[ANNOUNCE] GIT 1.5.6

[ANNOUNCE] GIT 1.5.6

[Junio C Hamano]

The latest feature release GIT 1.5.6 is available at the usual
places:

  http://www.kernel.org/pub/software/scm/git/

  git-1.5.6.tar.{gz,bz2}			(tarball)
  git-htmldocs-1.5.6.tar.{gz,bz2}		(preformatted docs)
  git-manpages-1.5.6.tar.{gz,bz2}		(preformatted docs)
  RPMS/$arch/git-*-1.5.6-1.$arch.rpm	(RPM)

As promised, this cycle was short and the release is with only relatively
small impact changes.

----------------------------------------------------------------
GIT v1.5.6 Release Notes
========================

Updates since v1.5.5
--------------------

(subsystems)

* Comes with updated gitk and git-gui.

(portability)

* git will build on AIX better than before now.

* core.ignorecase configuration variable can be used to work better on
  filesystems that are not case sensitive.

* "git init" now autodetects the case sensitivity of the filesystem and
  sets core.ignorecase accordingly.

* cpio is no longer used; neither "curl" binary (libcurl is still used).

(documentation)

* Many freestanding documentation pages have been converted and made
  available to "git help" (aka "man git<something>") as section 7 of
  the manual pages. This means bookmarks to some HTML documentation
  files may need to be updated (eg "tutorial.html" became
  "gittutorial.html").

(performance)

* "git clone" was rewritten in C.  This will hopefully help cloning a
  repository with insane number of refs.

* "git rebase --onto $there $from $branch" used to switch to the tip of
  $branch only to immediately reset back to $from, smudging work tree
  files unnecessarily.  This has been optimized.

* Object creation codepath in "git-svn" has been optimized by enhancing
  plumbing commands git-cat-file and git-hash-object.

(usability, bells and whistles)

* "git add -p" (and the "patch" subcommand of "git add -i") can choose to
  apply (or not apply) mode changes independently from contents changes.

* "git bisect help" gives longer and more helpful usage information.

* "git bisect" does not use a special branch "bisect" anymore; instead, it
  does its work on a detached HEAD.

* "git branch" (and "git checkout -b") can be told to set up
  branch.<name>.rebase automatically, so that later you can say "git pull"
  and magically cause "git pull --rebase" to happen.

* "git branch --merged" and "git branch --no-merged" can be used to list
  branches that have already been merged (or not yet merged) to the
  current branch.

* "git cherry-pick" and "git revert" can add a sign-off.

* "git commit" mentions the author identity when you are committing
  somebody else's changes.

* "git diff/log --dirstat" output is consistent between binary and textual
  changes.

* "git filter-branch" rewrites signed tags by demoting them to annotated.

* "git format-patch --no-binary" can produce a patch that lack binary
  changes (i.e. cannot be used to propagate the whole changes) meant only
  for reviewing.

* "git init --bare" is a synonym for "git --bare init" now.

* "git gc --auto" honors a new pre-auto-gc hook to temporarily disable it.

* "git log --pretty=tformat:<custom format>" gives a LF after each entry,
  instead of giving a LF between each pair of entries which is how
  "git log --pretty=format:<custom format>" works.

* "git log" and friends learned the "--graph" option to show the ancestry
  graph at the left margin of the output.

* "git log" and friends can be told to use date format that is different
  from the default via 'log.date' configuration variable.

* "git send-email" now can send out messages outside a git repository.

* "git send-email --compose" was made aware of rfc2047 quoting.

* "git status" can optionally include output from "git submodule
  summary".

* "git svn" learned --add-author-from option to propagate the authorship
  by munging the commit log message.

* new object creation and looking up in "git svn" has been optimized.

* "gitweb" can read from a system-wide configuration file.

(internal)

* "git unpack-objects" and "git receive-pack" is now more strict about
  detecting breakage in the objects they receive over the wire.


Fixes since v1.5.5
------------------

All of the fixes in v1.5.5 maintenance series are included in
this release, unless otherwise noted.

And there are too numerous small fixes to otherwise note here ;-)


----------------------------------------------------------------

Changes since v1.5.5 are as follows:

A Large Angry SCM (1):
      git-repack: re-enable parsing of -n command line option

Adam Roben (11):
      Add tests for git cat-file
      git-cat-file: Small refactor of cmd_cat_file
      git-cat-file: Make option parsing a little more flexible
      git-cat-file: Add --batch-check option
      git-cat-file: Add --batch option
      Move git-hash-object tests from t5303 to t1007
      Add more tests for git hash-object
      git-hash-object: Add --stdin-paths option
      Git.pm: Add command_bidi_pipe and command_close_bidi_pipe
      Git.pm: Add hash_and_insert_object and cat_blob
      git-svn: Speed up fetch

Adam Simpkins (15):
      Remove dead code: show_log() sep argument and diff_options.msg_sep
      log: print log entry terminator even if the message is empty
      revision API: split parent rewriting and parent printing options
      Add history graph API
      log and rev-list: add --graph option
      graph API: eliminate unnecessary indentation
      graph API: fix graph mis-alignment after uninteresting commits
      graph API: don't print branch lines for uninteresting merge parents
      log --graph --left-right: show left/right information in place of '*'
      Fix output of "git log --graph --boundary"
      get_revision(): honor the topo_order flag for boundary commits
      graph API: improve display of merge commits
      graph API: avoid printing unnecessary padding before some octopus merges
      graph API: fix "git log --graph --first-parent"
      git log --graph: print '*' for all commits, including merges

Alberto Bertogli (1):
      builtin-apply: Show a more descriptive error on failure when opening a patch

Alejandro Mery (1):
      git-am: head -1 is obsolete and doesn't work on some new systems

Alex Riesen (13):
      Use "=" instead of "==" in condition as it is more portable
      Fix use after free() in builtin-fetch
      Use the modern syntax of git-diff-files in t2002-checkout-cache-u.sh
      Improve reporting of errors in config file routines
      Make the exit code of add_file_to_index actually useful
      Extend interface of add_files_to_cache to allow ignore indexing errors
      Add --ignore-errors to git-add to allow it to skip files with read errors
      Add a test for git-add --ignore-errors
      Add a config option to ignore errors for git-add
      Ensure that a test is run in the trash directory
      Fix t6031 on filesystems without working exec bit
      Fix t3701 if core.filemode disabled
      Fix t5516 on cygwin: it does not like double slashes at the beginning of a path

Anders Waldenborg (1):
      gitweb: Convert string to internal form before chopping in chop_str

Andy Parkins (1):
      post-receive-email: fix accidental removal of a trailing space in signature line

Ariel Badichi (2):
      copy.c: copy_fd - correctly report write errors
      archive.c: format_subst - fixed bogus argument to memchr

Ask Bjørn Hansen (2):
      gitweb setup instruction: rewrite HEAD and root as well
      send-email: Allow the envelope sender to be set via configuration

Avery Pennarun (5):
      git-svn: add documentation for --use-log-author option.
      git-svn: Add --add-author-from option.
      git-svn: add documentation for --add-author-from option.
      git-svn: don't append extra newlines at the end of commit messages.
      git-svn: test that extra blank lines aren't inserted in commit messages.

Bart Trojanowski (1):
      make git-status use a pager

Björn Steinbrink (2):
      Fix section about backdating tags in the git-tag docs
      name-rev: Fix segmentation fault when using --all

Boyd Lynn Gerber (2):
      progress.c: avoid use of dynamic-sized array
      Port to 12 other Platforms.

Brandon Casey (8):
      filter-branch.sh: support nearly proper tag name filtering
      git-clone.txt: Adjust note to --shared for new pruning behavior of git-gc
      compat/fopen.c: avoid clobbering the system defined fopen macro
      repack: modify behavior of -A option to leave unreferenced objects unpacked
      git-gc: always use -A when manually repacking
      builtin-gc.c: deprecate --prune, it now really has no effect
      builtin-clone.c: Need to closedir() in copy_or_link_directory()
      t/Makefile: "trash" directory was renamed recently

Bryan Donlan (10):
      git-rebase.sh: Fix --merge --abort failures when path contains whitespace
      config.c: Escape backslashes in section names properly
      git-send-email.perl: Handle shell metacharacters in $EDITOR properly
      test-lib.sh: Add a test_set_editor function to safely set $VISUAL
      Use test_set_editor in t9001-send-email.sh
      test-lib.sh: Fix some missing path quoting
      lib-git-svn.sh: Fix quoting issues with paths containing shell metacharacters
      Don't use the 'export NAME=value' in the test scripts.
      Fix tests breaking when checkout path contains shell metacharacters
      Rename the test trash directory to contain spaces.

