin/ntracker/ via proxy: http://proxy.gslb.alcatel-lucent.com:8000 Bytes Cards Artifacts Deltas Sent: 866 17 0 0 Error: not authorized to write Received: 65 1 0 0 Total network traffic: 713 bytes sent, 0 bytes received Ron Georgia John 13:23 You seem to be pushing as an anonymous user who probably doesn't have write permission. You may want to try pushing as a user you've already setup. fossil push http://someusern...@carepo01.us.alcatel-lucent.com/cgi-bin/ntracker/ It should then prompt you for your password, afterwards you can just do fossil push and it will send the correct credentials. -- James Turner ja...@calminferno.net signature.asc Description: Digital signature ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] How should Fossil handle this merge conflict...
I would think if I merge a branch that deletes the file and I have local changes the file would still exist on disk but would be treated like a untracked file. Meaning, if I was to run fossil extras after the merge the file would show up. Does this make sense? On Dec 12, 2010, at 7:30 PM, Richard Hipp wrote: On Sun, Dec 12, 2010 at 7:24 PM, Joerg Sonnenberger jo...@britannica.bec.de wrote: On Sun, Dec 12, 2010 at 06:57:56PM -0500, Richard Hipp wrote: Suppose you have the trunk branch checked out and you have made changes to file xyz.txt locally, but have not checked them in. Then you do a merge of branch other-branch: fossil merge other-branch The file xyz.txt has been deleted in other-branch. What should fossil's response be? Should it retain the locally edited copy of xyz.txt, or should it delete file xyz.txt? Consider it a merge conflict. You mean refuse to do the merge? Certainly a conflict warning will be issued, but it seems like a flat-out refusal to do the merge is the wrong approach. In addition to the conflict warning, Fossil needs to do something with the file. As I ponder this question further, I'm leaning toward making a DELETE action override any EDIT or RENAME action. So if a file has been deleted in the local version, then a merge from a version where the file still exists keeps the file deleted. And if a file still exists in the local version, then a merge from a version where that file is deleted cause the file to become deleted locally. Any objections to this approach? Joerg ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] How should Fossil handle this merge conflict...
On Dec 12, 2010, at 8:15 PM, Richard Hipp wrote: On Sun, Dec 12, 2010 at 8:04 PM, James Turner ja...@calminferno.net wrote: I would think if I merge a branch that deletes the file and I have local changes the file would still exist on disk but would be treated like a untracked file. Meaning, if I was to run fossil extras after the merge the file would show up. Does this make sense? It makes sense in the sense that I understand what you are saying. But I don't understand the why. Please explain why you want the file to become untracked? To me, it seems like this would open you up to errors, since having untracked files laying around might cause your local tests to work, but then fail when somebody else checked out the same version that lacks those same files. FWIW, Fossil used to behave as you describe (it left untracked files laying around), but I changed it yesterday to actually delete the file from disk. I thought that change was a bug fix. Am I wrong? (Note: you can still recover the file using fossil undo) See http://www.fossil-scm.org/fossil/info/70744b76e6 For some reason I thought git handled merges this way, I guess I was seeing the behavior in previous fossil versions. It looks like git treats it as a merge conflict (not that this means fossil has to do the same) like Joerg has suggested. On Dec 12, 2010, at 7:30 PM, Richard Hipp wrote: On Sun, Dec 12, 2010 at 7:24 PM, Joerg Sonnenberger jo...@britannica.bec.de wrote: On Sun, Dec 12, 2010 at 06:57:56PM -0500, Richard Hipp wrote: Suppose you have the trunk branch checked out and you have made changes to file xyz.txt locally, but have not checked them in. Then you do a merge of branch other-branch: fossil merge other-branch The file xyz.txt has been deleted in other-branch. What should fossil's response be? Should it retain the locally edited copy of xyz.txt, or should it delete file xyz.txt? Consider it a merge conflict. You mean refuse to do the merge? Certainly a conflict warning will be issued, but it seems like a flat-out refusal to do the merge is the wrong approach. In addition to the conflict warning, Fossil needs to do something with the file. As I ponder this question further, I'm leaning toward making a DELETE action override any EDIT or RENAME action. So if a file has been deleted in the local version, then a merge from a version where the file still exists keeps the file deleted. And if a file still exists in the local version, then a merge from a version where that file is deleted cause the file to become deleted locally. Any objections to this approach? Joerg ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Settings patch
Attached is a patch that will check the correct checkboxes on the
settings screen based on the defaults in ctrlSettings now that they use
on and off.
