diff options
Diffstat (limited to 'Documentation/RelNotes')
| -rw-r--r-- | Documentation/RelNotes/2.4.0.txt | 514 | ||||
| -rw-r--r-- | Documentation/RelNotes/2.4.1.txt | 40 | ||||
| -rw-r--r-- | Documentation/RelNotes/2.5.0.txt | 247 |
3 files changed, 801 insertions, 0 deletions
diff --git a/Documentation/RelNotes/2.4.0.txt b/Documentation/RelNotes/2.4.0.txt new file mode 100644 index 0000000000..cde64be535 --- /dev/null +++ b/Documentation/RelNotes/2.4.0.txt @@ -0,0 +1,514 @@ +Git 2.4 Release Notes +===================== + +Backward compatibility warning(s) +--------------------------------- + +This release has a few changes in the user-visible output from +Porcelain commands. These are not meant to be parsed by scripts, but +users still may want to be aware of the changes: + + * The output from "git log --decorate" (and, more generally, the "%d" + format specifier used in the "--format=<string>" parameter to the + "git log" family of commands) has changed. It used to list "HEAD" + just like other branches; e.g., + + $ git log --decorate -1 master + commit bdb0f6788fa5e3cacc4315e9ff318a27b2676ff4 (HEAD, master) + ... + + This release changes the output slightly when HEAD refers to a + branch whose name is also shown in the output. The above is now + shown as: + + $ git log --decorate -1 master + commit bdb0f6788fa5e3cacc4315e9ff318a27b2676ff4 (HEAD -> master) + ... + + * The phrasing "git branch" uses to describe a detached HEAD has been + updated to agree with the phrasing used by "git status": + + - When HEAD is at the same commit as when it was originally + detached, they now both show "detached at <commit object name>". + + - When HEAD has moved since it was originally detached, they now + both show "detached from <commit object name>". + + Previously, "git branch" always used "from". + + +Updates since v2.3 +------------------ + +Ports + + * Our default I/O size (8 MiB) for large files was too large for some + platforms with smaller SSIZE_MAX, leading to read(2)/write(2) + failures. + + * We did not check the curl library version before using the + CURLOPT_PROXYAUTH feature, which did not exist in older versions of + the library. + + * We now detect number of CPUs on older BSD-derived systems. + + * Portability fixes and workarounds for shell scripts have been added + to help BSD-derived systems. + + +UI, Workflows & Features + + * The command usage info strings given by "git cmd -h" and in + documentation have been tweaked for consistency. + + * The "sync" subcommand of "git p4" now allows users to exclude + subdirectories like its "clone" subcommand does. + + * "git log --invert-grep --grep=WIP" will show only commits that do + not have the string "WIP" in their messages. + + * "git push" has been taught an "--atomic" option that makes a push + that updates more than one ref an "all-or-none" affair. + + * Extending the "push to deploy" feature that was added in 2.3, the + behaviour of "git push" when updating the branch that is checked + out can now be tweaked by a "push-to-checkout" hook. + + * HTTP-based transports now send Accept-Language when making + requests. The languages to accept are inferred from environment + variables on the client side (LANGUAGE, etc). + + * "git send-email" used to accept a mistaken "y" (or "yes") as an + answer to "What encoding do you want to use [UTF-8]?" without + questioning. Now it asks for confirmation when the answer looks too + short to be a valid encoding name. + + * When "git apply --whitespace=fix" fixed whitespace errors in the + common context lines, the command reports that it did so. + + * "git status" now allows the "-v" option to be given twice, in which + case it also shows the differences in the working tree that are not + staged to be committed. + + * "git cherry-pick" used to clean up the log message even when it is + merely replaying an existing commit. It now replays the message + verbatim unless you are editing the message of the resulting + commit. + + * "git archive" can now be told to set the 'text' attribute in the + resulting zip archive. + + * Output from "git log --decorate" now distinguishes between a + detached HEAD vs. a HEAD that points at a branch. + + This is a potentially backward-incompatible change; see above for + more information. + + * When HEAD was detached when at commit xyz and hasn't been moved + since it was detached, "git status" would report "detached at xyz" + whereas "git branch" would report "detached