{"thread":{"id":"14334","subject":"[GSoC] What is status of Git's Google Summer of Code 2008 projects?","startedAt":"2008-07-08T00:27:41Z","lastAt":"2008-08-22T23:03:39Z","messageCount":46,"participants":["Jakub Narebski","David Symonds","Stephan Beyer","Junio C Hamano","Lea Wiemann","Shawn O. Pearce","J.H.","Joshua Roys","Johannes Schindelin","Petr Baudis","Sam Vilain","Sverre Rabbelier","Miklos Vajna","Marek Zawirski"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"82508","messageId":"200807080227.43515.jnareb@gmail.com","threadId":"14334","inReplyTo":null,"subject":"[GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-08T00:27:41Z","receivedAt":"2008-07-08T00:27:41Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"I'd very much like to have (or perhaps to wrote) some sort of interim \nprogress report for Google Summer of Code 2008 projects on \nhttp://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to \nexpand and/or correct those mini-summaries below.\n\n(It would be, I guess, good preparation for GSoC 2008 mid-term \nevaluations, which according to GSoC 2008 timeline\n http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\nare to take place July 7 -- July 14.)\n\n\n1. GitTorrent (???)\n\nStudent: Joshua Roys\nMentor: Sam Vilain\n\nThere was short thread of me asking about project\n http://thread.gmane.org/gmane.comp.version-control.git/83611\nwhere I got gittorrent mailing list (no activity at least according to \nlist archive http://lists.utsl.gen.nz/pipermail/gittorrent/) and URL \nfor project repo / gitweb... which is currently down, so I cannot check \nif there is anything here.\n\nWhat is the status of this project, please?\n\nhttp://www.codinghorror.com/blog/archives/001134.html (\"Don't Got Dark\")\n\n\n2. git-statistics (?)\n\nStudent: Sverre Rabbelier\nMentor: David Symonds\n\nThere were some posts about how git-statistics can be used:\n http://thread.gmane.org/gmane.comp.version-control.git/81534\n http://thread.gmane.org/gmane.comp.version-control.git/82027\nbut it was mainly theoretical thread\n\nThe git-stats repository at http://repo.or.cz/w/git-stats.git (gitweb) \nhas some commits, but I don't remember any of them on git mailing list.\nNot ready?\n\n\n3. Gitweb caching\n\nStudent: Lea Wiemann\nMentor: John 'warthog' Hawley\n\nLea has chosen caching data and memcached as primary caching engine,\nand is working on object layer on top of Git.pm, namely Git::Repo and \nfriends, which will be used by gitweb. If I understand correctly \ncaching is to be done, or at least helped by this layer.\n\nThere were added tests for Git.pm (now in git.git I think) and Mechanize \ntest for gitweb (which detected a few errors).\n\nIs it correct?\n\n\n4. Eclipse plugin push support (!)\n\nStudent: Marek Zawirski\nMentor: Shawn O. Pearce\n\nI am not following egit/jgit development close enough, but if I remember \ncorrectly there is some code which provides very rudimentary support \nfor native generation of simplified packs, and IIRC also for push over \nsome protocols.\n\nAnd there is push support over SFTP and (encrypted) Amazon S3...\n\n\n5. git-merge builtin (!!!)\n\nStudent: Miklos Vajna\nMentor: Johannes Schindelin\n\nBuiltin merge, together with preparation patches, is now at n-th \niteration, as shown in the \"Build in merge\" thread:\n http://thread.gmane.org/gmane.comp.version-control.git/86584\n\nIn \"What's cooking in git.git (topics)\" Junio C Hamano wrote:\n\n It already is beginning to become clear what 1.6.0 will look like.\n [...]\n * git-merge will be rewritten in C.\n\n\n6. git-sequencer (!)\n\nStudent: Stephan Beyer\nMentor: Christian Couder, Daniel Barkalow\n\nIt started with discussion over TODO file format:\n http://thread.gmane.org/gmane.comp.version-control.git/84230\n\nNow there is prototype shell script implementation (which has some \nlimitations because it is prototype) in \"git sequencer prototype\"\n http://thread.gmane.org/gmane.comp.version-control.git/86985\n\nStephan Beyer wrote:\n I'm using sequencer-based git-am and git-rebase-i and also \n git-sequencer itself for around 2-3 weeks now. So, for me, it is\n reality-proven [...]\n\n-- \nJakub Narebski\nPoland\n"},{"id":"82511","messageId":"ee77f5c20807071743n350be8b5w68cd78e4977f879a@mail.gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"David Symonds","fromEmail":"dsymonds@gmail.com","sentAt":"2008-07-08T00:43:30Z","receivedAt":"2008-07-08T00:43:30Z","isPatch":false,"sender":{"key":"dsymonds@gmail.com","avatar":"https://gravatar.com/avatar/b22f5051cbfc11836e36cf7a690e6cde4e225d835e13295ff98d15c7a9ee3c0f?d=mp&s=160"},"body":"On Tue, Jul 8, 2008 at 10:27 AM, Jakub Narebski wrote:\n\n> I'd very much like to have (or perhaps to wrote) some sort of interim\n> progress report for Google Summer of Code 2008 projects on\n> http://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to\n> expand and/or correct those mini-summaries below.\n\n[.. snip ..]\n\n> 2. git-statistics (?)\n>\n> Student: Sverre Rabbelier\n> Mentor: David Symonds\n>\n> There were some posts about how git-statistics can be used:\n> http://thread.gmane.org/gmane.comp.version-control.git/81534\n> http://thread.gmane.org/gmane.comp.version-control.git/82027\n> but it was mainly theoretical thread\n>\n> The git-stats repository at http://repo.or.cz/w/git-stats.git (gitweb)\n> has some commits, but I don't remember any of them on git mailing list.\n> Not ready?\n\nSverre is in the process of writing up a progress report for the Git\nML; expect it by the end of this week.\n\n\n\nDave.\n"},{"id":"82515","messageId":"20080708010050.GD6726@leksak.fem-net","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-07-08T01:00:50Z","receivedAt":"2008-07-08T01:00:50Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\n> 6. git-sequencer (!)\n> \n> Student: Stephan Beyer\n> Mentor: Christian Couder, Daniel Barkalow\n> \n> It started with discussion over TODO file format:\n> http://thread.gmane.org/gmane.comp.version-control.git/84230\n> \n> Now there is prototype shell script implementation (which has some \n> limitations because it is prototype) in \"git sequencer prototype\"\n> http://thread.gmane.org/gmane.comp.version-control.git/86985\n> \n> Stephan Beyer wrote:\n> I'm using sequencer-based git-am and git-rebase-i and also \n> git-sequencer itself for around 2-3 weeks now. So, for me, it is\n> reality-proven [...]\n\nThat's a nice \"summary\".\nImho the prototype patchset needs some more review from others.\n(Well, I think the\n http://thread.gmane.org/gmane.comp.version-control.git/86985\nthread died in the last days, but I hope some responses will come and\nalso that Junio's patch for cherry-picking root commits will be\nincluded. I try to be patient...) ;-)\n\nYou may also add a little note, that I have started the builtin sequencer,\nbut I cannot provide really big news here :-)\n\nThanks for your overall git GSoC engagement,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"82518","messageId":"7vmyktxjlq.fsf@gitster.siamese.dyndns.org","threadId":"14334","inReplyTo":"20080708010050.GD6726@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-07-08T01:14:41Z","receivedAt":"2008-07-08T01:14:41Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Stephan Beyer writes:\n\n> Imho the prototype patchset needs some more review from others.\n\nYes, very much. Not just from others and not just from me.\n\n> (Well, I think the\n> http://thread.gmane.org/gmane.comp.version-control.git/86985\n> thread died in the last days, but I hope some responses will come and\n> also that Junio's patch for cherry-picking root commits will be\n> included. I try to be patient...) ;-)\n\nPlease don't be patient but actively re-review what you sent out.\n\nI _really_ wanted to merge the basic bits and rewrite of \"am\" at least to\npu tonight, but I had to drop them after noticing that it does not seem to\nhandle --rebasing at all (it parses to set $rebasing but after that where\ndoes that bit go? bash completion wants to see rebasing or applying\nmarkers in .dotest), which made it a non-starter especially I'll be\ncooking the other ORIG_HEAD in 'next' as well.\n\nAbout the \"rewrite rebase to use sequencer\" bits, because we've dropped\nthe older rebase-i change, I do not want your series to depend on it.\n"},{"id":"82523","messageId":"20080708014734.GE6726@leksak.fem-net","threadId":"14334","inReplyTo":"7vmyktxjlq.fsf@gitster.siamese.dyndns.org","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-07-08T01:47:34Z","receivedAt":"2008-07-08T01:47:34Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nJunio C Hamano wrote:\n> Stephan Beyer writes:\n> > (Well, I think the\n> > http://thread.gmane.org/gmane.comp.version-control.git/86985\n> > thread died in the last days, but I hope some responses will come and\n> > also that Junio's patch for cherry-picking root commits will be\n> > included. I try to be patient...) ;-)\n> \n> Please don't be patient but actively re-review what you sent out.\n> \n> I _really_ wanted to merge the basic bits and rewrite of \"am\" at least to\n> pu tonight, but I had to drop them after noticing that it does not seem to\n> handle --rebasing at all (it parses to set $rebasing but after that where\n> does that bit go?\n\nYes, you are right that am --rebasing is a no-op.\nThat option was a little mystery to me, because it seemed to do nothing\nspecial, but I'll check again (bash-completion etc) and do appropriate\nchanges.\n\n> About the \"rewrite rebase to use sequencer\" bits, because we've dropped\n> the older rebase-i change, I do not want your series to depend on it.\n\nYes, I made the mistake that I started developing on \"next\" and I did\nnot expect that js/rebase-i-sequencer could be dropped. (I somehow\nexpected the opposite, that it would be migrated to master some day.)\n\nThen I wanted to make it work on master, and thought this is simply\ndone by removing the \"additional features\" (-f and -p option using extended\ntodo list), but then I noticed that master also has a -p which works\nentirely different and I wanted to see that stuff on the list before\nJuly 1st and I had no idea what the status of js/rebase-i-sequencer is.\nSo I just tried and sent that last patch to the list in that way.\n\nBut now I have a clear statement.\n\nSo I'm going to send a new patchset (also with an EXAMPLES section in \nthe docs) to the list in a few days.\n\nBig thanks for the feedback,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"82538","messageId":"4872E858.4080607@gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Lea Wiemann","fromEmail":"lewiemann@gmail.com","sentAt":"2008-07-08T04:08:56Z","receivedAt":"2008-07-08T04:08:56Z","isPatch":false,"sender":{"key":"lewiemann@gmail.com","avatar":null},"body":"Jakub Narebski wrote:\n> 3. Gitweb caching\n\nI'll post a complete status update in the next few days. And three\nlarge patches (including the mechanize test). ;-)\n\n> Lea has chosen caching data and memcached as primary caching engine,\n> and is working on object layer on top of Git.pm, namely Git::Repo and \n> friends, which will be used by gitweb. If I understand correctly \n> caching is to be done, or at least helped by this layer.\n\nThat's correct, except that I'm not using Git.pm anywhere; Git::Repo is\nindependent of Git.pm. More about that later...\n"},{"id":"82542","messageId":"20080708041947.GA2542@spearce.org","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2008-07-08T04:19:47Z","receivedAt":"2008-07-08T04:19:47Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Jakub Narebski wrote:\n> I'd very much like to have (or perhaps to wrote) some sort of interim \n> progress report for Google Summer of Code 2008 projects on \n> http://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to \n> expand and/or correct those mini-summaries below.\n> \n> (It would be, I guess, good preparation for GSoC 2008 mid-term \n> evaluations, which according to GSoC 2008 timeline\n> http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> are to take place July 7 -- July 14.)\n\nYes, it is that time for evaluations. The evaluations are now open\nto students and mentors alike; please make sure you complete them\nby the deadline of July 14th.\n \n> 4. Eclipse plugin push support (!)\n> \n> Student: Marek Zawirski\n> Mentor: Shawn O. Pearce\n> \n> I am not following egit/jgit development close enough, but if I remember \n> correctly there is some code which provides very rudimentary support \n> for native generation of simplified packs, and IIRC also for push over \n> some protocols.\n> \n> And there is push support over SFTP and (encrypted) Amazon S3...\n\nMarek is on holiday right now, so I have to answer for him here.\nOtherwise I would have preferred to let him do it.