--- src/setup.c
+++ src/setup.c
@@ -877,11 +877,11 @@
login_insert_csrf_secret();
for(pSet=ctrlSettings; pSet-name!=0; pSet++){
if( pSet-width==0 ){
onoff_attribute(pSet-name, pSet-name,
pSet-var!=0 ? pSet-var : pSet-name,
- pSet-def[0]=='1');
+ strcmp(pSet-def,on)==0);
@ br /
}
}
@ /tdtd style=width: 30;/tdtd valign=top
for(pSet=ctrlSettings; pSet-name!=0; pSet++){
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Settings patch
Also just noticed, mtime-changes is set to off by default but the help
menu says it's on by default.
On Thu, Nov 04, 2010 at 03:00:32PM -0400, James Turner wrote:
Attached is a patch that will check the correct checkboxes on the
settings screen based on the defaults in ctrlSettings now that they use
on and off.
--- src/setup.c
+++ src/setup.c
@@ -877,11 +877,11 @@
login_insert_csrf_secret();
for(pSet=ctrlSettings; pSet-name!=0; pSet++){
if( pSet-width==0 ){
onoff_attribute(pSet-name, pSet-name,
pSet-var!=0 ? pSet-var : pSet-name,
- pSet-def[0]=='1');
+ strcmp(pSet-def,on)==0);
@ br /
}
}
@ /tdtd style=width: 30;/tdtd valign=top
for(pSet=ctrlSettings; pSet-name!=0; pSet++){
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Question about global settings
So looking through the source code, I can see why the autosync global doesn't work in my case. It would seem on repository creation, fossil checks to see if the global autosync setting is set, if it is, it sets the local config to whatever the global was. When a db_get is ran it checks the local config and only if it doesn't find a match does it check the global. Since I changed the global autosync setting after the fact, it doesn't get honored. So my question is, should the global config get checked before the local config? Or should local always win? I would think if after the fact I decided I want to turn autosync off globally I should be able to with out having to change autosync in each individual repository. Feel free to let me know if I've missed some obvious behavior. Thanks. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] New release of Fossil
On Mon, Nov 01, 2010 at 01:47:11PM -0400, Richard Hipp wrote: A new release build of Fossil is available. Precompiled binaries are available at: http://www.fossil-scm.org/download.html Release notes are available here: http://www.fossil-scm.org/fossil/event?name=d6ba73e22 You must run fossil all rebuild after updating to this release. Please send problem reports to this mailing list or directly to me. Tnx. -- D. Richard Hipp d...@sqlite.org On http://chiselapp.com we are running version f89d953a06 via cgi. If you go into Admin - Configuration on a newly created repository and change the Index Page to anything where the index-page gets set in the config table, when you attempt to access the index page you get stuck in a redirect loop. Even changing the Index Page back to /home the redirect loop stays until you delete the index-page record from the table. Running fossil using it's internal web server I don't see this problem. Is this a cgi only bug? ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] New release of Fossil
On Mon, Nov 01, 2010 at 03:19:46PM -0400, Richard Hipp wrote: On Mon, Nov 1, 2010 at 2:56 PM, James Turner ja...@calminferno.net wrote: On http://chiselapp.com we are running version f89d953a06 via cgi. If you go into Admin - Configuration on a newly created repository and change the Index Page to anything where the index-page gets set in the config table, when you attempt to access the index page you get stuck in a redirect loop. Even changing the Index Page back to /home the redirect loop stays until you delete the index-page record from the table. Running fossil using it's internal web server I don't see this problem. Is this a cgi only bug? The http://www.fossil-scm.org/ website is an instance of Fossil running in cgi mode. And Admin-Configuration-Index_Page is set /doc/tip/www/index.wiki. And it works fine. So something else must be going on Can you get me a dump of the HTTP traffic to chiselapp.com when the redirect loop is spinning? That might give a clue. -- D. Richard Hipp d...@sqlite.org I will try to get a dump of the HTTP traffic, just to add, if I set it to /doc/tip/index.wiki I get the below error instead of the redirect loop, but setting it back to /home the loop comes back. No such document: tip/index.wiki ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] New release of Fossil
