The logic to append a slash to the url if necessary in quote_ref_url
(added in 113106e0 "http.c: use strbuf API in quote_ref_url") has been
moved to a new function, end_url_with_slash.

The method takes a strbuf, the URL, and the path to be appended to the
URL. It first adds the URL to the strbuf. It then appends a slash
if the URL does not end with a slash.

The check on ref in quote_ref_url for a slash at the beginning has been
removed as a result of using end_url_with_slash. This check is not
needed, because slashes will be quoted anyway.
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Move RANGE_HEADER_SIZE to http.h.

Create no_pragma_header, the curl header list containing the header
"Pragma:" in http.[ch]. It is allocated in http_init, and freed in
http_cleanup. This replaces the no_pragma_header in http-push.c, and
the no_pragma_header member in walker_data in http-walker.c.

Create http_is_verbose. It is to be used by methods in http.c, and is
modified at the entry points of http.c's users, namely http-push.c
(when parsing options) and http-walker.c (in get_http_walker).
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NMike Hommey <mh@glandium.org>
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Since use_pack() will end up populating pack_size if it is not already set,
we can just adapt the code in verify_packfile() such that it doesn't require
pack_size to be set beforehand.

This allows callers not to have to set pack_size themselves, and we can thus
revert changes from 1c23d794 (Don't die in git-http-fetch when fetching packs).
Signed-off-by: NMike Hommey <mh@glandium.org>
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In a70c232e ("http-fetch: do not SEGV after fetching a bad pack idx
file"), changes were made to the setup_index method in http-fetch.c
(known in its present form as http-walker.c after 30ae764b ("Modularize
commit-walker")). Since http-push.c has similar similar code for
processing index files, these changes should apply to http-push.c's
implementation of setup_index as well.
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In the fetch_index implementations in http-push.c and http-walker.c,
the string returned by sha1_to_hex is assumed to stay immutable.

This patch ensures that hex stays immutable by copying the string
returned by sha1_to_hex (via xstrdup) and frees it subsequently. It
also refactors free()'s and fclose()'s with labels.
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In c17fb6ee ("Verify remote packs, speed up pending request queue"),
changes were made to index fetching in http-push.c, particularly the
methods fetch_index and setup_index. Since http-walker.c has similar
code for index fetching, these improvements should apply to
http-walker.c's fetch_index and setup_index.

Invocations of free() of string memory are reproduced as well.
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

- Use tabs to indent, instead of spaces.

- Do not use curly-braces around a single statement body in
  if/while statement;

- Do not start multi-line comment with description on the first
  line after "/*", i.e.

  /*
   * We prefer this over...
   */

  /* comments like
   * this (notice the first line)
   */
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

As it is breaking the build when USE_CURL_MULTI is not defined.
Signed-off-by: NChristian Couder <chriscool@tuxfamily.org>
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Previously, requests for remote files were simply added to the queue
(pointed to by request_queue_head) and no transfer actually takes
place (the fill function add_fill_function() is not added until line
2441), even though code that followed may rely on these remote files to
be present (eg. the setup_revisions invocation).

The code that sends out the requests on the request queue is refactored
into the method run_request_queue.

After the get_dav_remote_heads invocation (ie. after fetch requests are
added to the queue), the requests on the queue are sent out through an
invocation to run_request_queue.

This invocation to run_request_queue entails adding a fill function
before pushing checks take place, which may lead to accidental,
unwanted pushes previously.

The flag is_running_queue is introduced to prevent this from occurring.
fill_active_slot is made to check the flag is_running_queue before
the sending of the requests proceeds.
Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* rc/maint-http-local-slot-fix:
  http*: cleanup slot->local after fclose

Set slot->local to NULL after doing a fclose() on the file it points
to. This prevents the passing of a FILE* pointer to a fclose()'d file
to ftell() in http.c::run_active_slot().

This issue was raised by Clemens Buchacher on 30th May 2009:

  http://www.spinics.net/lists/git/msg104623.htmlSigned-off-by: NTay Ray Chuan <rctay89@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* np/push-delta:
  allow OFS_DELTA objects during a push

* ar/merge-one-file-diag:
  Clarify kind of conflict in merge-one-file helper

* ar/unlink-err:
  print unlink(2) errno in copy_or_link_directory
  replace direct calls to unlink(2) with unlink_or_warn
  Introduce an unlink(2) wrapper which gives warning if unlink failed

* jk/maint-add-empty:
  add: don't complain when adding empty project root

* js/add-edit:
  t3702: fix reliance on SHELL_PATH being '/bin/sh'
  git-add: introduce --edit (to edit the diff vs. the index)

* mh/diff-stat-color:
  diff: do not color --stat output like patch context

* mh/show-branch-color:
  bash completion: show-branch color support
  show-branch: color the commit status signs

