{"thread":{"id":"16860","subject":"[ANNOUNCE] GIT 1.6.1","startedAt":"2008-12-25T06:36:27Z","lastAt":"2008-12-27T10:31:01Z","messageCount":15,"participants":["Junio C Hamano","Christian MICHON","bill lam","René Scharfe","Heikki Orsila","Pau Garcia i Quiles","Sitaram Chamarty","Nicolas Pitre","Steffen Prohaska"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"98681","messageId":"7v7i5odams.fsf@gitster.siamese.dyndns.org","threadId":"16860","inReplyTo":null,"subject":"[ANNOUNCE] GIT 1.6.1","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-12-25T06:36:27Z","receivedAt":"2008-12-25T06:36:27Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"The latest feature release GIT 1.6.1 is available at the usual\nplaces:\n\n  http://www.kernel.org/pub/software/scm/git/\n\n  git-1.6.1.tar.{gz,bz2}\t\t\t(source tarball)\n  git-htmldocs-1.6.1.tar.{gz,bz2}\t\t(preformatted docs)\n  git-manpages-1.6.1.tar.{gz,bz2}\t\t(preformatted docs)\n\nThe RPM binary packages for a few architectures are also found in the\nvicinity.\n\n  RPMS/$arch/git-*-1.6.1-1.fc9.$arch.rpm\t(RPM)\n\n----------------------------------------------------------------\n\nGIT v1.6.1 Release Notes\n========================\n\nUpdates since v1.6.0\n--------------------\n\nWhen some commands (e.g. \"git log\", \"git diff\") spawn pager internally, we\nused to make the pager the parent process of the git command that produces\noutput.  This meant that the exit status of the whole thing comes from the\npager, not the underlying git command.  We swapped the order of the\nprocesses around and you will see the exit code from the command from now\non.\n\n(subsystems)\n\n* gitk can call out to git-gui to view \"git blame\" output; git-gui in turn\n  can run gitk from its blame view.\n\n* Various git-gui updates including updated translations.\n\n* Various gitweb updates from repo.or.cz installation.\n\n* Updates to emacs bindings.\n\n(portability)\n\n* A few test scripts used nonportable \"grep\" that did not work well on\n  some platforms, e.g. Solaris.\n\n* Sample pre-auto-gc script has OS X support.\n\n* Makefile has support for (ancient) FreeBSD 4.9.\n\n(performance)\n\n* Many operations that are lstat(3) heavy can be told to pre-execute\n  necessary lstat(3) in parallel before their main operations, which\n  potentially gives much improved performance for cold-cache cases or in\n  environments with weak metadata caching (e.g. NFS).\n\n* The underlying diff machinery to produce textual output has been\n  optimized, which would result in faster \"git blame\" processing.\n\n* Most of the test scripts (but not the ones that try to run servers)\n  can be run in parallel.\n\n* Bash completion of refnames in a repository with massive number of\n  refs has been optimized.\n\n* Cygwin port uses native stat/lstat implementations when applicable,\n  which leads to improved performance.\n\n* \"git push\" pays attention to alternate repositories to avoid sending\n  unnecessary objects.\n\n* \"git svn\" can rebuild an out-of-date rev_map file.\n\n(usability, bells and whistles)\n\n* When you mistype a command name, git helpfully suggests what it guesses\n  you might have meant to say.  help.autocorrect configuration can be set\n  to a non-zero value to accept the suggestion when git can uniquely\n  guess.\n\n* The packfile machinery hopefully is more robust when dealing with\n  corrupt packs if redundant objects involved in the corruption are\n  available elsewhere.\n\n* \"git add -N path...\" adds the named paths as an empty blob, so that\n  subsequent \"git diff\" will show a diff as if they are creation events.