Caio Marcelo de Oliveira Filho (1):
      git-format-patch: add --no-binary to omit binary changes in the patch.

Carlos Rica (2):
      Fix documentation syntax of optional arguments in short options.
      core-tutorial.txt: Fix showing the current behaviour.

Chris Frey (2):
      Documentation/git-prune.txt: document unpacked logic
      Documentation/git-repack.txt: document new -A behaviour

Chris Parsons (1):
      Updated status to show 'Not currently on any branch' in red

Chris Ridd (1):
      Improve sed portability

Christian Couder (32):
      bisect: add "git bisect help" subcommand to get a long usage string
      bisect: fix bad rev checking in "git bisect good"
      bisect: report bad rev better
      bisect: squelch "fatal: ref HEAD not a symref" misleading message
      git-bisect: make "start", "good" and "skip" succeed or fail atomically
      help: use man viewer path from "man.<tool>.path" config var
      documentation: help: add "man.<tool>.path" config variable
      help: use "man.<tool>.cmd" as custom man viewer command
      documentation: help: add info about "man.<tool>.cmd" config var
      documentation: web--browse: add a note about konqueror
      rev-parse: teach "--verify" to be quiet when using "-q" or "--quiet"
      rev-parse: fix --verify to error out when passed junk after a good rev
      Documentation: hooks: fix missing verb in pre-applypatch description
      Documentation: rename "hooks.txt" to "githooks.txt" and make it a man page
      Documentation: improve "add", "pull" and "format-patch" examples
      Documentation: bisect: add a few "git bisect run" examples
      bisect: print an error message when "git rev-list --bisect-vars" fails
      rev-parse: add test script for "--verify"
      rev-parse: fix using "--default" with "--verify"
      rev-parse --verify: do not output anything on error
      Documentation: rev-parse: add a few "--verify" and "--default" examples
      bisect: add test cases to check that "git bisect start" is atomic
      bisect: fix left over "BISECT_START" file when starting with junk rev
      bisect: trap critical errors in "bisect_start"
      bisect: use a detached HEAD to bisect
      Documentation: convert tutorials to man pages
      bisect: use "$GIT_DIR/BISECT_START" to check if we are bisecting
      Documentation: convert "glossary" and "core-tutorial" to man pages
      documentation: convert "diffcore" and "repository-layout" to man pages
      documentation: move git(7) to git(1)
      documentation: bisect: remove bits talking about a bisection branch
      Documentation: RelNotes-1.5.6: talk about renamed HTML files

Christian Engwer (1):
      git-svn fails in prop_walk if $self->{path} is not empty

Christian Stimming (3):
      git-gui: Update German translation
      gitk: Update German translation
      gitk: German translation again updated

Clemens Buchacher (2):
      Reset the signal being handled
      http-push: remove remote locks on exit signals

Clifford Caoile (2):
      Docs gitk: Explicitly mention the files that gitk uses (~/.gitk)
      git.el: Set process-environment instead of invoking env

Dan McGee (4):
      completion: allow 'git remote' subcommand completion
      completion: remove use of dashed git commands
      Allow cherry-pick (and revert) to add signoff line
      Remove 'header' from --signoff option description

Daniel Barkalow (14):
      Fix config key miscount in url.*.insteadOf
      Make walker.fetch_ref() take a struct ref.
      Make ls-remote http://... list HEAD, like for git://...
      Mark the list of refs to fetch as const
      Add a lockfile function to append to a file
      Add a library function to add an alternate to the alternates file
      Have a constant extern refspec for "--tags"
      Allow for having for_each_ref() list extra refs
      Add a function to set a non-default work tree
      Provide API access to init_db()
      Build in clone
      clone: fall back to copying if hardlinking fails
      Test that --reference actually suppresses fetching referenced objects
      Use nonrelative paths instead of absolute paths for cloned repositories

Dirk Suesserott (2):
      Documentation/git-request-pull: Fixed a typo ("send" -> "end")
      Documentation/git-mailsplit: Enhanced description of -o option

Dmitry Potapov (2):
      git-gc --prune is deprecated
      git-init: autodetect core.ignorecase

Dmitry V. Levin (1):
      builtin-fetch.c (store_updated_refs): Honor update_local_ref() return value

Dustin Sallings (3):
      Documentation/config.txt: Mention branch.<name>.rebase applies to "git pull"
      Allow tracking branches to set up rebase by default.
      Allow tracking branches to set up rebase by default.

Eric Wong (1):
      git-svn: fix cloning of HTTP URLs with '+' in their path

Flavio Poletti (1):
      git-instaweb: improve auto-discovery of httpd and call conventions.

Florian Ragwitz (1):
      filter-branch: Documentation fix.

Frank Lichtenheld (4):
      var: Don't require to be in a git repository.
      Git.pm: Don't require a repository instance for config
      Git.pm: Don't require repository instance for ident
      send-email: Don't require to be called in a repository

Fred Maranhão (1):
      fix typo in tutorial

Geoffrey Irving (1):
      doc: adding gitman.info and *.texi to .gitignore

Gerrit Pape (7):
      gitweb: fallback to system-wide config file if default config does not exist
      gitweb: fallback to system-wide config file (fixup)
      diff-options.txt: document the new "--dirstat" option
      gitk: Makefile/install: force permissions when installing files and dirs
      git-bisect.sh: don't accidentally override existing branch "bisect"
      Documentation/git-bundle.txt: fix synopsis
      commit --interactive: properly update the index before commiting

Govind Salinas (1):
      pretty.c: add %x00 format specifier.

Gustaf Hendeby (6):
      git-svn: Make create-ignore use git add -f
      Documentation: Add create-ignore to git svn manual
      Documentation/config.txt: Add git-gui options
      Documentation: Add missing git svn commands
      Documentation: Fix skipped section level
      Make git add -n and git -u -n output consistent

Heikki Orsila (8):
      Make core.sharedRepository more generic
      Document functions xmemdupz(), xread() and xwrite()
      Die for an early EOF in a file reading loop
      Make read_in_full() and write_in_full() consistent with xread() and xwrite()
      Cleanup xread() loops to use read_in_full()
      Add missing "short" alternative to --date in rev-list-options.txt
      Add log.date config variable
      Remove redundant code, eliminate one static variable

Horst H. von Brand (1):
      Fix recipient santitization

Ian Hilt (1):
      Documentation/git-describe.txt: make description more readable

Imran M Yousuf (1):
      Use '-f' option to point to the .gitmodules file

Jakub Narebski (7):
      gitweb: Fix 'history' view for deleted files with history
      gitweb: Use feed link according to current view
      gitweb: Remove gitweb/test/ directory
      gitweb: Fix "next" link on bottom of page
      gitweb: Add charset info to "raw" output of 'text/plain' blobs
      gitweb: Make it work with $GIT containing spaces
      Use 'trash directory' thoroughly in t/test-lib.sh

Jamis Buck (1):
      git-reset: honor -q and do not show progress message

Jeff King (30):
      add--interactive: ignore mode change in 'p'atch command
      add--interactive: allow user to choose mode update
      git-fetch: fix status output when not storing tracking ref
      git-fetch: always show status of non-tracking-ref fetches
      git-remote: show all remotes with "git remote show"
      Don't force imap.host to be set when imap.tunnel is set
      t5516: remove ambiguity test (1)
      doc/git-gc: add a note about what is collected
      push: allow unqualified dest refspecs to DWIM
      remote: create fetch config lines with '+'
      fix reflog approxidate parsing bug
      cvsimport: always pass user data to "system" as a list
      Documentation: point git-prune users to git-gc
      add merge.renamelimit config option
      bump rename limit defaults
      diff: make "too many files" rename warning optional
      checkout: don't rfc2047-encode oneline on detached HEAD
      doc: clarify definition of "update" for git-add -u
      fix bsd shell negation
      t5000: tar portability fix
      clone: bsd shell portability fix
      filter-branch: fix variable export logic
      doc/git-daemon: s/uploadarchive/uploadarch/
      git-am: fix typo in usage message
      send-email: specify content-type of --compose body
      send-email: rfc2047-quote subject lines with non-ascii characters
      clone: make sure we support the transport type
      Fix "git clone http://$URL" to check out the worktree when asked
      document --pretty=tformat: option
      clean up error conventions of remote.c:match_explicit