from xyz". Now the + output of "git branch" agrees with that of "git status". + + This is a potentially backward-incompatible change; see above for + more information. + + * "git -C '' subcmd" now works in the current directory (analogously + to "cd ''") rather than dying with an error message. + (merge 6a536e2 kn/git-cd-to-empty later to maint). + + * The versionsort.prereleaseSuffix configuration variable can be used + to specify that, for example, v1.0-pre1 comes before v1.0. + + * A new "push.followTags" configuration turns the "--follow-tags" + option on by default for the "git push" command. + + * "git log --graph --no-walk A B..." is a nonsensical combination of + options: "--no-walk" requests discrete points in the history, while + "--graph" asks to draw connections between these discrete points. + Forbid the use of these options together. + + * "git rev-list --bisect --first-parent" does not work (yet) and can + even cause SEGV; forbid it. "git log --bisect --first-parent" would + not be useful until "git bisect --first-parent" materializes, so + also forbid it for now. + + +Performance, Internal Implementation, Development Support etc. + + * Slightly change the implementation of the N_() macro to help us + detect mistakes. + + * Restructure the implementation of "reflog expire" to fit better + with the recently updated reference API. + + * The transport-helper did not pass transport options such as + verbosity, progress, cloning, etc. to import and export based + helpers, like it did for fetch and push based helpers, robbing them + of the chance to honor the wish of the end-users better. + + * The tests that wanted to see that a file becomes unreadable after + running "chmod a-r file", and the tests that wanted to make sure + that they are not run as root, used "can we write into the / + directory?" as a cheap substitute. But on some platforms that is + not a good heuristic. The tests and their prerequisites have been + updated to check what they really require. + (merge f400e51 jk/sanity later to maint). + + * Various issues around "reflog expire", e.g. using --updateref when + expiring a reflog for a symbolic reference, have been corrected + and/or made saner. + + * The documentation for the strbuf API had been split between the API + documentation and the header file. Consolidate the documentation in + strbuf.h. + + * The error handling functions and conventions are now documented in + the API manual (in api-error-handling.txt). + + * Optimize gitattribute look-up, mostly useful in "git grep" on a + project that does not use many attributes, by avoiding it when we + (should) know that the attributes are not defined in the first + place. + + * Typofix in comments. + (merge ef2956a ak/git-pm-typofix later to maint). + + * Code clean-up. + (merge 0b868f0 sb/hex-object-name-is-at-most-41-bytes-long later to maint). + (merge 5d30851 dp/remove-duplicated-header-inclusion later to maint). + + * Simplify the ref transaction API for verifying that "the ref should + be pointing at this object". + + * Simplify the code in "git daemon" that parses out and holds + hostnames used in request interpolation. + + * Restructure the "git push" codepath to make it easier to add new + configuration bits. + + * The run-command interface made it easy to make a pipe for us to + read from a process, wait for the process to finish, and then + attempt to read its output. But this pattern can lead to deadlock. + So introduce a helper to do this correctly (i.e., first read, and + then wait the process to finish) and also add code to prevent such + abuse in the run-command helper. + + * People often forget to chain the commands in their test together + with &&, letting a failure from an earlier command in the test go + unnoticed. The new GIT_TEST_CHAIN_LINT mechanism allows you to + catch such a mistake more easily. + + +Also contains various documentation updates and code clean-ups. + + +Fixes since v2.3 +---------------- + +Unless otherwise noted, all the fixes since v2.3 in the maintenance +track are contained in this release (see the maintenance releases' +notes for details). + + * "git blame HEAD -- missing" failed to correctly say "HEAD" when it + tried to say "No such path 'missing' in HEAD". + (merge a46442f jk/blame-commit-label later to maint). + + * "git rerere" (invoked internally from many mergy operations) did + not correctly signal errors when it attempted to update the working + tree files but failed