\n\n* \"git add\" gained a built-in synonym for people who want to say \"stage\n  changes\" instead of \"add contents to the staging area\" which amounts\n  to the same thing.\n\n* \"git apply\" learned --include=paths option, similar to the existing\n  --exclude=paths option.\n\n* \"git bisect\" is careful about a user mistake and suggests testing of\n  merge base first when good is not a strict ancestor of bad.\n\n* \"git bisect skip\" can take a range of commits.\n\n* \"git blame\" re-encodes the commit metainfo to UTF-8 from i18n.commitEncoding\n  by default.\n\n* \"git check-attr --stdin\" can check attributes for multiple paths.\n\n* \"git checkout --track origin/hack\" used to be a syntax error.  It now\n  DWIMs to create a corresponding local branch \"hack\", i.e. acts as if you\n  said \"git checkout --track -b hack origin/hack\".\n\n* \"git checkout --ours/--theirs\" can be used to check out one side of a\n  conflicting merge during conflict resolution.\n\n* \"git checkout -m\" can be used to recreate the initial conflicted state\n  during conflict resolution.\n\n* \"git cherry-pick\" can also utilize rerere for conflict resolution.\n\n* \"git clone\" learned to be verbose with -v\n\n* \"git commit --author=$name\" can look up author name from existing\n  commits.\n\n* output from \"git commit\" has been reworded in a more concise and yet\n  more informative way.\n\n* \"git count-objects\" reports the on-disk footprint for packfiles and\n  their corresponding idx files.\n\n* \"git daemon\" learned --max-connections=<count> option.\n\n* \"git daemon\" exports REMOTE_ADDR to record client address, so that\n  spawned programs can act differently on it.\n\n* \"git describe --tags\" favours closer lightweight tags than farther\n  annotated tags now.\n\n* \"git diff\" learned to mimic --suppress-blank-empty from GNU diff via a\n  configuration option.\n\n* \"git diff\" learned to put more sensible hunk headers for Python,\n  HTML and ObjC contents.\n\n* \"git diff\" learned to vary the a/ vs b/ prefix depending on what are\n  being compared, controlled by diff.mnemonicprefix configuration.\n\n* \"git diff\" learned --dirstat-by-file to count changed files, not number\n  of lines, when summarizing the global picture.\n\n* \"git diff\" learned \"textconv\" filters --- a binary or hard-to-read\n  contents can be munged into human readable form and the difference\n  between the results of the conversion can be viewed (obviously this\n  cannot produce a patch that can be applied, so this is disabled in\n  format-patch among other things).\n\n* \"--cached\" option to \"git diff has an easier to remember synonym \"--staged\",\n  to ask \"what is the difference between the given commit and the\n  contents staged in the index?\"\n\n* \"git for-each-ref\" learned \"refname:short\" token that gives an\n  unambiguously abbreviated refname.\n\n* Auto-numbering of the subject lines is the default for \"git\n  format-patch\" now.\n\n* \"git grep\" learned to accept -z similar to GNU grep.\n\n* \"git help\" learned to use GIT_MAN_VIEWER environment variable before\n  using \"man\" program.\n\n* \"git imap-send\" can optionally talk SSL.\n\n* \"git index-pack\" is more careful against disk corruption while\n  completing a thin pack.\n\n* \"git log --check\" and \"git log --exit-code\" passes their underlying diff\n  status with their exit status code.\n\n* \"git log\" learned --simplify-merges, a milder variant of --full-history;\n  \"gitk --simplify-merges\" is easier to view than with --full-history.\n\n* \"git log\" learned \"--source\" to show what ref each commit was reached\n  from.\n\n* \"git log\" also learned \"--simplify-by-decoration\" to show the\n  birds-eye-view of the topology of the history.