Johan Herland (5):
      Add a test for another combination of --reference
      Add test for cloning with "--reference" repo being a subset of source repo
      cpio is no longer used by git-clone
      Consistency: Use "libcurl" instead of "cURL library" and "curl"
      The "curl" executable is no longer required

Johannes Schindelin (12):
      Provide git_config with a callback-data parameter
      builtin-clone: fix initial checkout
      cvsexportcommit: chomp only removes trailing whitespace
      diff options: Introduce --ignore-submodules
      Teach update-index about --ignore-submodules
      Ignore dirty submodule states during rebase and stash
      cvsexportcommit: introduce -W for shared working trees (between Git and CVS)
      submodule update: add convenience option --init
      pull --rebase: exit early when the working directory is dirty
      mailsplit and mailinfo: gracefully handle NUL characters
      hg-to-git: add --verbose option
      merge-recursive: respect core.autocrlf when writing out the result

Johannes Sixt (11):
      Document option --only of git commit
      builtin-commit.c: Remove a redundant assignment.
      git-gui: Report less precise object estimates for database compression
      compat-util: avoid macro redefinition warning
      wt-status.h: declare global variables as extern
      rev-parse --symbolic-full-name: don't print '^' if SHA1 is not a ref
      t5700-clone-reference: Quote $U
      Revert "filter-branch: subdirectory filter needs --full-history"
      rebase --interactive: Compute upstream SHA1 before switching branches
      make_nonrelative_path: Use is_absolute_path()
      Remove exec bit from builtin-fast-export.c

John J. Franey (1):
      Clarify description of <repository> argument to pull/fetch for naming remotes.

Jon Loeliger (4):
      Clarify and fix English in "git-rm" documentation
      Add otherwise missing --strict option to unpack-objects summary.
      git-filter-branch: Clarify file removal example.
      git-show.txt: Not very stubby these days.

Jonas Fonseca (1):
      git-remote: reject adding remotes with invalid names

Junio C Hamano (80):
      Optimize rename detection for a huge diff
      t5300: add test for "unpack-objects --strict"
      unpack-objects: fix --strict handling
      rebase [--onto O] A B: omit needless checkout
      sha1-lookup: more memory efficient search in sorted list of SHA-1
      diff: make --dirstat binary-file safe
      sha1-lookup: make selection of 'middle' less aggressive
      log: teach "terminator" vs "separator" mode to "--pretty=format"
      Document -w option to shortlog
      Documentation/git-submodule: typofix
      git_config_bool_or_int()
      t7401: squelch garbage output
      write_index(): optimize ce_smudge_racily_clean_entry() calls with CE_UPTODATE
      diff-files: mark an index entry we know is up-to-date as such
      Fix git_config_bool_or_int
      rebase: do not munge commit log message
      git-am: minor cleanup
      am: POSIX portability fix
      GIT 1.5.5.1
      First batch of post 1.5.5 updates
      write-tree: properly detect failure to write tree objects
      clone: detect and fail on excess parameters
      fetch-pack: brown paper bag fix
      diff: a submodule not checked out is not modified
      diff-lib.c: rename check_work_tree_entity()
      is_racy_timestamp(): do not check timestamp for gitlinks
      git-svn: add test for --add-author-from and --use-log-author
      builtin-apply: typofix
      builtin-apply: accept patch to an empty file
      builtin-apply: do not declare patch is creation when we do not know it
      unpack-trees: allow Porcelain to give different error messages
      "git-add -n -u" should not add but just report
      tests: do not use implicit "git diff --no-index"
      diff-files: do not play --no-index games
      "git diff": do not ignore index without --no-index
      mailinfo: apply the same fix not to lose NULs in BASE64 and QP codepaths
      mailsplit: minor clean-up in read_line_with_nul()
      Update draft release notes for 1.5.6
      log --graph: do not accept log --graphbogus
      log --pretty: do not accept bogus "--prettyshort"
      Release Notes for 1.5.5.2
      Documentation/git.txt: link to 1.5.5.2 documentation.
      Makefile: fix dependency on wt-status.h
      show-branch --current: do not barf on detached HEAD
      git-diff: allow  --no-index semantics a bit more
      git diff --no-index: default to page like other diff frontends
      GIT 1.5.5.3
      t5100: Avoid filename "nul"
      Git::cat_blob: allow using an empty blob to fix git-svn breakage
      fix sha1_pack_index_name()
      Manual subsection to refer to other pages is SEE ALSO
      Documentation: git-cherry uses git-patch-id
      "git checkout -- paths..." should error out when paths cannot be written
      checkout: make reset_clean_to_new() not die by itself
      checkout: consolidate reset_{to_new,clean_to_new}()
      unpack_trees(): allow callers to differentiate worktree errors from merge errors
      checkout: "best effort" checkout
      commit: drop duplicated parents
      GIT v1.5.6-rc1
      t7502: do not globally unset GIT_COMMITTER_* environment variables
      t7502: tighten loosely written test sequence
      Documentation: git-log cannot use rev-list specific options
      t7502: honor SHELL_PATH
      GIT 1.5.5.4
      GIT 1.5.6-rc2
      http-push.c: remove duplicated code
      "remote prune": be quiet when there is nothing to prune
      Documentation/git-pull.txt: Use more standard [NOTE] markup
      Documentation: exclude @pxref{[REMOTES]} from texinfo intermediate output
      user-manual: describe how higher stages are set during a merge
      t4126: fix test that happened to work due to timing
      sha1_file.c: dead code removal
      GIT 1.5.6-rc3
      Makefile: update check-docs target
      Update RPM spec to drop curl executable requirement
      diff.c: fix emit_line() again not to add extra line
      create_tempfile: make sure that leading directories can be accessible by peers
      sha1_file.c: simplify parse_pack_index()
      builtin-rerere: fix a small leak
      GIT 1.5.6

Jörg Sommer (1):
      post-merge: Add it's not executed if merge failed.

Karl Hasselström (3):
      Add some tests for git update-ref -d
      Fix path duplication in git svn commit-diff
      Revert "git.el: Set process-environment instead of invoking env"

Kevin Ballard (1):
      Documentation/git-filter-branch.txt: Fix description of --commit-filter

Krzysztof Kowalczyk (1):
      alloc_ref_from_str(): factor out a common pattern of alloc_ref from string

Lars Hjemli (8):
      Add platform-independent .git "symlink"
      Teach resolve_gitlink_ref() about the .git file
      Teach git-submodule.sh about the .git file
      Teach GIT-VERSION-GEN about the .git file
      git-branch: add support for --merged and --no-merged
      git-branch.txt: compare --contains, --merged and --no-merged
      Add tests for `branch --[no-]merged`
      revision.c: really honor --first-parent

Lea Wiemann (13):
      gitweb: only display "next" links in logs if there is a next page
      t/test-lib.sh: resolve symlinks in working directory, for pathname comparisons
      Git.pm: fix documentation of hash_object
      glossary: improve a few links
      Git.pm: fix return value of config method
      cat-file --batch: flush stdout also when objects are missing
      git-for-each-ref.txt: minor improvements
      t1006-cat-file.sh: typo
      cat-file --batch / --batch-check: do not exit if hashes are missing
      Documentation/git-cat-file.txt: add missing line break
      t/.gitattributes: only ignore whitespace errors in test files
      gitweb: quote commands properly when calling the shell
      gitweb: remove unused parse_ref method