\n\nThus far Marek has completed generation of packs, including delta\nre-use from packs using either v1 or v2 index, including taking\nadvantage of the CRC inside the v2 index to accelerate a safe reuse.\nThis code permits jgit to write a valid pack.\n\nThe packing code does not (yet) search for a delta base, or create\na new delta for an undeltified object. Packing loose objects packs\nthem as whole objects in the pack file, resulting in little to no\nreduction over their loose object size. This is not a limitation\nof Java. Marek and I simply decided that protocol support was more\nimportant than really tight network transport at this point in time.\n\nAs a result of being able to create pack files Marek was able to\nimplement the client side of git-push for the native pack transfer\nservice, aka push over SSH, push to another local repository (by\nforking 'git receive-pack') and push over anonymous git://.\n\nUsing Marek's pack generation code I added support for push over\nthe dumb sftp:// and amazon-s3:// protocols, with the latter also\nsupporting transparent client side encryption.\n\nI chose to add these features to jgit partly as an exercise to prove\nthat Marek's code was built well enough to be reused for this task,\npartly because I wanted to backup some private personal repositories\nto Amazon S3, and partly to prove that multiple dumb transports\ncould implement push support.\n\nAll of the above is done in the non-Eclipse, BSD licensed jgit\nlibrary, making it available to any tool built on top of the Java\nplatform, even if said tool does not use the Eclipse platform or\nany other code from Eclipse.\n\n\nAt this point Marek's code is in the main egit.git tree's master\nbranch, and is in \"production\" use by myself and Robin, and maybe\na few others. I am quite happy with the work Marek has completed\nto date for the project.\n\nWhen Marek returns from his holiday he will be working on Eclipse\nUI features to expose this jgit push functionality to the end-user\nwithin the Eclipse workbench.\n\n-- \nShawn.\n"},{"id":"82564","messageId":"1215501638.1998.21.camel@localhost.localdomain","threadId":"14334","inReplyTo":"4872E858.4080607@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"J.H.","fromEmail":"warthog19@eaglescrag.net","sentAt":"2008-07-08T07:20:38Z","receivedAt":"2008-07-08T07:20:38Z","isPatch":false,"sender":{"key":"warthog19@eaglescrag.net","avatar":null},"body":">From a quick an cursory glance I think we are well on track with the\nGitweb stuff (though I'll let Lea do a full status update). It looks\nlike most of the code is in place, and there is a test site up and\nrunning on one of the kernel.org machines (though not anywhere near\nproduction yet - I still need to port some of my other changes to Lea's\nchange set so we can try this out for real on Kernel.org)\n\n- John 'Warthog9' Hawley\n\nOn Tue, 2008-07-08 at 06:08 +0200, Lea Wiemann wrote:\n> Jakub Narebski wrote:\n> > 3. Gitweb caching\n> \n> I'll post a complete status update in the next few days. And three\n> large patches (including the mechanize test). ;-)\n> \n> > Lea has chosen caching data and memcached as primary caching engine,\n> > and is working on object layer on top of Git.pm, namely Git::Repo and \n> > friends, which will be used by gitweb. If I understand correctly \n> > caching is to be done, or at least helped by this layer.\n> \n> That's correct, except that I'm not using Git.pm anywhere; Git::Repo is\n> independent of Git.pm. More about that later...\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at http://vger.kernel.org/majordomo-info.html\n"},{"id":"82565","messageId":"200807080939.31216.jnareb@gmail.com","threadId":"14334","inReplyTo":"20080708014734.GE6726@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-08T07:39:30Z","receivedAt":"2008-07-08T07:39:30Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Tue, 8 July 2008, Stephan Beyer wrote:\n> Junio C Hamano wrote:\n>> \n>> I _really_ wanted to merge the basic bits and rewrite of \"am\" at least to\n>> pu tonight, but I had to drop them after noticing that it does not seem to\n>> handle --rebasing at all (it parses to set $rebasing but after that where\n>> does that bit go?\n> \n> Yes, you are right that am --rebasing is a no-op.\n> That option was a little mystery to me, because it seemed to do nothing\n> special, but I'll check again (bash-completion etc) and do appropriate\n> changes.\n\nUndocumented option '--rebasing' to git-am is internal option changing\ngit-am behavior to be better used by git-rebase, namely it does not\nchange commit message even if it doesn't follow git commit message\nconvention, for example if it begins not with single line summary\nof commit, separated by empty line, but by multi-line paragraph.\nSee also t/t3405-rebase-malformed.sh\n\nAlthough I am not sure if when rebase is rewritten using git-sequencer\nimplementing \"git am --rebasing\" would be truly needed. On the other\nhand side it would be nice to have some _documented_ option which would\nallow to git-am mail messages with commits not following git commit\nmessages convention...\n\n-- \nJakub Narebski\nPoland\n"},{"id":"82605","messageId":"20080708144239.GJ6726@leksak.fem-net","threadId":"14334","inReplyTo":"200807080939.31216.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-07-08T14:42:39Z","receivedAt":"2008-07-08T14:42:39Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nJakub Narebski wrote:\n> > Yes, you are right that am --rebasing is a no-op.\n> > That option was a little mystery to me, because it seemed to do nothing\n> > special, but I'll check again (bash-completion etc) and do appropriate\n> > changes.\n> \n> Undocumented option '--rebasing' to git-am is internal option changing\n> git-am behavior to be better used by git-rebase, namely it does not\n> change commit message even if it doesn't follow git commit message\n> convention,\n\nAh yes, I've seen it now.\nIt is taking the commit message from the commit in the \"From .*\"\nline, does *not* change it in any way and then applies the changes using\nthreeway merge.\n\nKeeping that in mind what about dealing with --rebasing like that:\nif --rebasing is given, git am simply generates\n\tpick \nlines, instead of\n\tpatch -3 -k \nas it is now (and this is not enough, as it seems).\n\nDoes someone have strong objections against that?\n\nSpeed could be one point in the case that git-apply just works without\nneeding threeway-fallback, but in the case of the fallback this will be\nslower than pick, I think. So I'd not value that too high, but perhaps\nthere are opinions against my view.\nPerhaps I am missing another point, too?\n\n\nThe alternative for doing \"pick\" is teaching git-sequencer's \"patch\"\ninsn an option that emulates the --rebasing behavior.\n\nFor me this feels somehow unclean. But perhaps there are good reasons.\n\n> for example if it begins not with single line summary\n> of commit, separated by empty line, but by multi-line paragraph.\n> See also t/t3405-rebase-malformed.sh\n\nWell, I have a test script that runs\n for i in t0023* t3350* t340* t3901* t4014* t4150* t5520* t7402*\nand I run that script before I do a commit and after I rebased.\nAnd I ran the whole test suite before I posted the patchset to the list.\nWhat I want to say is: t3405 did not fail with my --rebasing no-op.\n\nThat's perhaps one reason why I forgot about implementing --rebasing\ncorrectly.\n\n> Although I am not sure if when rebase is rewritten using git-sequencer\n> implementing \"git am --rebasing\" would be truly needed.\n\nI didn't want to touch that behavior for several reasons.\n\nOf course, somehow I think that rebase and rebase-i should be merged,\ncalling sequencer directly, with the main difference that -i will\ninvoke an editor to allow editing of the TODO file.\nBut nobody is hurt, if I put such a change far far away.\n\n\nRegards,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"82606","messageId":"200807081812.15651.jnareb@gmail.com","threadId":"14334","inReplyTo":"20080708144239.GJ6726@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-08T16:12:14Z","receivedAt":"2008-07-08T16:12:14Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Stephan Beyer wrote:\n> Jakub Narebski wrote:\n\n>>> Yes, you are right that am --rebasing is a no-op.\n>>> That option was a little mystery to me, because it seemed to do nothing\n>>> special, but I'll check again (bash-completion etc) and do appropriate\n>>> changes.\n>> \n>> Undocumented option '--rebasing' to git-am is internal option changing\n>> git-am behavior to be better used by git-rebase, namely it does not\n>> change commit message even if it doesn't follow git commit message\n>> convention,\n> \n> Ah yes, I've seen it now.\n>\n> It is taking the commit message from the commit in the \"From .*\"\n> line, does *not* change it in any way and then applies the changes using\n> threeway merge.\n\nNot exactly. \"git am --rebasing\" still tries to first just *apply*\nthe patch, then (I think) it falls back on blob-id based 3way merge.\n \n> Keeping that in mind what about dealing with --rebasing like that:\n> if --rebasing is given, git am simply generates\n> \tpick \n> lines, instead of\n> \tpatch -3 -k \n> as it is now (and this is not enough, as it seems).\n\nIt is not.\n\nNevertheless it would be I think better for ordinary patch based rebase\nto fall back not on git-am 3way merge, but on cherry-pick based merge\n(i.e. on pick).\n\n> The alternative for doing \"pick\" is teaching git-sequencer's \"patch\"\n> insn an option that emulates the --rebasing behavior.\n> \n> For me this feels somehow unclean. But perhaps there are good reasons.\n\nWhy unclean?\n\nBut I agree that it would be nice to simplify '--rebasing' logic, for\nexample using patch or 2way merge to generate tree, and commit message\ntaken directly from commit, not via 'format-patch | am' pipeline.\n \n> Of course, somehow I think that rebase and rebase-i should be merged,\n> calling sequencer directly, with the main difference that -i will\n> invoke an editor to allow editing of the TODO file.\n> But nobody is hurt, if I put such a change far far away.\n\nrebase-m and rebase-i can be merged; ordinary rebase uses other\nmechanism: git-am pipeline, and not cherry-picking.\n\n-- \nJakub Narebski\nPoland\n"},{"id":"82607","messageId":"19b271a20807080931w75430148u8c9778117c9fb6cc@mail.gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Joshua Roys","fromEmail":"roysjosh@gmail.com","sentAt":"2008-07-08T16:31:45Z","receivedAt":"2008-07-08T16:31:45Z","isPatch":false,"sender":{"key":"roysjosh@gmail.com","avatar":null},"body":"Hello,\n\nOn Mon, Jul 7, 2008 at 8:27 PM, Jakub Narebski wrote:\n> 1. GitTorrent (???)\n>\n> Student: Joshua Roys\n> Mentor: Sam Vilain\n>\n> There was short thread of me asking about project\n> http://thread.gmane.org/gmane.comp.version-control.git/83611\n> where I got gittorrent mailing list (no activity at least according to\n> list archive http://lists.utsl.gen.nz/pipermail/gittorrent/) and URL\n> for project repo / gitweb... which is currently down, so I cannot check\n> if there is anything here.\n>\n> What is the status of this project, please?\n>\n> http://www.codinghorror.com/blog/archives/001134.html (\"Don't Got Dark\")\n>\n\nIt's going slower than I would like, in part due to a two-week period\nwhen I was out of the country. Other than that, it's going well, I\nthink. Sam had a lot of the framework stuff done at the start, and\nI've just been working my way through it. My schedule has cleared up\nfor the remaining time, so things should move a little faster now.\n\nThe gitweb randomly gives 500/internal server errors, not sure why.\n\nJoshua Roys\n"},{"id":"82608","messageId":"20080708163430.GK6726@leksak.fem-net","threadId":"14334","inReplyTo":"200807081812.15651.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-07-08T16:34:30Z","receivedAt":"2008-07-08T16:34:30Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nJakub Narebski wrote:\n> > It is taking the commit message from the commit in the \"From .*\"\n> > line, does *not* change it in any way and then applies the changes using\n> > threeway merge.\n> \n> Not exactly. \"git am --rebasing\" still tries to first just *apply*\n> the patch, then (I think) it falls back on blob-id based 3way merge.\n\nThat's of course totaly right and what I've meant, but unfortunately not\nwhat I've written ;-)\n\n> > Keeping that in mind what about dealing with --rebasing like that:\n> > if --rebasing is given, git am simply generates\n> > \tpick \n> > lines, instead of\n> > \tpatch -3 -k \n> > as it is now (and this is not enough, as it seems).\n> \n> It is not.\n> \n> Nevertheless it would be I think better for ordinary patch based rebase\n> to fall back not on git-am 3way merge, but on cherry-pick based merge\n> (i.e. on pick).