\n\n* \"git log --pretty=format:\" learned \"%d\" format element that inserts\n  names of tags that point at the commit.\n\n* \"git merge --squash\" and \"git merge --no-ff\" into an unborn branch are\n  noticed as user errors.\n\n* \"git merge -s $strategy\" can use a custom built strategy if you have a\n  command \"git-merge-$strategy\" on your $PATH.\n\n* \"git pull\" (and \"git fetch\") can be told to operate \"-v\"erbosely or\n  \"-q\"uietly.\n\n* \"git push\" can be told to reject deletion of refs with receive.denyDeletes\n  configuration.\n\n* \"git rebase\" honours pre-rebase hook; use --no-verify to bypass it.\n\n* \"git rebase -p\" uses interactive rebase machinery now to preserve the merges.\n\n* \"git reflog expire branch\" can be used in place of \"git reflog expire\n  refs/heads/branch\".\n\n* \"git remote show $remote\" lists remote branches one-per-line now.\n\n* \"git send-email\" can be given revision range instead of files and\n  maildirs on the command line, and automatically runs format-patch to\n  generate patches for the given revision range.\n\n* \"git submodule foreach\" subcommand allows you to iterate over checked\n  out submodules.\n\n* \"git submodule sync\" subcommands allows you to update the origin URL\n  recorded in submodule directories from the toplevel .gitmodules file.\n\n* \"git svn branch\" can create new branches on the other end.\n\n* \"gitweb\" can use more saner PATH_INFO based URL.\n\n(internal)\n\n* \"git hash-object\" learned to lie about the path being hashed, so that\n  correct gitattributes processing can be done while hashing contents\n  stored in a temporary file.\n\n* various callers of git-merge-recursive avoid forking it as an external\n  process.\n\n* Git class defined in \"Git.pm\" can be subclasses a bit more easily.\n\n* We used to link GNU regex library as a compatibility layer for some\n  platforms, but it turns out it is not necessary on most of them.\n\n* Some path handling routines used fixed number of buffers used alternately\n  but depending on the call depth, this arrangement led to hard to track\n  bugs.  This issue is being addressed.\n\n\nFixes since v1.6.0\n------------------\n\nAll of the fixes in v1.6.0.X maintenance series are included in this\nrelease, unless otherwise noted.\n\n* Porcelains implemented as shell scripts were utterly confused when you\n  entered to a subdirectory of a work tree from sideways, following a\n  symbolic link (this may need to be backported to older releases later).\n\n* Tracking symbolic links would work better on filesystems whose lstat()\n  returns incorrect st_size value for them.\n\n* \"git add\" and \"git update-index\" incorrectly allowed adding S/F when S\n  is a tracked symlink that points at a directory D that has a path F in\n  it (we still need to fix a similar nonsense when S is a submodule and F\n  is a path in it).\n\n* \"git am\" after stopping at a broken patch lost --whitespace, -C, -p and\n  --3way options given from the command line initially.\n\n* \"git diff --stdin\" used to take two trees on a line and compared them,\n  but we dropped support for such a use case long time ago.  This has\n  been resurrected.\n\n* \"git filter-branch\" failed to rewrite a tag name with slashes in it.\n\n* \"git http-push\" did not understand URI scheme other than opaquelocktoken\n  when acquiring a lock from the server (this may need to be backported to\n  older releases later).\n\n* After \"git rebase -p\" stopped with conflicts while replaying a merge,\n \"git rebase --continue\" did not work (may need to be backported to older\n  releases).