for whatever reason. + (merge 89ea903 jn/rerere-fail-on-auto-update-failure later to maint). + + * Setting diff.submodule to 'log' made "git format-patch" produce + broken patches. + (merge 339de50 dk/format-patch-ignore-diff-submodule later to maint). + + * After attempting and failing a password-less authentication (e.g., + Kerberos), libcURL refuses to fall back to password-based Basic + authentication without a bit of help/encouragement. + (merge 4dbe664 bc/http-fallback-to-password-after-krb-fails later to maint). + + * The "git push" documentation for the "--repo=<there>" option was + easily misunderstood. + (merge 57b92a7 mg/push-repo-option-doc later to maint). + + * Code to read a branch name from various files in the .git/ + directory would have overrun array limits if asked to read an empty + file. + (merge 66ec904 jk/status-read-branch-name-fix later to maint). + + * Remove a superfluous conditional that is always true. + (merge 94ee8e2 jk/remote-curl-an-array-in-struct-cannot-be-null later to maint). + + * The "git diff --raw" documentation incorrectly implied that C(opy) + and R(ename) are the only statuses that can be followed by a score + number. + (merge ac1c2d9 jc/diff-format-doc later to maint). + + * A broken pack .idx file in the receiving repository prevented the + dumb http transport from fetching a good copy of it from the other + side. + (merge 8b9c2dd jk/dumb-http-idx-fetch-fix later to maint). + + * The error message from "git commit", when a non-existing author + name was given as value to the "--author=" parameter, has been + reworded to avoid misunderstanding. + (merge 1044b1f mg/commit-author-no-match-malformed-message later to maint). + + * "git log --help" used to show rev-list options that are irrelevant + to the "log" command. + (merge 3cab02d jc/doc-log-rev-list-options later to maint). + + * "git apply --whitespace=fix" used to under-allocate memory when the + fix resulted in a longer text than the original patch. + (merge 407a792 jc/apply-ws-fix-expands later to maint). + + * The interactive "show a list and let the user choose from it" + interface used by "git add -i" unnecessarily prompted the user even + when the candidate list was empty, against which the only "choice" + the user could have made was to choose nothing. + (merge a9c4641 ak/add-i-empty-candidates later to maint). + + * The todo list created by "git rebase -i" did not fully honor + core.abbrev settings. + (merge edb72d5 ks/rebase-i-abbrev later to maint). + + * "git fetch" over a remote-helper that cannot respond to the "list" + command could not fetch from a symbolic reference (e.g., HEAD). + (merge 33cae54 mh/deref-symref-over-helper-transport later to maint). + + * "git push --signed" gave an incorrectly worded error message when + the other side did not support the capability. + + * The "git push --signed" protocol extension did not limit what the + "nonce" (a server-chosen string) could contain nor how long it + could be, which was unnecessarily lax. Limit both the length and + the alphabet to a reasonably small space that can still have enough + entropy. + (merge afcb6ee jc/push-cert later to maint). + + * The completion script (in contrib/) clobbered the shell variable $x + in the global shell namespace. + (merge 852ff1c ma/bash-completion-leaking-x later to maint). + + * We incorrectly formatted a "uintmax_t" integer that doesn't fit in + "int". + (merge d306f3d jk/decimal-width-for-uintmax later to maint). + + * The configuration parser used to be confused when reading + configuration from a blob object that ends with a lone CR. + (merge 1d0655c jk/config-no-ungetc-eof later to maint). + + * The pack bitmap support did not build with older versions of GCC. + (merge bd4e882 jk/pack-bitmap later to maint). + + * The documentation wasn't clear that "remote.<nick>.pushURL" and + "remote.