@matheusbsbs / matheus-bagliano / commits / 97be04077f

cat-file: only split on whitespace when %(rest) is used

Commit c334b87b (cat-file: split --batch input lines on whitespace, 2013-07-11) taught `cat-file --batch-check` to split input lines on the first whitespace, and stash everything after the first token into the %(rest) output format element. It claimed: Object names cannot contain spaces, so any input with spaces would have resulted in a "missing" line. But that is not correct. Refs, object sha1s, and various peeling suffixes cannot contain spaces, but some object names can. In particular: 1. Tree paths like "[<tree>]:path with whitespace" 2. Reflog specifications like "@{2 days ago}" 3. Commit searches like "rev^{/grep me}" or ":/grep me" To remain backwards compatible, we cannot split on whitespace by default, hence we will ship 1.8.4 with the commit reverted. Resurrect its attempt but in a weaker form; only do the splitting when "%(rest)" is used in the output format. Since that element did not exist at all before c334b87, old scripts cannot be affected. The existence of object names with spaces does mean that you cannot reliably do: echo ":path with space and other data" | git cat-file --batch-check="%(objectname) %(rest)" as it would split the path and feed only ":path" to get_sha1. But that command is nonsensical. If you wanted to see "and other data" in "%(rest)", git cannot possibly know where the filename ends and the "rest" begins. It might be more robust to have something like "-z" to separate the input elements. But this patch is still a reasonable step before having that. It makes the easy cases easy; people who do not care about %(rest) do not have to consider it, and the %(rest) code handles the spaces and newlines of "rev-list --objects" correctly. Hard cases remain hard but possible (if you might get whitespace in your input, you do not get to use %(rest) and must split and join the output yourself using more flexible tools). And most importantly, it does not preclude us from having different splitting rules later if a "-z" (or similar) option is added. So we can make the hard cases easier later, if we choose to. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Jeff King committed Aug 2, 2013 at 04:59 UTC 97be04077f9ed7ef6ca781cc85d5a0ed36530c04

3 files changed +55 -5

Documentation/git-cat-file.txt

+10 -4

@@ -86,10 +86,9 @@ BATCH OUTPUT

86 ------------

88 If `--batch` or `--batch-check` is given, `cat-file` will read objects

89 -from stdin, one per line, and print information about them.

90 -

91 -Each line is considered as a whole object name, and is parsed as if

92 -given to linkgit:git-rev-parse[1].

89 +from stdin, one per line, and print information about them. By default,

90 +the whole line is considered as an object, as if it were fed to

91 +linkgit:git-rev-parse[1].

93 You can specify the information shown for each object by using a custom

94 `<format>`. The `<format>` is copied literally to stdout for each

@@ -110,6 +109,13 @@ newline. The available atoms are:

109 The size, in bytes, that the object takes up on disk. See the

110 note about on-disk sizes in the `CAVEATS` section below.

111

112 +`rest`::

113 + If this atom is used in the output string, input lines are split

114 + at the first whitespace boundary. All characters before that

115 + whitespace are considered to be the object name; characters

116 + after that first run of whitespace (i.e., the "rest" of the

117 + line) are output in place of the `%(rest)` atom.

118 +

119 If no format is specified, the default format is `%(objectname)

120 %(objecttype) %(objectsize)`.

121

builtin/cat-file.c

+30 -1

@@ -119,6 +119,7 @@ struct expand_data {

119 enum object_type type;

120 unsigned long size;

121 unsigned long disk_size;

122 + const char *rest;

123

124 /*

125 * If mark_query is true, we do not expand anything, but rather

@@ -126,6 +127,13 @@ struct expand_data {

127 */

128 int mark_query;

129

130 + /*

131 + * Whether to split the input on whitespace before feeding it to

132 + * get_sha1; this is decided during the mark_query phase based on

133 + * whether we have a %(rest) token in our format.

134 + */

135 + int split_on_whitespace;

136 +

137 /*

138 * After a mark_query run, this object_info is set up to be

139 * passed to sha1_object_info_extended. It will point to the data

@@ -163,6 +171,11 @@ static void expand_atom(struct strbuf *sb, const char *atom, int len,

171 data->info.disk_sizep = &data->disk_size;

172 else

173 strbuf_addf(sb, "%lu", data->disk_size);

174 + } else if (is_atom("rest", atom, len)) {

175 + if (data->mark_query)

176 + data->split_on_whitespace = 1;

177 + else if (data->rest)

178 + strbuf_addstr(sb, data->rest);

179 } else

180 die("unknown format element: %.*s", len, atom);

181 }

@@ -273,7 +286,23 @@ static int batch_objects(struct batch_options *opt)

286 warn_on_object_refname_ambiguity = 0;

287

288 while (strbuf_getline(&buf, stdin, '\n') != EOF) {

276 - int error = batch_one_object(buf.buf, opt, &data);

289 + int error;

290 +

291 + if (data.split_on_whitespace) {

292 + /*

293 + * Split at first whitespace, tying off the beginning

294 + * of the string and saving the remainder (or NULL) in

295 + * data.rest.

296 + */

297 + char *p = strpbrk(buf.buf, " \t");

298 + if (p) {

299 + while (*p && strchr(" \t", *p))

300 + *p++ = '\0';

301 + }

302 + data.rest = p;

303 + }

304 +

305 + error = batch_one_object(buf.buf, opt, &data);

306 if (error)

307 return error;

308 }

t/t1006-cat-file.sh

+15

@@ -78,6 +78,13 @@ $content"

78 echo $sha1 | git cat-file --batch-check="%(objecttype) %(objectname)" >actual &&

79 test_cmp expect actual

80 '

81 +

82 + test_expect_success '--batch-check with %(rest)' '

83 + echo "$type this is some extra content" >expect &&

84 + echo "$sha1 this is some extra content" |

85 + git cat-file --batch-check="%(objecttype) %(rest)" >actual &&

86 + test_cmp expect actual

87 + '

88 }

90 hello_content="Hello World"

@@ -91,6 +98,14 @@ test_expect_success "setup" '

99 run_tests 'blob' $hello_sha1 $hello_size "$hello_content" "$hello_content"

100

101 +test_expect_success '--batch-check without %(rest) considers whole line' '

102 + echo "$hello_sha1 blob $hello_size" >expect &&

103 + git update-index --add --cacheinfo 100644 $hello_sha1 "white space" &&

104 + test_when_finished "git update-index --remove \"white space\"" &&

105 + echo ":white space" | git cat-file --batch-check >actual &&

106 + test_cmp expect actual

107 +'

108 +

109 tree_sha1=$(git write-tree)

110 tree_size=33

111 tree_pretty_content="100644 blob $hello_sha1 hello"