{"thread":{"id":"9416","subject":"'pu' branch for StGIT","startedAt":"2007-08-07T02:20:43Z","lastAt":"2007-08-12T22:47:52Z","messageCount":17,"participants":["Karl Hasselström","Shawn O. Pearce","Pavel Roskin","Catalin Marinas"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"50070","messageId":"20070807022043.GA8482@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":null,"subject":"'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-07T02:20:43Z","receivedAt":"2007-08-07T02:20:43Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"So I finally got my act together and published a 'pu'-like branch for\nStGIT. Get it at git://repo.or.cz/stgit/kha.git; gitweb at\nhttp://repo.or.cz/w/stgit/kha.git.\n\nThe idea is that I grab the StGIT patches that are posted to the list,\nput each series in its own topic branch (based on Catalin's master),\nand publish a merge of them. _After_ having made sure that the test\nsuite passes, naturally. It then becomes much easier for everyone to\ntest these patches.\n\nThis should lessen the pressure on Catalin to include patches in his\nmaster branch as soon as possible, which should in turn reduce the\nnumber of reverts in the official history.\n\nHere's what's currently in there:\n\nKarl Hasselström (8):\n      Teach StGIT about core.excludesfile\n      New test: make sure that popping doesn't change patch order\n      Verify patch status during the test\n      Make use of the get_patch() utility function\n      Compute patch appliedness from commit DAG\n      Test the new DAG appliedness machinery\n      Fix bash completion after the DAG appliedness patch\n      Speed up the appliedness test\n\nPavel Roskin (1):\n      Add support for SMTP over Transport Layer Security (TLS)\n\nYann Dirson (7):\n      Include contrib scripts in the release tarball.\n      Improve stg-fold-files-from doc.\n      New contrib scripts: stg-dispatch and stg-show.\n      Add -O flag to stg-fold-files-from.\n      Add a no-act flag to stg-dispatch and stg-fold-file-from.\n      Provide file completion for add/resolved/refresh based on status.\n      Fixed completion function hardcoding .git/.\n\n(Does anyone know the incantation for Junio's status mails, with topic\nbranches grouped nicely?)\n\nNOTE: The DAG appliedness patch series is in there. They will upgrade\nthe format of any stgit branch they touch, and there's no convenient\nway to change it back if you change your mind.\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50071","messageId":"20070807023836.GN9527@spearce.org","threadId":"9416","inReplyTo":"20070807022043.GA8482@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2007-08-07T02:38:36Z","receivedAt":"2007-08-07T02:38:36Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Karl Hasselstrm <kha@treskal.com> wrote:\n> (Does anyone know the incantation for Junio's status mails, with topic\n> branches grouped nicely?)\n\nLook at his todo branch; one of those scripts does the magic.\nStart with todo:WC, apparently it calls todo:git-topic.perl.\nJunio checks out the todo branch in a Meta/ subdirectory.  :-)\n\n-- \nShawn.\n"},{"id":"50172","messageId":"1186549433.2112.34.camel@dv","threadId":"9416","inReplyTo":"20070807022043.GA8482@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-08T05:03:53Z","receivedAt":"2007-08-08T05:03:53Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"On Tue, 2007-08-07 at 04:20 +0200, Karl Hasselström wrote:\n> So I finally got my act together and published a 'pu'-like branch for\n> StGIT. Get it at git://repo.or.cz/stgit/kha.git; gitweb at\n> http://repo.or.cz/w/stgit/kha.git.\n\nJust a word of warning.  This version converts all branches to version\n3, so that the standard StGIT won't work with them anymore.\n\nAlso, \"stg import\" commits the patch, which seems wrong to me.\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"50183","messageId":"20070808092027.GB7860@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"1186549433.2112.34.camel@dv","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-08T09:20:27Z","receivedAt":"2007-08-08T09:20:27Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-08 01:03:53 -0400, Pavel Roskin wrote:\n\n> On Tue, 2007-08-07 at 04:20 +0200, Karl Hasselström wrote:\n>\n> > So I finally got my act together and published a 'pu'-like branch\n> > for StGIT. Get it at git://repo.or.cz/stgit/kha.git; gitweb at\n> > http://repo.or.cz/w/stgit/kha.git.