all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Jonathan Ganc <jonganc@gmail.com>
To: Dmitry Gutov <dgutov@yandex.ru>,
	npostavs@users.sourceforge.net, 26066@debbugs.gnu.org,
	Stefan Monnier <monnier@IRO.UMontreal.CA>
Subject: bug#26066: 26.0.50; vc-git-status gives wrong result
Date: Mon, 10 Apr 2017 23:52:24 -0400	[thread overview]
Message-ID: <1a23269d-1ead-1771-1afd-e2c9c5840cb4@gmail.com> (raw)
In-Reply-To: <505388da-e642-8c98-35c0-261d09ff13e1@yandex.ru>

[-- Attachment #1: Type: text/plain, Size: 1424 bytes --]

I've incorporated Thien-Thi's suggestion, as well as an if to check 
status for nil (which indicates unregistered).


On 04/10/2017 08:07 PM, Dmitry Gutov wrote:
> You are right. But we could set it in vc-state-refresh.
>

On 04/10/2017 07:46 PM, Dmitry Gutov wrote:
>
> Thanks. I guess the main thing left is to decide whether vc-git-state 
> should bind default-directory. For example, vc-bzr-state doesn't, 
> while vc-hg-state does.
>
> Does this binding affect the command output, in this particular case?

The binding affects the output if default-directory is not set inside 
the file's repository.

Since, in principle, the vc functions should be agnostic to the choice 
of vcs, either a) vc-state documentation should state that 
default-directory should be set to get generally correct responses or b) 
it should be set in some function (and I agree that vc-state-refresh 
makes sense).

I think the overhead of setting the directory is rather low. In some 
admittedly rudimentary benchmarks, there is almost no difference in 
performance setting default-directory.

There's also the question of how to handle default-directory. You cannot 
simply do (file-name-directory file), because that fails if FILE is 
given without a directory. I think the correct one is 
(file-name-directory (expand-file-name file)) (which, surprisingly, is 
slighly faster than (file-name-directory (concat default-directory file)) ).

[-- Attachment #2: 0001-update-vc-git.patch --]
[-- Type: text/x-patch, Size: 4643 bytes --]

From 8892f98fc0a8a956bc1adefcc0394ac6cf5e47d3 Mon Sep 17 00:00:00 2001
From: Jonathan Ganc <jonganc@gmail.com>
Date: Mon, 10 Apr 2017 00:38:52 -0400
Subject: [PATCH] update vc-git

---
 lisp/vc/vc-git.el | 79 +++++++++++++++++++++++++++++++++++++------------------
 1 file changed, 53 insertions(+), 26 deletions(-)

diff --git a/lisp/vc/vc-git.el b/lisp/vc/vc-git.el
index 1a3f1bf..7c16125 100644
--- a/lisp/vc/vc-git.el
+++ b/lisp/vc/vc-git.el
@@ -231,34 +231,61 @@ (defun vc-git--state-code (code)
     (?U 'edited)     ;; FIXME
     (?T 'edited)))   ;; FIXME
 
+(defun vc-git--git-status-to-vc-state (code-list)
+  "Convert a list CODE-LIST of two-letter git status strings to a vc status.
+
+Each element of CODE-LIST comes from the first two characters of
+a line returned by 'git status' and should be passed in the order given by 'git status'.
+
+\(It is necessary to allow CODE-LIST to be a list because
+sometimes git status returns multiple lines, e.g. for a file that
+is removed from the index but is present in the HEAD and working
+tree.) "
+  (pcase code-list
+    ('nil 'up-to-date)
+    (`(,code)
+     (pcase code
+       ("!!" 'ignored)
+       ("??" 'unregistered)
+       ;; I have only seen this with a file that is only present in the
+       ;; index.  Let us call this `removed'
+       ("AD" 'removed)
+       (_ (cond
+           ((string-match-p "^[ RD]+$" code) 'removed)
+           ((string-match-p "^[ M]+$" code) 'edited)
+           ((string-match-p "^[ A]+$" code) 'added)
+           ((string-match-p "^[ U]+$" code) 'conflict)
+           (t 'edited)))))
+    ;;  I know of two times when git state returns more than one element,
+    ;;  in both cases returning '("D " "??")':
+    ;;  1. when a file is removed from the index but present in the
+    ;;     HEAD and working tree
+    ;;  2. when a file A is renamed to B in the index and then back to A
+    ;;     in the working tree
+    ;;  In both these instances, `unregistered' is a reasonable response.
+    (`("D " "??") 'unregistered)
+    ;;  In other cases, let us return `edited'.
+    (_ 'edited)))
+
 (defun vc-git-state (file)
   "Git-specific version of `vc-state'."
-  ;; FIXME: This can't set 'ignored or 'conflict yet
-  ;; The 'ignored state could be detected with `git ls-files -i -o
-  ;; --exclude-standard` It also can't set 'needs-update or
-  ;; 'needs-merge. The rough equivalent would be that upstream branch
-  ;; for current branch is in fast-forward state i.e. current branch
-  ;; is direct ancestor of corresponding upstream branch, and the file
-  ;; was modified upstream.  But we can't check that without a network
-  ;; operation.
-  ;; This assumes that status is known to be not `unregistered' because
-  ;; we've been successfully dispatched here from `vc-state', that
-  ;; means `vc-git-registered' returned t earlier once.  Bug#11757
-  (let ((diff (vc-git--run-command-string
-               file "diff-index" "-p" "--raw" "-z" "HEAD" "--")))
-    (if (and diff
-             (string-match ":[0-7]\\{6\\} [0-7]\\{6\\} [0-9a-f]\\{40\\} [0-9a-f]\\{40\\} \\([ADMUT]\\)\0[^\0]+\0\\(.*\n.\\)?"
-                           diff))
-        (let ((diff-letter (match-string 1 diff)))
-          (if (not (match-beginning 2))
-              ;; Empty diff: file contents is the same as the HEAD
-              ;; revision, but timestamps are different (eg, file
-              ;; was "touch"ed).  Update timestamp in index:
-              (prog1 'up-to-date
-                (vc-git--call nil "add" "--refresh" "--"
-                              (file-relative-name file)))
-            (vc-git--state-code diff-letter)))
-      (if (vc-git--empty-db-p) 'added 'up-to-date))))
+  (let* ((default-directory (file-name-directory (expand-file-name file)))
+         (status
+          (vc-git--run-command-string file "status" "--porcelain" "-z"
+                                      "--untracked-files" "--ignored" "--"))
+         code-list)
+    (if (null status)
+        ;; If status is nil, there was an error running git, likely because
+        ;; the file is not in a git repo.
+        'unregistered
+      ;; If this code is adapted to parse 'git status' for a directory,
+      ;; note that a renamed file takes up two null values and needs to be
+      ;; treated slightly more carefully.
+      (setq code-list
+            (mapcar (lambda (s)
+                      (substring s 0 2))
+                    (delete "" (split-string status "\0"))))
+      (vc-git--git-status-to-vc-state code-list))))
 