<nick>.URL" are there to name the same repository accessed + via different transports, not two separate repositories. + (merge 697f652 jc/remote-set-url-doc later to maint). + + * Older GnuPG implementations may not correctly import the keyring + material we prepare for the tests to use. + (merge 1f985d6 ch/new-gpg-drops-rfc-1991 later to maint). + + * The credential helper for Windows (in contrib/) used to mishandle + user names that contain an at-sign. + (merge 13d261e av/wincred-with-at-in-username-fix later to maint). + + * "diff-highlight" (in contrib/) used to show byte-by-byte + differences, which could cause multi-byte characters to be chopped + in the middle. It learned to pay attention to character boundaries + (assuming UTF-8). + (merge 8d00662 jk/colors later to maint). + + * Document longstanding configuration variable naming rules in + CodingGuidelines. + (merge 35840a3 jc/conf-var-doc later to maint). + + * An earlier workaround to squelch unhelpful deprecation warnings + from the compiler on OS X unnecessarily set a minimum required + version of the OS, which the user might want to raise (or lower) + for other reasons. + (merge 88c03eb es/squelch-openssl-warnings-on-macosx later to maint). + + * Certain older vintages of cURL give irregular output from + "curl-config --vernum", which confused our build system. + (merge 3af6792 tc/curl-vernum-output-broken-in-7.11 later to maint). + + * In v2.2.0, we broke "git prune" that runs in a repository that + borrows from an alternate object store. + (merge b0a4264 jk/prune-mtime later to maint). + + * "git submodule add" failed to squash "path/to/././submodule" to + "path/to/submodule". + (merge 8196e72 ps/submodule-sanitize-path-upon-add later to maint). + + * "git merge-file" did not work correctly when invoked in a + subdirectory. + (merge 204a8ff ab/merge-file-prefix later to maint). + + * "git blame" could die trying to free an uninitialized piece of + memory. + (merge e600592 es/blame-commit-info-fix later to maint). + + * "git fast-import" used to crash when it could not close and + finalize the resulting packfile cleanly. + (merge 5e915f3 jk/fast-import-die-nicely-fix later to maint). + + * "update-index --refresh" used to leak memory when an entry could + not be refreshed for whatever reason. + (merge bc1c2ca sb/plug-leak-in-make-cache-entry later to maint). + + * The "interpolated-path" option of "git daemon" inserted any string + the client declared on the "host=" capability request without + checking. Sanitize and limit %H and %CH to a saner and a valid DNS + name. + (merge b485373 jk/daemon-interpolate later to maint). + + * "git daemon" unnecessarily looked up the hostname even when "%CH" + and "%IP" interpolations were not requested. + (merge dc8edc8 rs/daemon-interpolate later to maint). + + * We relied on "--no-" prefix handling in Perl's Getopt::Long + package, even though that support didn't exist in Perl 5.8 (which + we still support). Manually add support to help people with older + Getopt::Long packages. + (merge f471494 km/send-email-getopt-long-workarounds later to maint). + + * "git apply" was not very careful about reading from, removing, + updating and creating paths outside the working tree (under + --index/--cached) or the current directory (when used as a + replacement for GNU patch). + (merge e0d201b jc/apply-beyond-symlink later to maint). + + * Correct a breakage in git-svn, introduced around the v2.2 era, that + can cause FileHandles to be closed prematurely. + (merge e426311 ew/svn-maint-fixes later to maint). + + * We did not parse usernames followed by literal IPv6 addresses + correctly in SSH transport URLs; e.g., + ssh://user@[2001:db8::1]:22/repo.git. + (merge 6b6c5f7 tb/connect-ipv6-parse-fix later to maint). + + * The configuration variable 'mailinfo.scissors' was hard to + discover in the documentation. + (merge afb5de7 mm/am-c-doc later to maint). + + * The interaction between "git submodule update" and the + submodule.*.update configuration was not clearly documented. + (merge 5c31acf ms/submodule-update-config-doc later to maint). + + * "git diff --shortstat" used together with "--dirstat=changes" or + "--dirstat=files" incorrectly output dirstat information twice. + (merge ab27389 mk/diff-shortstat-dirstat-fix later to maint). + + * The manpage for "git remote add" mentioned "--tags" and "--no-tags" + but did not explain what happens if neither option is provided. + (merge aaba0ab mg/doc-remote-tags-or-not later to maint). + + * The description of "--exclude-standard option" in the output of + "git grep -h" was phrased poorly. + (merge 77fdb8a nd/grep-exclude-standard-help-fix later to maint). + + * "git rebase -i" recently started to include the number of commits + in the todo list, but that output included extraneous whitespace on + a platform that prepends leading whitespaces to its "wc -l" output. + (merge 2185d3b es/rebase-i-count-todo later to maint). + + * The borrowed code in the kwset API did not follow our usual + convention to use "unsigned char" to store values that range from + 