\n\nHmm, if I get you right you _partly_ agree with me in choosing \"pick\" for\nam --rebasing... But cherry-pick should only be chosen if a simple git-apply\nfailed first. Right?\n\nI just got another idea which could easily be done and perhaps is the\nright thing :)\nGenerating\n\tpatch -C -3 \n\nThis takes authorship and message from and does the usual\nthreeway-fallback behavior.\n\nWhat do you think?\n\n> But I agree that it would be nice to simplify '--rebasing' logic, for\n> example using patch or 2way merge to generate tree, and commit message\n> taken directly from commit, not via 'format-patch | am' pipeline.\n\nThat's right, but that would require me to hack around in git-rebase\nwhich I tried to avoid for now. :)\n\nRegards,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"82609","messageId":"alpine.DEB.1.00.0807081745040.18205@racer","threadId":"14334","inReplyTo":"19b271a20807080931w75430148u8c9778117c9fb6cc@mail.gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2008-07-08T16:45:29Z","receivedAt":"2008-07-08T16:45:29Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Tue, 8 Jul 2008, Joshua Roys wrote:\n\n> Hello,\n> \n> On Mon, Jul 7, 2008 at 8:27 PM, Jakub Narebski wrote:\n> > 1. GitTorrent (???)\n> >\n> > Student: Joshua Roys\n> > Mentor: Sam Vilain\n> >\n> > There was short thread of me asking about project\n> > http://thread.gmane.org/gmane.comp.version-control.git/83611\n> > where I got gittorrent mailing list (no activity at least according to\n> > list archive http://lists.utsl.gen.nz/pipermail/gittorrent/) and URL\n> > for project repo / gitweb... which is currently down, so I cannot check\n> > if there is anything here.\n> >\n> > What is the status of this project, please?\n> >\n> > http://www.codinghorror.com/blog/archives/001134.html (\"Don't Got Dark\")\n> >\n> \n> It's going slower than I would like, in part due to a two-week period\n> when I was out of the country. Other than that, it's going well, I\n> think. Sam had a lot of the framework stuff done at the start, and\n> I've just been working my way through it. My schedule has cleared up\n> for the remaining time, so things should move a little faster now.\n> \n> The gitweb randomly gives 500/internal server errors, not sure why.\n\nI thought you were working on the torrent stuff? What is the status on \nthat?\n\nCiao,\nDscho\n"},{"id":"82611","messageId":"20080708170059.GA32184@machine.or.cz","threadId":"14334","inReplyTo":"19b271a20807080931w75430148u8c9778117c9fb6cc@mail.gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2008-07-08T17:00:59Z","receivedAt":"2008-07-08T17:00:59Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":" Hi,\n\nOn Tue, Jul 08, 2008 at 12:31:45PM -0400, Joshua Roys wrote:\n> On Mon, Jul 7, 2008 at 8:27 PM, Jakub Narebski wrote:\n> > 1. GitTorrent (???)\n> >\n> > Student: Joshua Roys\n> > Mentor: Sam Vilain\n> >\n> > There was short thread of me asking about project\n> > http://thread.gmane.org/gmane.comp.version-control.git/83611\n> > where I got gittorrent mailing list (no activity at least according to\n> > list archive http://lists.utsl.gen.nz/pipermail/gittorrent/) and URL\n> > for project repo / gitweb... which is currently down, so I cannot check\n> > if there is anything here.\n> >\n> > What is the status of this project, please?\n> >\n> > http://www.codinghorror.com/blog/archives/001134.html (\"Don't Got Dark\")\n> >\n> \n> It's going slower than I would like, in part due to a two-week period\n> when I was out of the country. Other than that, it's going well, I\n> think. Sam had a lot of the framework stuff done at the start, and\n> I've just been working my way through it. My schedule has cleared up\n> for the remaining time, so things should move a little faster now.\n> \n> The gitweb randomly gives 500/internal server errors, not sure why.\n\n there's now a mirror at\n\n\thttp://repo.or.cz/w/VCS-Git-Torrent.git\n\n\t\t\t\tPetr \"Pasky\" Baudis\n"},{"id":"82612","messageId":"200807081922.48010.jnareb@gmail.com","threadId":"14334","inReplyTo":"alpine.DEB.1.00.0807081745040.18205@racer","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-08T17:22:46Z","receivedAt":"2008-07-08T17:22:46Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Johannes Schindelin wrote:\n> On Tue, 8 Jul 2008, Joshua Roys wrote:\n>> On Mon, Jul 7, 2008 at 8:27 PM, Jakub Narebski wrote:\n>>> 1. GitTorrent (???)\n\n>>> There was short thread of me asking about project\n>>> http://thread.gmane.org/gmane.comp.version-control.git/83611\n>>> where I got gittorrent mailing list [...] and URL\n>>> for project repo / gitweb... which is currently down, so I cannot check\n>>> if there is anything here.\n\n[cut]\n>> \n>> The gitweb randomly gives 500/internal server errors, not sure why.\n\n(I think it was web server error, as error message didn't look like\nit was coming from gitweb; besides gitweb up to some time ago didn't\nuse \"500 Internal Server Error\" HTTP error status code.)\n\n> I thought you were working on the torrent stuff? What is the status on \n> that?\n\nI think Jushua was referring here to the fact that gitweb for\nGitTorrent project repository[1] is sometimes down (it was down\nwhen I was writing initial email in this thread).\n\n[1] http://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent\n-- \nJakub Narebski\nPoland\n"},{"id":"82614","messageId":"200807081931.31764.jnareb@gmail.com","threadId":"14334","inReplyTo":"20080708163430.GK6726@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-08T17:31:31Z","receivedAt":"2008-07-08T17:31:31Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Hi!\n\nOn Wed, 8 July 2008, Stephan Beyer wrote:\n> Jakub Narebski wrote:\n>> Stephan Beyer wrote:\n>>>\n>>> It is taking the commit message from the commit in the \"From .*\"\n>>> line, does *not* change it in any way and then applies the changes using\n>>> threeway merge.\n>> \n>> Not exactly. \"git am --rebasing\" still tries to first just *apply*\n>> the patch, then (I think) it falls back on blob-id based 3way merge.\n> \n> That's of course totaly right and what I've meant, but unfortunately not\n> what I've written ;-)\n> \n>>> Keeping that in mind what about dealing with --rebasing like that:\n>>> if --rebasing is given, git am simply generates\n>>> \tpick \n>>> lines, instead of\n>>> \tpatch -3 -k \n>>> as it is now (and this is not enough, as it seems).\n>> \n>> It is not.\n>> \n>> Nevertheless it would be I think better for ordinary patch based rebase\n>> to fall back not on git-am 3way merge, but on cherry-pick based merge\n>> (i.e. on pick).\n> \n> Hmm, if I get you right you _partly_ agree with me in choosing \"pick\" for\n> am --rebasing... But cherry-pick should only be chosen if a simple git-apply\n> failed first. Right?\n\nRight.\n\n> I just got another idea which could easily be done and perhaps is the\n> right thing :)\n> Generating\n> \tpatch -C -3 \n> \n> This takes authorship and message from and does the usual\n> threeway-fallback behavior.\n> \n> What do you think?\n\nVery good idea (I have proposed something similar either here on in\nanother thread). It would avoid some unnecessary \"marshalling\" and\n\"unmarshalling\" which is needed to transfer commit message [unchanged]\nthrough git-format-patch -> git-am pipeline, namely putting first\nparagraph into subject line, generating then parsing RFC-2822 date,\nusing quoted printable encoding for first paragraph / subject header\n(I think).\n\nIt would be still better to fallback to _pick_, not \"git am --3way\",\nas the latter IIRC use _shortened_ _blob_ identifiers for pre- and\npost-image to find common ancestor (merge base) for 3way merge.\nWhich is not necessary as we can find merge base and base commits\neasier.\n\n-- \nJakub Narebski\nPoland\n"},{"id":"82637","messageId":"4873DAFA.6060502@vilain.net","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sam Vilain","fromEmail":"sam@vilain.net","sentAt":"2008-07-08T21:24:10Z","receivedAt":"2008-07-08T21:24:10Z","isPatch":false,"sender":{"key":"sam@vilain.net","avatar":"https://gravatar.com/avatar/8fc840ca854dbf6f7065b4335e3b934951c1dca3b11db688e95e471901f8f4a8?d=mp&s=160"},"body":"Jakub Narebski wrote:\n> 1. GitTorrent (???)\n>\n> Student: Joshua Roys\n> Mentor: Sam Vilain\n>\n> There was short thread of me asking about project\n> http://thread.gmane.org/gmane.comp.version-control.git/83611\n> where I got gittorrent mailing list (no activity at least according to \n> list archive http://lists.utsl.gen.nz/pipermail/gittorrent/) and URL \n> for project repo / gitweb... which is currently down, so I cannot check \n> if there is anything here.\n>\n> What is the status of this project, please?\n>\n> http://www.codinghorror.com/blog/archives/001134.html (\"Don't Got Dark\")\n> \n\nHi, thanks for your interest. I don't know when it was you checked - it \nmay have been during a brief network outage our admins needed to do \n(unfortunately utsl.gen.nz is on a grace-and-favour service arrangement \nwith its hosts, Catalyst), but the gitweb \n(http://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent) is currently running, and \nI hope you can see the current activity.\n\nI announced before that I would be sending an update shortly after the \nmid-term deadline, and I still plan on doing this. Please be patient \nand wait until that happens in a week or so.\n\nCheers,\nSam.\n"},{"id":"82703","messageId":"bd6139dc0807090318r4cbcac93s907871c6b45c03ac@mail.gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sverre Rabbelier","fromEmail":"alturin@gmail.com","sentAt":"2008-07-09T10:18:41Z","receivedAt":"2008-07-09T10:18:41Z","isPatch":false,"sender":{"key":"alturin@gmail.com","avatar":null},"body":"[Once again I forgot to \"reply to all\", sorry Jakub ;)]\n\nHeya,\n\nOn Tue, Jul 8, 2008 at 2:27 AM, Jakub Narebski wrote:\n> I'd very much like to have (or perhaps to wrote) some sort of interim\n> progress report for Google Summer of Code 2008 projects on\n> http://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to\n> expand and/or correct those mini-summaries below.\n\nWould you like us to go to the wiki page and edit them ourselves, yes?\nIf so, I'll see about kicking up something interesting to put up there\nsoon.\n\n> (It would be, I guess, good preparation for GSoC 2008 mid-term\n> evaluations, which according to GSoC 2008 timeline\n> http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> are to take place July 7 -- July 14.)\n\nAgreed.\n\n> 2. git-statistics (?)\n>\n> Student: Sverre Rabbelier\n> Mentor: David Symonds\n>\n> There were some posts about how git-statistics can be used:\n> http://thread.gmane.org/gmane.comp.version-control.git/81534\n> http://thread.gmane.org/gmane.comp.version-control.git/82027\n> but it was mainly theoretical thread\n>\n> The git-stats repository at http://repo.or.cz/w/git-stats.git (gitweb)\n> has some commits, but I don't remember any of them on git mailing list.\n> Not ready?\n\nI am not yet sure how I should send in my code for peer review.\n$ git rev-list master | wc -l\n151\nA bunch of these patches introduce significant changes, a few are\nfixes, and some are only changes to the changelog [0] which is\nactually a pretty good way to see what I've been up to. Credit here\ngoes to David for suggesting I keep one. I am contemplating sending in\none patch per module since I've structured my code in such a way that\neach module is mostly a stand-alone file that can be reviewed in\nitself. I am not sure if there is any way to get such code reviewed,\nmost other projects had several distinct steps (e.g., modifying parts\nof existing code first) they had to go through before they could write\nnew code. My project has been \"new code\" from the get go, so any\nadvice as to how to send this in for peer review is very welcome.\n\nI have been working on GitStats, an attempt to gather metrics on a git\nrepository. For those who would like to know more about my goals,\nplease have a look at [1] or the more general [2]. Currently I have\nquite a few metrics done, most of them are aimed at detecting whether\na commit is a bugfix. A short listing of metrics done:\n* stats.py author -d: Shows file activity of a specific developer\nmeasured in how often they made a modification to a file and total\nlines added/removed (much like a diffstat, but now for a specific\ndeveloper instead of one commit).