Re: [fossil-users] Fossil vs. Windows
On Tue, Jun 12, 2012 at 9:15 PM, Mike Meyer m...@mired.org wrote: We're not talking about a repository that's being served by an HTTP server, but about the repository that's on his machine. However... once a developer clones the repo, if he has it on a public filesystem then all bets are off. Exactly. He wants it on a file system that he's where he's using ACLs to control access to the contents of the file system, and fossil is apparently refusing to work unless he opens access to those files up to everyone. That's not acceptable to him. The Fossil repository should only need to be accessable to the user ID under which Fossil runs. If he is using the Fossil command line, then it runs under his user ID, so it would have access to a repository he owned. Should be the same for using fossil ui. If he wants/needs to run fossil server on his PC, then Fossil should run as a different user ID with a repository owned by the same user ID - or at least the repository has permissions allowing the Fossil user ID read/write access. For him to access that same repository with the Fossile command line, the orthodox way would be to have the command line treat the Fossil server as if it was on another PC and access it the same as accessing any remote repository. If he really wants to directly access the same resposity, the repository would have to have permissions granting both him and the Fossil user ID read/write access. While I suppose there could be a bug in Fossil, in my own use of Fossil, repositories I own are only accessable to me, therefore only accessable to Fossil running under my user ID. Other users' repositories are likewise only accessable to them. To share changes, we each have a fossil server running under our own IDs that respond to push/pull/sync requests. While I don't know, I'd be surprized if Mercurial had any interface to ACLs, Windows, Linux, Posix or other. For what it is worth, as best I recall, last time I used git, it did not. Like most other apps, including Fossil, it depends on the OS to decide whether it has access to any given file. Ultimately, without knowing exactly what your boss was trying to do, it is unlikely to be possible to demonstrate Fossil working in a given scenario. I suppose he could be expecting Fossil to somehow use Windows ACLs to control access to artifacts stored in Fossil's database, but since Fossil's databse is contained in a single file, that is not practical and probably not even possible. ___ 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 vs. Windows
On 2012-06-12 00:36, Mike Meyer wrote: My boss just sent me mail that said, and I quote: Fossil sucks and is actually not compatible with Windows I'm pretty sure I've seen people here who use it no Windows, and there's a Windows distribution, which makes me think he's wrong. His problem is that he has lots of spaces in his directory and file names, and we (Windows is a third-line gaming platform for me, but I try...) couldn't figure out how to get such passed to fossil as file names, instead of broken up by command.com or whatever does that job these days. Quoting again: If I try adding parentheses or quotes then the ignorant application bombs out. So, anyone got advice on this? Maybe a GUI that works with the fossil Windows binary from fossil.org? Thanks, mike Mike, I think he doesn't want to move to fossil. And all the arguments you make will not help. He keep up making arguments why it sucks Better sit down with him and ask him what the problem is, maybe first a few pints of strong lager. Maybe you selected the tool without him participating (enough) in the decision. It is not a technical issue. -- Rene ___ 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 vs. Windows
2012/6/13 Rene renew...@xs4all.nl: On 2012-06-12 00:36, Mike Meyer wrote: My boss just sent me mail that said, and I quote: Fossil sucks and is actually not compatible with Windows Mike, ... Better sit down with him and ask him what the problem is, maybe first a few pints of strong lager. ... It is not a technical issue. -- Fully agree! Cheers, Jacek ___ 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] Update of checkout after push operations.
On Thu, Jun 7, 2012 at 5:54 PM, John Found johnfo...@evrocom.net wrote: I have one central repository, that I use as an archive for a web site. The web site root directory is actually checkout of the trunk branch of the repository. I am using lighttpd server on Linux and fossil as a CGI script in very standard manner. So, I need when I make a push (or commit with autosync) from a remote computer to the central repository, some script to make fossil update to the checkout directory and this way to update the content of the web site. Is it possible at all? Any Ideas? Last I knew, the closest Fossil has to hooks is an RSS feed. You can run a RSS monitor that looks for events of interest, then triggers actions based on those events. ___ 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] Update of checkout after push operations.
On 2012-06-13 19:07, Ron Wilson wrote: On Thu, Jun 7, 2012 at 5:54 PM, John Found johnfo...@evrocom.net wrote: I have one central repository, that I use as an archive for a web site. The web site root directory is actually checkout of the trunk branch of the repository. I am using lighttpd server on Linux and fossil as a CGI script in very standard manner. So, I need when I make a push (or commit with autosync) from a remote computer to the central repository, some script to make fossil update to the checkout directory and this way to update the content of the web site. Is it possible at all? Any Ideas? Last I knew, the closest Fossil has to hooks is an RSS feed. You can run a RSS monitor that looks for events of interest, then triggers actions based on those events. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users Hm wild idea I don't know if it is practical cgi1 script monitor if a push or commit command is given if so set a flag and connect to cgi2(your original script) after the operation do a checkout. -- Rene ___ 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] Update of checkout after push operations.
On 2012-06-13 19:07, Ron Wilson wrote: On Thu, Jun 7, 2012 at 5:54 PM, John Found johnfo...@evrocom.net wrote: I have one central repository, that I use as an archive for a web site. The web site root directory is actually checkout of the trunk branch of the repository. I am using lighttpd server on Linux and fossil as a CGI script in very standard manner. So, I need when I make a push (or commit with autosync) from a remote computer to the central repository, some script to make fossil update to the checkout directory and this way to update the content of the web site. Is it possible at all? Any Ideas? Last I knew, the closest Fossil has to hooks is an RSS feed. You can run a RSS monitor that looks for events of interest, then triggers actions based on those events. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users You could make a second cgi script that will do a checkout. and after a commit do http://yoursever/checkout.cgi It has the added benefit that you can save your work from location 1. Go to location 2 finalize the work and publish then. -- Rene ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