\n>\n> Just a word of warning. This version converts all branches to\n> version 3, so that the standard StGIT won't work with them anymore.\n\nI believe I said (or tried to say) as much towards the end of my mail.\nThanks for trying it out though! It seems my evil master plan is\nworking. :-)\n\n> Also, \"stg import\" commits the patch, which seems wrong to me.\n\nHmm, I hadn't noticed. That would be an unintended side-effect of the\nDAG patches, presumably. I'll look into it tonight.\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50244","messageId":"20070808213917.GA22521@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"20070808092027.GB7860@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-08T21:39:17Z","receivedAt":"2007-08-08T21:39:17Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-08 11:20:27 +0200, Karl Hasselström wrote:\n\n> On 2007-08-08 01:03:53 -0400, Pavel Roskin wrote:\n>\n> > Also, \"stg import\" commits the patch, which seems wrong to me.\n>\n> Hmm, I hadn't noticed. That would be an unintended side-effect of\n> the DAG patches, presumably. I'll look into it tonight.\n\nI can't reproduce. (Using master of git://repo.or.cz/stgit/kha.git.)\nThe last subtest of t1800 actually tests that a mailbox was imported\nas a series of patches, and that works fine for me (the other subtests\ndon't make sure that a patch was created; they just verify that we got\nthe right tree). Furthermore, the following works for me:\n\nkha@yoghurt:~/stgit/t> ./t1800-import.sh\n*   ok 1: Initialize the StGIT repository\n*   ok 2: Apply a patch created with \"git diff\"\n*   ok 3: Apply a patch created with GNU diff\n*   ok 4: Apply a patch created with \"stg export\"\n*   ok 5: Apply a patch from an 8bit-encoded e-mail\n*   ok 6: Apply a patch from a QP-encoded e-mail\n*   ok 7: Apply several patches from an mbox file\n* passed all 7 test(s)\nkha@yoghurt:~/stgit/t> cd trash/\nkha@yoghurt:~/stgit/t/trash> ls\nfoo.txt\nkha@yoghurt:~/stgit/t/trash> PATH=/home/kha/stgit:$PATH stg series -d\nkha@yoghurt:~/stgit/t/trash> PATH=/home/kha/stgit:$PATH stg import -m ../t1800-import/email-qp\nChecking for changes in the working directory ... done\nFinding uninteresting commits ... done\nImporting patch \"email-qp\" ... done\nNow at patch \"email-qp\"\nkha@yoghurt:~/stgit/t/trash> PATH=/home/kha/stgit:$PATH stg series -d\n> email-qp | test patch\n\nSo a patch is definitely created. Could you describe in more detail\nexactly what you do to make it fail?\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50252","messageId":"1186611514.7383.4.camel@dv","threadId":"9416","inReplyTo":"20070808213917.GA22521@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-08T22:18:34Z","receivedAt":"2007-08-08T22:18:34Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"Hello, Karl!\n\nOn Wed, 2007-08-08 at 23:39 +0200, Karl Hasselström wrote:\n\n> > Hmm, I hadn't noticed. That would be an unintended side-effect of\n> > the DAG patches, presumably. I'll look into it tonight.\n> \n> I can't reproduce.\n\nOK, it's trickier.  There are some bad patch names that don't get\nimported properly.  In particular, patches ending with \".diff\" are\ncommitted after import.\n\nTry changing this in the testsuite:\n\ndiff --git a/t/t1800-import.sh b/t/t1800-import.sh\nindex 8c8c9a0..6cd3cdb 100755\n--- a/t/t1800-import.sh\n+++ b/t/t1800-import.sh\n@@ -15,7 +15,7 @@ test_expect_success \\\n test_expect_success \\\n     'Apply a patch created with \"git diff\"' \\\n     '\n-    stg import ../t1800-import/git-diff &&\n+    stg import -n git.diff ../t1800-import/git-diff &&\n     [ $(git cat-file -p $(stg id) \\\n         | grep -c \"tree e96b1fba2160890ff600b675d7140d46b022b155\") = 1 ] &&\n     stg delete ..\n\nAnd now run the test\n\n$ ./t1800-import.sh -i -v\n...\nImporting patch \"git.diff\" ... done\nNo patches applied\n\nThe mainline StGIT is OK.