\n* stats.py author -f: Shows file activity of a specific file meaured\nin how often they made a modification to a file, could be extended to\nalso count changes like \"author -d\"\n* stats.py branch -b: Shows which branch a specific file belongs to,\nfor more information on this metric see below\n* stats.py commit -v: Shows all commits that are reverted by the\nspecified commit, will be extended to allow detection of partial\nreverts\n* stats.py diff -e: Shows whether two specific commits introduce the\nsame changes\n* stats.py diff -e -n: ditto, but ignores what changes were made, only\nlooks at the changed lines\n* stats.py diff -e -i: ditto, but inverts the comparison, instead of\ncomparing addition with addition and deletions with deletions, the\nadditions of the first diff are compared with the deletions of the\nsecond diff, and vise versa. This way a revert is easily detected.\n* stats.py index -t: Shows which commits touched the same paths as the\nstaged changes\n\nI am working now on getting the 'is a bugfix' metric going, it's\nsub-metrics are mostly done ('branch contains' , 'is revert', 'commit\nmsg matches', 'commit diff matches') with the exception of a few\nsimple ones ('is partial revert', 'modifies same lines') that are\nalready almost done. As a sneak peek into what I've been up to I'll\nzoom in a bit on the development of the 'belongs to' metric.\n\nOne of the hardest things to tackle was the 'belongs to'\nmetric. It's goal is to determine how much a certain commit\n'belongs to' a specific branch. When aggregating this\nmetric one can pick the best fit. This is useful when\ntrying to figure out if a commit was made on a maintenance\nbranch, and as such whether it should be marked as a 'fix'.\nThe definition of \"belonging to\" a branch was made as follows: \"Each\nbranch that contains the target commit begins with a 'dilution' of 0,\nfor each merge on the way between it and the commit, increase the\ndilution by one if it was not the primary parent of the merge. This\nmeans that if a commit was \"merged into\" a branch instead of having\nbeen made on that branch, it will have a higher dilution than if the\ncommit was made on that branch. This way, the lower the dilution, the\nmore a commit belongs on a branch.\n\nThe problem with this metric was not in writing it, but in\ngetting it to not be terribly slow on large repositories.\nA few of the major problems include:\n* Do not call git-rev-list on each commit, but instead\n gather all 'parentage' information beforehand. This\n avoids a lot of calls to the git binary, which is\n expensive. This simple optimization resulted in a huge\n speed increase, even on the small tests.\n With this in place it runs in under a second for the test\n repository, but the git.git repo still takes ages.\n* Do not revisit commits that have already been checked\n _and had a lower or equal dilution_ in the previous\n check.\n Especially in the git.git repository this\n optimization made the algorithm go from 'unusable' to\n useful.\n With this in place it runs in under a second for recent\n commits. It takes under 20 seconds for the first commit\n in the repository.\n* Keep a global memory too, that is, when multiple branches\n contain the commit, remember that we have already visited\n a commit in a global memory as well. When visiting\n a commit we check if we already visited it before but\n with a lower dilution, if so, this means that another\n branch is always going to be better than the current\n branch at this point, as such, we stop walking there.\n With this in place it still runs in under a second for\n recent commits. It takes about 7 seconds for the first\n commit in the repository.\n* Do a rev-list on all branches we are interested in once,\n instead of once for each branch. In order to cut down the\n output by not listing 'subsets', the rev-lists of each\n individual branch was needed. By not doing this filtering\n a lot of time is saved.\n With this in place it still runs in under a second for\n recent commits. It takes a little over 5 seconds for the\n first commit in the repository.\n\nThe above timings were made on a rather outdated version of git.git,\nafter updating to today's latest the timings are as follows:\n\n$ time stats.py branch -b e83c516331\nMatching branches:\npu\nnext\nmaster\nmaint\noffcuts\n\nreal 0m6.360s\nuser 0m6.228s\nsys 0m0.244s\n\nIt is plausible that there is no way to do this any faster\nwith the current approach. Walking all commits, examining\nthem, and calculating the dilution, all the way to the root\ncommit (on the git.git repo) just takes that long. The\nalgorithm itself is almost instant, most of the time is\nspent waiting for git rev-list to return.\n\nFor a recent commit on the maint branch we can find the following\ninformation (the -d flag was included to include 'debug' information,\nso that we can see what other information was found, but left out in\nthe regular report):\n\nsverre@Laptop-Sverre:~/code/git$ time stats.py branch -b 2b2828b -d\nRetreiving branches that contain the commit...\nGathering information on the history...\nDone.\n\nChecking branches now:\norigin/next\nmaster\norigin/maint\norigin/pu\nDone.\n\nListing found metrics:\nBranch next, dilution: 2.\nBranch master, dilution: 1.\nBranch maint, dilution: 0.\nDone.\n\nMatching branches:\nThe minimal dilution is: 0\nmaint\n\nreal 0m6.431s\nuser 0m6.164s\nsys 0m0.260s\n\n>From the above one can see that master has recently merged in that\ncommit, but that next did not include it until after it merged in\nmaster. In this way a commit can cascade through multiple merge, earch\nmerge increasing it's dilution by one.\n\nI am very interested to hear comments on my progress so far, but also\non what is thought to be \"important to work on next\". It would be\nawesome if a few people could give it a test drive. I recommend using\nthe setupRepo.py script in 'src/scripts' which will create a\nmetricsrepo in /tmp/ that is very well suited to experiment with the\n'belongs to' metric. The testrepo also created in /tmp is better\nsuited to test some of the other metrics. Both repositories are used\nby the testcases under 'src/t', which should all pass :). My\nrepository can be found at [3].\n\nThank you for reading, and I'm looking forward to review/comments.\n\n[0] http://repo.or.cz/w/git-stats.git?a=blob;f=doc/changelog.txt\n[1] http://alturin.googlepages.com/Use_cases.html\n[2] http://alturin.googlepages.com/gsoc2008\n[3] http://repo.or.cz/w/git-stats.git\n\n--\nCheers,\n\nSverre Rabbelier\n"},{"id":"82705","messageId":"20080709105623.GP10347@genesis.frugalware.org","threadId":"14334","inReplyTo":"bd6139dc0807090318r4cbcac93s907871c6b45c03ac@mail.gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Miklos Vajna","fromEmail":"vmiklos@frugalware.org","sentAt":"2008-07-09T10:56:23Z","receivedAt":"2008-07-09T10:56:23Z","isPatch":false,"sender":{"key":"vmiklos@frugalware.org","avatar":"https://gravatar.com/avatar/401c1cbbb3a5d13e650c691a2c71d6fd0b80df1a01bc74d9f1972675dd58f2bd?d=mp&s=160"},"body":"On Wed, Jul 09, 2008 at 12:18:41PM +0200, Sverre Rabbelier wrote:\n> I am not yet sure how I should send in my code for peer review.\n> $ git rev-list master | wc -l\n> 151\n\nWhat I did squashing fixes regularly, but I keep a 'history' branch\nwhere I always reference to the old HEAD, so others can still have a\nlook at the individual commits.\n\nIt's like:\n\n$ cat rebase-builtin-merge.sh\n#!/bin/sh -e\n\n[ -n \"$*\" ]\n\nold_head=$(git rev-parse HEAD)\ngit rebase $*\ngit update-ref refs/heads/rebase-history \\\n $(echo \"Rebased with 'git rebase $*'\" | \\\n git commit-tree HEAD^{tree} -p rebase-history -p $old_head -p HEAD)\n\nIt will not work properly if you get conflicts, but you got the idea.\n"},{"id":"82707","messageId":"200807091336.16965.jnareb@gmail.com","threadId":"14334","inReplyTo":"bd6139dc0807090318r4cbcac93s907871c6b45c03ac@mail.gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-09T11:36:16Z","receivedAt":"2008-07-09T11:36:16Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Wed, 9 July 2008, Sverre Rabbelier wrote:\n> On Tue, Jul 8, 2008 at 2:27 AM, Jakub Narebski wrote:\n> >\n> > I'd very much like to have (or perhaps to wrote) some sort of interim\n> > progress report for Google Summer of Code 2008 projects on\n> > http://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to\n> > expand and/or correct those mini-summaries below.\n> \n> Would you like us to go to the wiki page and edit them ourselves, yes?\n\nYes, thanks in advance.\n\n> If so, I'll see about kicking up something interesting to put up there\n> soon.\n\nI'd rather students or mentors did write short summary on mentioned\nwiki page, but I'll try to come up with some short summary for each\nGSoC2008 project (if it won't be there already) after July 14.\n\n\n[thanks for summary]\n-- \nJakub Narebski\nPoland\n"},{"id":"84084","messageId":"200807210029.31543.jnareb@gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-20T22:29:29Z","receivedAt":"2008-07-20T22:29:29Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Tue, 8 July 2008, Jakub Narebski wrote:\n\n> I'd very much like to have (or perhaps to wrote) some sort of interim \n> progress report for Google Summer of Code 2008 projects on \n> http://git.or.cz/gitwiki/SoC2008Projects. Therefore I'd like you to \n> expand and/or correct those mini-summaries below.\n\nHere is a bit late summary of this thread (and of information gathered\nelsewhere). I'll try to add this information to wiki page in approx\ntwo days from now... of course unless project student or mentor wouldn't\ndo it first.\n \n\n1. GitTorrent\n \nStudent: Joshua Roys\nMentor: Sam Vilain\n\nI never got more response than \"it is going slower than I would like, \n[...] Other than that, it's going well, I think.\" from Joshua Roys.\n\nMailing list archives for gittorrent mailing list doesn't show anything \ninteresting, either (last post is from 2007).\n http://lists.utsl.gen.nz/pipermail/gittorrent/\n\nBesides canonical repository gitweb\n http://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent\nthere is also mirror at\n http://repo.or.cz/w/VCS-Git-Torrent.git\n\nThere is some activity there... but no summary of it anywhere I could \nfind. (I wonder if this was the project Johannes and Shawn were talking \nabout of \"going dark\" in GSoC 2008 podcast 018...)\n\n\n2. git-statistics\n \nStudent: Sverre Rabbelier\nMentor: David Symonds\n \nThere were some posts about how git-statistics can be used:\n http://thread.gmane.org/gmane.comp.version-control.git/81534\n http://thread.gmane.org/gmane.comp.version-control.git/82027\nThere is post with link to different git.git statictics:\n \"[GitStats] Bling bling or some statistics on the git.git repository\"\n http://thread.gmane.org/gmane.comp.version-control.git/88174\n\nA short listing of metrics done:\n* stats.py author -d: Shows file activity of a specific developer\n measured in how often they made a modification to a file and total\n lines added/removed (much like a diffstat, but now for a specific\n developer instead of one commit).\n* stats.py author -f: Shows file activity of a specific file measured\n in how often they made a modification to a file, could be extended to\n also count changes like \"author -d\"\n* stats.py branch -b: Shows which branch a specific file belongs to,\n for more information on this metric see below\n* stats.py commit -v: Shows all commits that are reverted by the\n specified commit, will be extended to allow detection of partial\n reverts\n* stats.py diff -e: Shows whether two specific commits introduce the\n same changes\n* stats.py diff -e -n: ditto, but ignores what changes were made, only\n looks at the changed lines\n* stats.py diff -e -i: ditto, but inverts the comparison, instead of\n comparing addition with addition and deletions with deletions, the\n additions of the first diff are compared with the deletions of the\n second diff, and vise versa. This way a revert is easily detected.\n* stats.py index -t: Shows which commits touched the same paths as the\n staged changes\n\n\n3. Gitweb caching\n \nStudent: Lea Wiemann\nMentor: John 'warthog' Hawley\n\nLea has chosen caching data and memcached as example (primary) caching \nengine, wrote Git::Repo object-oriented Perl interface to git, used\nit in gitweb, and added caching to gitweb.\n\nAdditionally tests for old Git.pm (simple), Git::Repo and friends, and\nMechanize based gitweb test were added. Mechanize tests detected a few\nbugs in current gitweb code: using Git::Repo and adding caching didn't \ncreate any new errors.\n\nCurrently first round of patches were send, and second version \nincorporating comments from is in progress. There is a test site\n(live demo) up and running on one of the kernel.org machines:\n http://odin3.kernel.org/git-lewiemann/\n\nYou can find first version of patches in 'pu' branch.\n\n\n4. Eclipse plugin push support\n \nStudent: Marek Zawirski\nMentor: Shawn O. Pearce\n\nThus far Marek has completed generation of packs, including delta\nre-use from packs using either v1 or v2 index, including taking\nadvantage of the CRC inside the v2 index to accelerate a safe reuse.