Re: [fossil-users] Ticket 305143bd876f693f446f78d12dbef143c46eec58 [SOLVED]
Thanks for all the friendly help I got on this issue and for the near saintlike patience Richard showed. The problem has been solved. I blame China. (I'm only being a little bit facetious in this.) Something Richard asked -- about a proxy that filters anything with timeline in the URL -- got me thinking. The pattern I was seeing was anything involving the timeline view was getting me flagged as not being logged in. This only happened with Google Chrome (at first) and not with Firefox or Opera. Further it didn't happen on Windows' version of Google Chrome (along with Firefox and Opera). It was only with my Ubuntu version of Chrome and only with remote repositories, not local. Now for the interesting part. I routinely use an SSH-based SOCKS5 proxy as a back door through the Great Firewall because I'm not afraid of the harmful effects of seeing the Internet in its entirety. On my Windows machine I just have the proxy running all the time and use it for all access (for Chrome and Firefox at any rate -- Opera doesn't support SOCKS5 proxies for some bizarre reason). When I log in to a repository from Windows I'm either always going through the proxy (Chrome/Firefox) or I'm never going through a proxy (Opera). On Ubuntu I have a slightly more sophisticated setup. Because my SSH tunnel is painfully slow, I only use it when I'm forced to. I've built a sizable script to decide on an URL-by-URL basis whether it should go through my proxy or not. This is irrelevant for Opera, of course, because Opera sucks for proxy support. Both Firefox and Chrome, however, support proxies quite well. What Chrome DOESN'T do well, however, is quickly turning proxy support on and off. My Firefox configuration has a button for it, though. So sometimes I run with the proxy script in line and sometimes I don't. Initially I didn't have it turned on, so I was just like Opera -- running without a proxy. Sometime in the few days I was playing I turned it on for accessing some site the Chinese government decided my tender eyes couldn't cope with (Youtube) and then, a day later, long after I'd forgotten that I'd done it, I tried accessing my repositories with Firefox and the bit rot happened. This doesn't explain the odd specificity of the problem, however. What does is the URL patterns I use. Usually I give URL patterns for domains (* youtube.com*) or by path elements known to be problematical (*/blog/*). For safety's sake I always try to locate it either with a full domain name or with path elements in the pattern. But I made a mistake at some point and when I meant to type *time.com* I accidentally wound up inserting *time*. The relevance of that pattern to the timeline issue is, I think, obvious. I did grep on timeline after Richard suggested this possibility but it never occurred to me until later in the shower (I do a lot of my thinking there) that there might be other fragments (time or line) that could be to blame. I went over my (literally hundreds of) filters with a fine-toothed comb and found the offending script entry. So why did this not cause a problem for local repositories? Why did this not cause a problem when I went through Richard's sockettee thing? Well, I automatically, at the head of my script, return the direct link for any access to localhost, to my local network and a few other such entries where I know I can't physically ever be blocked by the Great Firewall. Since sockettee goes through localhost, and since fossil server/ui also goes through localhost there was never any kind of filtering. TL;DR summary: if you're using complex scripts to work around people with the mentality of small, frightened children controlling your Internet, sometimes it'll do bizarre, undiagnosable things. Fossil is blameless. The stupid bastards in Beijing are another matter. (And my own intellect is apparently somewhat questionable as well.) -- Perhaps people don't believe this, but throughout all of the discussions of entering China our focus has really been what's best for the Chinese people. It's not been about our revenue or profit or whatnot. --Sergey Brin, demonstrating the emptiness of the don't be evil mantra. ___ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
[fossil-users] index.fossil?
I suppose I could file a ticket requesting this, but I'd like to discuss it first as I feel odd about filing tickets for feature requests that may not be desirable. I have a Fossil-hosted project at http://spielproject.info. In this instance, the project owns the entire domain, and I'd like for the URLs to reflect that, which they currently do. However, Spiel has spun off a few subprojects, which I'd like hosted in their own repositories and with their own websites. I'd like to host these on spielproject.info at URLs like http://spielproject.info/bazaar, http://spielproject.info/accessibility-helper, etc. First, I know that I can accomplish this via web server configuration and rewrites, so please don't take this in that direction. When Fossil is pointed to a directory rather than a repository, I'm wondering if anyone else would find it useful to serve up index.fossil as the project that is automatically selected if none is given? So URLs like: http://spielproject.info/wiki?name=faq would automatically load the given page in index.fossil, while: http://spielproject.info/bazaar/wiki?name=faq would load the page with the same name in the bazaar repository? This would also simplify configuration some. I currently have two distinct Fossil configurations, one which points to a directory of all my various projects hosted at http://dev.thewordnerd.info, and others pointing to distinct repository names hosted at various TLDs. With support for index.fossil, I'd only need the directory case pointing at different directories. TLDs could just replace the index.fossil, and are then free to spin off subprojects in different repositories simply by dropping in another fossil. Note that this would preclude you from naming projects wiki, timeline and other reserved words for the directory case with an index.fossil present. Those seem like such specific cases that I don't think they'll be encountered often enough. Does this seem like a good idea, or does it make Fossil too complicated? Again, I know I can do this with web server configuration, but I like the fact that I can just drop in a fossil and have it served up by the CGI/server instance, and supporting this case would seem to simplify that setup even further. Thanks for reading. ___ 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] index.fossil?
On Thu, Mar 24, 2011 at 10:32 AM, Nolan Darilek no...@thewordnerd.infowrote: When Fossil is pointed to a directory rather than a repository, I'm wondering if anyone else would find it useful to serve up index.fossil as the project that is automatically selected if none is given? So URLs like: http://spielproject.info/wiki?name=faq would automatically load the given page in index.fossil, while: http://spielproject.info/bazaar/wiki?name=faq would load the page with the same name in the bazaar repository? Does this seem like a good idea, or does it make Fossil too complicated? I think it would work better to provide a default repo, and also to redirect to the default repo rather than use it directly, so that there is never any chance of a webpage colliding with another repository name. -- 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] Fossil on Hammer Principle
Hello! Hammer Principle is a whimsical site where people can rank contentious things on various axes, and it then generates overall scores for stuff. They added version control systems lately, and Fossil's one of them: http://versioncontrol.hammerprinciple.com/ However, as of the time of writing, only two people have expressed opinions on Fossil (and I'm one of them). Let's fix that :-) Enjoy, 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
Re: [fossil-users] Fossil on Hammer Principle
It doesn't have ClearCase!!! OMG -Original Message- From: Stephen De Gabrielle stephen.degabrie...@acm.org To: fossil-users@lists.fossil-scm.org fossil-users@lists.fossil-scm.org Sent: Fri, Mar 25, 2011 1:59 am Subject: Re: [fossil-users] Fossil on Hammer Principle Cute. It seems pretty accurate. IMHOS.On Thursday, March 24, 2011, Alaric Snell-Pym ala...@snell-pym.org.uk wrote: Hello! Hammer Principle is a whimsical site where people can rank contentious things on various axes, and it then generates overall scores for stuff. They added version control systems lately, and Fossil's one of them: http://versioncontrol.hammerprinciple.com/ However, as of the time of writing, only two people have expressed opinions on Fossil (and I'm one of them). Let's fix that :-) Enjoy, 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-- --Stephen De Gabriellestephen.degabrielle@acm.orgTelephone +44 (0)20 85670911Mobile+44 (0)79 85189045http://www.degabrielle.name/stephen__ _fossil-users mailing listfossil-users@lists.fossil-scm.orghttp://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