From: "Chris Torek via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Chris Torek <chris.torek@gmail.com>
Subject: [PATCH v4 0/3] improve git-diff documentation and A...B handling
Date: Fri, 12 Jun 2020 16:19:57 +0000 [thread overview]
Message-ID: <pull.804.v4.git.git.1591978801.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.804.v3.git.git.1591888511.gitgitgadget@gmail.com>
git diff -h help is succinct, but perhaps too much so.
The symmetric-diff syntax, git diff A...B, is defined by the documentation
to compare the merge base of A and B to commit B. It does so just fine when
there is a merge base. It compares A and B directly if there is no merge
base, and it is overly forgiving of bad arguments after which it can produce
nonsensical diffs. It also behaves badly with other odd/incorrect usages,
such as git diff A...B C..D.
The first patch simply adjusts a test that will fail if the second patch is
accepted. The second patch adds special handling for the symmetric and range
diff syntax so that the option parsing works, plus a small test suite. The
third patch updates the documentation, including adding a section for
combined commits, and makes the help output more verbose (to match the
SYNOPSIS and provide common diff options like git-diff-files, for instance).
Changes since v3:
* correct > / >= goof
* fix test nit per Philip Oakley
Chris Torek (3):
t/t3430: avoid undefined git diff behavior
git diff: improve range handling
Documentation: usage for diff combined commits
Documentation/git-diff.txt | 20 ++++--
builtin/diff.c | 132 +++++++++++++++++++++++++++++++++----
t/t3430-rebase-merges.sh | 2 +-
t/t4068-diff-symmetric.sh | 91 +++++++++++++++++++++++++
4 files changed, 226 insertions(+), 19 deletions(-)
create mode 100755 t/t4068-diff-symmetric.sh
base-commit: 20514004ddf1a3528de8933bc32f284e175e1012
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-git-804%2Fchris3torek%2Fcleanup-diff-v4
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-git-804/chris3torek/cleanup-diff-v4
Pull-Request: https://github.com/git/git/pull/804
Range-diff vs v3:
1: 2ccaad645ff = 1: 2ccaad645ff t/t3430: avoid undefined git diff behavior
2: 60aed3f9d65 ! 2: 4fa6fba33b3 git diff: improve range handling
@@ builtin/diff.c: static int builtin_diff_files(struct rev_info *revs, int argc, c
+ basecount++;
+ break; /* do mark all bases */
+ case REV_CMD_LEFT:
-+ if (lpos > 0)
++ if (lpos >= 0)
+ usage(builtin_diff_usage);
+ lpos = i;
+ if (obj->flags & SYMMETRIC_LEFT) {
@@ builtin/diff.c: static int builtin_diff_files(struct rev_info *revs, int argc, c
+ }
+ continue;
+ case REV_CMD_RIGHT:
-+ if (rpos > 0)
++ if (rpos >= 0)
+ usage(builtin_diff_usage);
+ rpos = i;
+ continue; /* don't mark B */
@@ t/t4068-diff-symmetric.sh (new)
+. ./test-lib.sh
+
+# build these situations:
-+# - normal merge with one merge base (b1...b2);
-+# - criss-cross merge ie 2 merge bases (b1...master);
-+# - disjoint subgraph (orphan branch, b3...master).
++# - normal merge with one merge base (br1...b2r);
++# - criss-cross merge ie 2 merge bases (br1...master);
++# - disjoint subgraph (orphan branch, br3...master).
+#
+# B---E <-- master
+# / \ /
3: a7da92cd635 = 3: d7bc9aca44b Documentation: usage for diff combined commits
--
gitgitgadget
next prev parent reply other threads:[~2020-06-12 16:20 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-09 0:03 [PATCH 0/3] improve git-diff documentation and A...B handling Chris Torek via GitGitGadget
2020-06-09 0:03 ` [PATCH 1/3] t/t3430: avoid undocumented git diff behavior Chris Torek via GitGitGadget
2020-06-09 5:18 ` Junio C Hamano
2020-06-09 0:03 ` [PATCH 2/3] git diff: improve A...B merge-base handling Chris Torek via GitGitGadget
2020-06-09 5:40 ` Junio C Hamano
2020-06-12 13:38 ` Philip Oakley
2020-06-12 17:06 ` Junio C Hamano
2020-06-09 0:03 ` [PATCH 3/3] Documentation: tweak git diff help slightly Chris Torek via GitGitGadget
2020-06-09 5:45 ` Junio C Hamano
2020-06-09 19:00 ` [PATCH v2 0/3] improve git-diff documentation and A...B handling Chris Torek via GitGitGadget
2020-06-09 19:00 ` [PATCH v2 1/3] t/t3430: avoid undefined git diff behavior Chris Torek via GitGitGadget
2020-06-09 19:00 ` [PATCH v2 2/3] git diff: improve A...B merge-base handling Chris Torek via GitGitGadget
2020-06-09 22:52 ` Junio C Hamano
2020-06-09 19:00 ` [PATCH v2 3/3] Documentation: tweak git diff help slightly Chris Torek via GitGitGadget
2020-06-09 23:01 ` Junio C Hamano
2020-06-11 15:15 ` [PATCH v3 0/3] improve git-diff documentation and A...B handling Chris Torek via GitGitGadget
2020-06-11 15:15 ` [PATCH v3 1/3] t/t3430: avoid undefined git diff behavior Chris Torek via GitGitGadget
2020-06-11 15:15 ` [PATCH v3 2/3] git diff: improve range handling Chris Torek via GitGitGadget
2020-06-11 15:51 ` Chris Torek
2020-06-11 15:15 ` [PATCH v3 3/3] Documentation: usage for diff combined commits Chris Torek via GitGitGadget
2020-06-12 16:19 ` Chris Torek via GitGitGadget [this message]
2020-06-12 16:19 ` [PATCH v4 1/3] t/t3430: avoid undefined git diff behavior Chris Torek via GitGitGadget
2020-06-12 16:19 ` [PATCH v4 2/3] git diff: improve range handling Chris Torek via GitGitGadget
2020-06-12 18:31 ` Junio C Hamano
2020-06-12 19:25 ` Chris Torek
2020-06-12 16:20 ` [PATCH v4 3/3] Documentation: usage for diff combined commits Chris Torek via GitGitGadget
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=pull.804.v4.git.git.1591978801.gitgitgadget@gmail.com \
--to=gitgitgadget@gmail.com \
--cc=chris.torek@gmail.com \
--cc=git@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.