Linus Torvalds (22):
      Make unpack_trees_options bit flags actual bitfields
      Move name hashing functions into a file of its own
      Make "index_name_exists()" return the cache_entry it found
      Make hash_name_lookup able to do case-independent lookups
      Add 'core.ignorecase' option
      Make branch merging aware of underlying case-insensitive filsystems
      Make unpack-tree update removed files before any updated files
      When adding files to the index, add support for case-independent matches
      Make git-add behave more sensibly in a case-insensitive environment
      Ignore leading empty lines while summarizing merges
      git-am: cope better with an empty Subject: line
      Optimize match_pathspec() to avoid fnmatch()
      fetch-pack: do not stop traversing an already parsed commit
      Avoid some unnecessary lstat() calls
      Optimize symlink/directory detection
      Make pack creation always fsync() the result
      Remove now unnecessary 'sync()' calls
      Consolidate SHA1 object file close
      Avoid cross-directory renames and linking on object creation
      Make loose object file reading more careful
      Simplify and rename find_sha1_file()
      write_loose_object: don't bother trying to read an old object

Liu Yubao (1):
      Documentation on --git-dir and --work-tree

Luciano Rocha (1):
      git-init: accept --bare option

Marcel Koeppen (2):
      Replace in-place sed in t7502-commit
      Fix prepare-commit-msg hook and replace in-place sed

Marius Storm-Olsen (3):
      Clearify the documentation for core.ignoreStat
      Add shortcut in refresh_cache_ent() for marked entries.
      Add testcase for merging in a CRLF repo

Mark Hills (1):
      Be more careful with objects directory permissions on clone

Mark Levedahl (2):
      git-submodule - possibly use branch name to describe a module
      git-submodule - Fix errors regarding resolve_relative_url

Martin Koegler (3):
      unpack-objects: prevent writing of inconsistent objects
      receive-pack: allow using --strict mode for unpacking objects
      t5300: add test for "index-pack --strict"

Matt Graham (1):
      Linked glossary from cvs-migration page

Matthew Ogilvie (4):
      gitattributes: Fix subdirectory attributes specified from root directory
      git-cvsserver: add mechanism for managing working tree and current directory
      implement gitcvs.usecrlfattr
      git-cvsserver: add ability to guess -kb from contents

Matthias Kestenholz (1):
      Use color.ui variable in scripts too

Matthieu Moy (2):
      Document that WebDAV doesn't need git on the server, and works over SSL
      git-svn: detect and fail gracefully when dcommitting to a void

Michael Dressel (1):
      describe: match pattern for lightweight tags too

Michael Weber (1):
      svn-git: Use binmode for reading/writing binary rev maps

Michele Ballabio (6):
      revision.c: make --date-order overriddable
      gitk: Disable "Reset %s branch to here" when on a detached head
      gitk: Move es.po where it belongs
      builtin-cat-file.c: use parse_options()
      change quoting in test t1006-cat-file.sh
      Documentation: fix graph in git-rev-parse.txt

Mikael Magnusson (1):
      Typo in RelNotes.

Mike Hommey (1):
      Don't allocate too much memory in quote_ref_url

Mike Ralphson (1):
      Makefile: update the default build options for AIX

Miklos Vajna (19):
      git-gc --auto: add pre-auto-gc hook
      Documentation/hooks: add pre-auto-gc hook
      contrib/hooks: add an example pre-auto-gc hook
      diff options documentation: refer to --diff-filter in --name-status
      git checkout: add -t alias for --track
      git-format-patch: add a new format.cc configuration variable
      git-send-email: add a new sendemail.cc configuration variable
      Add tests for sendemail.cc configuration variable
      INSTALL: add a note about GNU interactive tools has been renamed
      git-fast-import: rename cmd_*() functions to parse_*()
      git-merge: exclude unnecessary options from OPTIONS_SPEC
      CodingGuidelines: Add a note to avoid assignments inside if()
      Revision walking documentation: document most important functions
      Strbuf documentation: document most functions
      Remove unused code in parse_commit_buffer()
      git-rebase -i: mention the short command aliases in the todo list
      git-read-tree: document -v option.
      run-command documentation: fix "memset()" parameter
      path-list documentation: document all functions and data structures

Nicolas Pitre (10):
      pack-objects: small cleanup
      pack-objects: remove some double negative logic
      pack-objects: simplify the condition associated with --all-progress
      pack-objects: clean up write_object() a bit
      pack-objects: move compression code in a separate function
      pack-objects: allow for early delta deflating
      pack-objects: fix early eviction for max depth delta objects
      add a force_object_loose() function
      let pack-objects do the writing of unreachable objects as loose objects
      make verify-pack a bit more useful with bad packs

Olivier Marin (5):
      remote show: fix the -n option
      builtin-remote: split show_or_prune() in two separate functions
      remote prune: print the list of pruned branches
      remote show: list tracked remote branches with -n
      Fix approxidate("never") to always return 0

Paolo Bonzini (3):
      Add a remote.*.mirror configuration option
      add special "matching refs" refspec
      rollback lock files on more signals than just SIGINT

Paul Mackerras (41):
      gitk: Use git log without --topo-order and reorganize the commits ourselves
      gitk: Fix bug in assigning row numbers to arcs
      gitk: Fix bug in parsing multiple revision arguments
      gitk: Compute row numbers and order tokens lazily
      gitk: Fix a couple of bugs
      gitk: Fix more bugs resulting in Tcl "no such element in array" errors
      gitk: More bug fixes and cleanups
      gitk: Implement date mode in the new framework
      gitk: Fix another collection of bugs
      gitk: Don't try to show local changes from a head that isn't shown
      gitk: Keep the same commits visible as other commits come in
      gitk: Fix some corner cases in the targetid/targetrow stuff
      gitk: Fix a couple of bugs in the find function
      gitk: Fix potential bug with fake commit IDs in renumbervarc
      gitk: Index [fnvr]highlights by id rather than row
      gitk: Fix handling of flag arguments
      gitk: Fix a bug in make_disporder
      gitk: Select head of current branch by default
      gitk: Select something appropriate on cherry-pick, branch reset and checkout
      gitk: Fix bug where editing an existing view would cause an infinite loop
      gitk: Fix bug causing Tcl error when no commits are selected
      gitk: Fix cherry-picking to insert a real row not a fake row
      gitk: Cope better with getting commits that we have already seen
      gitk: Fix bug where arcs could get lost
      gitk: Handle updating with path limiting better
      gitk: Fix problems with target row stuff
      gitk: Don't filter view arguments through git rev-parse
      gitk: Correct a few strings and comments to say "git log"
      gitk: Fix some corner cases in computing vrowmod and displayorder
      gitk: Avoid a crash in selectline if commitinfo($id) isn't set
      gitk: Fix problem with target row not being in scroll region
      gitk: Reorganize processing of arguments for git log
      gitk: Fix handling of tree file list with special chars in names
      gitk: Make updates go faster
      gitk: Synchronize highlighting in file view for 'f' and 'b' commands
      gitk: Show current row number and total number of rows
      gitk: Add a progress bar for checking out a head
      gitk: Fix "wrong # coordinates" error on reload
      gitk: Fix bug where current row number display stops working
      gitk: Fix bug introduced by "gitk: Fix "wrong # coordinates" error on reload"
      gitk: Handle detached heads better

Paul Oliver (1):
      Make git-cvsimport remove ['s from tags, as bad_ref_char doesn't allow them.

Pedro Melo (1):
      Force the medium pretty format on calls to git log

Peter Karlsson (1):
      gitk: Initial Swedish translation.

Philippe Bruhat (BooK) (1):
      git-cvsimport: do not fail when CVSROOT is /

Pierre Habouzit (1):
      Make git reflog expire honour core.sharedRepository.