\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"50266","messageId":"20070808232349.GA23172@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"1186611514.7383.4.camel@dv","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-08T23:23:49Z","receivedAt":"2007-08-08T23:23:49Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-08 18:18:34 -0400, Pavel Roskin wrote:\n\n> On Wed, 2007-08-08 at 23:39 +0200, Karl Hasselström wrote:\n>\n> > I can't reproduce.\n>\n> OK, it's trickier. There are some bad patch names that don't get\n> imported properly. In particular, patches ending with \".diff\" are\n> committed after import.\n\nAh, sneaky. And it turns out to be not a problem with import, but a\nproblem with dots in patch names. It's just that import is the only\nplace those are typically created.\n\nIt was all due to a sloppy regexp. This is the fix:\n\ndiff --git a/stgit/stack.py b/stgit/stack.py\nindex 4186ba9..c403f51 100644\n--- a/stgit/stack.py\n+++ b/stgit/stack.py\n@@ -391,11 +391,11 @@ def read_refs(branch):\n     given branch. The patches are listed by name; the branch head is\n     None.\"\"\"\n     refs = {}\n-    patchpat = re.compile(r'^refs/patches/%s/([^\\.]+)$' % branch)\n+    patchpat = re.compile(r'^refs/patches/%s/(.+)$' % branch)\n     for line in git._output_lines('git-show-ref'):\n         sha1, ref = line.split()\n         m = re.match(patchpat, ref)\n-        if m:\n+        if m and not m.group(1).endswith('.log'):\n             refs[m.group(1)] = sha1\n         elif ref == 'refs/heads/%s' % branch:\n             refs[None] = sha1\n\nThanks for taking the time to track this down -- with the detailed\nsymptoms you gave, I found it in no time. I've pushed an updated\nseries (as well as the other patches I've posted tonight) to\ngit://repo.or.cz/stgit/kha.git.\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50272","messageId":"20070808201003.nm90u2s5ny888wcc@webmailbeta.spamcop.net","threadId":"9416","inReplyTo":"20070808232349.GA23172@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-09T00:10:03Z","receivedAt":"2007-08-09T00:10:03Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"Quoting Karl Hasselström <kha@treskal.com>:\n\n> It was all due to a sloppy regexp. This is the fix:\n\nThank you!\n\nAnd by the way, please fix the description of the commit where you  \nadded --smtp-server option.  You added it to \"stg mail\", not to \"stg  \nadd\".\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"50294","messageId":"20070809073801.GA31482@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"20070808201003.nm90u2s5ny888wcc@webmailbeta.spamcop.net","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-09T07:38:01Z","receivedAt":"2007-08-09T07:38:01Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-08 20:10:03 -0400, Pavel Roskin wrote:\n\n> Quoting Karl Hasselström <kha@treskal.com>:\n>\n> > It was all due to a sloppy regexp. This is the fix:\n>\n> Thank you!\n\nYou did all the work on this one.\n\n> And by the way, please fix the description of the commit where you\n> added --smtp-server option. You added it to \"stg mail\", not to \"stg\n> add\".\n\nMmm, thanks. Will fix tonight.\n\nI take it this all means you're actually using my branch? What's your\nopinion on its usefulness?\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50318","messageId":"1186665883.28228.31.camel@dv","threadId":"9416","inReplyTo":"20070809073801.GA31482@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-09T13:24:43Z","receivedAt":"2007-08-09T13:24:43Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"On Thu, 2007-08-09 at 09:38 +0200, Karl Hasselström wrote:\n\n> I take it this all means you're actually using my branch? What's your\n> opinion on its usefulness?\n\nWell, I tried it, and then ran a script to update all local\nrepositories.  It converted everything to \"version 3\", so I'm sort of\nstuck with it.  If the \"version 3\" code is not committed to the mainline\nStGIT, I'll have to convert my repositories back or even re-fetch them.\n\nI have noticed two problems so far, but I cannot tell is they are\nspecific to the \"pu\" branch.\n\n1) Undead patches.\n\nStGIT finds a patch I deleted long ago and shows it as unapplied.  It\ncannot be deleted by \"stg delete\" because some files are missing.