\n\n* \"git revert\" records relative to which parent a revert was made when\n  reverting a merge.  Together with new documentation that explains issues\n  around reverting a merge and merging from the updated branch later, this\n  hopefully will reduce user confusion (this may need to be backported to\n  older releases later).\n\n* \"git rm --cached\" used to allow an empty blob that was added earlier to\n  be removed without --force, even when the file in the work tree has\n  since been modified.\n\n* \"git push --tags --all $there\" failed with generic usage message without\n  telling saying these two options are incompatible.\n\n* \"git log --author/--committer\" match used to potentially match the\n  timestamp part, exposing internal implementation detail.  Also these did\n  not work with --fixed-strings match at all.\n\n* \"gitweb\" did not mark non-ASCII characters imported from external HTML fragments\n  correctly.\n"},{"id":"98685","messageId":"46d6db660812250032h2e04b209mb3abd3bb747ddcf9@mail.gmail.com","threadId":"16860","inReplyTo":"7v7i5odams.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Christian MICHON","fromEmail":"christian.michon@gmail.com","sentAt":"2008-12-25T08:32:32Z","receivedAt":"2008-12-25T08:32:32Z","isPatch":false,"sender":{"key":"christian.michon@gmail.com","avatar":"https://gravatar.com/avatar/8a7c327b21187fbcab5c27640a49450eec72e0355dc292501197f27a5a744ec4?d=mp&s=160"},"body":"On Thu, Dec 25, 2008 at 7:36 AM, Junio C Hamano <gitster@pobox.com> wrote:\n> The latest feature release GIT 1.6.1 is available at the usual\n> places:\n>\n\na nice christmas gift indeed! thanks for making this release possible.\n\nmy best wishes to the list for xmas and incoming 2009!\n\n-- \nChristian\n--\nhttp://detaolb.sourceforge.net/, a linux distribution for Qemu with Git inside !\n"},{"id":"98686","messageId":"20081225100033.GA8451@b2j","threadId":"16860","inReplyTo":"7v7i5odams.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"bill lam","fromEmail":"cbill.lam@gmail.com","sentAt":"2008-12-25T10:00:33Z","receivedAt":"2008-12-25T10:00:33Z","isPatch":false,"sender":{"key":"cbill.lam@gmail.com","avatar":null},"body":"On ubuntu64 8.10, the NO_UINTMAX_T seems cause some trouble, I have to\ncomment out the block (in the Makefile),\n\nfdef NO_UINTMAX_T\n BASIC_CFLAGS += -Duintmax_t=uint32_t\nndif\n\nto make a successful compilation. Does it happen to me only?\n\n-- \nregards,\n====================================================\nGPG key 1024D/4434BAB3 2008-08-24\ngpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3\n唐詩282 白居易  後宮詞\n    淚濕羅巾夢不成  夜深前殿按歌聲  紅顏未老恩先斷  斜倚薰籠坐到明\n"},{"id":"98687","messageId":"7v3agca7fp.fsf@gitster.siamese.dyndns.org","threadId":"16860","inReplyTo":"20081225100033.GA8451@b2j","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-12-25T10:13:46Z","receivedAt":"2008-12-25T10:13:46Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"bill lam <cbill.lam@gmail.com> writes:\n\n> On ubuntu64 8.10, the NO_UINTMAX_T seems cause some trouble, I have to\n> comment out the block (in the Makefile),\n>\n> fdef NO_UINTMAX_T\n>  BASIC_CFLAGS += -Duintmax_t=uint32_t\n> ndif\n>\n> to make a successful compilation. Does it happen to me only?\n\nNobody reported it since that was added about a month and half ago.\n\nWhy are you building with NO_UINTMAX_T to begin with?  Isn't ubuntu 8.10 a\nrecent enough platform that ships with modern enough header files that\ndefine ANSI uintmax_t type?\n"},{"id":"98688","messageId":"20081225102527.GB8451@b2j","threadId":"16860","inReplyTo":"7v3agca7fp.