[fossil-users] Basic FossilHub functionality works
At this point, the repository also serves as a demo (that will probably change). You can see a basic timeline of the repositories the account is following (currently only one, but I'll add some more) if you click on the Demo Account link at the top of the home page. Here's the repository link: http://tinyconcepts.com/fs.pl/hub.fsl This allows you to make a Fossil repository that functions as your account and register repositories with it that you are watching (I don't have a GUI yet to register repositories -- the data is in a JSON wiki page attachment). The followed page shows aggregated timelines from all of the repositories you are following, currently sorted by time, most recent first (interleaving the entries from all of the watched repositories). This can be expanded to support more social networking functionality. I plan to add management, so you can easily add and remove followed repositories, along with some optional mods you can make to project repositories so they can display which accounts are following them, etc. In order to watch a repository, that repository has to a small (1K) HTML asset which provides access to the timeline feed (using HTML5's window.postMessage) -- this can just be attached to a wiki page; it doesn't have to be in a project branch. The main body of code, however, is in the account repository, in JavaScript and served up using a Fossil doc URL. Again, this did not require any changes to the Fossil executable. Bill ___ 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] Basic FossilHub functionality works
On 03/28/11 07:43, Bill Burdick wrote: In order to watch a repository, that repository has to a small (1K) HTML asset which provides access to the timeline feed (using HTML5's window.postMessage) -- this can just be attached to a wiki page; it doesn't have to be in a project branch. The main body of code, however, is in the account repository, in JavaScript and served up using a Fossil doc URL. Again, this did not require any changes to the Fossil executable. It's awesome that you can do this without modifying Fossil. I dunno if the social networking stuff will take off or not... but the fact it can be implemented like this is amazing! Well done to you and to Richard for making Fossil flexible through simplicity ;-) [What a geeky answer. Dunno about the social stuff, what an awesome gadget ;-)] Bill ABS -- Alaric Snell-Pym http://www.snell-pym.org.uk/alaric/ ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] svn to fossil migration script
I followed someone's advice to pipe svn-fe's output into fossil's git importer, and the result is just awful. I blame svn-fe for not handling the standard svn repository layout properly. What I have are svn dump files, and lots of them. The steps I've found that will import all branches and tags properly require git-svn, and are as follows: # If you have a dump file... svnadmin create svn_myrepos svnadmin load svn_myrepos myrepos.dmp # Now that you have a repository directory... mkdir git_myrepos cd git_myrepos git svn init -s file:///home/myuser/svn_myrepos git svn fetch # at this point, it's the standard import git fast-export --all | fossil import --git myrepos.fossil I've attached a script to do this in batch, along with setting a standard password on all of the imported repositories (I needed it). mightyhe#!/usr/bin/python import os import re import getpass def main(): ***Warning*** - This script works on Unix only ***Warning*** - This script requires git-svn to work, try yum install svn-git For those of us who read the subversion manual, and did a standard layout, we find that svn-fe path_to_repository | fossil import --git repos.fossil just isn't good enough, because svn-fe doesn't properly handle branches and tags. To handle things properly, we need to use git-svn -s somewhwere along the data transfer path. This script will take a directory of subversion dmp files with standard layout (trunk, root, tags) and turn them into fossil repositories if you don't have a dmp file, you can either svnadmin dump /path_to_repos repos.dmp or alter the script to use repository paths directly (jump to git svn init) cwd = os.getcwd() defuser = getpass.getuser() defpasswd = getpass.getpass('Default Fossil Password: ') for fname in os.listdir('.'): mo = re.match(r'(.+)\.dmp', fname) if not mo: continue print processing +os.path.join(cwd,fname)+... base = mo.group(1) dmpfile = os.path.join(cwd, fname) svndir = os.path.join(cwd,'svn_'+base) gitdir = os.path.join(cwd,'git_'+base) fossilfile = os.path.join(cwd,base+'.fossil') os.system('svnadmin create '+svndir) os.system('svnadmin load '+svndir+' '+dmpfile) os.mkdir(gitdir) os.chdir(gitdir) os.system('git svn init -s file://'+svndir) os.system('git svn fetch') os.system('git fast-export --all | fossil import --git '+fossilfile) os.chdir(cwd) os.system('rm -Rf '+gitdir) os.system('rm -Rf '+svndir) os.system('fossil user --repository '+fossilfile+' password '+defuser+' '+defpasswd) main()___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Generated wiki links and name parameter?
