---
- branch: pkgsrc-2017Q3
date: Sat Nov 25 08:49:05 UTC 2017
files:
- new: 1.61.4.1
old: '1.61'
path: pkgsrc/devel/git/Makefile.version
pathrev: pkgsrc/devel/git/Makefile.version@1.61.4.1
type: modified
- new: 1.45.4.1
old: '1.45'
path: pkgsrc/devel/git-base/Makefile
pathrev: pkgsrc/devel/git-base/Makefile@1.45.4.1
type: modified
- new: 1.70.4.1
old: '1.70'
path: pkgsrc/devel/git-base/distinfo
pathrev: pkgsrc/devel/git-base/distinfo@1.70.4.1
type: modified
id: 20171125T084905Z.98dd06fe578ea0c684c69fce32ec4031eeaed4ee
log: "Pullup ticket #5646 - requested by he\ndevel/git-base: security fix\n\nRevisions
pulled up:\n- devel/git-base/Makefile 1.46\n-
devel/git-base/distinfo 1.71-1.72\n- devel/git/Makefile.version
\ 1.62-1.63\n\n---\n Module Name:\tpkgsrc\n
\ Committed By:\tadam\n Date:\t\tWed Sep 27 06:37:47 UTC 2017\n\n Modified
Files:\n \tpkgsrc/devel/git: Makefile.version\n \tpkgsrc/devel/git-base: Makefile
distinfo\n\n Log Message:\n git: update to 2.14.2\n\n Fixes since v2.14.1\n\n
\ * Because recent Git for Windows do come with a real msgfmt, the\n build
procedure for git-gui has been updated to use it instead of a\n hand-rolled
substitute.\n\n * \"%C(color name)\" in the pretty print format always produced
ANSI\n color escape codes, which was an early design mistake. They now\n
\ honor the configuration (e.g. \"color.ui = never\") and also tty-ness\n
\ of the output medium.\n\n * The http.{sslkey,sslCert} configuration variables
are to be\n interpreted as a pathname that honors \"~[username]/\" prefix,
but\n weren't, which has been fixed.\n\n * Numerous bugs in walking of
reflogs via \"log -g\" and friends have\n been fixed.\n\n * \"git commit\"
when seeing an totally empty message said \"you did not\n edit the message\",
which is clearly wrong. The message has been\n corrected.\n\n * When
a directory is not readable, \"gitweb\" fails to build the\n project list.
\ Work this around by skipping such a directory.\n\n * A recently added test
for the \"credential-cache\" helper revealed\n that EOF detection done around
the time the connection to the cache\n daemon is torn down were flaky. This
was fixed by reacting to\n ECONNRESET and behaving as if we got an EOF.\n\n
\ * Some versions of GnuPG fail to kill gpg-agent it auto-spawned\n and
such a left-over agent can interfere with a test. Work it\n around by attempting
to kill one before starting a new test.\n\n * \"git log --tag=no-such-tag\"
showed log starting from HEAD, which\n has been fixed---it now shows nothing.\n\n
\ * The \"tag.pager\" configuration variable was useless for those who\n actually
create tag objects, as it interfered with the use of an\n editor. A new
mechanism has been introduced for commands to enable\n pager depending on
what operation is being carried out to fix this,\n and then \"git tag -l\"
is made to run pager by default.\n\n * \"git push --recurse-submodules $there
HEAD:$target\" was not\n propagated down to the submodules, but now it is.\n\n
\ * Commands like \"git rebase\" accepted the --rerere-autoupdate option\n from
the command line, but did not always use it. This has been\n fixed.\n\n
\ * \"git clone --recurse-submodules --quiet\" did not pass the quiet\n option
down to submodules.\n\n * \"git am -s\" has been taught that some input may
end with a trailer\n block that is not Signed-off-by: and it should refrain
from adding\n an extra blank line before adding a new sign-off in such a
case.\n\n * \"git svn\" used with \"--localtime\" option did not compute the
tz\n offset for the timestamp in question and instead always used the\n current
time, which has been corrected.\n\n * Memory leaks in a few error codepaths
have been plugged.\n\n * bash 4.4 or newer gave a warning on NUL byte in command\n
\ substitution done in \"git stash\"; this has been squelched.\n\n * \"git
grep -L\" and \"git grep --quiet -L\" reported different exit\n codes; this
has been corrected.\n\n * When handshake with a subprocess filter notices that
the process\n asked for an unknown capability, Git did not report what program\n
\ the offending subprocess was running. This has been corrected.\n\n *
\"git apply\" that is used as a better \"patch -p1\" failed to apply a\n taken
from a file with CRLF line endings to a file with CRLF line\n endings. The
root cause was because it misused convert_to_git()\n that tried to do \"safe-crlf\"
processing by looking at the index\n entry at the same path, which is a nonsense---in
that mode, \"apply\"\n is not working on the data in (or derived from) the
index at all.\n This has been fixed.\n\n * Killing \"git merge --edit\"
before the editor returns control left\n the repository in a state with MERGE_MSG
but without MERGE_HEAD,\n which incorrectly tells the subsequent \"git commit\"
that there was\n a squash merge in progress. This has been fixed.\n\n *
\"git archive\" did not work well with pathspecs and the\n export-ignore
attribute.\n\n * \"git cvsserver\" no longer is invoked by \"git daemon\" by
default,\n as it is old and largely unmaintained.\n\n * Various Perl scripts
did not use safe_pipe_capture() instead of\n backticks, leaving them susceptible
to end-user input. They have\n been corrected.\n\n---\n Module Name:\tpkgsrc\n
\ Committed By:\tadam\n Date:\t\tTue Oct 24 06:43:24 UTC 2017\n\n Modified
Files:\n \tpkgsrc/devel/git: Makefile.version\n \tpkgsrc/devel/git-base: distinfo\n\n
\ Log Message:\n git: updated to 2.14.3\n\n Git v2.14.3 Release Notes\n\n
\ Fixes since v2.14.2\n\n * A helper function to read a single whole line
into strbuf\n mistakenly triggered OOM error at EOF under certain conditions,\n
\ which has been fixed.\n\n * In addition to \"cc: # cruft\",
\"cc: a@dd.re.ss # cruft\"\n was taught to \"git send-email\" as a valid
way to tell it that it\n needs to also send a carbon copy to
in the trailer\n section.\n\n * Fix regression to \"gitk --bisect\" by
a recent update.\n\n * Unlike \"git commit-tree < file\", \"git commit-tree
-F file\" did not\n pass the contents of the file verbatim and instead completed
an\n incomplete line at the end, if exists. The latter has been updated\n
\ to match the behaviour of the former.\n\n * \"git archive\", especially
when used with pathspec, stored an empty\n directory in its output, even
though Git itself never does so.\n This has been fixed.\n\n * API error-proofing
which happens to also squelch warnings from GCC.\n\n * \"git gc\" tries to
avoid running two instances at the same time by\n reading and writing pid/host
from and to a lock file; it used to\n use an incorrect fscanf() format when
reading, which has been\n corrected.\n\n * The test linter has been taught
that we do not like \"echo -e\".\n\n * Code cmp.std.c nitpick.\n\n * \"git
describe --match\" learned to take multiple patterns in v2.13\n series, but
the feature ignored the patterns after the first one\n and did not work at
all. This has been fixed.\n\n * \"git cat-file --textconv\" started segfaulting
recently, which\n has been corrected.\n\n * The built-in pattern to detect
the \"function header\" for HTML did\n not match