 (defun vc-git-working-revision (_file)
   "Git-specific version of `vc-working-revision'."
-- 
2.9.3


  reply	other threads:[~2017-04-11  3:52 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-12  2:42 bug#26066: 26.0.50; vc-git-status gives wrong result Jonathan Ganc
2017-03-14 13:45 ` npostavs
2017-03-16  0:42 ` npostavs
2017-03-16  0:48   ` Dmitry Gutov
2017-03-16  2:40     ` Jonathan Ganc
2017-03-21  9:19       ` Dmitry Gutov
2017-03-21 16:10         ` Jonathan Ganc
2017-03-22 12:11           ` Michael Heerdegen
2017-03-22 16:20           ` Dmitry Gutov
2017-03-23  2:18             ` Jonathan Ganc
2017-03-31  3:16               ` Jonathan Ganc
2017-04-10  2:16                 ` Jonathan Ganc
2017-04-10  2:58                 ` npostavs
2017-04-10  3:26                 ` Dmitry Gutov
2017-04-10  4:41                   ` Jonathan Ganc
2017-04-10  6:05                     ` Eli Zaretskii
2017-04-10  7:35                     ` Thien-Thi Nguyen
2017-04-10 23:46                     ` Dmitry Gutov
2017-04-10  4:43                   ` Jonathan Ganc
2017-04-10 23:26                     ` Dmitry Gutov
2017-04-11  0:07               ` Dmitry Gutov
2017-04-11  3:52                 ` Jonathan Ganc [this message]
2017-04-11 13:08                   ` Jonathan Ganc
2017-04-11 23:27                     ` Dmitry Gutov
2017-04-11 23:36                       ` Dmitry Gutov
2017-04-14  0:42                       ` Jonathan Ganc
2017-04-23 18:21                         ` Jonathan Ganc
2017-04-23 22:28                           ` Noam Postavsky
2017-04-23 22:45                             ` Stefan Monnier
2017-04-24  0:50                             ` Jonathan Ganc
2017-04-24  9:20                               ` Andreas Schwab
2017-04-25 15:38                                 ` Jonathan Ganc
2017-04-25 23:49                                   ` Dmitry Gutov
2017-04-26  3:18                                     ` Jonathan Ganc
2017-05-01  1:43                                       ` Dmitry Gutov
2019-06-24 23:01                                         ` Lars Ingebrigtsen
2020-08-10 15:05                                           ` Lars Ingebrigtsen
2017-05-01  1:57                         ` Dmitry Gutov
2017-05-01 19:22                           ` Jonathan Ganc
2017-05-01 19:42                             ` Eli Zaretskii
2017-05-01 22:13                             ` Dmitry Gutov
2017-03-18  2:38     ` Jonathan Ganc

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=1a23269d-1ead-1771-1afd-e2c9c5840cb4@gmail.com \
    --to=jonganc@gmail.com \
    --cc=26066@debbugs.gnu.org \
    --cc=dgutov@yandex.ru \
    --cc=monnier@IRO.UMontreal.CA \
    --cc=npostavs@users.sourceforge.net \
    /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.
Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

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.