0-255. + (merge 189c860 bw/kwset-use-unsigned later to maint). + + * A corrupt input to "git diff -M" used to cause it to segfault. + (merge 4d6be03 jk/diffcore-rename-duplicate later to maint). + + * Certain builds of GPG triggered false breakages in a test. + (merge 3f88c1b mg/verify-commit later to maint). + + * "git imap-send" learned to optionally talk with an IMAP server via + libcURL. Because there is no other option when Git is built with + the NO_OPENSSL option, use libcURL by default in that case. + (merge dcd01ea km/imap-send-libcurl-options later to maint). + + * "git log --decorate" did not reset colors correctly around the + branch names. + (merge 5ee8758 jc/decorate-leaky-separator-color later to maint). + + * The code that reads from the ctags file in the completion script + (in contrib/) did not spell ${param/pattern/string} substitution + correctly, which happened to work with bash but not with zsh. + (merge db8d750 js/completion-ctags-pattern-substitution-fix later to maint). + + * The transfer.hiderefs support did not quite work for smart-http + transport. + (merge 8ddf3ca jk/smart-http-hide-refs later to maint). + + * In the "git tag -h" output, move the documentation for the + "--column" and "--sort" options to the "Tag listing options" + section. + (merge dd059c6 jk/tag-h-column-is-a-listing-option later to maint). + + * "git prune" used to largely ignore broken refs when deciding which + objects are still being used, which could cause reference + corruption to lead to object loss. + (merge ea56c4e jk/prune-with-corrupt-refs later to maint). + + * The split-index mode introduced in v2.3.0-rc0~41 was broken in the + codepath to protect us against a broken reimplementation of Git + that writes an invalid index with duplicated index entries, etc. + (merge 03f15a7 tg/fix-check-order-with-split-index later to maint). + + * "git fetch", when fetching a commit using the + allow-tip-sha1-in-want extension, could have failed to fetch all of + the requested refs. + (merge 32d0462 jk/fetch-pack later to maint). + + * An failure early in the "git clone" that started creating the + working tree and repository could have resulted in the failure to + clean up some directories and files. + (merge 16eff6c jk/cleanup-failed-clone later to maint). + + * Recommend format-patch and send-email for those who want to submit + patches to this project. + (merge b25c469 jc/submitting-patches-mention-send-email later to maint). + + * Do not spawn the pager when "git grep" is run with "--quiet". + (merge c2048f0 ws/grep-quiet-no-pager later to maint). + + * The prompt script (in contrib/) did not show the untracked sign + when working in a subdirectory without any untracked files. + (merge 9bdc517 ct/prompt-untracked-fix later to maint). + + * An earlier update to the URL parser broke an address that contains + a colon but an empty string for the port number, like + ssh://example.com:/path/to/repo. + (merge 6b6c5f7 tb/connect-ipv6-parse-fix later to maint). + + * Code cleanups and documentation updates. + (merge 2ce63e9 rs/simple-cleanups later to maint). + (merge 33baa69 rj/no-xopen-source-for-cygwin later to maint). + (merge 817d03e jc/diff-test-updates later to maint). + (merge eb32c66 ak/t5516-typofix later to maint). + (merge bcd57cb mr/doc-clean-f-f later to maint). + (merge 0d6accc mg/doc-status-color-slot later to maint). + (merge 53e53c7 sg/completion-remote later to maint). + (merge 8fa7975 ak/git-done-help-cleanup later to maint). + (merge 9a6f128 rs/deflate-init-cleanup later to maint). + (merge 6f75d45 rs/use-isxdigit later to maint). + (merge 376e4b3 jk/test-annoyances later to maint). + (merge 7032054 nd/doc-git-index-version later to maint). + (merge e869c5e tg/test-index-v4 later to maint). + (merge 599d223 jk/simplify-csum-file-sha1fd-check later to maint). + (merge 260d585 sg/completion-gitcomp-nl-for-refs later to maint). + (merge 777c55a jc/report-path-error-to-dir later to maint). + (merge fddfaf8 ph/push-doc-cas later to maint). + (merge d50d31e ss/pull-rebase-preserve later to maint). + (merge c8c3f1d pt/enter-repo-comment-fix later to maint). + (merge d7bfb9e jz/gitweb-conf-doc-fix later to maint). + (merge f907282 jk/cherry-pick-docfix later to maint). + (merge d3c0811 iu/fix-parse-options-h-comment later to maint). + (merge 6c3b2af jg/cguide-we-cannot-count later to maint). + (merge 2b8bd44 jk/pack-corruption-post-mortem later to maint). + (merge 9585cb8 jn/doc-fast-import-no-16-octopus-limit