On Mon, Nov 01, 2010 at 03:27:13PM -0400, James Turner wrote: On Mon, Nov 01, 2010 at 03:19:46PM -0400, Richard Hipp wrote: On Mon, Nov 1, 2010 at 2:56 PM, James Turner ja...@calminferno.net wrote: On http://chiselapp.com we are running version f89d953a06 via cgi. If you go into Admin - Configuration on a newly created repository and change the Index Page to anything where the index-page gets set in the config table, when you attempt to access the index page you get stuck in a redirect loop. Even changing the Index Page back to /home the redirect loop stays until you delete the index-page record from the table. Running fossil using it's internal web server I don't see this problem. Is this a cgi only bug? The http://www.fossil-scm.org/ website is an instance of Fossil running in cgi mode. And Admin-Configuration-Index_Page is set /doc/tip/www/index.wiki. And it works fine. So something else must be going on Can you get me a dump of the HTTP traffic to chiselapp.com when the redirect loop is spinning? That might give a clue. -- D. Richard Hipp d...@sqlite.org I will try to get a dump of the HTTP traffic, just to add, if I set it to /doc/tip/index.wiki I get the below error instead of the redirect loop, but setting it back to /home the loop comes back. No such document: tip/index.wiki Attached is a dump from the live headers firefox extension. As you can see when index-page is set it just keep redirecting to the same page. Is this the kind of output you were looking for? http://chiselapp.com/user/james/repository/test GET /user/james/repository/test HTTP/1.1 Host: chiselapp.com User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 115 Connection: keep-alive HTTP/1.1 302 Moved Temporarily Date: Mon, 01 Nov 2010 20:18:11 GMT Server: Apache Cache-Control: no-cache, no-store Location: http://chiselapp.com/user/james/repository/test/home Content-Length: 87 Keep-Alive: timeout=15, max=100 Connection: Keep-Alive Content-Type: text/html; charset=utf-8 -- http://chiselapp.com/user/james/repository/test/home GET /user/james/repository/test/home HTTP/1.1 Host: chiselapp.com User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 115 Connection: keep-alive HTTP/1.1 302 Moved Temporarily Date: Mon, 01 Nov 2010 20:18:11 GMT Server: Apache Cache-Control: no-cache, no-store Location: http://chiselapp.com/user/james/repository/test/home Content-Length: 87 Keep-Alive: timeout=15, max=99 Connection: Keep-Alive Content-Type: text/html; charset=utf-8 -- http://chiselapp.com/user/james/repository/test/home GET /user/james/repository/test/home HTTP/1.1 Host: chiselapp.com User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 115 Connection: keep-alive HTTP/1.1 302 Moved Temporarily Date: Mon, 01 Nov 2010 20:18:11 GMT Server: Apache Cache-Control: no-cache, no-store Location: http://chiselapp.com/user/james/repository/test/home Content-Length: 87 Keep-Alive: timeout=15, max=98 Connection: Keep-Alive Content-Type: text/html; charset=utf-8 -- http://chiselapp.com/user/james/repository/test/home GET /user/james/repository/test/home HTTP/1.1 Host: chiselapp.com User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101026 Firefox/3.6.12 Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8 Accept-Language: en-us,en;q=0.5 Accept-Encoding: gzip,deflate Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7 Keep-Alive: 115 Connection: keep-alive HTTP/1.1 302 Moved Temporarily Date: Mon, 01 Nov 2010 20:18:11 GMT Server: Apache Cache-Control: no-cache, no-store Location: http://chiselapp.com/user/james/repository/test/home Content-Length: 87 Keep-Alive: timeout=15, max=97 Connection: Keep-Alive Content-Type: text/html; charset=utf-8 -- http://chiselapp.com/user/james/repository/test/home GET /user/james/repository/test/home HTTP/1.1 Host: chiselapp.com User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.12) Gecko/20101026
Re: [fossil-users] New release of Fossil
On Mon, Nov 01, 2010 at 05:08:16PM -0400, James Turner wrote:
On Mon, Nov 01, 2010 at 03:27:13PM -0400, James Turner wrote:
On Mon, Nov 01, 2010 at 03:19:46PM -0400, Richard Hipp wrote:
On Mon, Nov 1, 2010 at 2:56 PM, James Turner ja...@calminferno.net
wrote:
On http://chiselapp.com we are running version f89d953a06 via cgi. If
you go into Admin - Configuration on a newly created repository and
change the Index Page to anything where the index-page gets
set in the config table, when you attempt to access the
index page you get stuck in a redirect loop. Even changing the Index
Page back to /home the redirect loop stays until you delete
the index-page record from the table. Running fossil using it's
internal web server I don't see this problem. Is this a cgi only bug?