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"bill lam","fromEmail":"cbill.lam@gmail.com","sentAt":"2008-12-25T10:25:27Z","receivedAt":"2008-12-25T10:25:27Z","isPatch":false,"sender":{"key":"cbill.lam@gmail.com","avatar":null},"body":"> Why are you building with NO_UINTMAX_T to begin with?  Isn't ubuntu 8.10 a\n> recent enough platform that ships with modern enough header files that\n> define ANSI uintmax_t type?\n\nNo, I did not do anything on that, \n\nmake clean\n./configure --prefix=/usr\nmake \n\nand this is the error logged.\n \nGIT_VERSION = 1.6.1\n    * new build flags or prefix\nIn file included from /usr/include/netinet/in.h:24,\n                 from git-compat-util.h:78,\n                 from builtin.h:4,\n                 from fast-import.c:142:\n/usr/include/stdint.h:136: error: conflicting types for ‘uint32_t’\n/usr/include/stdint.h:52: error: previous declaration of ‘uint32_t’ was here\nfast-import.c: In function ‘parse_progress’:\nfast-import.c:2339: warning: ignoring return value of ‘fwrite’, declared with attribute warn_unused_result\nmake: *** [fast-import.o] Error 1\n\nversion of gcc:\ngcc (Ubuntu 4.3.2-1ubuntu11) 4.3.2\n\n-- \nregards,\n====================================================\nGPG key 1024D/4434BAB3 2008-08-24\ngpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3\n唐詩105 杜甫  月夜\n    今夜鄜州月  閨中只獨看  遙憐小兒女  未解憶長安\n    香霧雲鬟濕  清輝玉臂寒  何時倚虛幌  雙照淚痕乾\n"},{"id":"98689","messageId":"49537197.1000805@lsrfire.ath.cx","threadId":"16860","inReplyTo":"20081225102527.GB8451@b2j","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"René Scharfe","fromEmail":"rene.scharfe@lsrfire.ath.cx","sentAt":"2008-12-25T11:42:15Z","receivedAt":"2008-12-25T11:42:15Z","isPatch":false,"sender":{"key":"l.s.r@web.de","avatar":"https://avatars.githubusercontent.com/u/26122331?v=4"},"body":"bill lam schrieb:\n>> Why are you building with NO_UINTMAX_T to begin with?  Isn't ubuntu 8.10 a\n>> recent enough platform that ships with modern enough header files that\n>> define ANSI uintmax_t type?\n> \n> No, I did not do anything on that, \n> \n> make clean\n> ./configure --prefix=/usr\n> make \n> \n> and this is the error logged.\n>  \n> GIT_VERSION = 1.6.1\n>     * new build flags or prefix\n> In file included from /usr/include/netinet/in.h:24,\n>                  from git-compat-util.h:78,\n>                  from builtin.h:4,\n>                  from fast-import.c:142:\n> /usr/include/stdint.h:136: error: conflicting types for ‘uint32_t’\n> /usr/include/stdint.h:52: error: previous declaration of ‘uint32_t’ was here\n> fast-import.c: In function ‘parse_progress’:\n> fast-import.c:2339: warning: ignoring return value of ‘fwrite’, declared with attribute warn_unused_result\n> make: *** [fast-import.o] Error 1\n> \n> version of gcc:\n> gcc (Ubuntu 4.3.2-1ubuntu11) 4.3.2\n\nI don't get any error on Ubuntu 8.10 on x86_64 (but several of those \nwarnings about ignored return values), neither with the tar file nor a \ncloned repository (in the former case I had to add a \"make configure\" \nstep, though).\n\nLine 52 of /usr/include/stdint.h:\n\ttypedef unsigned int\t\tuint32_t;\n\nLines 134-136 of /usr/include/stdint.h:\n\t#if __WORDSIZE == 64\n\ttypedef long int\t\tintmax_t;\n\ttypedef unsigned long int\tuintmax_t;\n\nIf you get to line 136, you probably are on a 64 bit installation, too, \ncorrect?\n\nconfigure seems to be confused.  What happens if you take it out of the \nequation, i.e. run the following commands?\n\n\t$ make distclean\n\t$ make prefix=/usr\n\nRené\n"},{"id":"98690","messageId":"20081225114443.GD6115@zakalwe.fi","threadId":"16860","inReplyTo":"7v7i5odams.