@matheusbsbs / matheus-bagliano / commits / 9472935d81

add: introduce "--renormalize"

Make it safer to normalize the line endings in a repository. Files that had been commited with CRLF will be commited with LF. The old way to normalize a repo was like this: # Make sure that there are not untracked files $ echo "* text=auto" >.gitattributes $ git read-tree --empty $ git add . $ git commit -m "Introduce end-of-line normalization" The user must make sure that there are no untracked files, otherwise they would have been added and tracked from now on. The new "add --renormalize" does not add untracked files: $ echo "* text=auto" >.gitattributes $ git add --renormalize . $ git commit -m "Introduce end-of-line normalization" Note that "git add --renormalize <pathspec>" is the short form for "git add -u --renormalize <pathspec>". While at it, document that the same renormalization may be needed, whenever a clean filter is added or changed. Helped-By: Junio C Hamano <gitster@pobox.com> Signed-off-by: Torsten Bögershausen <tboegi@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Torsten Bögershausen committed Nov 16, 2017 at 17:38 UTC 9472935d81eaf9faed771878c9df0216ae0d9045

7 files changed +102 -18

Documentation/git-add.txt

+8 -1

@@ -10,7 +10,7 @@ SYNOPSIS

10 [verse]

12 [--edit | -e] [--[no-]all | --[no-]ignore-removal | [--update | -u]]

13 - [--intent-to-add | -N] [--refresh] [--ignore-errors] [--ignore-missing]

13 + [--intent-to-add | -N] [--refresh] [--ignore-errors] [--ignore-missing] [--renormalize]

14 [--chmod=(+|-)x] [--] [<pathspec>...]

16 DESCRIPTION

@@ -175,6 +175,13 @@ for "git add --no-all <pathspec>...", i.e. ignored removed files.

175 warning (e.g., if you are manually performing operations on

176 submodules).

177

178 +--renormalize::

179 + Apply the "clean" process freshly to all tracked files to

180 + forcibly add them again to the index. This is useful after

181 + changing `core.autocrlf` configuration or the `text` attribute

182 + in order to correct files added with wrong CRLF/LF line endings.

183 + This option implies `-u`.

184 +

185 --chmod=(+|-)x::

186 Override the executable bit of the added files. The executable

187 bit is only changed in the index, the files on disk are left

Documentation/gitattributes.txt

+4 -2

@@ -232,8 +232,7 @@ From a clean working directory:

232

233 -------------------------------------------------

234 $ echo "* text=auto" >.gitattributes

235 -$ git read-tree --empty # Clean index, force re-scan of working directory

236 -$ git add .

235 +$ git add --renormalize .

236 $ git status # Show files that will be normalized

237 $ git commit -m "Introduce end-of-line normalization"

238 -------------------------------------------------

@@ -328,6 +327,9 @@ You can declare that a filter turns a content that by itself is unusable

327 into a usable content by setting the filter.<driver>.required configuration

328 variable to `true`.

329

330 +Note: Whenever the clean filter is changed, the repo should be renormalized:

331 +$ git add --renormalize .

332 +

333 For example, in .gitattributes, you would assign the `filter`

334 attribute for paths.

335

builtin/add.c

+26 -2

@@ -26,6 +26,7 @@ static const char * const builtin_add_usage[] = {

26 };

27 static int patch_interactive, add_interactive, edit_interactive;

28 static int take_worktree_changes;

29 +static int add_renormalize;

31 struct update_callback_data {

32 int flags;

@@ -123,6 +124,25 @@ int add_files_to_cache(const char *prefix,

124 return !!data.add_errors;

125 }

126

127 +static int renormalize_tracked_files(const struct pathspec *pathspec, int flags)

128 +{

129 + int i, retval = 0;

130 +

131 + for (i = 0; i < active_nr; i++) {

132 + struct cache_entry *ce = active_cache[i];

133 +

134 + if (ce_stage(ce))

135 + continue; /* do not touch unmerged paths */

136 + if (!S_ISREG(ce->ce_mode) && !S_ISLNK(ce->ce_mode))

137 + continue; /* do not touch non blobs */

138 + if (pathspec && !ce_path_match(ce, pathspec, NULL))

139 + continue;

140 + retval |= add_file_to_cache(ce->name, flags | HASH_RENORMALIZE);