Pieter de Bie (2):
      builtin-fast-export: Only output a single parent per line
      git-send-email: allow whitespace in addressee list

Ping Yin (6):
      git-submodule: Avoid 'fatal: cannot describe' message
      git-submodule summary: --for-status option
      builtin-status: submodule summary support
      builtin-status: Add tests for submodule summary
      t4027: test diff for submodule with empty directory
      Add t7506 to test submodule related functions for git-status

Rafael Garcia-Suarez (1):
      Spelling fixes in the gitweb documentation

René Scharfe (2):
      git-archive: ignore prefix when checking file attribute
      Ignore .gitattributes in bare repositories

Richard Quirk (2):
      bash: Add completion for gitk --merge
      Documentation gitk: Describe what --merge does

SZEDER Gábor (8):
      doc: moved merge.* config variables into separate merge-config.txt
      merge, pull: introduce '--(no-)stat' option
      add 'merge.stat' config variable
      fmt-merge-msg: add '--(no-)log' options and 'merge.log' config variable
      merge, pull: add '--(no-)log' command line option
      git add: add long equivalents of '-u' and '-f' options
      completion: add more 'git add' options
      diff: reset color before printing newline

Sam Vilain (1):
      Amend git-push refspec documentation

Santi Béjar (3):
      Preparation to call determine_author_info from prepare_to_commit
      commit: Show author if different from committer
      commit: Show committer if automatic

Santiago Gala (1):
      gitk: Spanish translation of gitk

Scott Collins (1):
      Clarify documentation of git-cvsserver, particularly in relation to git-shell

Sebastian Schuberth (1):
      mergetool: Make ECMerge use the settings as specified by the user in the GUI

Seth Falcon (1):
      Add a --dry-run option to git-svn rebase

Shawn Bohrer (2):
      git clean: Don't automatically remove directories when run within subdirectory
      git clean: Add test to verify directories aren't removed with a prefix

Shawn O. Pearce (13):
      git-gui: Don't use '$$cr master' with aspell earlier than 0.60
      git-gui: Setup branch.remote,merge for shorthand git-pull
      git-gui: Delete branches with 'git branch -D' to clear config
      git-gui: Add a --trace command line option
      git-gui: Handle workdir detection when CYGWIN=nowinsymlinks
      Clarify repack -n documentation
      Don't diff empty tree on branch creation in paranoid update hook
      Don't load missing ACL files in paranoid update hook
      Ignore no-op changes in paranoid update hook
      Remove unused remote_prefix member in builtin-remote
      Make "git-remote prune" delete refs according to fetch specs
      Make "git-remote rm" delete refs acccording to fetch specs
      fast-export: Correctly generate initial commits with no parents

Sitaram Chamarty (1):
      builtin-commit.c: add -u as short name for --untracked-files

Steffen Prohaska (4):
      t0050: Test autodetect core.ignorecase
      t0050: Set core.ignorecase case to activate case insensitivity
      t0050: Add test for case insensitive add
      t0050: Fix merge test on case sensitive file systems

Stephan Beyer (9):
      builtin-apply.c: use git_config_string() to get apply_default_whitespace
      Add test cases for git-am
      Merge t4150-am-subdir.sh and t4151-am.sh into t4150-am.sh
      git-commit.txt: Correct option alternatives
      git-commit.txt: Add missing long/short options
      Docs: Use "-l::\n--long\n" format in OPTIONS sections
      Docs: add some long/short options
      git-describe.txt: document --always
      git-name-rev.txt: document --no-undefined and --always

Stephen R. van den Berg (2):
      Simplify and fix --first-parent implementation
      git-svn: Same default as cvsimport when using --use-log-author

Steven Grimm (1):
      Add svn-compatible "blame" output format to git-svn

Teemu Likonen (3):
      bash: Add completion for git diff --base --ours --theirs
      Documentation/git-web--browse.txt: fix small typo
      Print info about "git help COMMAND" on git's main usage pages

Thomas Arcila (1):
      gitk: Allow users to view diffs in external diff viewer

Thomas Guyot-Sionnest (1):
      git-svn bug with blank commits and author file

Trent Piepho (1):
      cvsexportcommit: Create config option for CVS dir

Twiinz (1):
      git-gui: Vertically align textboxes with labels

martin f. krafft (2):
      Escape project name in regexp
      Escape project names before creating pathinfo URLs

A note from the maintainer

[Junio C Hamano]

Welcome to git community.

This message talks about how git.git is managed, and how you can work
with it.

* IRC and Mailing list

Many active members of development community hang around on #git
IRC channel on Freenode.  Its log is available at:

        http://colabti.org/irclogger/irclogger_log/git

The development however is primarily done on the git mailing list
(git@vger.kernel.org).  If you have patches, please send them to the
list, following Documentation/SubmittingPatches.

I usually try to read all patches posted to the list, and follow
almost all the discussions on the list, unless the topic is about an
obscure corner that I do not personally use.  But I am obviously not
perfect.  If you sent a patch that you did not hear from anybody for
three days, that is a very good indication that it was dropped on the
floor --- please do not hesitate to remind me.

The list archive is available at a few public sites as well:

        http://news.gmane.org/gmane.comp.version-control.git/
        http://marc.theaimsgroup.com/?l=git
        http://www.spinics.net/lists/git/

and some people seem to prefer to read it over NNTP:

        nntp://news.gmane.org/gmane.comp.version-control.git

When you point at a message in a mailing list archive, using
gmane is often the easiest to follow by readers, like this:

        http://thread.gmane.org/gmane.comp.version-control.git/27/focus=217

as it also allows people who subscribe to the mailing list as
gmane newsgroup to "jump to" the article.

* Repositories, branches and documentation.

My public git.git repository is at:

        git://git.kernel.org/pub/scm/git/git.git/

Immediately after I publish to the primary repository at kernel.org, I
also push into an alternate here:

        git://repo.or.cz/alt-git.git/

Impatient people might have better luck with the latter one.

Their gitweb interfaces are found at:

        http://git.kernel.org/?p=git/git.git
        http://repo.or.cz/w/alt-git.git

There are three branches in git.git repository that are not
about the source tree of git: "todo", "html" and "man".  The
first one was meant to contain TODO list for me, but I am not
good at maintaining such a list and it is not as often updated as
it could/should be.  It also contains some helper scripts I use
to maintain git.

The "html" and "man" are autogenerated documentation from the
tip of the "master" branch; the tip of "html" is extracted to be
visible at kernel.org at:

        http://www.kernel.org/pub/software/scm/git/docs/

The above URL is the top-level documentation page, and it has
links to documentation of older releases.

The script to maintain these two documentation branches are
found in "todo" branch as dodoc.sh, if you are interested.  It
is a good demonstration of how to use an update hook to automate
a task.

There are four branches in git.git repository that track the
source tree of git: "master", "maint", "next", and "pu".  I may
add more maintenance branches (e.g. "maint-1.5.4") if we have
hugely backward incompatible feature updates in the future to keep
an older release alive; I may not, but the distributed nature of
git means any volunteer can run a stable-tree like that herself.

The "master" branch is meant to contain what are very well
tested and ready to be used in a production setting.  There
could occasionally be minor breakages or brown paper bag bugs
but they are not expected to be anything major, and more
importantly quickly and trivially fixable.  Every now and
then, a "feature release" is cut from the tip of this branch and
they typically are named with three dotted decimal digits.  The
last such release was 1.5.6 done on Jun 18th this year.  You
can expect that the tip of the "master" branch is always more
stable than any of the released versions.

Whenever a feature release is made, "maint" branch is forked off
from "master" at that point.  Obvious, safe and urgent fixes
after a feature release are applied to this branch and
maintenance releases are cut from it.  The maintenance releases
are named with four dotted decimal, named after the feature
release they are updates to; the last such release was 1.5.5.4.
New features never go to this branch.  This branch is also
merged into "master" to propagate the fixes forward.

A trivial and safe enhancement goes directly on top of "master".
A new development, either initiated by myself or more often by
somebody who found his or her own itch to scratch, does not
usually happen on "master", however.  Instead, a separate topic
branch is forked from the tip of "master", and it first is
tested in isolation; I may make minimum fixups at this point.
Usually there are a handful such topic branches that are running
ahead of "master" in git.git repository.  I do not publish the
tip of these branches in my public repository, however, partly
to keep the number of branches that downstream developers need
to worry about low, and primarily because I am lazy.

The quality of topic branches are judged primarily by the mailing list
discussions.  Some of them start out as "good idea but obviously is
broken in some areas (e.g. breaks the existing testsuite)" and then
with some more work (either by the original contributor's effort or
help from other people on the list) becomes "more or less done and can
now be tested by wider audience".  Luckily, most of them start out in
the latter, better shape.