Why do Fossil-generated links to wiki pages include the ?name=pagename parameter when /wiki/pagename seems to work as well as /wiki?name=pagename? 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] Basic FossilHub functionality works
On Mon, Mar 28, 2011 at 10:24 AM, Stephan Beal sgb...@googlemail.comwrote: On Mon, Mar 28, 2011 at 8:43 AM, Bill Burdick bill.burd...@gmail.comwrote: This allows you to make a Fossil repository that functions as your account and register repositories with it that you are watching (I don't have a GUI yet to register repositories -- the data is in a JSON wiki page attachment). The followed page shows aggregated Great stuff! Regarding the JSON bits... i recently added some basic JSON support to my copy of fossil, but haven't done anything interesting with it. The current code can output the timeline in JSON, but currently no other areas use JSON (for lack of energy and real use cases, not the difficulty of solving the problem). If you have concrete ideas about how having JSON output come directly from Fossil might help you to implement your hub, i'd be interested in discussing that with you and potentially adding the JSON output support. Well, I'm parsing the XML timeline RSS feed in JavaScript, right now, but JSON would certainly be more convenient :). Bill ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] Moving multiple files
Hello everyone, can fossil recognize moved files automatically? Thank you, Louis ___ 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] Hierarchy of tickets in fossil
On Sun, Mar 27, 2011 at 8:48 PM, v...@lavabit.com wrote: I know it's possible to create custom ticket reports, but reports are still flat lists, not hierarchies. Are there any plans to make ticket management easier in fossil, by making it possible to group them into trees Part of what you are asking can (probably) be accomplished in the SQL query using the GROUP BY and ORDER BY clauses. THis would result in each ticket being followed by its children. Unfortunately, I have only rudimentary knowledge of SQL, so I don't know how to do this. To get the list to be rendered in a tree structure, you could then use Javascript. Most likely there is existing Javascript available that can be adapted (or, possibly, simply configured) to work with the HTML table generated from the SQL query. The basic logic for an outline view of the tickets would be to check the parent-id field of each ticket in turn. If the tcket has the same parent as the previous ticket, then set the indent level the same as the previous ticket. If the parent is the previous ticket, then incement the indent level. If the parent is some earlier ticket, then set the indent level to 1 plus the parent's indent level. If the parent ticket has never been seen, then reset the indent level to 0. I hope this is somewhat helpful. ___ 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] Generated wiki links and name parameter?
