The simplify_merges() function needs to look at all history chain to
find the closest ancestor that is relevant after the simplification,
but after --first-parent traversal, side parents haven't been marked
for relevance (they are irrelevant by definition due to the nature
of first-parent-only traversal) nor culled from the parents list of
resulting commits.

We cannot simply remove these side parents from the parents list, as
the output phase still wants to see the parents.  Instead, teach
simplify_one() and its callees to ignore the later parents.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Among the three similar-looking loops that walk singly linked
commit_list, the first one is only peeking and the same list is
later used for real work.  Leave a comment not to mistakenly
free its elements there.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The code internally runs sort_in_topo_order() already; it is more clear
to spell it out in the option parsing phase, instead of adding a special
case in simplify_merges() function.

Finishing touches...

* mm/api-credentials-doc:
  docs: fix cross-directory linkgit references

Most of our documentation is in a single directory, so using
linkgit:git-config[1] just generates a relative link in the
same directory. However, this is not the case with the API
documentation in technical/*, which need to refer to
git-config from the parent directory.

We can fix this by passing a special prefix attribute when building
in a subdirectory, and respecting that prefix in our linkgit
definitions.

We only have to modify the html linkgit definition.  For
manpages, we can ignore this for two reasons:

  1. we do not generate actual links to the file in
     manpages, but instead just give the name and section of
     the linked manpage

  2. we do not currently build manpages for subdirectories,
     only html
Signed-off-by: NJeff King <peff@peff.net>
Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* mm/api-credentials-doc:
  api-credentials.txt: add "see also" section
  api-credentials.txt: mention credential.helper explicitly
  api-credentials.txt: show the big picture first
  doc: fix xref link from api docs to manual pages

* rr/maint-t3510-cascade-fix:
  t3510 (cherry-pick-sequence): add missing '&&'

Regression fix for people with libsvn between 1.6.12 and 1.6.15, on
which we tried to use the non-working platform auth providers.

* jc/svn-auth-providers-unusable-at-1.6.12:
  git-svn: platform auth providers are working only on 1.6.15 or newer

A remote helper that acts as a proxy and caches ssl session for the
https:// transport is added to the contrib/ area.

By Colby Ranger
* cr/persistent-https:
  Add persistent-https to contrib

German and Chinese translation updates.

By Ralf Thielow (4) and others
via Jiang Xin (1) and Tran Ngoc Quan (1)
* git://github.com/git-l10n/git-po:
  l10n: Update translation for Vietnamese
  l10n: de.po: add additional newline
  l10n: de.po: translate 2 new, 3 fuzzy messages
  l10n: de.po: translate 41 new messages
  l10n: de.po: translate 265 new messages
  l10n: zh_CN.po: translate 2 new, 3 fuzzy messages
  l10n: Update git.pot (5 new, 3 removed messages)

Breaks in a test assertion's && chain can potentially hide failures
from earlier commands in the chain.  Fix an instance of this in the
setup.
Signed-off-by: NRamkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The name of the configuration variable was mentioned only at the very
end of the explanation, in a place specific to a specific rule, hence it
was not very clear what the specification was about.
Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The API documentation targets two kinds of developers: those using the
C API, and those writing remote-helpers. The document was not clear
about which part was useful to which category, and for example, the C API
could be mistakenly thought as an API for writting remote helpers.
Based-on-patch-by: NJeff King <peff@peff.net>
Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

They are one-level above, so refer them as linkgit:../git-foo[n] with "../"
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Matthijs Kooijman reports that the cut-off point 082afee6 (git-svn:
use platform specific auth providers, 2012-04-26) set at 1.6.12 to
use this feature safely was incorrect, and it is 1.6.15 instead:

    http://svn.apache.org/repos/asf/subversion/trunk/CHANGES
    Version 1.6.15
       * improve some swig parameter mapping (r984565, r1035745)
Signed-off-by: NJunio C Hamano <gitster@pobox.com>
Acked-by: NEric Wong <normalperson@yhbt.net>