The "next" branch is to merge and test topic branches in the
latter category.  In general, the branch always contains the tip
of "master".  It might not be quite rock-solid production ready,
but is expected to work more or less without major breakage.  I
usually use "next" version of git for my own work, so it cannot
be _that_ broken to prevent me from pushing the changes out.
The "next" branch is where new and exciting things take place.

The two branches "master" and "maint" are never rewound, and
"next" usually will not be either (this automatically means the
topics that have been merged into "next" are usually not
rebased, and you can find the tip of topic branches you are
interested in from the output of "git log next"). You should be
able to safely track them.

After a feature release is made from "master", however, "next"
will be rebuilt from the tip of "master" using the surviving
topics.  The commit that replaces the tip of the "next" will
have the identical tree, but it will have different ancestry
from the tip of "master".  An announcement will be made to warn
people about such a rebasing.

The "pu" (proposed updates) branch bundles all the remainder of
topic branches.  The "pu" branch, and topic branches that are
only in "pu", are subject to rebasing in general.  By the above
definition of how "next" works, you can tell that this branch
will contain quite experimental and obviously broken stuff.

When a topic that was in "pu" proves to be in testable shape, it
graduates to "next".  I do this with:

        git checkout next
        git merge that-topic-branch

Sometimes, an idea that looked promising turns out to be not so
good and the topic can be dropped from "pu" in such a case.

A topic that is in "next" is expected to be tweaked and fixed to
perfection before it is merged to "master" (that's why "master"
can be expected to stay very stable).  Similarly to the above, I
do it with this:

        git checkout master
        git merge that-topic-branch
        git branch -d that-topic-branch

Note that being in "next" is not a guarantee to appear in the
next release (being in "master" is such a guarantee, unless it
is later found seriously broken and reverted), or even in any
future release.  There even were cases that topics needed
reverting a few commits in them before graduating to "master",
or a topic that already was in "next" were entirely reverted
from "next" because fatal flaws were found in them later.

Starting from v1.5.0, "master" and "maint" have release notes
for the next release in Documentation/RelNotes-* files, so that
I do not have to run around summarizing what happened just
before the release.


* Other people's trees, trusted lieutenants and credits.

Documentation/SubmittingPatches outlines who your changes should
be sent to.  As described in contrib/README, I would delegate
fixes and enhancements in contrib/ area to primary contributors
of them.

Although the following are included in git.git repository, they
have their own authoritative repository and maintainers:

 - git-gui/ comes from Shawn Pearce's git-gui project:

        git://repo.or.cz/git-gui.git

 - gitk-git/ comes from Paul Mackerras's gitk project:

        git://git.kernel.org/pub/scm/gitk/gitk.git

I would like to thank everybody who helped to raise git into the
current shape.  Especially I would like to thank the git list
regulars whose help I have relied on and expect to continue
relying on heavily:

 - Linus on general design issues.

 - Linus, Shawn Pearce, Johannes Schindelin, Nicolas Pitre,
   René Scharfe and Jeff King on general implementation issues.

 - Shawn and Nicolas Pitre on pack issues.

 - Martin Langhoff and Frank Lichtenheld on cvsserver and cvsimport.

 - Paul Mackerras on gitk.

 - Eric Wong on git-svn.

 - Simon Hausmann on git-p4.

 - Jakub Narebski, Petr Baudis, and Luben Tuikov on gitweb.

 - J. Bruce Fields on documentaton issues.

 - Johannes Schindelin, Johannes Sixt and others for their effort
   to move things forward on the Windows front.  Although my
   repository does not have much from the effort of MinGW team,
   I expect a merge into mainline will happen so that everybody
   can work from the same codebase.

 - People on non-Linux platforms for keeping their eyes on
   portability; especially, Randal Schwartz, Theodore Ts'o,
   Jason Riedy, Thomas Glanzmann, but countless others as well.

* This document

The latest copy of this document is found in git.git repository,
on 'todo' branch, as MaintNotes.

'next' will be rewound shortly

[Junio C Hamano]

Following the new tradition we began a few releases ago, I'll rewind the
tip of 'next' shortly.  Two topics will be kicked back to 'pu' while we
are at it (jc/send-pack-tell-me-more and js/rebase-i-sequencer).

This is just an advance warning.

Re: 'next' will be rewound shortly

[Stephan Beyer]

Hi Junio,

> Following the new tradition we began a few releases ago, I'll rewind the
> tip of 'next' shortly.  Two topics will be kicked back to 'pu' while we
> are at it (jc/send-pack-tell-me-more and js/rebase-i-sequencer).

I wonder why the bugfix commit

	aec7e9cd	Don't append default merge message to -m message

has disappeared or why/if this belongs to js/rebase-i-sequencer.

I noticed this because a test case of sequencer failed during rebasing
to pure "master"/"next" without js/rebase-i-sequencer.


I also have a question:
my development branch for the sequencer prototype is based on next,
then:
	* Merge js/rebase-i-sequencer
	* ... development ...

The only reason that makes js/rebase-i-sequencer important (besides
aec7e9cd which is mentioned above), is for the last patch
("Migrate rebase-i to use sequencer") in the patchset that I want to
send to the list.  (Otherwise a lot of work of Joerg Sommer would be
annotated to me.)
So I wanted to
 1. send a patchset based on "master"/"next" without the rebase-i
    feature extentions of Joerg Sommer, and
 2. resend the last patch (the one about rebase-i) based on "pu",
    where js/rebase-i-sequencer is still merged into.
Is this sane?

(The other variant could be that I send the "Merge js/rebase-i-sequencer"
 commit as a patch to the list, but this sounds insane to me.)

Regards,
  Stephan

-- 
Stephan Beyer <s-beyer@gmx.net>, PGP 0x6EDDD207FCC5040F

Re: 'next' will be rewound shortly

[Miklos Vajna]

[-- Attachment #1: Type: text/plain, Size: 627 bytes --]

On Fri, Jun 27, 2008 at 06:12:20PM +0200, Stephan Beyer <s-beyer@gmx.net> wrote:
> I wonder why the bugfix commit
> 
> 	aec7e9cd	Don't append default merge message to -m message
> 
> has disappeared or why/if this belongs to js/rebase-i-sequencer.

I think it its current form that is not a bugfix. The user may want to
prepend a custom message and may want to replace the original message
with a custom one, and I would not consider either of them as "buggy".

Funny enough, I just sent a patch:

http://thread.gmane.org/gmane.comp.version-control.git/86584/focus=86589

that tests the original behaviour. ;-)

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: 'next' will be rewound shortly

[Stephan Beyer]

[-- Attachment #1: Type: text/plain, Size: 1015 bytes --]

Hi,

I just looked how you solved that and wanted to start a discussion but
you've swooped in first, find. ;-)

> I think it its current form that is not a bugfix. The user may want to
> prepend a custom message and may want to replace the original message
> with a custom one, and I would not consider either of them as "buggy".

Well, when I do -m <msg>, I expect that my commit message is exactly
<msg>, and not <msg> with appended stuff.
Of course, it doesn't matter what I expect, but I think what the
documentation says matters.
This is (in "master" and in "builtin-merge" of vmiklos.git):

-m <msg>::
	The commit message to be used for the merge commit (in case
	it is created). The `git-fmt-merge-msg` script can be used
	to give a good default for automated `git-merge` invocations.

So it is not mentioned that a standard message is appended, and thus the
original behavior is somehow "buggy" :)

Regards,
  Stephan

-- 
Stephan Beyer <s-beyer@gmx.net>, PGP 0x6EDDD207FCC5040F

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 'next' will be rewound shortly

[Miklos Vajna]

[-- Attachment #1: Type: text/plain, Size: 715 bytes --]

On Fri, Jun 27, 2008 at 07:19:48PM +0200, Stephan Beyer <s-beyer@gmx.net> wrote:
> -m <msg>::
> 	The commit message to be used for the merge commit (in case
> 	it is created). The `git-fmt-merge-msg` script can be used
> 	to give a good default for automated `git-merge` invocations.
> 
> So it is not mentioned that a standard message is appended, and thus the
> original behavior is somehow "buggy" :)

Ah, OK. Then the code and the documentation differs and that's a bug,
sure.