\n\n$ stg delete at76_usb\nTraceback (most recent call last):\n  File \"/home/proski/bin/stg\", line 43, in <module>\n    main()\n  File \"home/proski/lib/python2.5/site-packages/stgit/main.py\", line 284, in main\n  File \"home/proski/lib/python2.5/site-packages/stgit/commands/delete.py\", line 76, in func\n  File \"home/proski/lib/python2.5/site-packages/stgit/stack.py\", line 1227, in delete_patch\n  File \"home/proski/lib/python2.5/site-packages/stgit/stack.py\", line 1209, in delete_patch_data\n  File \"home/proski/lib/python2.5/site-packages/stgit/stack.py\", line 160, in delete\nOSError: [Errno 2] No such file or directory: '.git/patches/wireless-dev/patches/at76_usb'\n\nThat's Linux repository, and I'm on wireless-dev branch.  There is a\nfile .git/patches/wireless-dev/trash/at76_usb containing \"None\".  There\nare two other files in that directory, but they have some SHA1 hashes.\n\nThere is also a file .git/patches/wireless-dev/patchorder, which\ncontains \"at76_usb\".\n\nI was updating the repository by \"stg pull\", there were two patches,\n\"at76_usb\" being first.  It couldn't be merged, so I deleted it.  I\ndeleted the other patch as well, since I new it was applied upstream.\nAfter another \"stg pull\" at76_usb became \"undead\".\n\nI cannot reproduce it on another repository.\n\n2) Invisible branches.\n\nStGIT stopped showing other branches.  It's always showing the same\nbranch, although it can switch to other branches:\n\n[proski@dv linux-2.6]$ stg branch --list\nAvailable branches:\n> s     wireless-dev  | \n[proski@dv linux-2.6]$ stg branch wireless-2.6\nChecking for changes in the working directory ... done\nSwitching to branch \"wireless-2.6\" ... done\n[proski@dv linux-2.6]$ stg branch --list\nAvailable branches:\n  s     wireless-dev  | \n[proski@dv linux-2.6]$ stg branch linus       \nChecking for changes in the working directory ... done\nSwitching to branch \"linus\" ... done\n[proski@dv linux-2.6]$ stg branch --list\nAvailable branches:\n  s     wireless-dev  | \n[proski@dv linux-2.6]$ stg init\nstg init: linus already initialized\n[proski@dv linux-2.6]$\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"50324","messageId":"20070809141848.GA6342@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"1186665883.28228.31.camel@dv","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-09T14:18:48Z","receivedAt":"2007-08-09T14:18:48Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-09 09:24:43 -0400, Pavel Roskin wrote:\n\n> On Thu, 2007-08-09 at 09:38 +0200, Karl Hasselström wrote:\n>\n> > I take it this all means you're actually using my branch? What's\n> > your opinion on its usefulness?\n>\n> Well, I tried it, and then ran a script to update all local\n> repositories. It converted everything to \"version 3\", so I'm sort of\n> stuck with it. If the \"version 3\" code is not committed to the\n> mainline StGIT, I'll have to convert my repositories back or even\n> re-fetch them.\n\nThanks for the vote of confidence. :-)\n\nYou should be able to do something like\n\n  $ stg applied > .git/patches/branch/applied\n  $ stg unapplied > .git/patches/branch/unapplied\n\nand then manually change the version from 3 to 2, and be ready to go.\nI haven't tested this, though!\n\n> I have noticed two problems so far, but I cannot tell is they are\n> specific to the \"pu\" branch.\n>\n> 1) Undead patches.\n\nI saw the same problem today. I haven't had time to look into it, but\nI believe it's due to stgit trying to directly modify files under\n.git/refs instead of using git-update-ref, which breaks with packed\nrefs. The DAG patches rely much more on the refs, so the bug is more\nsevere in that case.\n\nhttps://gna.org/bugs/?9710\n\n> There is also a file .git/patches/wireless-dev/patchorder, which\n> contains \"at76_usb\".\n\nThe patchorder file should be harmless. It's only used to determine\npatch order for those cases where the DAG information isn't\nsufficient. (That is, for unapplied patches.) It's strictly advisory,\nand _not_ used to determine which patches exist.\n\n> I was updating the repository by \"stg pull\", there were two patches,\n> \"at76_usb\" being first. It couldn't be merged, so I deleted it. I\n> deleted the other patch as well, since I new it was applied\n> upstream. After another \"stg pull\" at76_usb became \"undead\".\n\nUntil this is fixed, you can use git-show-ref and git-update-ref to\nmanually delete the offending ref. That fixed the problem for me.