The http://www.fossil-scm.org/ website is an instance of Fossil running in
cgi mode. And Admin-Configuration-Index_Page is set
/doc/tip/www/index.wiki. And it works fine.
So something else must be going on
Can you get me a dump of the HTTP traffic to chiselapp.com when the
redirect
loop is spinning? That might give a clue.
--
D. Richard Hipp
d...@sqlite.org
I will try to get a dump of the HTTP traffic, just to add, if I set it
to /doc/tip/index.wiki I get the below error instead of the redirect
loop, but setting it back to /home the loop comes back.
No such document: tip/index.wiki
So I think I found the issue, just don't now how to fix it. In the
home_page function in wiki.c on line 91, the zIndexPage does not get set
to 0 since PATH_INFO is not returning the correct item. I assume it
should return /home after the first redirect, but it's not which causes
line 93 to pass and put it in the redirect loop. I can't tell you what
PATH_INFO actually returns right now, but I should note I'm running this
behind apache.
If this doesn't spark anything, I'll get you the actual PATH_INFO value
when I get home this evening. Thanks.
Fix attached. It seems PATH_INFO doesn't contain the leading / hence the
compare fails every time. I moved up a line of code that removed the
leading / from the zIndexPage. Tested with index-page set and without.
Index: src/wiki.c
===
--- src/wiki.c
+++ src/wiki.c
@@ -86,14 +86,14 @@
if( !g.okRdWiki ){
cgi_redirectf(%s/login?g=%s/home, g.zBaseURL, g.zBaseURL);
}
if( zIndexPage ){
const char *zPathInfo = P(PATH_INFO);
+while( zIndexPage[0]=='/' ) zIndexPage++;
if( strcmp(zIndexPage, zPathInfo)==0 ) zIndexPage = 0;
}
if( zIndexPage ){
-while( zIndexPage[0]=='/' ) zIndexPage++;
cgi_redirectf(%s/%s, g.zBaseURL, zIndexPage);
}
if( zPageName ){
login_check_credentials();
g.zExtra = zPageName;
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] fossil build issues
I'm running fossil version 9ea47d4953 built from the experimental branch on OpenBSD. Using this version I just did a fresh clone of fossil-scm.org. Ran fossil open, then fossil checkout experimental. I then tried to run gmake and get the below error: gmake: *** No rule to make target `src/../manifest.uuid', needed by `VERSION.h'. Stop. Since the uuid are no longer present, I ran fossil setting manifest on within the opened fossil repo. However, they are still not present and fossil will not build. Is there a way to force fossil to generate the uuid and manifest file? Thanks. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil build issues
On Sat, Oct 23, 2010 at 03:31:17PM -0400, Richard Hipp wrote: On Sat, Oct 23, 2010 at 2:40 PM, James Turner ja...@calminferno.net wrote: I'm running fossil version 9ea47d4953 built from the experimental branch on OpenBSD. Using this version I just did a fresh clone of fossil-scm.org. Ran fossil open, then fossil checkout experimental. I then tried to run gmake and get the below error: gmake: *** No rule to make target `src/../manifest.uuid', needed by `VERSION.h'. Stop. Since the uuid are no longer present, I ran fossil setting manifest on within the opened fossil repo. However, they are still not present and fossil will not build. Is there a way to force fossil to generate the uuid and manifest file? Thanks. Please try running fossil update trunk; fossil update experimental After you have run fossil setting manifest on. That might clear it. I need to fix it so that the manifest setting is download automatically when the repository is cloned -- D. Richard Hipp d...@sqlite.org Thank you, that worked. After updating to trunk the two files were present and switching back to experimental I was able to build again. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Fossil enhancements: Please test
On Tue, Oct 19, 2010 at 04:59:11PM -0400, Richard Hipp wrote: [snip] Thanks for testing out the recent Fossil changes. Please be sure to report either success or failure (whichever you encounter) with the new Fossil to this mailing list, or directly to me at d...@sqlite.org. -- D. Richard Hipp d...@sqlite.org Everything seems to work fine under OpenBSD. Both delta and baseline seem to do as advertised. I tested with the chiselapp.com repository, which is tiny. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Updated: Enhancement: ssh:// sync method
Nice work on adding ssh support, I always prefer it when cloning over my local network. I am able to clone a current repository just fine from my OpenBSD server using the following command: fossil clone ssh://ja...@10.0.1.2/fossil/chisel.fossil chisel.fossil I see Closing SSH tunnel: Killed by signal 2. in the output, but the clone seems to work so I figure this is normal. However, when I run fossil pull from within an opened repository it asks me for a password even though I have public keys setup and it didn't ask me when I did the initial clone and the server seems to remove the slash between the host and path. ssh://ja...