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Heikki Orsila","fromEmail":"shdl@zakalwe.fi","sentAt":"2008-12-25T11:44:43Z","receivedAt":"2008-12-25T11:44:43Z","isPatch":false,"sender":{"key":"shdl@zakalwe.fi","avatar":null},"body":"On Wed, Dec 24, 2008 at 10:36:27PM -0800, Junio C Hamano wrote:\n> * The packfile machinery hopefully is more robust when dealing with\n>   corrupt packs if redundant objects involved in the corruption are\n>   available elsewhere.\n\nHas anyone written a summary of how Git's redundancy operates?\n\n* What would be the probability for a single bit flip to corrupt the \nrepository?\n\n* And what is the situation where a single bit flip can not corrupt the \ndatabase?\n\n* When (which commands/functions) is error detection done?\n\nHeikki\n"},{"id":"98691","messageId":"20081225140912.GA7110@b2j","threadId":"16860","inReplyTo":"49537197.1000805@lsrfire.ath.cx","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"bill lam","fromEmail":"cbill.lam@gmail.com","sentAt":"2008-12-25T14:09:12Z","receivedAt":"2008-12-25T14:09:12Z","isPatch":false,"sender":{"key":"cbill.lam@gmail.com","avatar":null},"body":"On Thu, 25 Dec 2008, René Scharfe wrote:\n> configure seems to be confused.  What happens if you take it out of the  \n> equation, i.e. run the following commands?\n>\n> \t$ make distclean\n> \t$ make prefix=/usr\n\nThanks René!  I followed your installation, deleted the old configure,\nand it compiled successfully,\n\nHappy holidays to everyone on mailing list!\n\n-- \nregards,\n====================================================\nGPG key 1024D/4434BAB3 2008-08-24\ngpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3\n唐詩104 李白  夜泊牛渚懷古\n    牛渚西江夜  青天無片雲  登舟望秋月  空憶謝將軍\n    余亦能高詠  斯人不可聞  明朝挂帆席  楓葉落紛紛\n"},{"id":"98696","messageId":"7vvdt884xu.fsf@gitster.siamese.dyndns.org","threadId":"16860","inReplyTo":"20081225102527.GB8451@b2j","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2008-12-25T18:50:37Z","receivedAt":"2008-12-25T18:50:37Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"bill lam <cbill.lam@gmail.com> writes:\n\n>> Why are you building with NO_UINTMAX_T to begin with?  Isn't ubuntu 8.10 a\n>> recent enough platform that ships with modern enough header files that\n>> define ANSI uintmax_t type?\n>\n> No, I did not do anything on that, \n>\n> make clean\n> ./configure --prefix=/usr\n> make \n\nI do not use configure myself (use of configure is entirely optional, and\nit is not tested often and core developers do not touch that part very\nmuch --- in a sense, use of autoconf is a second-class citizen in our\nbuild process); it is plausible that it has broken checks for detecting\nthe need of NO_UINTMAX_T.\n\nRelevant part of configure.ac reads like this:\n\n    # Define NO_UINTMAX_T if your platform does not have uintmax_t\n    AC_CHECK_TYPE(uintmax_t,\n    [NO_UINTMAX_T=],\n    [NO_UINTMAX_T=YesPlease],[\n    #include <inttypes.h>\n    ])\n    AC_SUBST(NO_UINTMAX_T)\n\nand I do not see anything suspicious there...\n\nRunning \"./configure --verbose\" might leave some clues in config.log; for\nme on my primary development box (Debian on x86_64), the relevant part\npasses the test (iow, inclusion of inttypes.h does give a working\nuintmax_t type) like this:\n\n    configure:5709: checking for uintmax_t\n    configure:5742: cc -c -g -O2  conftest.c >&5\n    configure:5748: $? = 0\n    configure:5763: result: yes\n"},{"id":"98697","messageId":"3af572ac0812251053q3b3c27e1i109a780fa657dd1f@mail.gmail.com","threadId":"16860","inReplyTo":"20081225100033.GA8451@b2j","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Pau Garcia i Quiles","fromEmail":"pgquiles@elpauer.org","sentAt":"2008-12-25T18:53:58Z","receivedAt":"2008-12-25T18:53:58Z","isPatch":false,"sender":{"key":"pgquiles@elpauer.org","avatar":null},"body":"On Thu, Dec 25, 2008 at 11:00 AM, bill lam <cbill.lam@gmail.com> wrote:\n> On ubuntu64 8.10, the NO_UINTMAX_T seems cause some trouble, I have to\n> comment out the block (in the Makefile),\n>\n> fdef NO_UINTMAX_T\n>  BASIC_CFLAGS += -Duintmax_t=uint32_t\n> ndif\n>\n> to make a successful compilation. Does it happen to me only?