/wiki/pagename does work just as well for wiki pages, but it doesn't handle attachments -- you could use a different URL prefix for that, if you wanted, like wiki-attachment. There's always a tradeoff between URL path components and named parameters; it's a matter of taste. Personally, I'd say that if you have exactly one required parameter that has a path structure, then using the URL path might make sense, like with doc links, but there's no clear-cut rule that says when you should choose one representation or the other. Using parameters makes it clear that */wiki is a service. Bill On Mon, Mar 28, 2011 at 10:14 AM, Nolan Darilek no...@thewordnerd.infowrote: Why do Fossil-generated links to wiki pages include the ?name=pagename parameter when /wiki/pagename seems to work as well as /wiki?name=pagename? Thanks. ___ 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] File dates
I've raised this issue before with little success, although a few people responded in support of the idea. The issue is the file dates being always today's date whenever one downloads a copy from the repository or simply opens a local copy. I'd really like to either see the file dates being the date of the last update or at least have an option of saving the update date in the filename so I can use that info and force the file date myself. I figure that all of the necessary information is already in the database and the only step that's missing is a call to the file system for the file's date/time update after it is re-created from the database. Does anybody else feel this is a useful feature/option? thanks for fossil, I love it! Vl@dimir ___ 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] File dates
On 03/28/2011 08:17 PM, Volodya Savastiouk wrote: Does anybody else feel this is a useful feature/option? I do, yes. thanks for fossil, I love it! Me too! ___ 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] File dates
On Mon, Mar 28, 2011 at 1:17 PM, Volodya Savastiouk volo...@io3.ca wrote: I've raised this issue before with little success, although a few people responded in support of the idea. The issue is the file dates being always today's date whenever one downloads a copy from the repository or simply opens a local copy. I'd really like to either see the file dates being the date of the last update or at least have an option of saving the update date in the filename so I can use that info and force the file date myself. I figure that all of the necessary information is already in the database and the only step that's missing is a call to the file system for the file's date/time update after it is re-created from the database. Does anybody else feel this is a useful feature/option? thanks for fossil, I love it! Oh, wow, I didn't notice that -- haven't been using Fossil for that long. Yes, that's definitely important! Bill ___ 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] File dates
On Mon, Mar 28, 2011 at 02:17:10PM -0400, Volodya Savastiouk wrote: The issue is the file dates being always today's date whenever one downloads a copy from the repository or simply opens a local copy. This was discussed a while ago and I think the agreement was: (1) Initial open may use the time of last modification for a file. One problem is that finding this can be quite expensive. (2) All changes due to updated etc should use the current time. This is the same behavior as other VCS and required to not break make etc. Noone has stepped up yet to implement (1) :) Joerg ___ 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 add somefile.rs = Error
Hi, I get the following error attempting to add a file with the Ampersand() in the file extension. Fossil wrote: 's' is not recognized as an internal or external command, operable program or batch file. Please advise, Steve ___ 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 add somefile.rs = Error
On Mon, Mar 28, 2011 at 04:33:04PM -0400, sky5w...@gmail.com wrote: Hi, I get the following error attempting to add a file with the Ampersand() in the file extension. Fossil wrote: 's' is not recognized as an internal or external command, operable program or batch file. Isn't it your shell, telling that? '' stands for launch the left process in background and run that of the right. ___ 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 add somefile.rs = Error
On Mon, Mar 28, 2011 at 10:33 PM, sky5w...@gmail.com wrote: Hi, I get the following error attempting to add a file with the Ampersand() in the file extension. Fossil wrote: 's' is not recognized as an internal or external command, operable program or batch file That message comes from your shell, not fossil, because your command does not properly escape the '' character, which is special in unix-like shells. You are running two commands: fossil add somefile.r is run in the BACKGROUND because of the character. then you try to run: s The simplest workaround is to quote the filename: fossil add 'somefile.rs' -- - stephan beal http://wanderinghorse.net/home/stephan/ ___ 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 add somefile.rs = Error