@10.0.1.2fossil/clean.fossil Although I don't get any other error besides the Closing SSH tunnel: Killed by signal 2. I have a feeling the pull failed? Also if I create a new repository on the server and then try to clone it on my client which is Mac OS X it just hangs. Both are running the latest fossil. Seems strange an old fossil repository would clone just fine but a fresh one doesn't. On Aug 27, 2010, at 5:02 PM, Richard Hipp wrote: The ssh:// code in the Fossil self-hosting repository is now reasonably complete, I believe. It now works correctly on unix and windows. Please try it out. Once I get 4 or 5 works for me! replies, I'll do a new release build. Here is how I use ssh:// under windows. (1) Get plink.exe. Rename it to ssh.exe (2) Run ssh usern...@host.org at least once to get past the remote host accept key message (3) fossil clone ssh://username:*...@host.org:22/path/to/repo.fossil clone.fossil Notice the password is specified in the URL as *. Whenever Fossil sees a password that begins with * it will assume that password is a placeholder and prompt for the real password on each connection attempt. You can specify the complete password in the URL, if you want, but that would mean that your password appears in plain sight on your screen. If you omit the password, then Fossil assumes that you are using public-key authentication using pageant or the equivalent. If you do not have a public key set up, be sure to puts * as your password in the URL or else Fossil will not know to prompt for the real password. The previous two paragraphs apply for windows. On unix, any password you supply in the URL is ignored. Unix will prompt if and only if a password is needed, whether you supply a password or not. If you are running your ssh server on a non-standard port, then you can specify an alternative TCP port where the :22 is shown in the example above. 22 is the default can can be omitted. If the name of the Fossil command on the remote is not fossil, then add a query parameter to specify the actual name. For example: ssh://host.org/path/to/repo.fossil?fossil=/home/me/bin/fossil-ex If your local SSH command is not named ssh, then specify a new one using fossil setting ssh-command 'NEW-COMMAND'. The default command on unix is ssh -e none -T. The default command on windows is ssh -T. I will entertain arguments for and against changing the default windows command to plink -T. Known issues: (1) When things go wrong, the error output is generally not very helpful. I'm not sure what can be done about this. (2) Documentation is missing. (3) Comments in the code to explain how this is all implemented could be much better. Thanks for all the suggestions. Please continue to provide feedback! Note I have tested ssh:// for windows on the client side only (windows client talking to a unix server.) Can anyone suggest a reasonable sshd implementation for windows so that I can test out windows acting as an ssh server? On Wed, Aug 25, 2010 at 4:25 PM, Richard Hipp d...@sqlite.org wrote: The latest version of Fossil (in the self-hosting Fossil repository - not the precompiled binaries which are a little too old) supports a new method of pushing, pulling, cloning, and syncing using SSH. Examples: fossil clone ssh://usern...@hostname.com/local/path/repo.fossil ex1.fossil fossil clone ssh://usern...@hostname.com//full/path/name/repo.fossil ex2.fossil Notice that with a single / between the hostname and the beginning of the repository path, the repository path is relative to the home directory of the user. With two // characters, the pathname to the repository is an absolute pathname. This new feature currently only works on unix. As part of the implementation, I needed a bidirectional popen() function. (The standard library popen() only works in one direction.) I implemented this for unix in the popen.c source file. But I do not know how to do the same on windows. If someone cares to contribute ideas on how to implement a bidirection popen() for windows, that will help me get the new ssh:// functionality working on windows. On the other hand, no many windows machines that I have seen support ssh. So
Re: [fossil-users] Updated: Enhancement: ssh:// sync method
The second url should be ssh://ja...@10.0.1.2fossil/chisel.fossil, not clean.fossil. The missing slash still stands. On Aug 27, 2010, at 10:51 PM, James Turner wrote: Nice work on adding ssh support, I always prefer it when cloning over my local network. I am able to clone a current repository just fine from my OpenBSD server using the following command: fossil clone ssh://ja...@10.0.1.2/fossil/chisel.fossil chisel.fossil I see Closing SSH tunnel: Killed by signal 2. in the output, but the clone seems to work so I figure this is normal. However, when I run fossil pull from within an opened repository it asks me for a password even though I have public keys setup and it didn't ask me when I did the initial clone and the server seems to remove the slash between the host and path. ssh://ja...@10.0.1.2fossil/clean.fossil Although I don't get any other error besides the Closing SSH tunnel: Killed by signal 2. I have a feeling the pull failed? Also if I create a new repository on the server and then try to clone it on my client which is Mac OS X it just hangs. Both are running the latest fossil. Seems strange an old fossil repository would clone just fine but a fresh one doesn't. On Aug 27, 2010, at 5:02 PM, Richard Hipp wrote: The ssh:// code in the Fossil self-hosting repository is now reasonably complete, I believe. It now works correctly on unix and windows. Please try it out. Once I get 4 or 5 works for me! replies, I'll do a new release build. Here is how I use ssh:// under windows. (1) Get plink.exe. Rename it to ssh.exe (2) Run ssh usern...