\n\n> 2) Invisible branches.\n\nI haven't seen this problem at all -- in my repositories, \"stg branch\n-l\" just works. Will try to reproduce (hopefully tonight). Do you have\na recepie on how to reproduce this from scratch?\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50325","messageId":"20070809142440.GB6342@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"20070809141848.GA6342@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-09T14:24:40Z","receivedAt":"2007-08-09T14:24:40Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-09 16:18:48 +0200, Karl Hasselström wrote:\n\n> I saw the same problem today. I haven't had time to look into it,\n> but I believe it's due to stgit trying to directly modify files\n> under .git/refs instead of using git-update-ref, which breaks with\n> packed refs. The DAG patches rely much more on the refs, so the bug\n> is more severe in that case.\n\ngit-gc started packing refs by default in late May. That's probably\nwhat's caused it to surface.\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50330","messageId":"1186677210.31394.28.camel@dv","threadId":"9416","inReplyTo":"20070809141848.GA6342@diana.vm.bytemark.co.uk","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-09T16:33:30Z","receivedAt":"2007-08-09T16:33:30Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"On Thu, 2007-08-09 at 16:18 +0200, Karl Hasselström wrote:\n\n> You should be able to do something like\n> \n>   $ stg applied > .git/patches/branch/applied\n>   $ stg unapplied > .git/patches/branch/unapplied\n> \n> and then manually change the version from 3 to 2, and be ready to go.\n> I haven't tested this, though!\n\nThat seems to work.  Thank you!  \"branch\" should be substituted with the\ncurrent branch, of course.\n\n> > I have noticed two problems so far, but I cannot tell is they are\n> > specific to the \"pu\" branch.\n> >\n> > 1) Undead patches.\n> \n> I saw the same problem today. I haven't had time to look into it, but\n> I believe it's due to stgit trying to directly modify files under\n> .git/refs instead of using git-update-ref, which breaks with packed\n> refs. The DAG patches rely much more on the refs, so the bug is more\n> severe in that case.\n> \n> https://gna.org/bugs/?9710\n\nI've attached the test case to that bug.  You are right, git-gc is involved.\n\n> > 2) Invisible branches.\n> \n> I haven't seen this problem at all -- in my repositories, \"stg branch\n> -l\" just works. Will try to reproduce (hopefully tonight). Do you have\n> a recepie on how to reproduce this from scratch?\n\nIt's a problem with git-gc too!  Just clone some repository and run \"stg\nbranch -l\" in it.  It with show master.  Run git-gc, and \"stg branch -l\"\nwill show \"No branches\".\n\nI see that in my Linux repository there are files\nin .git/refs/remotes/wireless-dev but not in other directories\nunder .git/refs/remotes/\n\n-- \nRegards,\nPavel Roskin\n"},{"id":"50342","messageId":"20070809203957.GA14441@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"1186677210.31394.28.camel@dv","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-09T20:39:57Z","receivedAt":"2007-08-09T20:39:57Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-09 12:33:30 -0400, Pavel Roskin wrote:\n\n> On Thu, 2007-08-09 at 16:18 +0200, Karl Hasselström wrote:\n>\n> > You should be able to do something like\n> >\n> >   $ stg applied > .git/patches/branch/applied\n> >   $ stg unapplied > .git/patches/branch/unapplied\n> >\n> > and then manually change the version from 3 to 2, and be ready to\n> > go. I haven't tested this, though!\n>\n> That seems to work. Thank you!\n\nWhat? You mean I didn't forget any step? ;-)\n\n> \"branch\" should be substituted with the current branch, of course.\n\nI was just too lazy to type the brackets. Or maybe I'm just evil and\n_want_ the lock-in effect. Who knows? :-)\n\n> > I saw the same problem today.\n> >\n> > https://gna.org/bugs/?9710\n>\n> I've attached the test case to that bug. You are right, git-gc is\n> involved.\n\nThanks.\n\n> > I haven't seen this problem at all -- in my repositories, \"stg\n> > branch -l\" just works. Will try to reproduce (hopefully tonight).