\nThis code permits jgit to write a valid pack.\n\nThe packing code does not (yet) search for a delta base, or create\na new delta for an undeltified object. Packing loose objects packs\nthem as whole objects in the pack file, resulting in little to no\nreduction over their loose object size. This is not a limitation\nof Java. Marek and I simply decided that protocol support was more\nimportant than really tight network transport at this point in time.\n\nAs a result of being able to create pack files Marek was able to\nimplement the client side of git-push for the native pack transfer\nservice, aka push over SSH, push to another local repository (by\nforking 'git receive-pack') and push over anonymous git://.\n\nUsing Marek's pack generation code Shawn added support for push over\nthe dumb sftp:// and amazon-s3:// protocols, with the latter also\nsupporting transparent client side encryption.\n\n\n5. git-merge builtin\n\nStudent: Miklos Vajna\nMentor: Johannes Schindelin\n\n> In \"What's cooking in git.git (topics)\" Junio C Hamano wrote:\n> \n> It already is beginning to become clear what 1.6.0 will look like.\n> [...]\n> * git-merge will be rewritten in C.\n\nIt is already in git as 1c7b76b (Build in merge). \nIn Documentation/RelNotes-1.6.0.txt you can find the following:\n \"* git-merge has been reimplemented in C.\"\n \n\n6. git-sequencer\n\nStudent: Stephan Beyer\nMentor: Christian Couder, Daniel Barkalow\n \nIt started with discussion over TODO file format:\n http://thread.gmane.org/gmane.comp.version-control.git/84230\n \nNow there is prototype shell script implementation (which has some \nlimitations because it is prototype) in \"git sequencer prototype\"\n http://thread.gmane.org/gmane.comp.version-control.git/86985\n http://thread.gmane.org/gmane.comp.version-control.git/88754\n\nThe latter thread includes migration of git-am, git-rebase, and \ninteractive rebase to sequencer. \n\n> Stephan Beyer wrote:\n> I'm using sequencer-based git-am and git-rebase-i and also \n> git-sequencer itself for around 2-3 weeks now. So, for me, it is\n> reality-proven [...]\n\nThere were some problems with sequencer based implementation of\n\"git am --rebasing\", or sequencer based patch application driven\nordinary rebase, but I think there were resolved.\n\nStephen have started the builtin sequencer (but till now no patches\nwere sent to list: seems to be work in progress). \n\nSome performance benchmarks:\n * applying 45 patches with git-am \n - 3 seconds using the original \n - 8 seconds using the (scripted) sequencer-based one\n * rebasing 100 commits\n - 4.8 seconds using the original\n - 10.1 seconds using the (scripted) sequencer-based one\n - 1.7 seconds using builtin sequencer\n\n\nSUMMARY:\n========\nFrom those projects, \"git-merge builtin\" did what it was meant to do \nalready. \"Eclipse plugin push support\" and \"git-statistics\" did \nminimum what it was meant to do already, and it looks like it would be \nfinished before August 11. \"Gitweb caching\" is after first round of \npatches, \"git-sequencer\" looks like already done; I don't know what is \nthe state of \"GitTorrent\" project.\n\nPlease correct any mistakes in this summary / writeup. Thanks in \nadvance.\n-- \nJakub Narebski\nPoland\n"},{"id":"84089","messageId":"bd6139dc0807201543r6fd718d2rfbc8ad100cc75f2c@mail.gmail.com","threadId":"14334","inReplyTo":"200807210029.31543.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sverre Rabbelier","fromEmail":"alturin@gmail.com","sentAt":"2008-07-20T22:43:16Z","receivedAt":"2008-07-20T22:43:16Z","isPatch":false,"sender":{"key":"alturin@gmail.com","avatar":null},"body":"On Mon, Jul 21, 2008 at 12:29 AM, Jakub Narebski wrote:\n> Here is a bit late summary of this thread (and of information gathered\n> elsewhere). I'll try to add this information to wiki page in approx\n> two days from now... of course unless project student or mentor wouldn't\n> do it first.\n\nI've been feeling guilty about not writing a summary yet, sorry for that :(.\n\n> 2. git-statistics\n>\n> Student: Sverre Rabbelier\n> Mentor: David Symonds\n>\n> There were some posts about how git-statistics can be used:\n> http://thread.gmane.org/gmane.comp.version-control.git/81534\n> http://thread.gmane.org/gmane.comp.version-control.git/82027\n> There is post with link to different git.git statictics:\n> \"[GitStats] Bling bling or some statistics on the git.git repository\"\n> http://thread.gmane.org/gmane.comp.version-control.git/88174\n>\n> A short listing of metrics done:\n> * stats.py author -d: Shows file activity of a specific developer\n> measured in how often they made a modification to a file and total\n> lines added/removed (much like a diffstat, but now for a specific\n> developer instead of one commit).\n> * stats.py author -f: Shows file activity of a specific file measured\n> in how often they made a modification to a file, could be extended to\n> also count changes like \"author -d\"\n\n* stats.py bug -t: Calculates a 'score' for a specific commit,\nrepresenting how likely it is to be a bugfix. There are four metrics\nused to determine this: \"Commits x reverts another commit y\", \"Commit\nx belongs to one of the specified branches (e.g., 'maint')\", \"The diff\nfor commit x contains a specific phrase\", \"The msg for commit x\ncontains a specific phrase\". A rating can be given to each metric by\nthe user.\n* stats.py bug -a: Aggregates the 'bug -t' metric over a range of\ncommits. The default is the last 10 commits on HEAD. This routine is\noptimized to not recalculate metrics/to not redo work that was\nalready done in a 'bug -t' call for another commit. As such there is a\nset setup time to determine the type of the first commit, after which\ncalculation time increases at a much lower pace. (So 'bug -a' on 10\ncommits might take 4 seconds, and running it over 11 commits might\ntake only 4.5\".)\n\n> * stats.py branch -b: Shows which branch a specific file belongs to,\n\na commit 'belongs to' a branch when the commit is made on that branch.\n\n> * stats.py commit -v: Shows all commits that are reverted by the\n> specified commit, will be extended to allow detection of partial\n> reverts\n\nThis has been moved to 'diff -r'.\n\n> * stats.py diff -e: Shows whether two specific commits introduce the\n> same changes\n> * stats.py diff -e -n: ditto, but ignores what changes were made, only\n> looks at the changed lines\n> * stats.py diff -e -i: ditto, but inverts the comparison, instead of\n> comparing addition with addition and deletions with deletions, the\n> additions of the first diff are compared with the deletions of the\n> second diff, and vise versa. This way a revert is easily detected.\n> * stats.py index -t: Shows which commits touched the same paths as the\n> staged changes\n\nI think the rest is a nice summary of what I've been doing :).\n\n\n> SUMMARY:\n> ========\n> [...] \"git-statistics\" did a minimum what it was meant to do already, and it looks like it would be finished before August 11. [..]\n\nMy deadline is August 1 since my vacation starts then and I can't work\nduring my vacation (all hail American tax laws), but David is\nconfident that I will have a finished product before then, and I plan\nnot to let him down on that expectation.\n\n> Please correct any mistakes in this summary / writeup.\n\nI tried to as best as I could :).\n\n> Thanks in advance.\n\nNo, thank you! Thanks for writing up this summary, very nicely done.\n\n-- \nCheers,\n\nSverre Rabbelier\n"},{"id":"84093","messageId":"20080720225722.GC5950@leksak.fem-net","threadId":"14334","inReplyTo":"200807210029.31543.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-07-20T22:57:22Z","receivedAt":"2008-07-20T22:57:22Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nwhat you write is quite detailed :)\n\nJakub Narebski wrote:\n> 6. git-sequencer\n> \n[...]\n> There were some problems with sequencer based implementation of\n> \"git am --rebasing\", or sequencer based patch application driven\n> ordinary rebase, but I think there were resolved.\n\nThey were resolved, but there is a new problem that occured in the\nam --abort thread of Junio: the sequencer-based git-am does not work\non dirty working tree.\n\nI've also fixed some other minor issues and have not yet sent this to\nthe list (because I think I stumble over even more while writing the\nbuiltin-sequencer.)\n\n> Stephen have started the builtin sequencer (but till now no patches\n> were sent to list: seems to be work in progress).\n\nRight.\n\n> Some performance benchmarks:\n> * applying 45 patches with git-am \n> - 3 seconds using the original \n> - 8 seconds using the (scripted) sequencer-based one\n> * rebasing 100 commits\n> - 4.8 seconds using the original\n> - 10.1 seconds using the (scripted) sequencer-based one\n> - 1.7 seconds using builtin sequencer\n\n:)\nI think I'm going to format-patch the same 100 test commits and then I\nchange the \"applying 45 patches with git-am\" part on the Wiki.\n\nRegards.\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"84119","messageId":"1216601739.6523.48.camel@maia.lan","threadId":"14334","inReplyTo":"200807210029.31543.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sam Vilain","fromEmail":"sam@vilain.net","sentAt":"2008-07-21T00:55:39Z","receivedAt":"2008-07-21T00:55:39Z","isPatch":false,"sender":{"key":"sam@vilain.net","avatar":"https://gravatar.com/avatar/8fc840ca854dbf6f7065b4335e3b934951c1dca3b11db688e95e471901f8f4a8?d=mp&s=160"},"body":"On Mon, 2008-07-21 at 00:29 +0200, Jakub Narebski wrote:\n> 1. GitTorrent\n> \n> Student: Joshua Roys\n> Mentor: Sam Vilain\n> \n> I never got more response than \"it is going slower than I would like, \n> [...] Other than that, it's going well, I think.\" from Joshua Roys.\n\n> Mailing list archives for gittorrent mailing list doesn't show anything \n> interesting, either (last post is from 2007).\n> http://lists.utsl.gen.nz/pipermail/gittorrent/\n\nThat's a valid complaint. I've posted a summary of the project status\nthere, and will keep as much related discussion as appropriate on-list\nfrom here.\n\n> Besides canonical repository gitweb\n> http://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent\n> there is also mirror at\n> http://repo.or.cz/w/VCS-Git-Torrent.git\n> \n> There is some activity there... but no summary of it anywhere I could \n> find.\n\ngit-log | git-shortlog? ;)\n\nSam.\n"},{"id":"84121","messageId":"alpine.DEB.1.00.0807210300210.3305@eeepc-johanness","threadId":"14334","inReplyTo":"1216601739.6523.48.camel@maia.lan","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2008-07-21T01:05:18Z","receivedAt":"2008-07-21T01:05:18Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Mon, 21 Jul 2008, Sam Vilain wrote:\n\n> On Mon, 2008-07-21 at 00:29 +0200, Jakub Narebski wrote:\n> > 1. GitTorrent\n> > \n> > [...]\n> >\n> > Besides canonical repository gitweb\n> > http://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent\n> > there is also mirror at\n> > http://repo.or.cz/w/VCS-Git-Torrent.git\n> > \n> > There is some activity there... but no summary of it anywhere I could \n> > find.\n> \n> git-log | git-shortlog? ;)\n\nPlease note that one of the reasons why last year's libgit-thin project \nhas not been merged, or even been useful to anybody else, because it has \nbeen developed in too private a manner. Sad.\n\nI would appreciate it, therefore, to keep the progress way more public \nthan it is now. That is, either you or your student should be not \ninvisible.\n\nCiao,\nDscho\n"},{"id":"84132","messageId":"20080721032203.GB2275@spearce.org","threadId":"14334","inReplyTo":"200807210029.31543.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2008-07-21T03:22:03Z","receivedAt":"2008-07-21T03:22:03Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Jakub Narebski wrote:\n> 1. GitTorrent\n> \n> Student: Joshua Roys\n> Mentor: Sam Vilain\n...\n> There is some activity there... but no summary of it anywhere I could \n> find. (I wonder if this was the project Johannes and Shawn were talking \n> about of \"going dark\" in GSoC 2008 podcast 018...)\n\nYes, this is the project I referred to in the podcast about going dark.\n \n> 4. Eclipse plugin push support\n> \n> Student: Marek Zawirski\n> Mentor: Shawn O. Pearce\n> \n> [...] Marek and I simply decided that protocol support was more\n> important than really tight network transport at this point in time.\n\nCorrection, the \"I\" in \"Marek and I\" isn't Jakub, its Shawn. This\nis just an editing mistake due to copy and paste from earlier thread.\nApparently my original paragraph here was already a nice summary of\nthe projects decisions thus far.