later to maint). + (merge 5dcd1b1 ps/grep-help-all-callback-arg later to maint). + (merge f1f4c84 va/fix-git-p4-tests later to maint). diff --git a/Documentation/RelNotes/2.4.1.txt b/Documentation/RelNotes/2.4.1.txt new file mode 100644 index 0000000000..a65a6c5829 --- /dev/null +++ b/Documentation/RelNotes/2.4.1.txt @@ -0,0 +1,40 @@ +Git v2.4.1 Release Notes +======================== + +Fixes since v2.4 +---------------- + + * The usual "git diff" when seeing a file turning into a directory + showed a patchset to remove the file and create all files in the + directory, but "git diff --no-index" simply refused to work. Also, + when asked to compare a file and a directory, imitate POSIX "diff" + and compare the file with the file with the same name in the + directory, instead of refusing to run. + + * The default $HOME/.gitconfig file created upon "git config --global" + that edits it had incorrectly spelled user.name and user.email + entries in it. + + * "git commit --date=now" or anything that relies on approxidate lost + the daylight-saving-time offset. + + * "git cat-file bl $blob" failed to barf even though there is no + object type that is "bl". + + * Teach the codepaths that read .gitignore and .gitattributes files + that these files encoded in UTF-8 may have UTF-8 BOM marker at the + beginning; this makes it in line with what we do for configuration + files already. + + * Access to objects in repositories that borrow from another one on a + slow NFS server unnecessarily got more expensive due to recent code + becoming more cautious in a naive way not to lose objects to pruning. + + * We avoid setting core.worktree when the repository location is the + ".git" directory directly at the top level of the working tree, but + the code misdetected the case in which the working tree is at the + root level of the filesystem (which arguably is a silly thing to + do, but still valid). + +Also contains typofixes, documentation updates and trivial code +clean-ups. diff --git a/Documentation/RelNotes/2.5.0.txt b/Documentation/RelNotes/2.5.0.txt new file mode 100644 index 0000000000..fb7aa970cb --- /dev/null +++ b/Documentation/RelNotes/2.5.0.txt @@ -0,0 +1,247 @@ +Git 2.5 Release Notes +===================== + +Updates since v2.4 +------------------ + +Ports + + +UI, Workflows & Features + + * "git p4" now detects the filetype (e.g. binary) correctly even when + the files are opened exclusively. + + * git p4 attempts to better handle branches in Perforce. + + * "git p4" learned "--changes-block-size <n>" to read the changes in + chunks from Perforce, instead of making one call to "p4 changes" + that may trigger "too many rows scanned" error from Perforce. + + * "git show-branch --topics HEAD" (with no other arguments) did not + do anything interesting. Instead, contrast the given revision + against all the local branches by default. + + * A replacement for contrib/workdir/git-new-workdir that does not + rely on symbolic links and make sharing of objects and refs safer + by making the borrowee and borrowers aware of each other. + + * Tweak the sample "store" backend of the credential helper to honor + XDG configuration file locations when specified. + + * A heuristic to help the "git <cmd> <revs> <pathspec>" command line + convention to catch mistyped paths is to make sure all the non-rev + parameters in the later part of the command line are names of the + files in the working tree, but that means "git grep $str -- \*.c" + must always be disambiguated with "--", because nobody sane will + create a file whose name literally is asterisk-dot-see. Loosen the + heuristic to declare that with a wildcard string the user likely + meant to give us a pathspec. + + * "git merge FETCH_HEAD" learned that the previous "git fetch" could + be to create an Octopus merge, i.e. recording multiple branches + that are not marked as "not-for-merge"; this allows us to lose an + old style invocation "git merge <msg> HEAD $commits..." in the + implementation of "git pull" script; the old style syntax can now + be deprecated. + + * Help us to find broken test script that splits the body part of the + test by mistaken use of wrong kind of quotes. + (merge d93d5d5 jc/test-prereq-validate later to maint). + + * Developer support to automatically detect broken &&-chain in the + test scripts is now turned on by default. + (merge 92b269f jk/test-chain-lint later to maint). + + +Performance, Internal Implementation, Development