On 03/28/2011 04:33 PM, sky5walk wrote: I get the following error attempting to add a file with the Ampersand() in the file extension. Try with single or double quotes around the file name. -- 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
Re: [fossil-users] fossil add somefile.rs = Error
DOH! fossil add somefile.rs does the trick. This shows how little I am in shell land. :(( Thanks, Steve On Mon, Mar 28, 2011 at 4:36 PM, Stephan Beal sgb...@googlemail.com wrote: On Mon, Mar 28, 2011 at 10:33 PM, sky5w...@gmail.com wrote: Hi, I get the following error attempting to add a file with the Ampersand() in the file extension. Fossil wrote: 's' is not recognized as an internal or external command, operable program or batch file That message comes from your shell, not fossil, because your command does not properly escape the '' character, which is special in unix-like shells. You are running two commands: fossil add somefile.r is run in the BACKGROUND because of the character. then you try to run: s The simplest workaround is to quote the filename: fossil add 'somefile.rs' -- - stephan beal http://wanderinghorse.net/home/stephan/ ___ 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 add somefile.rs = Error
On Mon, Mar 28, 2011 at 10:36 PM, Stephan Beal sgb...@googlemail.comwrote: The simplest workaround is to quote the filename: fossil add 'somefile.rs' That said, special characters in filenames (spaces, quotation marks, and the like) often cause more grief than simply renaming the files. Lots of shell scripts are not careful to quote filenames properly and may choke in unexpected ways when encountering files with special characters. e.g. if we have a filename 1a b c (with spaces but no quotes), the following will not work: for i in a*; do mv $i $i.bak; done that will try to move 5 files (if i'm not mis-counting) to a single file names c.bak. stephan@tiny:~$ touch '1a b c' stephan@tiny:~$ for i in 1a*; do echo mv $i $i.bak; done mv 1a b c 1a b c.bak As a general rule, to avoid compatibility problems, never use spaces or any characters which might be prohibited or problematic on some systems. e.g. having the character ':' in a filename will work on Unix but not Windows. -- - stephan beal http://wanderinghorse.net/home/stephan/ ___ 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 social networking
On Mar 22, 2011, at 21:55 , Ron Wilson wrote: Some have already critisized Fossil for including issue tracking and wiki fuctionality, Right now, I think the synergy of the 3 core functions is great. I do not see sufficient increase in synergy by adding other major functions directly to Fossil. The full capabilities of Fossil can be utilized by an extension - especially once hook support is implemented. Especially is not the right word... For many of my fancy ideas hooks are the only reasonable way. Kind regards, Remigiusz Modrzejewski ___ 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 add somefile.rs = Error
put the filename in quotes. -Original Message- From: Lluís Batlle i Rossell virik...@gmail.com Sent: Monday, March 28, 2011 22:36:10 Subject: Re: [fossil-users] fossil add somefile.rs = Error On Mon, Mar 28, 2011 at 04:33:04PM -0400, sky5w...@gmail.com wrote: Hi, I get the following error attempting to add a file with the Ampersand() in the file extension. Fossil wrote: 's' is not recognized as an internal or external command, operable program or batch file. Isn't it your shell, telling that? '' stands for launch the left process in background and run that of the right. ___ 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] Cloning question
Q: Does cloning (or sync or push or pull) work with CGI? If Fossil server is running on my server: Fossil open myrepo.fossil Fossil server Then cloning works as expected: Fossil clone http://myserver:8080/ mycopy.fossil However, I've got multiple repositories so I'm using a Perl script to process CGI requests: http://myserver/Fossil.pl?repository=\Path\myrepo.fossil The script looks for repository=\Path\myrepo.fossil and steers Fossil to the repository when browsing. However, this doesn't even invoke the Perl t: Fossil clone http://myserver/Fossil.pl?repository=\Path\myrepo.fossil mycopy.fossil Where should the requests go? Note: This is CGI on IIS. [cid:image001.jpg@01CBED68.5AEBDDE0] TONY PEROVIC tpero...@compumation.commailto:tpero...@compumation.com www.compumation.com 205 W. Grand Ave., Ste. 121 Bensenville, IL 60106 630-860-1921 Phone 630-860-1928 Fax inline: image001.jpg___ 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 social networking