\n\nWorks for me both in Hardy and Intrepid. If you have trouble building\nfrom source, there are binary package for Hardy and Intrepid in my\nPPA:\nhttp://launchpad.net/~pgquiles/+archive\n\n-- \nPau Garcia i Quiles\nhttp://www.elpauer.org\n(Due to my workload, I may need 10 days to answer)\n"},{"id":"98706","messageId":"slrngl8dc8.klo.sitaramc@sitaramc.homelinux.net","threadId":"16860","inReplyTo":"20081225114443.GD6115@zakalwe.fi","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Sitaram Chamarty","fromEmail":"sitaramc@gmail.com","sentAt":"2008-12-26T01:38:16Z","receivedAt":"2008-12-26T01:38:16Z","isPatch":false,"sender":{"key":"sitaramc@gmail.com","avatar":"https://avatars.githubusercontent.com/u/43316?v=4"},"body":"[\"Followup-To:\" header set to gmane.comp.version-control.git.]\nOn 2008-12-25, Heikki Orsila <shdl@zakalwe.fi> wrote:\n> * What would be the probability for a single bit flip to corrupt the \n> repository?\n\n> * And what is the situation where a single bit flip can not corrupt the \n> database?\n\nThese two questions are not specific to git; whatever\nanswers apply to sha1 also apply here.  Sha1 will always be\nstrong enough to detect any combination of random errors.\nAs for deliberate attacks, it is considered strong enough to\nresist all but the most computationally intensive attacks\n(the kind that require a worldwide effort or maybe NSA type\nfacilities).\n\n> * When (which commands/functions) is error detection done?\n\n\"git fsck --full\" will do it of course, but I'm not sure\nwhat other operations also start off or end up doing a\ncheck.\n"},{"id":"98731","messageId":"20081226143339.GB7103@b2j","threadId":"16860","inReplyTo":"3af572ac0812251053q3b3c27e1i109a780fa657dd1f@mail.gmail.com","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"bill lam","fromEmail":"cbill.lam@gmail.com","sentAt":"2008-12-26T14:33:39Z","receivedAt":"2008-12-26T14:33:39Z","isPatch":false,"sender":{"key":"cbill.lam@gmail.com","avatar":null},"body":"On Thu, 25 Dec 2008, Pau Garcia i Quiles wrote:\n> Works for me both in Hardy and Intrepid. If you have trouble building\n> from source, there are binary package for Hardy and Intrepid in my\n> PPA:\n> http://launchpad.net/~pgquiles/+archive\n\nThanks you head up. No problem now, but I'll bookmark your page.\n\n-- \nregards,\n====================================================\nGPG key 1024D/4434BAB3 2008-08-24\ngpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3\n唐詩281 劉禹錫  春詞\n    新妝宜面下朱樓  深鎖春光一院愁  行到中庭數花朵  蜻蜓飛上玉搔頭\n"},{"id":"98732","messageId":"20081226144250.GC7103@b2j","threadId":"16860","inReplyTo":"7vvdt884xu.fsf@gitster.siamese.dyndns.org","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"bill lam","fromEmail":"cbill.lam@gmail.com","sentAt":"2008-12-26T14:42:50Z","receivedAt":"2008-12-26T14:42:50Z","isPatch":false,"sender":{"key":"cbill.lam@gmail.com","avatar":null},"body":"On Thu, 25 Dec 2008, Junio C Hamano wrote:\n> build process); it is plausible that it has broken checks for detecting\n> the need of NO_UINTMAX_T.\n> \n> Relevant part of configure.ac reads like this:\n> \n>     # Define NO_UINTMAX_T if your platform does not have uintmax_t\n>     AC_CHECK_TYPE(uintmax_t,\n>     [NO_UINTMAX_T=],\n>     [NO_UINTMAX_T=YesPlease],[\n>     #include <inttypes.h>\n>     ])\n>     AC_SUBST(NO_UINTMAX_T)\n> \n> and I do not see anything suspicious there...