From git-merge.sh:

# All the rest are the commits being merged; prepare
# the standard merge summary message to be appended to
# the given message.

I did builtin-merge based on git-merge.sh, not the manpage. ;-)

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: 'next' will be rewound shortly

[Junio C Hamano]

Miklos Vajna <vmiklos@frugalware.org> writes:

> On Fri, Jun 27, 2008 at 07:19:48PM +0200, Stephan Beyer <s-beyer@gmx.net> wrote:
>> -m <msg>::
>> 	The commit message to be used for the merge commit (in case
>> 	it is created). The `git-fmt-merge-msg` script can be used
>> 	to give a good default for automated `git-merge` invocations.
>> 
>> So it is not mentioned that a standard message is appended, and thus the
>> original behavior is somehow "buggy" :)
>
> Ah, OK. Then the code and the documentation differs and that's a bug,
> sure.
>
> From git-merge.sh:
>
> # All the rest are the commits being merged; prepare
> # the standard merge summary message to be appended to
> # the given message.
>
> I did builtin-merge based on git-merge.sh, not the manpage. ;-)

Following git tradition, manpage came after the command's behaviour has
been long established.  It will be a behaviour change, and it is open to
debate if the new behaviour is better or if the proposed change of
behaviour hurts existing users.

Re: 'next' will be rewound shortly

[Miklos Vajna]

[-- Attachment #1: Type: text/plain, Size: 504 bytes --]

On Fri, Jun 27, 2008 at 02:28:29PM -0700, Junio C Hamano <gitster@pobox.com> wrote:
> Following git tradition, manpage came after the command's behaviour has
> been long established.  It will be a behaviour change, and it is open to
> debate if the new behaviour is better or if the proposed change of
> behaviour hurts existing users.

If my opinion counts, I like the current ("prepend") one, and I think
the best would be to add a new option (and/or make it configurable) for
the new ("replace") one.

[-- Attachment #2: Type: application/pgp-signature, Size: 197 bytes --]

Re: 'next' will be rewound shortly

[Stephan Beyer]

Hi,

> <gitster@pobox.com> wrote:
> > Following git tradition, manpage came after the command's behaviour has
> > been long established.  It will be a behaviour change, and it is open to
> > debate if the new behaviour is better or if the proposed change of
> > behaviour hurts existing users.
> 
> If my opinion counts, I like the current ("prepend") one, and I think
> the best would be to add a new option (and/or make it configurable) for
> the new ("replace") one.

Well, perhaps I am different, but I sometimes have temporary branches
named like "first-silly-experiment" and I do not expect having a
    Merge branch 'another-silly-experiment' into 'first-silly-experiment'
appended, when I do a
    git merge -m "Merge a lot of useful stuff... blabla" another-silly-experiment.

(btw, I don't *really* name my branches like this..it's just an example.)

Well, I see this from a "sequencer author point of view", where
     merge silly-experiment
will invoke an editor,
     merge --standard silly-experiment
generates some kind of the typical standard (or default) message,
and
     merge --message "Merge blabla" silly-experiment
does the "obvious". (For me this is the obvious since I've never
experienced another behavior. All my merges have been using the
now disappeared commit.)

So I'd vote for a "replace" behavior by default on -m, and an 
"append standard message" option, but if there is *one* person
who relies on the prepend feature, I'd also take the "prepend"
default and would like to vote for an option that does the
replacement.

For the current state of the art, it seems that I have to merge
with whatever message and then do a commit -m "..." --amend.

Regards,
  Stephan Beyer

PS: Currently using webmail. So sorry for any too long lines
or whatever.
-- 
GMX startet ShortView.de. Hier findest Du Leute mit Deinen Interessen!
Jetzt dabei sein: http://www.shortview.de/wasistshortview.php?mc=sv_ext_mf@gmx

Re: 'next' will be rewound shortly

[Junio C Hamano]

Miklos Vajna <vmiklos@frugalware.org> writes:

> On Fri, Jun 27, 2008 at 02:28:29PM -0700, Junio C Hamano <gitster@pobox.com> wrote:
>> Following git tradition, manpage came after the command's behaviour has
>> been long established.  It will be a behaviour change, and it is open to
>> debate if the new behaviour is better or if the proposed change of
>> behaviour hurts existing users.
>
> If my opinion counts, I like the current ("prepend") one,...

Well, I do not think you are alone --- otherwise the original behaviour
would not be such ;-)

In any case, what is more important is that the proposed change is a
change in behaviour and the burden of proof that it does not hurt people's
existing scripts is on the party that wants to change it.

A note from the maintainer

[Junio C Hamano]

Welcome to git community.

This message talks about how git.git is managed, and how you can work
with it.

* IRC and Mailing list

Many active members of development community hang around on #git
IRC channel on Freenode.  Its log is available at:

        http://colabti.org/irclogger/irclogger_log/git

The development however is primarily done on the git mailing list
(git@vger.kernel.org).  If you have patches, please send them to the
list, following Documentation/SubmittingPatches.

I usually try to read all patches posted to the list, and follow
almost all the discussions on the list, unless the topic is about an
obscure corner that I do not personally use.  But I am obviously not
perfect.  If you sent a patch that you did not hear from anybody for
three days, that is a very good indication that it was dropped on the
floor --- please do not hesitate to remind me.

The list archive is available at a few public sites as well:

        http://news.gmane.org/gmane.comp.version-control.git/
        http://marc.theaimsgroup.com/?l=git
        http://www.spinics.net/lists/git/

and some people seem to prefer to read it over NNTP:

        nntp://news.gmane.org/gmane.comp.version-control.git

When you point at a message in a mailing list archive, using
gmane is often the easiest to follow by readers, like this:

        http://thread.gmane.org/gmane.comp.version-control.git/27/focus=217

as it also allows people who subscribe to the mailing list as
gmane newsgroup to "jump to" the article.

* Repositories, branches and documentation.

My public git.git repository is at:

        git://git.kernel.org/pub/scm/git/git.git/

Immediately after I publish to the primary repository at kernel.org, I
also push into an alternate here:

        git://repo.or.cz/alt-git.git/

Impatient people might have better luck with the latter one.

Their gitweb interfaces are found at:

        http://git.kernel.org/?p=git/git.git
        http://repo.or.cz/w/alt-git.git

There are three branches in git.git repository that are not
about the source tree of git: "todo", "html" and "man".  The
first one was meant to contain TODO list for me, but I am not
good at maintaining such a list and it is not as often updated as
it could/should be.  It also contains some helper scripts I use
to maintain git.

The "html" and "man" are autogenerated documentation from the
tip of the "master" branch; the tip of "html" is extracted to be
visible at kernel.org at:

        http://www.kernel.org/pub/software/scm/git/docs/

The above URL is the top-level documentation page, and it has
links to documentation of older releases.

The script to maintain these two documentation branches are
found in "todo" branch as dodoc.sh, if you are interested.  It
is a good demonstration of how to use an update hook to automate
a task.

There are four branches in git.git repository that track the
source tree of git: "master", "maint", "next", and "pu".  I may
add more maintenance branches (e.g. "maint-1.5.4") if we have
hugely backward incompatible feature updates in the future to keep
an older release alive; I may not, but the distributed nature of
git means any volunteer can run a stable-tree like that herself.

The "master" branch is meant to contain what are very well
tested and ready to be used in a production setting.  There
could occasionally be minor breakages or brown paper bag bugs
but they are not expected to be anything major, and more
importantly quickly and trivially fixable.  Every now and
then, a "feature release" is cut from the tip of this branch and
they typically are named with three dotted decimal digits.  The
last such release was 1.5.6 done on Jun 18th this year.  You
can expect that the tip of the "master" branch is always more
stable than any of the released versions.

Whenever a feature release is made, "maint" branch is forked off
from "master" at that point.  Obvious, safe and urgent fixes
after a feature release are applied to this branch and
maintenance releases are cut from it.  The maintenance releases
are named with four dotted decimal, named after the feature
release they are updates to; the last such release was 1.5.6.3.
New features never go to this branch.  This branch is also
merged into "master" to propagate the fixes forward.