@host.org at least once to get past the remote host accept key message (3) fossil clone ssh://username:*...@host.org:22/path/to/repo.fossil clone.fossil Notice the password is specified in the URL as *. Whenever Fossil sees a password that begins with * it will assume that password is a placeholder and prompt for the real password on each connection attempt. You can specify the complete password in the URL, if you want, but that would mean that your password appears in plain sight on your screen. If you omit the password, then Fossil assumes that you are using public-key authentication using pageant or the equivalent. If you do not have a public key set up, be sure to puts * as your password in the URL or else Fossil will not know to prompt for the real password. The previous two paragraphs apply for windows. On unix, any password you supply in the URL is ignored. Unix will prompt if and only if a password is needed, whether you supply a password or not. If you are running your ssh server on a non-standard port, then you can specify an alternative TCP port where the :22 is shown in the example above. 22 is the default can can be omitted. If the name of the Fossil command on the remote is not fossil, then add a query parameter to specify the actual name. For example: ssh://host.org/path/to/repo.fossil?fossil=/home/me/bin/fossil-ex If your local SSH command is not named ssh, then specify a new one using fossil setting ssh-command 'NEW-COMMAND'. The default command on unix is ssh -e none -T. The default command on windows is ssh -T. I will entertain arguments for and against changing the default windows command to plink -T. Known issues: (1) When things go wrong, the error output is generally not very helpful. I'm not sure what can be done about this. (2) Documentation is missing. (3) Comments in the code to explain how this is all implemented could be much better. Thanks for all the suggestions. Please continue to provide feedback! Note I have tested ssh:// for windows on the client side only (windows client talking to a unix server.) Can anyone suggest a reasonable sshd implementation for windows so that I can test out windows acting as an ssh server? On Wed, Aug 25, 2010 at 4:25 PM, Richard Hipp d...@sqlite.org wrote: The latest version of Fossil (in the self-hosting Fossil repository - not the precompiled binaries which are a little too old) supports a new method of pushing, pulling, cloning, and syncing using SSH. Examples: fossil clone ssh://usern...@hostname.com/local/path/repo.fossil ex1.fossil fossil clone ssh://usern...@hostname.com//full/path/name/repo.fossil ex2.fossil Notice that with a single / between the hostname and the beginning of the repository path, the repository path is relative to the home directory of the user. With two // characters, the pathname to the repository is an absolute pathname. This new feature currently only works on unix. As part of the implementation, I needed a bidirectional popen() function. (The standard library popen() only works in one direction.) I implemented this for unix in the popen.c source file. But I do not know how to do the same on windows. If someone cares to contribute ideas on how to implement
Re: [fossil-users] Hosted fossil solution
Alright you can now set a default password at repo creation, it will generate the correct sha1 based on fossil id + username + password. I also tweaked the creation page by splitting the two types into separate pages. On Aug 1, 2010, at 2:31 PM, Joshua Paine wrote: What about give me the option to enter a password on repo creation? Depending on the form layout, my browser will fill in the same one I've used on the site automatically. Joshua Paine LetterBlock: Web applications built with joy http://letterblock.com/ 301-576-1920 ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Hosted fossil solution
On Jul 31, 2010, at 9:10 AM, Joshua Paine wrote: On 07/31/2010 08:09 AM, Dig412 wrote: Are you planning on releasing the code? This could be really useful for development shops to run on their own servers to manage projects. If James wants to share his code of course more power to him, but it seems like it would be more useful to competing services than development shops. If you've got the server access and know-how to install James' app, then surely you can write a 2-line CGI script and drop all your .fossil files in an arbitrary directory. See Serving multiple repositories with one script on http://www.fossil-scm.org/index.html/doc/tip/www/server.wiki. (This is not to diminish James' work--it admirably removes a barrier for people who don't have appropriate server access or just want to try things out.) If someone built a repo management tool that could manage users across multiple repos and/or do ticket or timeline aggregation, that would be helpful for us development shops. -- Joshua Paine LetterBlock: Web applications built with joy http://letterblock.com/ 301-576-1920 Like Joshua said, Chisel is really just a simple way to spawn new repositories and keep them