By Ralf Thielow
via Ralf Thielow
* ralfth/git-po-de/master:
  l10n: de.po: add additional newline
  l10n: de.po: translate 2 new, 3 fuzzy messages
  l10n: de.po: translate 41 new messages
  l10n: de.po: translate 265 new messages

 * Updated 5 strings for v1.7.11-rc0-100-g5498c
 * Retranslated about 16 strings
Signed-off-by: NTran Ngoc Quan <vnwildman@gmail.com>

Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* maint:
  Git 1.7.10.4

Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When "git rebase" was given a bad commit to replay the history on,
its error message did not correctly give the command line argument
it had trouble parsing.

By Erik Faye-Lund
* ef/maint-rebase-error-message:
  rebase: report invalid commit correctly

The translation of "builtin/gc.c:224" was missing of
a newline which made the second part of the message
quite long. We simply add a newline.
Signed-off-by: NRalf Thielow <ralf.thielow@googlemail.com>

Translate 2 new and 3 fuzzy messages came from git.pot update
in 75f7b4b9 (l10n: Update git.pot (5 new, 3 removed messages)).
Signed-off-by: NRalf Thielow <ralf.thielow@googlemail.com>

Signed-off-by: NRalf Thielow <ralf.thielow@googlemail.com>

Signed-off-by: NRalf Thielow <ralf.thielow@googlemail.com>

Translate 2 new and 3 fuzzy messages came from git.pot update
in 75f7b4b9 (l10n: Update git.pot (5 new, 3 removed messages)).
Signed-off-by: NJiang Xin <worldhello.net@gmail.com>

Generate po/git.pot from v1.7.11-rc0-100-g5498c:

 * 5 new l10n messages at lines:
   635, 639, 1203, 1208, 3946

 * 3 removed l10n messages at lines:
   1194, 3158, 3936
Signed-off-by: NJiang Xin <worldhello.net@gmail.com>

Signed-off-by: NJunio C Hamano <gitster@pobox.com>

By Erik Faye-Lund
* ef/maint-rebase-error-message:
  rebase: report invalid commit correctly

* nh/empty-rebase:
  cherry-pick: regression fix for empty commits

"git rebase -p" used to pay attention to rebase.autosquash which was
wrong.  "git rebase -p -i" should, but "git rebase -p" by itself
should not.

By Vincent van Ravesteijn
* vr/rebase-autosquash-does-not-imply-i:
  Do not autosquash in case of an implied interactive rebase

"git tags" used to suggest "git stage" which was nonsense; it should
have favored "git tag".  Tweak the cost of deletion to correct it.

By Matthieu Moy
* mm/levenstein-penalize-deletion-less:
  Reduce cost of deletion in levenstein distance (4 -> 3)

"git submodule init" used to report "registered for path ..."  even
for submodules that were registered earlier.

By Jens Lehmann
* jl/submodule-report-new-path-once:
  submodules: print "registered for path" message only once

Signed-off-by: NJunio C Hamano <gitster@pobox.com>

A minor compilation fix.

By Erik Faye-Lund
* ef/http-o-depends-on-gvf:
  Makefile: add missing GIT-VERSION-FILE dependency

By Jiang Xin (4) and others
via Jiang Xin
* git://github.com/git-l10n/git-po:
  l10n: Set nplurals of zh_CN.po from 1 to 2
  l10n: zh_CN.po: translate 323 new messages
  l10n: zh.CN.po: update by msgmerge git.pot
  First release translation for Vietnamese
  Init translation for Vietnamese
  l10n: New it.po file with 504 translations
  Update Swedish translation (728t0f0u)
  l10n: Update git.pot (41 new messages)

"git grep -e '$pattern'", unlike the case where the patterns are read from
a file, did not treat individual lines in the given pattern argument as
separate regular expressions as it should.

By René Scharfe
* rs/maint-grep-F:
  grep: stop leaking line strings with -f
  grep: support newline separated pattern list
  grep: factor out do_append_grep_pat()
  grep: factor out create_grep_pat()