\n\nI'm likely to be offline much of the rest of this week (I got lucky\nand found an open access point just now) but Marek is actively\nworking on user interface for push support, and I think if he finds\nthe time is considering adding delta generation. That will be a\nlot more time consuming as I think he needs to go back to original\nacademic papers to learn an LCS algorithm and then implement that.\n\nCopyright and licenses around libxdiff and delta-diff.c won't allow\nus to directly port the diff code to Java and our BSD license. No,\nI don't want to start a BSD-GPL license war. Our decision to go\nBSD in jgit may be a thorn in our side in areas such as this, but it\nis probably better for our long-term goals of working more directly\nwith the Eclipse Foundation and perhaps also the NetBeans folks.\n\n> SUMMARY:\n> ========\n> From those projects, \"git-merge builtin\" did what it was meant to do \n> already. \"Eclipse plugin push support\" and \"git-statistics\" did \n> minimum what it was meant to do already, and it looks like it would be \n> finished before August 11. \"Gitweb caching\" is after first round of \n> patches, \"git-sequencer\" looks like already done; I don't know what is \n> the state of \"GitTorrent\" project.\n> \n> Please correct any mistakes in this summary / writeup. Thanks in \n> advance.\n\nI think this is a pretty good summary. I want to go through the\nmid-term evaluations and summarize those for the mailing list but\nI have not had a chance to do that yet. With network being spotty\nfor the rest of this week it probably won't happen until the weekend.\n\nI think the quick summary is our students and our mentors think\ntheir projects are going well. Jakub's summary above suggests\nvery much the same thing. Its hard to claim a GSoC project isn't\nmeeting its goals when the code is already merged, or is at least\nunder active patch review. ;-)\n\n-- \nShawn.\n"},{"id":"84171","messageId":"200807211223.45707.jnareb@gmail.com","threadId":"14334","inReplyTo":"1216601739.6523.48.camel@maia.lan","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-07-21T10:23:45Z","receivedAt":"2008-07-21T10:23:45Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Mon, 21 July 2008, Sam Vilain wrote:\n> On Mon, 2008-07-21 at 00:29 +0200, Jakub Narebski wrote:\n> >\n> > 1. GitTorrent\n> > \n> > Student: Joshua Roys\n> > Mentor: Sam Vilain\n> > \n> > I never got more response than \"it is going slower than I would like, \n> > [...] Other than that, it's going well, I think.\" from Joshua Roys.\n> \n> > Mailing list archives for gittorrent mailing list doesn't show anything \n> > interesting, either (last post is from 2007).\n> > http://lists.utsl.gen.nz/pipermail/gittorrent/\n> \n> That's a valid complaint. I've posted a summary of the project status\n> there, and will keep as much related discussion as appropriate on-list\n> from here.\n\nYou can find it on GMane, too:\n http://thread.gmane.org/gmane.comp.version-control.git.gittorrent/1\n\nIt would be nice if it was send to git mailing list too, perhaps\ndirecting replies to gittorrent mailing list only by default.\n\nA short summary of this thread (please correct me if I am wrong):\n\n * Tracker: representing \"metainfo\" files, working tracker (test)\n is in progress\n\n * Core infrastructure: \"Commit Reel\" sorting algorithm implemented[1]\n B-tree index for fast querying.\n\n To be done is determining minimal set of edge objects which define\n reel (that can be passed to \"git rev-list --objects-edge\" to get thin\n pack representing reel).\n\n * Peer to Peer: handshake and two trivial messages[2] implemented.\n To be done are (minimally) \"References\", \"Blocks\", \"Request\" and\n \"Play\" messages.\n\nFootnotes:\n==========\n[*1*] \"the current implementation is quite slow (requiring two calls to\n'git-cat-file' for each object)\" <-- why you don't use '--batch' or\n'--batch-check' options to git cat file: see also Git::Repo and friends\nimplementation send by Lea Wiemann to git mailing list as part of\n\"Gitweb caching\" project? BTW. by keeping discussion off the list,\nyou are off the knowledge of git community, too.\n\n[*2*] _What_ are those \"two trivial messages\"?\n\n-- \nJakub Narebski\nPoland\n"},{"id":"84172","messageId":"20080721104032.GI32184@machine.or.cz","threadId":"14334","inReplyTo":"200807211223.45707.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2008-07-21T10:40:32Z","receivedAt":"2008-07-21T10:40:32Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"On Mon, Jul 21, 2008 at 12:23:45PM +0200, Jakub Narebski wrote:\n> [*1*] \"the current implementation is quite slow (requiring two calls to\n> 'git-cat-file' for each object)\" <-- why you don't use '--batch' or\n> '--batch-check' options to git cat file: see also Git::Repo and friends\n> implementation send by Lea Wiemann to git mailing list as part of\n> \"Gitweb caching\" project? BTW. by keeping discussion off the list,\n> you are off the knowledge of git community, too.\n\nI concur. Only now I realized that this project might be an important\nGit Perl API user. Can you please reconsider having a separate mailing\nlist for discussions etc.? The traffic seems to be very low anyway, and\nI'm not sure what benefits does it actually have at all.\n\n\t\t\t\tPetr \"Pasky\" Baudis\n"},{"id":"84191","messageId":"48848DCF.30803@gmail.com","threadId":"14334","inReplyTo":"20080721104032.GI32184@machine.or.cz","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Joshua Roys","fromEmail":"roysjosh@gmail.com","sentAt":"2008-07-21T13:23:27Z","receivedAt":"2008-07-21T13:23:27Z","isPatch":false,"sender":{"key":"roysjosh@gmail.com","avatar":null},"body":"Petr Baudis wrote:\n> On Mon, Jul 21, 2008 at 12:23:45PM +0200, Jakub Narebski wrote:\n>> [*1*] \"the current implementation is quite slow (requiring two calls to\n>> 'git-cat-file' for each object)\" <-- why you don't use '--batch' or\n>> '--batch-check' options to git cat file: see also Git::Repo and friends\n>> implementation send by Lea Wiemann to git mailing list as part of\n>> \"Gitweb caching\" project? BTW. by keeping discussion off the list,\n>> you are off the knowledge of git community, too.\n> \n> I concur. Only now I realized that this project might be an important\n> Git Perl API user. Can you please reconsider having a separate mailing\n> list for discussions etc.? The traffic seems to be very low anyway, and\n> I'm not sure what benefits does it actually have at all.\n> \n> \t\t\t\tPetr \"Pasky\" Baudis\n\nThank you for that suggestion! I made it call cat-file only once in \nbatch mode and it goes quite a bit faster now.\n\nJosh\n"},{"id":"87141","messageId":"200808140457.56464.jnareb@gmail.com","threadId":"14334","inReplyTo":"200807080227.43515.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-08-14T02:57:54Z","receivedAt":"2008-08-14T02:57:54Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Now that according to Google Summer of Code 2008 timeline\n http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\nwe are in the middle of the suggested 'pencils down' date (August 11 --\n-- August 18), I and perhaps other on git mailing list would like\nto know what is the current status of all git GSoC 2008 projects.\nI think that writing it down would help GSoC manager and co-manager...\n\nBelow there are my impressions about state of various Git's Google\nSummer of Code 2008 projects, in the sequence the are on Git Wiki\nhttp://git.or.cz/gitwiki/SoC2008Projects (I guess it would be nice\nto have final information there after final 'pencils down' date at\n18 August 2008).\n\n1. GitTorrent\n \nStudent: Joshua Roys\nMentor: Sam Vilain\n\nIf I remember correctly at midterm it was deemed to be somewhat late;\nmetainfo was done, tracker was in works, some core infrastructure\nand beginnings of peer to peer:\n http://thread.gmane.org/gmane.comp.version-control.git.gittorrent/1\n\nUnfortunately this project, even that is as much tied with git as StGIT,\nor egit/jgit, or git-gui or gitk, all of which use git mailing list for \ndiscussion and patches, choose to have its own separate mailing list; \nmoreover I think most of discussion was kept private.\n\nStatus: I have no idea how close GitTorrent is to completion (where by \ncompletion I mean ready, tested and benchmarked code running e.g. on \nkernel.org). I'm not sure if it is meant to be incorporated in git, \neven in contrib, or remain separate like StGIT, TopGit or jgit.\n\nStudent retention: One of reasons in Git application for participating \nin GSoC was having \"fresh blood\", new long-time contributors. I have \nvague notion that Joshua is quite active not only with GitTorrent, and \nwould stay git contributor...\n\n\n2. git-statistics\n\nStudent: Sverre Rabbelier\nMentor: David Symonds\n\nGitStat development is finished at least with respect to GSoC 2008, see\nhttp://thread.gmane.org/gmane.comp.version-control.git/90691 (for the \ntax reasons), although I guess its development will continue. \n\nStatus: Finished, I think also accepted: what is left is to put it \neventually in 'contrib/gitstat' or 'contrib/stats'\n\nStudent retention: Sverre has said if I remember correctly that he wants \nto work on improving '--follow', which now works only for very simple \nhistories, for GitStats to be better among others.\n\n\n3. Gitweb caching\n\nStudent: Lea Wiemann\nMentor: John 'warthog' Hawley\n\nThere are new tests for gitweb (to check if caching would not break \nanything new: it did caught a few breakages), new object Perl API to \ngit, and gitweb caching implemented using caching data at the level \nslightly above calling git commands. But full code (tests, Perl API \nand changes to gitweb) are only after first, maybe second round of \nreview. In short, it looks like it is 90% done, as in: \"The first \nninety percent of the task takes ninety percent of the time, and the \nlast ten percent takes the other ninety percent.\"\n\nThere is a test site (live demo) up and running on one of the kernel.org \nmachines:\n http://odin3.kernel.org/git-lewiemann/\n\nStatus: Seems to done, but: not yet merged in, neither used at \nkernel.org or repo.or.cz (i.e. not as fork of gitweb), and no \nbenchmarks.\n\nStudent retention: unknown.\n\n\n4. Eclipse plugin push support\n \nStudent: Marek Zawirski\nMentor: Shawn O. Pearce\n\nJGit can now create (suboptimal) packs: it can reuse delta, but cannot \ncurrently create one. This is used to implement push support in \njgit/egit.\n\nStatus: Done, merged in, and used as example to add for example Amazon's \nS3 support and SFTP transport support.\n\nStudent retention: I don't know, but I guess it is likely.\n\n\n5. git-merge builtin (!!!)\n\nStudent: Miklos Vajna\nMentor: Johannes Schindelin\n\nIn \"What's cooking in git.git (topics)\" Junio C Hamano wrote:\n\n It already is beginning to become clear what 1.6.0 will look like.\n [...]\n * git-merge will be rewritten in C.\n\nStatus: Done and merged in.\nStudent retention: Most likely.\n\n \n6. git-sequencer\n\nStudent: Stephan Beyer\nMentor: Christian Couder, Daniel Barkalow\n\nThere was discussion about TODO file format, there is prototype shell \nscript implementation, and reimplementing git-rebase--interactive and \nother using it; there is built-in sequencer done or almost done, but I \ndon't remember it being sent to git mailing list for review. \nBenchmarks show performance improvements for built-in sequencer \nversions.\n\nStatus: AFAIK close to be done. I don't know about it being merged-in.\nStudent retention: Likely.\n\n-- \nJakub Narebski\nPoland\n"},{"id":"87187","messageId":"1218717743.11530.99.camel@maia.lan","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sam Vilain","fromEmail":"sam@vilain.net","sentAt":"2008-08-14T12:42:23Z","receivedAt":"2008-08-14T12:42:23Z","isPatch":false,"sender":{"key":"sam@vilain.net","avatar":"https://gravatar.com/avatar/8fc840ca854dbf6f7065b4335e3b934951c1dca3b11db688e95e471901f8f4a8?d=mp&s=160"},"body":"On Thu, 2008-08-14 at 04:57 +0200, Jakub Narebski wrote:\n> 1. GitTorrent\n> \n> Student: Joshua Roys\n> Mentor: Sam Vilain\n> \n> If I remember correctly at midterm it was deemed to be somewhat late;\n> metainfo was done, tracker was in works, some core infrastructure\n> and beginnings of peer to peer:\n> http://thread.gmane.org/gmane.comp.version-control.git.gittorrent/1\n> \n> Unfortunately this project, even that is as much tied with git as\n> StGIT,\n> or egit/jgit, or git-gui or gitk, all of which use git mailing list\n> for \n> discussion and patches, choose to have its own separate mailing list; \n> moreover I think most of discussion was kept private.\n> \n> Status: I have no idea how close GitTorrent is to completion (where\n> by \n> completion I mean ready, tested and benchmarked code running e.g. on \n> kernel.org). I'm not sure if it is meant to be incorporated in git, \n> even in contrib, or remain separate like StGIT, TopGit or jgit.