Support etc. + + * "unsigned char [20]" used thoughout the code to represent object + names are being converted into a semi-opaque "struct object_id". + This effort is expected to interfere with other topics in flight, + but hopefully will give us one extra level of abstraction in the + end, when completed. + + * Catch a programmer mistake to feed a pointer not an array to + ARRAY_SIZE() macro, by using a couple of GCC extensions. + (merge 89c855e ep/do-not-feed-a-pointer-to-array-size later to maint). + + * Some error messages in "git config" were emitted without calling + the usual error() facility. + + * When "add--interactive" splits a hunk into two overlapping hunks + and then let the user choose only one, it sometimes feeds an + incorrect patch text to "git apply". Add tests to demonstrate + this. + + I have a slight suspicion that this may be $gmane/87202 coming back + and biting us (I seem to have said "let's run with this and see + what happens" back then). + + * More line-ending tests. + + * An earlier rewrite to use strbuf_getwholeline() instead of fgets(3) + to read packed-refs file revealed that the former is unacceptably + inefficient. + + * The refs API uses ref_lock struct which had its own "int fd", even + though the same file descriptor was in the lock struct it contains. + Clean-up the code to lose this redundant field. + + * Add the "--allow-unknown-type" option to "cat-file" to allow + inspecting loose objects of an experimental or a broken type. + + * Many long-running operations show progress eye-candy, even when + they are later backgrounded. Hide the eye-candy when the process + is sent to the background instead. + (merge 9a9a41d lm/squelch-bg-progress later to maint). + + +Also contains various documentation updates and code clean-ups. + + +Fixes since v2.4 +---------------- + +Unless otherwise noted, all the fixes since v2.4 in the maintenance +track are contained in this release (see the maintenance releases' +notes for details). + + * Memory usage of "git index-pack" has been trimmed by tens of + per-cent. + (merge c6458e6 nd/slim-index-pack-memory-usage later to maint). + + * "git rev-list --objects $old --not --all" to see if everything that + is reachable from $old is already connected to the existing refs + was very inefficient. + (merge b6e8a3b jk/still-interesting later to maint). + + * "hash-object --literally" introduced in v2.2 was not prepared to + take a really long object type name. + (merge 1427a7f jc/hash-object later to maint). + + * "git rebase --quiet" was not quite quiet when there is nothing to + do. + (merge 22946a9 jk/rebase-quiet-noop later to maint). + + * The completion for "log --decorate=" parameter value was incorrect. + (merge af16bda sg/complete-decorate-full-not-long later to maint). + + * "filter-branch" corrupted commit log message that ends with an + incomplete line on platforms with some "sed" implementations that + munge such a line. Work it around by avoiding to use "sed". + (merge df06201 jk/filter-branch-use-of-sed-on-incomplete-line later to maint). + + * "git daemon" fails to build from the source under NO_IPV6 + configuration (regression in 2.4). + (merge d358f77 jc/daemon-no-ipv6-for-2.4.1 later to maint). + + * Some time ago, "git blame" (incorrectly) lost the convert_to_git() + call when synthesizing a fake "tip" commit that represents the + state in the working tree, which broke folks who record the history + with LF line ending to make their project portabile across + platforms while terminating lines in their working tree files with + CRLF for their platform. + (merge 4bf256d tb/blame-resurrect-convert-to-git later to maint). + + * We avoid setting core.worktree when the repository location is the + ".git" directory directly at the top level of the working tree, but + the code misdetected the case in which the working tree is at the + root level of the filesystem (which arguably is a silly thing to + do, but still valid). + (merge 84ccad8 jk/init-core-worktree-at-root later to maint). + + * "git commit --date=now" or anything that relies on approxidate lost + the daylight-saving-time offset. + (merge f6e6362 jc/epochtime-wo-tz later to maint). + + * Access to objects in repositories that borrow from another one on a + slow NFS server unnecessarily got more expensive due to recent code + becoming more cautious in a naive way not to lose objects to pruning. + (merge ee1c6c3 jk/prune-mtime later to maint). + + * The