\n\nThanks, I guess the configure is not part of the repository and I use\na legacy configure from earlier version. After I run the make\nconfigure again, it is now ok.\n\n-- \nregards,\n====================================================\nGPG key 1024D/4434BAB3 2008-08-24\ngpg --keyserver subkeys.pgp.net --recv-keys 4434BAB3\n唐詩093 駱賓王  在獄詠蟬\n    西路蟬聲唱  南冠客思侵  那堪玄鬢影  來對白頭吟\n    露重飛難進  風多響易沉  無人信高潔  誰為表予心\n"},{"id":"98769","messageId":"alpine.LFD.2.00.0812262025320.30266@xanadu.home","threadId":"16860","inReplyTo":"20081225114443.GD6115@zakalwe.fi","subject":"Re: [ANNOUNCE] GIT 1.6.1","fromName":"Nicolas Pitre","fromEmail":"nico@cam.org","sentAt":"2008-12-27T01:56:07Z","receivedAt":"2008-12-27T01:56:07Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 25 Dec 2008, Heikki Orsila wrote:\n\n> On Wed, Dec 24, 2008 at 10:36:27PM -0800, Junio C Hamano wrote:\n> > * The packfile machinery hopefully is more robust when dealing with\n> >   corrupt packs if redundant objects involved in the corruption are\n> >   available elsewhere.\n> \n> Has anyone written a summary of how Git's redundancy operates?\n> \n> * What would be the probability for a single bit flip to corrupt the \n> repository?\n\nThis is not something that git itself could answer.  The probability \ndepends on the quality of your hardware.  Once that probability has \noccurred though, it is clear that your repository is then corrupted as \nthere is hardly any redundant bits in a git repository.\n\n> * And what is the situation where a single bit flip can not corrupt the \n> database?\n\nThe database can be resilient against most kind of corruptions if you \nhave a redundant copy of the affected object.  It will still be \ncorrupted, but git is now able to detect corruptions gracefully and \nfunction correctly with some fallback objects.  Those objects must exist \nin some other related repository though, and copied over to the affected \nrepository manually. It's then possible and recommended to \"fix\" the \ncorruption simply by repacking the repository at that point.\n\nSo there is no magic involved: you need to have some kind of backups, \neither using traditional backup solutions, or by simply having your \nrepository cloned somewhere else.  The idea of having a repository fixed \nwith redundant objects is for those cases where you need to salvage new \nwork that has no corresponding backup, but although corresponding \nobjects are not corrupted, they could be delta objects which base is \nagainst old objects which happen to be corrupted.\n\n> * When (which commands/functions) is error detection done?\n\nError detection is performed all the time.  When it's not the more \nexpensive SHA1 checksum, at least the zlib CRC32 is verified.  What the \nlatest git version does amongst other corruption related things is to \nclose some small holes where some specific kind of corruptions could \nhave been undetected and propagated from one pack to another when \nrepacking.\n\n\nNicolas\n"},{"id":"98785","messageId":"E60BAAB7-2C39-4932-BCD0-B880E97EE235@zib.de","threadId":"16860","inReplyTo":"7v7i5odams.fsf@gitster.siamese.dyndns.org","subject":"[ANNOUNCE] MSYSGIT 1.6.1","fromName":"Steffen Prohaska","fromEmail":"prohaska@zib.de","sentAt":"2008-12-27T10:31:01Z","receivedAt":"2008-12-27T10:31:01Z","isPatch":false,"sender":{"key":"prohaska@zib.de","avatar":"https://avatars.githubusercontent.com/u/217580?v=4"},"body":"\nOn Dec 25, 2008, at 7:36 AM, Junio C Hamano wrote:\n\n> The latest feature release GIT 1.6.1 is available at the usual\n> places:\n\n\nThe msysgit installer is available at\n\n     http://code.google.com/p/msysgit/downloads\n\n\tSteffen\n"}]}