\n\nThe scope outlined in the GitTorrent proposal was a little bit more of a\nresearch project; being ready for production use on kernel.org or having\ncode ready to merge to git was not directly a deliverable.\n\nThe approved idea at the outset of *this* project was to try out the RFC\nprotocol design as it stands, iron out the weaknesses and see how it\nperforms. As it was pointed out along the way, there are other simpler\ndesigns possible, but I tried to explain how the fully denormalized\ndesign of the current RFC gives it more legs compared to other\napproaches. ie the goal was to prove the proposed network design, and\ntest it and show it to be as resource efficient as the more pragmatic\ndesigns proposed by people like Johannes (or, indeed, earlier versions\nfrom Jonas which were much closer to bittorrent in design).\n\nI was certainly hoping that at or shortly after the end it would at that\npoint also be good enough to be useful for production sites, even\nkernel.org sized (perhaps requiring the odd piece rewritten in C). We\njust need to crawl before we can walk, so to speak.\n\nI think saying \"we're on track to prove the protocol\" might be a little\nof an exaggeration at this point. But we have got a test script that is\nperforming a replication between two test directories, transmitting data\nover sockets using GTP.\n\nOnce Joshua has wrapped up this work to make it \"less hackish\" to\ndrive ;) then we can review the implementation and sign off on\ncompletion of what was technically set out to complete. At that point\nthe design becomes open to suggestions from the floor again.\n\nAnyway, that's the current status at this time. Thanks for the prod to\nupdate, Jakub. Come the actual completion date of the program, there\nwill be complete news.\n\nCheers,\nSam.\n\n---\nand now the rant:\n\nPlease bear in mind that some of the \"why Didn't You Just...\" type\ncomments are really \"20/20 hindsight\" now. And please, comments about\nwhat was made public and what not are also not useful. The initial\ndesign decisions between Jonas and myself are mostly in the RFC in\ncomments and in history from gittorrent.utsl.gen.nz. Sure we chatted on\nIRC about it here and there as well. But the RFC has had a public\nprofile since at least last year's GSoC when our last student was\naccepted to build it, but few people found time to comment on it. I\nhave made every effort to keep everything useful in the public arena and\nas far as I know never turned down the opportunity to answer questions.\nAnd snide remarks about my choice to press a few buttons on my list\nserver to make a new mailing list for a project I was undertaking I\nsimply have no time for.\n"},{"id":"87265","messageId":"alpine.DEB.1.00.0808150103430.24820@pacific.mpi-cbg.de.mpi-cbg.de","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2008-08-14T23:04:44Z","receivedAt":"2008-08-14T23:04:44Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Thu, 14 Aug 2008, Jakub Narebski wrote:\n\n> 5. git-merge builtin (!!!)\n> \n> Student: Miklos Vajna\n> Mentor: Johannes Schindelin\n> \n> In \"What's cooking in git.git (topics)\" Junio C Hamano wrote:\n> \n> It already is beginning to become clear what 1.6.0 will look like.\n> [...]\n> * git-merge will be rewritten in C.\n> \n> Status: Done and merged in.\n> Student retention: Most likely.\n\nI could not put it better, IOW 'nuff said :-)\n\nCiao,\nDscho\n"},{"id":"87269","messageId":"20080814231749.GE10544@machine.or.cz","threadId":"14334","inReplyTo":"1218717743.11530.99.camel@maia.lan","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Petr Baudis","fromEmail":"pasky@suse.cz","sentAt":"2008-08-14T23:17:49Z","receivedAt":"2008-08-14T23:17:49Z","isPatch":false,"sender":{"key":"pasky@ucw.cz","avatar":"https://avatars.githubusercontent.com/u/18439?v=4"},"body":"On Fri, Aug 15, 2008 at 12:42:23AM +1200, Sam Vilain wrote:\n> I was certainly hoping that at or shortly after the end it would at that\n> point also be good enough to be useful for production sites, even\n> kernel.org sized (perhaps requiring the odd piece rewritten in C). We\n> just need to crawl before we can walk, so to speak.\n\nFWIW, whenever you have any working code, let me know and I can deploy\nit on repo.or.cz for some heavier testing. ;-)\n\n\t\t\t\tPetr \"Pasky\" Baudis\n"},{"id":"87270","messageId":"200808150123.33546.jnareb@gmail.com","threadId":"14334","inReplyTo":"1218717743.11530.99.camel@maia.lan","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-08-14T23:23:31Z","receivedAt":"2008-08-14T23:23:31Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Thu, 14 Aug 2008, Sam Vilain wrote:\n> On Thu, 2008-08-14 at 04:57 +0200, Jakub Narebski wrote:\n>>\n>> 1. GitTorrent\n>> \n>> Student: Joshua Roys\n>> Mentor: Sam Vilain\n>> \n>> If I remember correctly at midterm it was deemed to be somewhat late;\n>> metainfo was done, tracker was in works, some core infrastructure\n>> and beginnings of peer to peer:\n>> http://thread.gmane.org/gmane.comp.version-control.git.gittorrent/1\n[...]\n\n>> Status: I have no idea how close GitTorrent is to completion (where by \n>> completion I mean ready, tested and benchmarked code running e.g. on \n>> kernel.org). I'm not sure if it is meant to be incorporated in git, \n>> even in contrib, or remain separate like StGIT, TopGit or jgit.\n> \n> The scope outlined in the GitTorrent proposal was a little bit more of a\n> research project; being ready for production use on kernel.org or having\n> code ready to merge to git was not directly a deliverable.\n> \n> The approved idea at the outset of *this* project was to try out the RFC\n> protocol design as it stands, iron out the weaknesses and see how it\n> performs. [...]\n\nThat is a very good idea, but neither \nhttp://git.or.cz/gitwiki/SoC2008Projects#GitTorrent nor\nhttp://code.google.com/soc/2008/git/appinfo.html?csaid=F544F0DAA82AFDFC\ntells me that it is more about prototype implementation, or even\npre-prototype implementation (testing protocol design), than actually\nimplementing it in the state for it to be ready to use.\n\nWhat I (and I guess also GSoC 2008 admins) would know is how far this\nresearch on GitTorrent Protocol went; what milestones were achieved and\nwhat were missed...\n\n-- \nJakub Narebski\nPoland\n"},{"id":"87344","messageId":"48A5DB4D.5060906@gmail.com","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Lea Wiemann","fromEmail":"lewiemann@gmail.com","sentAt":"2008-08-15T19:38:53Z","receivedAt":"2008-08-15T19:38:53Z","isPatch":false,"sender":{"key":"lewiemann@gmail.com","avatar":null},"body":"Jakub Narebski wrote:\n> 3. Gitweb caching\n> \n> Student: Lea Wiemann\n> Mentor: John 'warthog' Hawley\n> \n> There are new tests for gitweb (to check if caching would not break \n> anything new: it did caught a few breakages), new object Perl API to \n> git, and gitweb caching implemented using caching data at the level \n> slightly above calling git commands. But full code (tests, Perl API \n> and changes to gitweb) are only after first, maybe second round of \n> review.\n\nCorrect. I'm planning to post the next round of patches tonight or\ntomorrow. The remaining to-do items are:\n\n1. We'll still have to have some discussion wrt. what to do about the\nAPI and what status it should have -- Petr (and you as well?) complained\nabout missing command-line calling functionality.\n\n2. I'll run benchmarks under various conditions, to measure how much\nperformance we gain with caching, and under what conditions it is most\nbeneficial.\n\n3. Optional, but with some real performance benefits: Add support for\nLast-Modified/If-Modified-Since.\n\n4. Deploy on kernel.org (though John will have to do some/most of this,\nand it'll take time beyond the pencils-down deadline I expect).\n\n> Student retention: unknown.\n\nWell, on the downside, I'm expecting to be pretty busy with college, so\nthere won't be much time to do substantial work on git or gitweb. On\nthe upside, I feel perfectly comfortable with contributing to git (i.e.,\nmaintaining my own patch queue, sending patches, etc.), so it's very\nmuch possible that at some point I'll be hacking git or gitweb again.\n\n-- Lea\n"},{"id":"87349","messageId":"200808152236.56479.jnareb@gmail.com","threadId":"14334","inReplyTo":"48A5DB4D.5060906@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-08-15T20:36:54Z","receivedAt":"2008-08-15T20:36:54Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Fri, 15 August 2008, Lea Wiemann wrote:\n> Jakub Narebski wrote:\n> > 3. Gitweb caching\n> > \n> > Student: Lea Wiemann\n> > Mentor: John 'warthog' Hawley\n> > \n> > There are new tests for gitweb (to check if caching would not break \n> > anything new: it did caught a few breakages), new object Perl API to \n> > git, and gitweb caching implemented using caching data at the level \n> > slightly above calling git commands. But full code (tests, Perl API \n> > and changes to gitweb) are only after first, maybe second round of \n> > review.\n> \n> Correct. I'm planning to post the next round of patches tonight or\n> tomorrow. [...]\n\nPlease remember that according to timeline in GSoC 2008 FAQ:\n http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\nAugust 18: ~12 noon PDT / 19:00 UTC is \"Firm 'pencils down' date.\"\n(suggested pencils down date was August 11). So you don't have much\ntime, and you don't leave much time for review, so I guess evaluation\nwould be \"done, but nor merged in\" or something like that.\n\n> 2. I'll run benchmarks under various conditions, to measure how much\n> performance we gain with caching, and under what conditions it is most\n> beneficial.\n\nAnd to compare with plain, vanilla gitweb, with kernel.org fork (both\nperformance and disk space used I guess) and repo.or.cz fork (only caches\nprojects list view, IIRC), I guess.\n\nBenchmarking gitweb caching could be difficult, as you would have somehow\n(fio?) replay conditions of I/O pressure as on kernel.org and repo.or.cz,\nas from anegdotical evidence gitweb is IO bound, not CPU bound (so\nordinary speed benchmarks could give wrong results).\n\n> > Student retention: unknown.\n> \n> Well, on the downside, I'm expecting to be pretty busy with college, so\n> there won't be much time to do substantial work on git or gitweb. On\n> the upside, I feel perfectly comfortable with contributing to git (i.e.,\n> maintaining my own patch queue, sending patches, etc.), so it's very\n> much possible that at some point I'll be hacking git or gitweb again.\n\nIt would be nice, even if you would be a \"weekend contributor\".\n\nBut I guess that making you into gitweb maintainer, or git.kernel.org\nadmin is out of the question... ;-)\n\n-- \nJakub Narebski\nPoland\n"},{"id":"87355","messageId":"20080816011633.GC17121@leksak.fem-net","threadId":"14334","inReplyTo":"200808152236.56479.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-08-16T01:16:33Z","receivedAt":"2008-08-16T01:16:33Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nJakub Narebski wrote:\n> Please remember that according to timeline in GSoC 2008 FAQ:\n> http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> August 18: ~12 noon PDT / 19:00 UTC is \"Firm 'pencils down' date.\"\n> (suggested pencils down date was August 11). So you don't have much\n> time, and you don't leave much time for review, so I guess evaluation\n> would be \"done, but nor merged in\" or something like that.\n\nI'm wondering about \"and you don't leave much time for review\".\nA comment on this:\nLH (from the Google Open Source Team) clarified that the state of\nAug 18 is not the date where everything should be reviewed and merged\nin. It is the date where the (following) evaluation (done by the mentor)\nof the code is based on.\n\nSee also\nhttp://groups.google.com/group/google-summer-of-code-announce/browse_thread/thread/df7278c6e027dee1\n\nRegards,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"},{"id":"87356","messageId":"20080816012250.GB17399@spearce.org","threadId":"14334","inReplyTo":"20080816011633.GC17121@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2008-08-16T01:22:50Z","receivedAt":"2008-08-16T01:22:50Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Stephan Beyer wrote:\n> Jakub Narebski wrote:\n> > Please remember that according to timeline in GSoC 2008 FAQ:\n> > http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> > August 18: ~12 noon PDT / 19:00 UTC is \"Firm 'pencils down' date.\"\n> > (suggested pencils down date was August 11). So you don't have much\n> > time, and you don't leave much time for review, so I guess evaluation\n> > would be \"done, but nor merged in\" or something like that.