141 + }

142 +

143 + return retval;

144 +}

145 +

146 static char *prune_directory(struct dir_struct *dir, struct pathspec *pathspec, int prefix)

147 {

148 char *seen;

@@ -276,6 +296,7 @@ static struct option builtin_add_options[] = {

296 OPT_BOOL('e', "edit", &edit_interactive, N_("edit current diff and apply")),

297 OPT__FORCE(&ignored_too, N_("allow adding otherwise ignored files")),

298 OPT_BOOL('u', "update", &take_worktree_changes, N_("update tracked files")),

299 + OPT_BOOL(0, "renormalize", &add_renormalize, N_("renormalize EOL of tracked files (implies -u)")),

300 OPT_BOOL('N', "intent-to-add", &intent_to_add, N_("record only the fact that the path will be added later")),

301 OPT_BOOL('A', "all", &addremove_explicit, N_("add changes from all tracked and untracked files")),

302 { OPTION_CALLBACK, 0, "ignore-removal", &addremove_explicit,

@@ -406,7 +427,7 @@ int cmd_add(int argc, const char **argv, const char *prefix)

427 chmod_arg[1] != 'x' || chmod_arg[2]))

428 die(_("--chmod param '%s' must be either -x or +x"), chmod_arg);

429

409 - add_new_files = !take_worktree_changes && !refresh_only;

430 + add_new_files = !take_worktree_changes && !refresh_only && !add_renormalize;

431 require_pathspec = !(take_worktree_changes || (0 < addremove_explicit));

432

433 hold_locked_index(&lock_file, LOCK_DIE_ON_ERROR);

@@ -500,7 +521,10 @@ int cmd_add(int argc, const char **argv, const char *prefix)

521

522 plug_bulk_checkin();

523

503 - exit_status |= add_files_to_cache(prefix, &pathspec, flags);

524 + if (add_renormalize)

525 + exit_status |= renormalize_tracked_files(&pathspec, flags);

526 + else

527 + exit_status |= add_files_to_cache(prefix, &pathspec, flags);

528

529 if (add_new_files)

530 exit_status |= add_files(&dir, flags);

cache.h

@@ -686,6 +686,7 @@ extern int ie_modified(const struct index_state *, const struct cache_entry *, s

686

687 #define HASH_WRITE_OBJECT 1

688 #define HASH_FORMAT_CHECK 2

689 +#define HASH_RENORMALIZE 4

690 extern int index_fd(struct object_id *oid, int fd, struct stat *st, enum object_type type, const char *path, unsigned flags);

691 extern int index_path(struct object_id *oid, const char *path, struct stat *st, unsigned flags);

692

read-cache.c

+19 -11

@@ -631,13 +631,17 @@ int add_to_index(struct index_state *istate, const char *path, struct stat *st,

631 {

632 int size, namelen, was_same;

633 mode_t st_mode = st->st_mode;

634 - struct cache_entry *ce, *alias;

634 + struct cache_entry *ce, *alias = NULL;

635 unsigned ce_option = CE_MATCH_IGNORE_VALID|CE_MATCH_IGNORE_SKIP_WORKTREE|CE_MATCH_RACY_IS_DIRTY;

636 int verbose = flags & (ADD_CACHE_VERBOSE | ADD_CACHE_PRETEND);

637 int pretend = flags & ADD_CACHE_PRETEND;

638 int intent_only = flags & ADD_CACHE_INTENT;

639 int add_option = (ADD_CACHE_OK_TO_ADD|ADD_CACHE_OK_TO_REPLACE|

640 (intent_only ? ADD_CACHE_NEW_ONLY : 0));

641 + int newflags = HASH_WRITE_OBJECT;

642 +

643 + if (flags & HASH_RENORMALIZE)

644 + newflags |= HASH_RENORMALIZE;

645

646 if (!S_ISREG(st_mode) && !S_ISLNK(st_mode) && !S_ISDIR(st_mode))

647 return error("%s: can only add regular files, symbolic links or git-directories", path);

@@ -678,19 +682,23 @@ int add_to_index(struct index_state *istate, const char *path, struct stat *st,

682 if (ignore_case) {

683 adjust_dirname_case(istate, ce->name);

684 }

685 + if (!(flags & HASH_RENORMALIZE)) {

686 + alias = index_file_exists(istate, ce->name,

687 + ce_namelen(ce), ignore_case);