under a user structure. At this time I don't plan on releasing the source code, as you might imagine it really isn't all that impressive. However, I do plan on adding new features to chisel, like better repository management. I've been thinking about a better way to link up repository credentials, so you can change your user password and have it be reflective across all repositories, however I haven't come up with a good way to set the password on newly created repositories without storing user password in cleartext, which I really don't want to do. Right now it's a one way hash using a key to compare the hash I have and what you enter at login time. Right now I'm just having fun using fossil more and more, as you might imagine all chisel development is housed in a fossil repository! ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Hosted fossil solution
Your right, the sha1 is of the fossil repo id + username + password. So I guess here is my question for you guys. Would you rather have the random initial password and be forced to change it within the fossil ui on creation or would you rather I store your chisel password in cleartext, then created the sha1 for your fossil repo and set it so you can use the same username/password? On Jul 31, 2010, at 10:34 AM, Joshua Paine wrote: On 07/31/2010 10:11 AM, James Turner wrote: Alright, I think I have a decent enough solution. I will just mimic fossil's way of storing passwords, which I believe is either in cleartext or a sha1 of the password. The repo id (and maybe the username?--can't recall) is hashed together with the password. You should look at the fossil source for this part. -- Joshua Paine LetterBlock: Web applications built with joy http://letterblock.com/ 301-576-1920 ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Hosted fossil solution
On Jul 27, 2010, at 10:42 PM, Richard Hipp wrote: On Tue, Jul 27, 2010 at 10:15 PM, James Turner ja...@calminferno.net wrote: Alright well I finally managed to sit down over the weekend and get something up and running. You can check out the web app at: http://chiselapp.com You can currently create an account, create new repositories as well as clone existing ones. Repositories are served up like: https://chiselapp.com/user/james/repository/fossil Currently I'm limiting 5 repositories per account, everything should be considered in an alpha state, but so far everything seems to be running smoothly. Let me know if you run into any issues or have suggestions. Thanks. Very nice. Thanks for this! I added a link to Chisel from the Fossil homepage. Will the mirror of Fossil that you have on Chisel automatically sync at some point? Do you have a cron job that does that? How does that work? If you clone another repository if added an option to have auto pull's happen every hour if desired, like you said this is good for repositories that just mirror other ones. Thanks again for the suggestion. snip -- - D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Hosted fossil solution
On Wed, Jul 28, 2010 at 02:19:13AM -0400, saulgo...@flashingtwelve.brickfilms.com wrote: Quoting James Turner ja...@calminferno.net: Currently I'm limiting 5 repositories per account, everything should be considered in an alpha state, but so far everything seems to be running smoothly. Let me know if you run into any issues or have suggestions. Thanks. I have registered and am attempting to learn how to work all this (I have used Fossil on remote servers on my home network but not over the Internet). One issue encountered was that my e-mail domain was flagged as invalid. I believe this was owing to it containing more than a single period (flashingtwelve.brickfilms.com). You may want to check your validation algorithm. I will fix this validation rule asap. I screwed up and did not properly note the password assigned after creating my first repository (https://chiselapp.com/user/saulgoode/repository/script-fu/) and can not log in as administrator now. If you would be so kind as to remove it when you get the chance, it would be much appreciated. When you log into chiselapp.com you can remove repositories you created from the Dashboard. I created a second repository (https://chiselapp.com/user/saulgoode/repository/script-fu-sg/) and have been able to clone it successfully; however, when I tried to PUSH some changes back (I'd disabled autosync), I received a 302 Re-direct error. I think it may be trying to re-direct from HTTP to HTTPS but I am not sure (if I use the HTTPS URL then Fossil reports that it does not support HTTPS). It is quite possible that I am doing something wrong at this point as I am inexperienced at using some of Fossil's multi-user features. I will look into this issue today as well, funny thing, I never tested pushing to a cloned repository, did I mention this was still alpha :) Thank you for the opportunity to use your hosting. I will spend more time experimenting tomorrow but if there is anything you wish me to try, please ask. Regards. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] Hosted fossil solution