\n> \n> I'm wondering about \"and you don't leave much time for review\".\n> A comment on this:\n> LH (from the Google Open Source Team) clarified that the state of\n> Aug 18 is not the date where everything should be reviewed and merged\n> in. It is the date where the (following) evaluation (done by the mentor)\n> of the code is based on.\n> \n> See also\n> http://groups.google.com/group/google-summer-of-code-announce/browse_thread/thread/df7278c6e027dee1\n\nLH is correct. A student doesn't have to get their code merged\njust to have a successful project. Perhaps the overall project is\nin a frozen release candidate period and doesn't want to take _any_\nnew features until after the release is made. That may be a full\nmonth after the \"pencils down\" date.\n\nSuppose even further that the project is on SVN, and has just one\nbranch, /trunk, such that even checking in the student's project\ncould break that release.\n\nClearly you cannot require the student to have their code merged\njust to pass them.\n\nHowever, I think it is reasonable to require that the student\nhas at least posted their patches for review and discussion, and\nthat the student has made a good-faith effort towards making those\npatches something the community _might_ accept. IOW it would be\ngood enough that Junio would consider slating it into \"pu\", even\nthough some final cleanup changes may be necessary.\n\n-- \nShawn.\n"},{"id":"87361","messageId":"200808160510.31005.jnareb@gmail.com","threadId":"14334","inReplyTo":"20080816011633.GC17121@leksak.fem-net","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2008-08-16T03:10:29Z","receivedAt":"2008-08-16T03:10:29Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Sat, 16 Aug 2008, Stephan Beyer wrote:\n\n> Jakub Narebski wrote:\n> >\n> > Please remember that according to timeline in GSoC 2008 FAQ:\n> > http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> > August 18: ~12 noon PDT / 19:00 UTC is \"Firm 'pencils down' date.\"\n> > (suggested pencils down date was August 11). So you don't have much\n> > time, and you don't leave much time for review, so I guess evaluation\n> > would be \"done, but nor merged in\" or something like that.\n> \n> I'm wondering about \"and you don't leave much time for review\".\n\nI mean here that we won't have time for a round of reviews and\ncorrections, it means that the code to be sent would be final\nversion to base GSoC evaluation. No time for comments and corrections.\n\nBuilt-in merge and push support for Eclipse went through many, many\ncycles...\n\n> A comment on this:\n> LH (from the Google Open Source Team) clarified that the state of\n> Aug 18 is not the date where everything should be reviewed and merged\n> in. It is the date where the (following) evaluation (done by the mentor)\n> of the code is based on.\n\nI agree that is a good idea. Especially that project schedule (feature\nfreeze and like) not necesssary agree with GSoC timeline.\n\n-- \nJakub Narebski\nPoland\n"},{"id":"87409","messageId":"bd6139dc0808162226mc98de9bg38b44f64ea4969dd@mail.gmail.com","threadId":"14334","inReplyTo":"200807210029.31543.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sverre Rabbelier","fromEmail":"alturin@gmail.com","sentAt":"2008-08-17T05:26:07Z","receivedAt":"2008-08-17T05:26:07Z","isPatch":false,"sender":{"key":"alturin@gmail.com","avatar":null},"body":"[Reply to all this time, instead of reply, thanks Jakub for poking me\nabout that.]\n\nOn Wed, Aug 13, 2008 at 21:57, Jakub Narebski wrote:\n> 2. git-statistics\n>\n> Student: Sverre Rabbelier\n> Mentor: David Symonds\n>\n> GitStat development is finished at least with respect to GSoC 2008, see\n> http://thread.gmane.org/gmane.comp.version-control.git/90691 (for the\n> tax reasons), although I guess its development will continue.\n\nYup, it will, although at the moment I am on vacation in the US, and\nI'm having a hard time finding the time to do any coding at all. I\nwill try to send in a patch to add GitStats to /contrib sometime soon,\nbut between not having internet/little time, it could take a bit.\n\n> Status: Finished, I think also accepted: what is left is to put it\n> eventually in 'contrib/gitstat' or 'contrib/stats'\n\nAye, as said above, might take a bit, but I will submit that patch.\n\n> Student retention: Sverre has said if I remember correctly that he wants\n> to work on improving '--follow', which now works only for very simple\n> histories, for GitStats to be better among others.\n\nYup, that is correct, I will work on that and on improving GitStats in\nmy free time.\n\n--\nCheers,\n\nSverre Rabbelier\n"},{"id":"87495","messageId":"48A88EC1.5020503@gmail.com","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Marek Zawirski","fromEmail":"marek.zawirski@gmail.com","sentAt":"2008-08-17T20:49:05Z","receivedAt":"2008-08-17T20:49:05Z","isPatch":false,"sender":{"key":"marek.zawirski@gmail.com","avatar":null},"body":"Jakub Narebski wrote:\n> Now that according to Google Summer of Code 2008 timeline\n> http://code.google.com/opensource/gsoc/2008/faqs.html#0.1_timeline\n> we are in the middle of the suggested 'pencils down' date (August 11 --\n> -- August 18), I and perhaps other on git mailing list would like\n> to know what is the current status of all git GSoC 2008 projects.\n> I think that writing it down would help GSoC manager and co-manager...\n(...)\n> 4. Eclipse plugin push support\n> \n> Student: Marek Zawirski\n> Mentor: Shawn O. Pearce\n> \n> JGit can now create (suboptimal) packs: it can reuse delta, but cannot \n> currently create one. This is used to implement push support in \n> jgit/egit.\n> \n> Status: Done, merged in, and used as example to add for example Amazon's \n> S3 support and SFTP transport support.\n\nI'm just putting in the background patches with my 2nd big task for \nGSoC: (mostly) GUI for push. This code uses prior work - the one \ndescribed by Jakub.\n\nIt took me a long time to do it, I was expecting some more time for \nother things, but it was probably better to focus on 1 thing at time. I \nthink that we can say now that it's usable: it's possible to push from \nEclipse in some (hopefully) sensible way.\n\nBasing on created components, we can also add fetch GUI or remotes \nconfiguration editor easier than before.\n\n> Student retention: I don't know, but I guess it is likely.\n\nNice guess! ;) That's quite true. I'm sure that I will no longer have \nthat much time for coding jgit or egit, but when possible, I'll try to \ncontribute in some way. Now, I'm starting to work on fetch GUI.\n\nHaving this opportunity, I want to say: thank you Shawn for being that \ngreat mentor! :) And thanks to other folks for given help. I've learned \nsome new things here and it was (is?) nice experience to code for egit.\n\n-- \nMarek Zawirski [zawir]\nmarek.zawirski@gmail.com\n"},{"id":"87516","messageId":"20080818055133.GD7376@spearce.org","threadId":"14334","inReplyTo":"48A88EC1.5020503@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2008-08-18T05:51:33Z","receivedAt":"2008-08-18T05:51:33Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Marek Zawirski wrote:\n>\n>> Student retention: I don't know, but I guess it is likely.\n\nYea, I'm hoping Marek is going to be around for a bit. He turned\ninto a really good contributor this summer. Its been a pleasure\nworking with him on push support.\n\n> Nice guess! ;) That's quite true. I'm sure that I will no longer have \n> that much time for coding jgit or egit, but when possible, I'll try to \n> contribute in some way. Now, I'm starting to work on fetch GUI.\n\nYay, fetch UI! ;-)\n\n-- \nShawn.\n"},{"id":"87635","messageId":"19b271a20808181825mf2d0de8lc83ba067ed40f4e8@mail.gmail.com","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Joshua Roys","fromEmail":"roysjosh@gmail.com","sentAt":"2008-08-19T01:25:55Z","receivedAt":"2008-08-19T01:25:55Z","isPatch":false,"sender":{"key":"roysjosh@gmail.com","avatar":null},"body":"Hello all.\n\nAs of today, here is the status of GitTorrent. We can only transfer\nrepositories under extremely limited circumstances, which is not what\nI would consider success, unfortunately. I apologize for that. I\nwould like to continue working on it, and other git things, in the\nfuture though. This was a really great experience, and in fact my\nfirst actually working (closely) with the Open Source community.\nPreviously I had only sent in the random patch here and there.\n\nA big 'thank you' to Sam Vilain and the other git folk who helped out.\n\nFor the curious that wish to look at the code:\nhttp://utsl.gen.nz/gitweb/?p=VCS-Git-Torrent;a=summary\nand a repo.or.cz mirror (since the above randomly fails):\nhttp://repo.or.cz/w/VCS-Git-Torrent.git?a=shortlog\n\nJoshua Roys\n\nOn Wed, Aug 13, 2008 at 10:57 PM, Jakub Narebski wrote:\n> 1. GitTorrent\n>\n> Student: Joshua Roys\n> Mentor: Sam Vilain\n>\n> Student retention: One of reasons in Git application for participating\n> in GSoC was having \"fresh blood\", new long-time contributors. I have\n> vague notion that Joshua is quite active not only with GitTorrent, and\n> would stay git contributor...\n>\n"},{"id":"87796","messageId":"1219213161.4417.12.camel@maia.lan","threadId":"14334","inReplyTo":"19b271a20808181825mf2d0de8lc83ba067ed40f4e8@mail.gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Sam Vilain","fromEmail":"sam@vilain.net","sentAt":"2008-08-20T06:19:21Z","receivedAt":"2008-08-20T06:19:21Z","isPatch":false,"sender":{"key":"sam@vilain.net","avatar":"https://gravatar.com/avatar/8fc840ca854dbf6f7065b4335e3b934951c1dca3b11db688e95e471901f8f4a8?d=mp&s=160"},"body":"On Mon, 2008-08-18 at 21:25 -0400, Joshua Roys wrote:\n> Hello all.\n> \n> As of today, here is the status of GitTorrent. We can only transfer\n> repositories under extremely limited circumstances, which is not what\n> I would consider success, unfortunately. I apologize for that. I\n> would like to continue working on it, and other git things, in the\n> future though. This was a really great experience, and in fact my\n> first actually working (closely) with the Open Source community.\n> Previously I had only sent in the random patch here and there.\n> \n\nHi Joshua,\n\nThanks for saying this, it takes courage to face failure.\n\nThat being said, the code base is very close to the point where it can\nbe used to assess the success of the protocol design. So while the GSoC\nproject is officially failed, I don't think we have to let the community\ndown. Let's see if we can finish off the remaining bits and pieces and\nput together such a report soon.\n\nCheers!\nSam.\n"},{"id":"88225","messageId":"20080822230339.GA14684@leksak.fem-net","threadId":"14334","inReplyTo":"200808140457.56464.jnareb@gmail.com","subject":"Re: [GSoC] What is status of Git's Google Summer of Code 2008 projects?","fromName":"Stephan Beyer","fromEmail":"s-beyer@gmx.net","sentAt":"2008-08-22T23:03:39Z","receivedAt":"2008-08-22T23:03:39Z","isPatch":false,"sender":{"key":"s-beyer@gmx.net","avatar":"https://avatars.githubusercontent.com/u/143889?v=4"},"body":"Hi,\n\nJakub Narebski wrote:\n> 6. git-sequencer\n> \n> Student: Stephan Beyer\n\nThe Google Open Source blog had an article about this thread[1]\nand I noted that I forgot to inform you about my project here.\n\n 1. http://google-opensource.blogspot.com/2008/08/git-and-google-summer-of-code.html\n\nUntil Sunday, Aug 17, I have been on testing, code cleanup and\nbugfixing of the builtin-sequencer. Then I've sent this version\nof the patchset to my mentors for a final review before the real\nhard one (on the list *g*) and then I went away for three days.\nThat's perhaps why I forgot about this thread :)\n\nThat's also why there's not yet a patchset on the list.\n\nI know there are a few people that already use (or at least used) the\n(builtin) sequencer. For the other curious people: the branches\n\"seq-builtin-dev\" (development branch) or \"seq-builtin-rfc\" (rebased\npatchset) on repo.or.cz[2] can be used, because I currently do not\nknow how long it will take until I send it to the list but I hope\nit will be soon.\n\n 2. http://repo.or.cz/w/git/sbeyer.git (seq-builtin-rfc, seq-builtin-dev)\n\n(The last commit in seq-builtin-rfc is a test and I don't really like\nit atm. It's reverted in the -dev branch.)\n\nWhat else to say?\nI will have not *much* time for git each day until the end of September\nbecause of some other work and, especially, an important exam. But I\nalso try to take some time for git.\nI also intend to keep contributing (at least small stuff) after the\nbuiltin sequencer got into git. (I expect this is still some hard work\nuntil the last one who has doubts is somehow satisfied.)\n\nKind regards,\n Stephan\n\n-- \nStephan Beyer , PGP 0x6EDDD207FCC5040F\n"}]}