Conflicts:
	contrib/completion/git-completion.bash

* ac/graph-horizontal-line:
  graph API: Use horizontal lines for more compact graphs

* ae/anon-fetch-info:
  fetch: Strip usernames from url's before storing them

* maint:
  show-branch: Fix die message in parse_reflog_param()

Commit 76a44c5c (show-branch --reflog: show the reflog message at the
top, 2007-01-19) introduced parse_reflog_param(). The die() call was
incorrectly passed arg + 9, when it should have been passed arg.
Signed-off-by: NStephen Boyd <bebarino@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* maint:
  completion: add missing options to show-branch and show
  dir.c: clean up handling of 'path' parameter in read_directory_recursive()
  Fix type-punning issues

Add --oneline and --abbrev-commit to show and --sparse to show-branch.
Signed-off-by: NStephen Boyd <bebarino@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Right now we pass two different pathnames ('path' and 'base') down to
read_directory_recursive(), and the only real reason for that is that we
want to allow an empty 'base' parameter, but when we do so, we need the
pathname to "opendir()" to be "." rather than the empty string.

And rather than handle that confusion in the caller, we can just fix
read_directory_recursive() to handle the case of an empty path itself,
by just passing opendir() a "." ourselves if the path is empty.

This would allow us to then drop one of the pathnames entirely from the
calling convention, but rather than do that, we'll start separating them
out as a "filesystem pathname" (the one we use for filesystem accesses)
and a "git internal base name" (which is the name that we use for git
internally).

That will eventually allow us to do things like handle different
encodings (eg the filesystem pathnames might be Latin1, while git itself
would use UTF-8 for filename information).
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In these two places we are casting part of our unsigned char sha1 array into
an unsigned int, which violates GCCs strict-aliasing rules (and probably
other compilers).
Signed-off-by: NDan McGee <dpmcgee@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* maint:
  test: checkout shouldn't say that HEAD has moved if it didn't
  completion: enhance "current branch" display
  completion: simplify "current branch" in __git_ps1()
  completion: fix PS1 display during a merge on detached HEAD
  builtin-checkout: Don't tell user that HEAD has moved before it has
  pre-commit.sample: don't print incidental SHA1
  tests: Add tests for missing format-patch long options
  api-parse-options.txt: use 'func' instead of 'funct'
  Turn on USE_ST_TIMESPEC for OpenBSD
  ls-tree manpage: output of ls-tree is compatible with update-index
  ls-tree manpage: use "unless" instead of "when ... is not"

Signed-off-by: Nしらいしななこ <nanako3@lavabit.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Introduce GIT_PS1_DESCRIBE option you can set to "contains", "branch", or
"describe" to tweak the way how a detached HEAD is described.

The default behaviour is to describe only exact match with some tag
(otherwise use the first 7 hexdigits) as before.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

As I very often work on a detached HEAD, I found it pretty confusing
when __git_ps1() said 'some-name'.  Did I create a branch with that name
by mistake, or do I happen to be on a commit with that exact tag?

This patch fixes the issue by enclosing non branch names in a pair of
parentheses when used to substitute %s token in __git_ps1() argument.

It also fixes a small bug where the branch part is left empty when
.git/HEAD is unreadable for whatever reason.  The output now says
"(unknown)".
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

If your merge stops in a conflict while on a detached HEAD, recent
completion code fails to show anything.  This was because various cases
added to support the operation-in-progress markers (e.g. REBASE, MERGING)
forgot that they need to set the variable "b" to something for the result
they computed to be displayed at all.

Probably not many people make trial merges on a detached HEAD (which is
tremendously useful feature of git, by the way), and that may be why this
was not noticed for a long time.
Acked-By: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Strip leading and trailing spaces off guessed target directory, and
replace sequences of whitespace and 'control' characters with one
space character.

User still can have any name by specifying it explicitely after url.
Signed-off-by: NAlex Riesen <raa.lkml@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

An old iconv (GNU libiconv 1.11) does not know about utf8, it does know
UTF-8 though, which is also understood by all newer iconv implementations.
Signed-off-by: NAlex Riesen <raa.lkml@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Previously, checkout would tell the user this message before moving HEAD,
without regard to whether the upcoming move will result in success.
If the move failed, this causes confusion.

Show the message after the move, unless the move failed.
Signed-off-by: NDaniel Cordero <theappleman@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Make the sample pre-commit hook script discard
all git-rev-parse output, not just stderr.
Otherwise, it would print an SHA1.
Signed-off-by: NJim Meyering <meyering@redhat.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Exercise format-patch's --signoff, --in-reply-to and --start-number long
options.
Signed-off-by: NStephen Boyd <bebarino@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>