On Tue, Mar 22, 2011 at 3:55 PM, Ron Wilson ronw.m...@gmail.com wrote: On Tue, Mar 22, 2011 at 3:28 PM, Bill Burdick bill.burd...@gmail.com wrote: Has anyone talked about adding Github-like social networking to Fossil? Just as github is a serperate project built on top of git, any Fossil based social network would be better implemented as a seperate project built on top of Fossil. Some have already critisized Fossil for including issue tracking and wiki fuctionality, Right now, I think the synergy of the 3 core functions is great. I do not see sufficient increase in synergy by adding other major functions directly to Fossil. The full capabilities of Fossil can be utilized by an extension - especially once hook support is implemented. Can you link me some info on the hooks? I didn't need them for the social networking stuff, but something like server-side scripting with scripts in the repository would be very nice. TH1 is good for templating, but personally, I'd like a full language -- a complete TCL or Lua or something like that. Bill ___ 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] Cloning question
On Mon, Mar 28, 2011 at 5:52 PM, Tony Perovic tpero...@compumation.comwrote: Q: Does cloning (or sync or push or pull) work with CGI? There is a discussion of this in the Fossil documentation describing ways to do this. As far as I know, this should work on IIS, but I have no experience setting up IIS. However, I’ve got multiple repositories so I’m using a Perl script to process CGI requests: http://myserver/Fossil.pl?repository=\Path\myrepo.fossil The script looks for repository=\Path\myrepo.fossil and steers Fossil to the repository when browsing. Even with IIS, I think you have to use / instead of \ ___ 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] File dates
On 2:59 PM, Joerg Sonnenberger wrote: On Mon, Mar 28, 2011 at 02:17:10PM -0400, Volodya Savastiouk wrote: The issue is the file dates being always today's date whenever one downloads a copy from the repository or simply opens a local copy. This was discussed a while ago and I think the agreement was: (1) Initial open may use the time of last modification for a file. One problem is that finding this can be quite expensive. (2) All changes due to updated etc should use the current time. This is the same behavior as other VCS and required to not break make etc. Noone has stepped up yet to implement (1) :) Joerg OK, thanks to Joerg for clarifying the situation. I will have a look at the Fossil code to assess if I can contribute. Vl@dimir ___ 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] File dates
On Mon, 28 Mar 2011 14:17:10 -0400 Volodya Savastiouk volo...@io3.ca wrote: I've raised this issue before with little success, although a few people responde The issue is the file dates being always today's date whenever one downloads a copy from the repository or simply opens a local copy. I'd really like to either see the file dates being the date of the last update or at least have an option of saving the update date in the filename so I can use that info and force the file date myself. I figure that all of the necessary information is already in the database and the only step that's missing is a call to the file system for the file's date/time update after it is re-created from the database. Does anybody else feel this is a useful feature/option? As noted, this breaks build systems that compare file dates to see if a file needs to be recompiled. As such, this feature is dangerous. mike -- Mike Meyer m...@mired.org http://www.mired.org/consulting.html Independent Software developer/SCM consultant, email for more information. O ascii ribbon campaign - stop html mail - www.asciiribbon.org ___ 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] File dates
On 2:59 PM, Mike Meyer wrote: As noted, this breaks build systems that compare file dates to see if a file needs to be recompiled. As such, this feature is dangerous. mike I appreciate that Mike. I'm thinking of having the update information being part of the filename on request (an option). Vl@dimir ___ 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] Sub-repositories?
On Mon, Mar 28, 2011 at 10:37 PM, Nolan Darilek no...@thewordnerd.infowrote: More important, what *are* subrepositories? Entirely separate fossils linked to the main repository, or separate namespaces for tags and such in a single fossil? Or the ability to open a nested repository in another? Wondering if I'll need to migrate some of my multi-repository fossils to a new format somehow. On 03/28/2011 09:36 PM, Michael Richter wrote: I just saw an intriguing message in the timeline. Leaf: Merge the sub-repo capability into trunk. I can't find anywhere in the help, the fossil docs nor in the wiki that talks about this. How does one use this new sub-repo capability? Here's my conjecture, based on how Git does submodules, without having looked at the Fossil subrepository code... I'm guessing that in the repository, a submodule might be represented by an artifact that contains an identification of the subrepository, plus the identifier of the checked-out tree for that repository, with a path in a manifest identifying the location of the subrepository in the directory tree. A subrepository could be identified by a Fossil URL or maybe requiring it to be a sibling of the current repository -- a URL is nice, because it provides a way to retrieve it when you clone the parent repository. On disk, a subrepository might be checked-out as a subdirectory of the current check-out (corresponding to the entry in the manifest). Bill ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users