\n> > Do you have a recepie on how to reproduce this from scratch?\n>\n> It's a problem with git-gc too! Just clone some repository and run\n> \"stg branch -l\" in it. It with show master. Run git-gc, and \"stg\n> branch -l\" will show \"No branches\".\n\nWell, well! In that case, I'm off to kill two birds with one stone!\n\n> I see that in my Linux repository there are files in\n> .git/refs/remotes/wireless-dev but not in other directories under\n> .git/refs/remotes/\n\nPresumably the only nonpacked refs are the ones that have been updated\nafter the last ref packing. Or however it works.\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50347","messageId":"b0943d9e0708091431k57ac095asa940cca73ad95b13@mail.gmail.com","threadId":"9416","inReplyTo":"1186665883.28228.31.camel@dv","subject":"Re: 'pu' branch for StGIT","fromName":"Catalin Marinas","fromEmail":"catalin.marinas@gmail.com","sentAt":"2007-08-09T21:31:09Z","receivedAt":"2007-08-09T21:31:09Z","isPatch":false,"sender":{"key":"catalin.marinas@gmail.com","avatar":null},"body":"On 09/08/2007, Pavel Roskin <proski@gnu.org> wrote:\n> On Thu, 2007-08-09 at 09:38 +0200, Karl Hasselström wrote:\n>\n> > I take it this all means you're actually using my branch? What's your\n> > opinion on its usefulness?\n>\n> Well, I tried it, and then ran a script to update all local\n> repositories.  It converted everything to \"version 3\", so I'm sort of\n> stuck with it.  If the \"version 3\" code is not committed to the mainline\n> StGIT, I'll have to convert my repositories back or even re-fetch them.\n\nI'll merge many of Karl's patches for the next stable release (which I\nhope to be sooner than the current 0.13) but I'm still on holiday for\none more week and can't do much work (busy with DIY).\n\nThanks to Karl for setting up this experimental branch. BTW, what does\n'pu' mean?\n\n-- \nCatalin\n"},{"id":"50361","messageId":"20070810003014.GA18294@diana.vm.bytemark.co.uk","threadId":"9416","inReplyTo":"b0943d9e0708091431k57ac095asa940cca73ad95b13@mail.gmail.com","subject":"Re: 'pu' branch for StGIT","fromName":"Karl Hasselström","fromEmail":"kha@treskal.com","sentAt":"2007-08-10T00:30:14Z","receivedAt":"2007-08-10T00:30:14Z","isPatch":false,"sender":{"key":"kha@treskal.com","avatar":"https://gravatar.com/avatar/f0120c734b5279b345075a28521e1ac66acb20c9913ffe9bf6ae97e53f7f3f13?d=mp&s=160"},"body":"On 2007-08-09 22:31:09 +0100, Catalin Marinas wrote:\n\n> I'll merge many of Karl's patches for the next stable release (which\n> I hope to be sooner than the current 0.13) but I'm still on holiday\n> for one more week and can't do much work (busy with DIY).\n\nGreat! Would you like me to select the \"safe\" topics and give you a\nsingle integration branch to pull, or do you want to be the one that\ndoes the selecting?\n\n> Thanks to Karl for setting up this experimental branch. BTW, what\n> does 'pu' mean?\n\nProposed Updates, as I recall. But I just used that to explain how my\nbranch is going to behave (meaning: it'll behave like Junio's pu\nbranch); it isn't actually called \"pu\". Not so pedagogical if no one\nhas heard of Junio's pu branch, I guess. :-)\n\n-- \nKarl Hasselström, kha@treskal.com\n      www.treskal.com/kalle\n"},{"id":"50603","messageId":"20070812184752.wi4l5womoosk0004@webmailbeta.spamcop.net","threadId":"9416","inReplyTo":"b0943d9e0708091431k57ac095asa940cca73ad95b13@mail.gmail.com","subject":"Re: 'pu' branch for StGIT","fromName":"Pavel Roskin","fromEmail":"proski@gnu.org","sentAt":"2007-08-12T22:47:52Z","receivedAt":"2007-08-12T22:47:52Z","isPatch":false,"sender":{"key":"proski@gnu.org","avatar":null},"body":"Quoting Catalin Marinas <catalin.marinas@gmail.com>:\n\n> I'll merge many of Karl's patches for the next stable release (which I\n> hope to be sooner than the current 0.13) but I'm still on holiday for\n> one more week and can't do much work (busy with DIY).\n\nI'm on vacation for two more weeks, on a poor modem connection and  \naway from anything POSIX-compatible (and pulling StGIT is too much for  \nmy modem anyway), so please don't wait for my replies.\n\n> Thanks to Karl for setting up this experimental branch. BTW, what does\n> 'pu' mean?\n\nI guess it's \"pick-up\", modeled after git's \"pu\" branch.\n\n-- \nRegards,\nPavel Roskin\n"}]}