A trivial and safe enhancement goes directly on top of "master".
A new development, either initiated by myself or more often by
somebody who found his or her own itch to scratch, does not
usually happen on "master", however.  Instead, a separate topic
branch is forked from the tip of "master", and it first is
tested in isolation; I may make minimum fixups at this point.
Usually there are a handful such topic branches that are running
ahead of "master" in git.git repository.  I do not publish the
tip of these branches in my public repository, however, partly
to keep the number of branches that downstream developers need
to worry about low, and primarily because I am lazy.

The quality of topic branches are judged primarily by the mailing list
discussions.  Some of them start out as "good idea but obviously is
broken in some areas (e.g. breaks the existing testsuite)" and then
with some more work (either by the original contributor's effort or
help from other people on the list) becomes "more or less done and can
now be tested by wider audience".  Luckily, most of them start out in
the latter, better shape.

The "next" branch is to merge and test topic branches in the
latter category.  In general, the branch always contains the tip
of "master".  It might not be quite rock-solid production ready,
but is expected to work more or less without major breakage.  I
usually use "next" version of git for my own work, so it cannot
be _that_ broken to prevent me from pushing the changes out.
The "next" branch is where new and exciting things take place.

The two branches "master" and "maint" are never rewound, and
"next" usually will not be either (this automatically means the
topics that have been merged into "next" are usually not
rebased, and you can find the tip of topic branches you are
interested in from the output of "git log next"). You should be
able to safely track them.

After a feature release is made from "master", however, "next"
will be rebuilt from the tip of "master" using the surviving
topics.  The commit that replaces the tip of the "next" will
have the identical tree, but it will have different ancestry
from the tip of "master".  An announcement will be made to warn
people about such a rebasing.

The "pu" (proposed updates) branch bundles all the remainder of
topic branches.  The "pu" branch, and topic branches that are
only in "pu", are subject to rebasing in general.  By the above
definition of how "next" works, you can tell that this branch
will contain quite experimental and obviously broken stuff.

When a topic that was in "pu" proves to be in testable shape, it
graduates to "next".  I do this with:

        git checkout next
        git merge that-topic-branch

Sometimes, an idea that looked promising turns out to be not so
good and the topic can be dropped from "pu" in such a case.

A topic that is in "next" is expected to be tweaked and fixed to
perfection before it is merged to "master" (that's why "master"
can be expected to stay very stable).  Similarly to the above, I
do it with this:

        git checkout master
        git merge that-topic-branch
        git branch -d that-topic-branch

Note that being in "next" is not a guarantee to appear in the
next release (being in "master" is such a guarantee, unless it
is later found seriously broken and reverted), or even in any
future release.  There even were cases that topics needed
reverting a few commits in them before graduating to "master",
or a topic that already was in "next" were entirely reverted
from "next" because fatal flaws were found in them later.

Starting from v1.5.0, "master" and "maint" have release notes
for the next release in Documentation/RelNotes-* files, so that
I do not have to run around summarizing what happened just
before the release.


* Other people's trees, trusted lieutenants and credits.

Documentation/SubmittingPatches outlines who your changes should
be sent to.  As described in contrib/README, I would delegate
fixes and enhancements in contrib/ area to primary contributors
of them.

Although the following are included in git.git repository, they
have their own authoritative repository and maintainers:

 - git-gui/ comes from Shawn Pearce's git-gui project:

        git://repo.or.cz/git-gui.git

 - gitk-git/ comes from Paul Mackerras's gitk project:

        git://git.kernel.org/pub/scm/gitk/gitk.git

I would like to thank everybody who helped to raise git into the
current shape.  Especially I would like to thank the git list
regulars whose help I have relied on and expect to continue
relying on heavily:

 - Linus on general design issues.

 - Linus, Shawn Pearce, Johannes Schindelin, Nicolas Pitre,
   René Scharfe and Jeff King on general implementation issues.

 - Shawn and Nicolas Pitre on pack issues.

 - Martin Langhoff and Frank Lichtenheld on cvsserver and cvsimport.

 - Paul Mackerras on gitk.

 - Eric Wong on git-svn.

 - Simon Hausmann on git-p4.

 - Jakub Narebski, Petr Baudis, and Luben Tuikov on gitweb.

 - J. Bruce Fields on documentaton issues.

 - Johannes Schindelin, Johannes Sixt and others for their effort
   to move things forward on the Windows front.  Most of the fruits
   from their porting efforts have been merged to the mainline git.git
   repository in 1.6.0 release.

 - People on non-Linux platforms for keeping their eyes on
   portability; especially, Randal Schwartz, Theodore Ts'o,
   Jason Riedy, Thomas Glanzmann, but countless others as well.

* This document

The latest copy of this document is found in git.git repository,
on 'todo' branch, as MaintNotes.

Re: [ANNOUNCE] GIT 1.5.6

[Steffen Prohaska]

On Jun 19, 2008, at 1:24 AM, Junio C Hamano wrote:

> The latest feature release GIT 1.5.6 is available at the usual
> places:


The msysgit Windows installer is available at

    http://code.google.com/p/msysgit/downloads

	Steffen

[ANNOUNCE] GIT 1.5.6.1

[Junio C Hamano]

The latest maintenance release GIT 1.5.6.1 is available at the
usual places:

  http://www.kernel.org/pub/software/scm/git/

  git-1.5.6.1.tar.{gz,bz2}			(tarball)
  git-htmldocs-1.5.6.1.tar.{gz,bz2}		(preformatted docs)
  git-manpages-1.5.6.1.tar.{gz,bz2}		(preformatted docs)
  RPMS/$arch/git-*-1.5.6.1-1.$arch.rpm	(RPM)

GIT v1.5.6.1 Release Notes
==========================

Fixes since v1.5.6
------------------

* Last minute change broke loose object creation on AIX.

* (performance fix) We used to make $GIT_DIR absolute path early in the
  programs but keeping it relative to the current directory internally
  gives 1-3 per-cent performance boost.

* bash completion knows the new --graph option to git-log family.


* git-diff -c/--cc showed unnecessary "deletion" lines at the context
  boundary.

* git-for-each-ref ignored %(object) and %(type) requests for tag
  objects.

* git-merge usage had a typo.

* Rebuilding of git-svn metainfo database did not take rewriteRoot
  option into account.

* Running "git-rebase --continue/--skip/--abort" before starting a
  rebase gave nonsense error messages.

----------------------------------------------------------------

Changes since v1.5.6 are as follows:

Brandon Casey (2):
      git-merge.sh: fix typo in usage message: sucesses --> succeeds
      t7502-commit.sh: test_must_fail doesn't work with inline environment variables

Dan McGee (1):
      completion: add --graph to log command completion

Dmitry Potapov (1):
      fix update-hook-example to work with packed tag references

Jan Krüger (2):
      Documentation: fix formatting in git-svn
      git-svn: make rebuild respect rewriteRoot option

Jeff King (2):
      for-each-ref: implement missing tag values
      clone: create intermediate directories of destination repo

Junio C Hamano (2):
      diff -c/--cc: do not include uninteresting deletion before leading context
      GIT 1.5.6.1

Linus Torvalds (1):
      Make git_dir a path relative to work_tree in setup_work_tree()

Michele Ballabio (1):
      parse-options.c: fix documentation syntax of optional arguments

Patrick Higgins (1):
      Workaround for AIX mkstemp()

Stephan Beyer (4):
      git-rebase.sh: Add check if rebase is in progress
      api-builtin.txt: update and fix typo
      api-parse-options.txt: Introduce documentation for parse options API
      Extend parse-options test suite

Re: [ANNOUNCE] GIT 1.5.6.1

[Steffen Prohaska]

On Jun 26, 2008, at 8:21 AM, Junio C Hamano wrote:

> The latest maintenance release GIT 1.5.6.1 is available at the
> usual places:


The msysgit release is available at

    http://code.google.com/p/msysgit/downloads

Besides GIT 1.5.6.1, the installer also brings an updated
msys-1.0.dll that works on Vista.

	Steffen