codepaths that read .gitignore and .gitattributes files have been + taught that these files encoded in UTF-8 may have UTF-8 BOM marker at + the beginning; this makes it in line with what we do for configuration + files already. + (merge 27547e5 cn/bom-in-gitignore later to maint). + + * a few helper scripts in the test suite did not report errors + correcty. + (merge de248e9 ep/fix-test-lib-functions-report later to maint). + + * The default $HOME/.gitconfig file created upon "git config --global" + that edits it had incorrectly spelled user.name and user.email + entries in it. + (merge 7e11052 oh/fix-config-default-user-name-section later to maint). + + * "git cat-file bl $blob" failed to barf even though there is no + object type that is "bl". + (merge b7994af jk/type-from-string-gently later to maint). + + * The usual "git diff" when seeing a file turning into a directory + showed a patchset to remove the file and create all files in the + directory, but "git diff --no-index" simply refused to work. Also, + when asked to compare a file and a directory, imitate POSIX "diff" + and compare the file with the file with the same name in the + directory, instead of refusing to run. + (merge 0615173 jc/diff-no-index-d-f later to maint). + + * "git rebase -i" moved the "current" command from "todo" to "done" a + bit too prematurely, losing a step when a "pick" did not even start. + (merge 8cbc57c ph/rebase-i-redo later to maint). + + * The connection initiation code for "ssh" transport tried to absorb + differences between the stock "ssh" and Putty-supplied "plink" and + its derivatives, but the logic to tell that we are using "plink" + variants were too loose and falsely triggered when "plink" appeared + anywhere in the path (e.g. "/home/me/bin/uplink/ssh"). + (merge baaf233 bc/connect-plink later to maint). + + * "git stash pop/apply" forgot to make sure that not just the working + tree is clean but also the index is clean. The latter is important + as a stash application can conflict and the index will be used for + conflict resolution. + (merge ed178ef jk/stash-require-clean-index later to maint). + + * We have prepended $GIT_EXEC_PATH and the path "git" is installed in + (typically "/usr/bin") to $PATH when invoking subprograms and hooks + for almost eternity, but the original use case the latter tried to + support was semi-bogus (i.e. install git to /opt/foo/git and run it + without having /opt/foo on $PATH), and more importantly it has + become less and less relevant as Git grew more mainstream (i.e. the + users would _want_ to have it on their $PATH). Stop prepending the + path in which "git" is installed to users' $PATH, as that would + interfere the command search order people depend on (e.g. they may + not like versions of programs that are unrelated to Git in /usr/bin + and want to override them by having different ones in /usr/local/bin + and have the latter directory earlier in their $PATH). + (merge a0b4507 jk/git-no-more-argv0-path-munging later to maint). + + * core.excludesfile (defaulting to $XDG_HOME/git/ignore) is supposed + to be overridden by repository-specific .git/info/exclude file, but + the order was swapped from the beginning. This belatedly fixes it. + (merge 099d2d8 jc/gitignore-precedence later to maint). + + * After "git add -N", the path appeared in output of "git diff HEAD" + and "git diff --cached HEAD", leading "git status" to classify it + as "Changes to be committed". Such a path, however, is not yet to + be scheduled to be committed. "git diff" showed the change to the + path as modification, not as a "new file", in the header of its + output. + + Treat such paths as "yet to be added to the index but Git already + know about them"; "git diff HEAD" and "git diff --cached HEAD" + should not talk about them, and "git diff" should show them as new + files yet to be added to the index. + (merge d95d728 nd/diff-i-t-a later to maint). + + * Code cleanups and documentation updates. + (merge 0269f96 mm/usage-log-l-can-take-regex later to maint). + (merge 64f2589 nd/t1509-chroot-test later to maint). + (merge f86a374 sb/test-bitmap-free-at-end later to maint). + (merge 05bfc7d sb/line-log-plug-pairdiff-leak later to maint). + (merge 846e5df pt/xdg-config-path later to maint). + (merge 1154aa4 jc/plug-fmt-merge-msg-leak later to maint). + (merge 319b678 jk/sha1-file-reduce-useless-warnings later to maint). + (merge 9a35c14 fg/document-commit-message-stripping later to maint). |