688 + if (alias &&

689 + !ce_stage(alias) &&

690 + !ie_match_stat(istate, alias, st, ce_option)) {

691 + /* Nothing changed, really */

692 + if (!S_ISGITLINK(alias->ce_mode))

693 + ce_mark_uptodate(alias);

694 + alias->ce_flags |= CE_ADDED;

695

682 - alias = index_file_exists(istate, ce->name, ce_namelen(ce), ignore_case);

683 - if (alias && !ce_stage(alias) && !ie_match_stat(istate, alias, st, ce_option)) {

684 - /* Nothing changed, really */

685 - if (!S_ISGITLINK(alias->ce_mode))

686 - ce_mark_uptodate(alias);

687 - alias->ce_flags |= CE_ADDED;

688 -

689 - free(ce);

690 - return 0;

696 + free(ce);

697 + return 0;

698 + }

699 }

700 if (!intent_only) {

693 - if (index_path(&ce->oid, path, st, HASH_WRITE_OBJECT)) {

701 + if (index_path(&ce->oid, path, st, newflags)) {

702 free(ce);

703 return error("unable to index file %s", path);

704 }

sha1_file.c

+14 -2

@@ -74,6 +74,18 @@ static struct cached_object *find_cached_object(const unsigned char *sha1)

74 return NULL;

75 }

77 +

78 +static enum safe_crlf get_safe_crlf(unsigned flags)

79 +{

80 + if (flags & HASH_RENORMALIZE)

81 + return SAFE_CRLF_RENORMALIZE;

82 + else if (flags & HASH_WRITE_OBJECT)

83 + return safe_crlf;

84 + else

85 + return SAFE_CRLF_FALSE;

86 +}

87 +

88 +

89 int mkdir_in_gitdir(const char *path)

90 {

91 if (mkdir(path, 0777)) {

@@ -1680,7 +1692,7 @@ static int index_mem(unsigned char *sha1, void *buf, size_t size,

1692 if ((type == OBJ_BLOB) && path) {

1693 struct strbuf nbuf = STRBUF_INIT;

1694 if (convert_to_git(&the_index, path, buf, size, &nbuf,

1683 - write_object ? safe_crlf : SAFE_CRLF_FALSE)) {

1695 + get_safe_crlf(flags))) {

1696 buf = strbuf_detach(&nbuf, &size);

1697 re_allocated = 1;

1698 }

@@ -1714,7 +1726,7 @@ static int index_stream_convert_blob(unsigned char *sha1, int fd,

1726 assert(would_convert_to_git_filter_fd(path));

1727

1728 convert_to_git_filter_fd(&the_index, path, fd, &sbuf,

1717 - write_object ? safe_crlf : SAFE_CRLF_FALSE);

1729 + get_safe_crlf(flags));

1730

1731 if (write_object)

1732 ret = write_sha1_file(sbuf.buf, sbuf.len, typename(OBJ_BLOB),

t/t0025-crlf-renormalize.sh new

+30

@@ -0,0 +1,30 @@

1 +#!/bin/sh

2 +

3 +test_description='CRLF renormalization'

4 +

5 +. ./test-lib.sh

6 +

7 +test_expect_success setup '

8 + git config core.autocrlf false &&

9 + printf "LINEONE\nLINETWO\nLINETHREE\n" >LF.txt &&

10 + printf "LINEONE\r\nLINETWO\r\nLINETHREE\r\n" >CRLF.txt &&

11 + printf "LINEONE\r\nLINETWO\nLINETHREE\n" >CRLF_mix_LF.txt &&

12 + git add . &&

13 + git commit -m initial

14 +'

15 +

16 +test_expect_success 'renormalize CRLF in repo' '

17 + echo "*.txt text=auto" >.gitattributes &&

18 + git add --renormalize "*.txt" &&

19 + cat >expect <<-\EOF &&

20 + i/lf w/crlf attr/text=auto CRLF.txt

21 + i/lf w/lf attr/text=auto LF.txt

22 + i/lf w/mixed attr/text=auto CRLF_mix_LF.txt

23 + EOF

24 + git ls-files --eol |

25 + sed -e "s/ / /g" -e "s/ */ /g" |

26 + sort >actual &&

27 + test_cmp expect actual

28 +'

29 +

30 +test_done