Alright well I finally managed to sit down over the weekend and get something up and running. You can check out the web app at: http://chiselapp.com You can currently create an account, create new repositories as well as clone existing ones. Repositories are served up like: https://chiselapp.com/user/james/repository/fossil Currently I'm limiting 5 repositories per account, everything should be considered in an alpha state, but so far everything seems to be running smoothly. Let me know if you run into any issues or have suggestions. Thanks. On Jul 21, 2010, at 1:08 PM, Richard Hipp wrote: On Wed, Jul 21, 2010 at 12:53 PM, James Turner ja...@calminferno.net wrote: I'm currently playing around with the idea and code for a hosted fossil solution. Since each fossil repository already has everything you would need, the solution would be more of a fossil repository management tool on the web, that would let you create/manage repositories and provide a mechanism to serve them (still working on the best solution for this part). I was just wondering if this is a service that people might find useful? As you have observed, Fossil strives to be a hosted-solution-in-a-box. Just add the host and you are ready to roll. So I'm thinking that the hosted Fossil idea is not nearly as useful as GitHub, since the distance between raw Fossil and your hosted solution is far less than the distance from raw git to GitHub. That said, even Fossil requires a host. So if you don't already have a host sitting around (as many people don't) I think such a service would be quite useful. Please keep us posted of your progress! I'm still hashing out my ideas and currently have very basic code up and running (creating/deleting/account mgmt). Since everything is distributed within fossil there isn't any kind of lockin, so it's more of a way to remove the small overhead of setting up a fossil repository somewhere. Feel free to let me know what you all think, or maybe point me to a solution that already exists? Thanks. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users -- - D. Richard Hipp d...@sqlite.org ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Logo url fix
Diff added $baseurl to the front of /logo. src_style.diff Description: Binary data ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] 64-bit warnings
I've been working on a fossil port for OpenBSD and noticed when compiling under OpenBSD amd64 I see a couple 'warning: cast from pointer to integer of different size' warnings attached are three patches that I was including in my port and was wondering if this is something worth changing upstream? Since there is no configure scripts provided I'm not sure how you feel about requiring stdint.h but I think it's a pretty standard library. Thanks. patch-src_main_mk Description: Binary data patch-src_makeheaders_c Description: Binary data patch-src_th_lang_c Description: Binary data ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] 64-bit warnings
I too haven't seen any problems running fossil on 64-bit systems without my patches. One of the OpenBSD port developers asked me if I could look into fixing the warnings and suggested I check upstream. Thanks for your feedback. On Sep 23, 2009, at 11:22 AM, D. Richard Hipp wrote: On Sep 23, 2009, at 9:11 AM, James Turner wrote: I've been working on a fossil port for OpenBSD and noticed when compiling under OpenBSD amd64 I see a couple 'warning: cast from pointer to integer of different size' warnings attached are three patches that I was including in my port and was wondering if this is something worth changing upstream? Since there is no configure scripts provided I'm not sure how you feel about requiring stdint.h but I think it's a pretty standard library. Thanks. I've seen the warnings too. But I use fossil daily on 64-bit linux (SuSE 10.1) and it works fine there. So I'm not overly worried about them. Most of the warnings come from sqlite3.c. And we are very sure that it works correctly on 64-bit systems. I'll work on reducing the number of warnings. But it isn't a priority. patch-src_main_mkpatch-src_makeheaders_cpatch- src_th_lang_c___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil- users D. Richard Hipp d...@hwaci.com ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
Re: [fossil-users] fossil + nostromo
I've adding the output of test_env to the default page, it looks like when I'm running nostromo in chroot the starting / gets stripped from the PATH_INFO environment setting. no chroot: PATH_INFO = /home chroot: PATH_INFO = home Now to just figure out why this is happening. On Sep 19, 2009, at 7:27 PM, James Turner wrote: I'm trying to setup a fossil cgi script under nostromo [0] on OpenBSD. When I run nhttpd in non-chroot mode it displays the repository as one would think. However when I run it in chroot mode (nhttpd -r) it only renders the homepage and all other pages render the homepage as well. You can see it here [1]. However when ran in chroot apache it displays everything correctly. Both chroots contain the fossil binary and libc and libz. I know this mostly likely isn't a fossil issue but rather an issue with nhttpd chroot, but I was wonder if anyone could point me in a direction or at least give a clue to explain the behavior I am seeing. Thanks. [0] http://dns.nazgul.ch/dev_nostromo.html [1] http://bsdgroup.org:8080/cgi-bin/sre/ ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
