color-words: take an optional regular expression describing words
authorJohannes Schindelin <johannes.schindelin@gmx.de>
Sat, 17 Jan 2009 16:29:45 +0000 (17:29 +0100)
committerJunio C Hamano <gitster@pobox.com>
Sat, 17 Jan 2009 18:43:08 +0000 (10:43 -0800)
In some applications, words are not delimited by white space.  To
allow for that, you can specify a regular expression describing
what makes a word with

git diff --color-words='[A-Za-z0-9]+'

Note that words cannot contain newline characters.

As suggested by Thomas Rast, the words are the exact matches of the
regular expression.

Note that a regular expression beginning with a '^' will match only
a word at the beginning of the hunk, not a word at the beginning of
a line, and is probably not what you want.

This commit contains a quoting fix by Thomas Rast.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Documentation/diff-options.txt
diff.c
diff.h
t/t4034-diff-words.sh

index 43793d75005a7af875a055d377a8ab6a13510682..2c1fa4b102ecb8b274450c71171a478782488a95 100644 (file)
@@ -91,8 +91,12 @@ endif::git-format-patch[]
        Turn off colored diff, even when the configuration file
        gives the default to color output.
 
---color-words::
+--color-words[=regex]::
        Show colored word diff, i.e. color words which have changed.
++
+Optionally, you can pass a regular expression that tells Git what the
+words are that you are looking for; The default is to interpret any
+stretch of non-whitespace as a word.
 
 --no-renames::
        Turn off rename detection, even when the configuration
diff --git a/diff.c b/diff.c
index 37c886a81588c556c105665d9926f25da377d3e5..9fb3d0df31895e81a02d6bb67b9c4e0c13640155 100644 (file)
--- a/diff.c
+++ b/diff.c
@@ -333,12 +333,14 @@ static void diff_words_append(char *line, unsigned long len,
        len--;
        memcpy(buffer->text.ptr + buffer->text.size, line, len);
        buffer->text.size += len;
+       buffer->text.ptr[buffer->text.size] = '\0';
 }
 
 struct diff_words_data {
        struct diff_words_buffer minus, plus;
        const char *current_plus;
        FILE *file;
+       regex_t *word_regex;
 };
 
 static void fn_out_diff_words_aux(void *priv, char *line, unsigned long len)
@@ -382,17 +384,49 @@ static void fn_out_diff_words_aux(void *priv, char *line, unsigned long len)
        diff_words->current_plus = plus_end;
 }
 
+/* This function starts looking at *begin, and returns 0 iff a word was found. */
+static int find_word_boundaries(mmfile_t *buffer, regex_t *word_regex,
+               int *begin, int *end)
+{
+       if (word_regex && *begin < buffer->size) {
+               regmatch_t match[1];
+               if (!regexec(word_regex, buffer->ptr + *begin, 1, match, 0)) {
+                       char *p = memchr(buffer->ptr + *begin + match[0].rm_so,
+                                       '\n', match[0].rm_eo - match[0].rm_so);
+                       *end = p ? p - buffer->ptr : match[0].rm_eo + *begin;
+                       *begin += match[0].rm_so;
+                       return *begin >= *end;
+               }
+               return -1;
+       }
+
+       /* find the next word */
+       while (*begin < buffer->size && isspace(buffer->ptr[*begin]))
+               (*begin)++;
+       if (*begin >= buffer->size)
+               return -1;
+
+       /* find the end of the word */
+       *end = *begin + 1;
+       while (*end < buffer->size && !isspace(buffer->ptr[*end]))
+               (*end)++;
+
+       return 0;
+}
+
 /*
  * This function splits the words in buffer->text, stores the list with
  * newline separator into out, and saves the offsets of the original words
  * in buffer->orig.
  */
-static void diff_words_fill(struct diff_words_buffer *buffer, mmfile_t *out)
+static void diff_words_fill(struct diff_words_buffer *buffer, mmfile_t *out,
+               regex_t *word_regex)
 {
        int i, j;
+       long alloc = 0;
 
        out->size = 0;
-       out->ptr = xmalloc(buffer->text.size);
+       out->ptr = NULL;
 
        /* fake an empty "0th" word */
        ALLOC_GROW(buffer->orig, 1, buffer->orig_alloc);
@@ -400,11 +434,8 @@ static void diff_words_fill(struct diff_words_buffer *buffer, mmfile_t *out)
        buffer->orig_nr = 1;
 
        for (i = 0; i < buffer->text.size; i++) {
-               if (isspace(buffer->text.ptr[i]))
-                       continue;
-               for (j = i + 1; j < buffer->text.size &&
-                               !isspace(buffer->text.ptr[j]); j++)
-                       ; /* find the end of the word */
+               if (find_word_boundaries(&buffer->text, word_regex, &i, &j))
+                       return;
 
                /* store original boundaries */
                ALLOC_GROW(buffer->orig, buffer->orig_nr + 1,
@@ -414,6 +445,7 @@ static void diff_words_fill(struct diff_words_buffer *buffer, mmfile_t *out)
                buffer->orig_nr++;
 
                /* store one word */
+               ALLOC_GROW(out->ptr, out->size + j - i + 1, alloc);
                memcpy(out->ptr + out->size, buffer->text.ptr + i, j - i);
                out->ptr[out->size + j - i] = '\n';
                out->size += j - i + 1;
@@ -443,9 +475,10 @@ static void diff_words_show(struct diff_words_data *diff_words)
 
        memset(&xpp, 0, sizeof(xpp));
        memset(&xecfg, 0, sizeof(xecfg));
-       diff_words_fill(&diff_words->minus, &minus);
-       diff_words_fill(&diff_words->plus, &plus);
+       diff_words_fill(&diff_words->minus, &minus, diff_words->word_regex);
+       diff_words_fill(&diff_words->plus, &plus, diff_words->word_regex);
        xpp.flags = XDF_NEED_MINIMAL;
+       /* as only the hunk header will be parsed, we need a 0-context */
        xecfg.ctxlen = 0;
        xdi_diff_outf(&minus, &plus, fn_out_diff_words_aux, diff_words,
                      &xpp, &xecfg, &ecb);
@@ -484,6 +517,7 @@ static void free_diff_words_data(struct emit_callback *ecbdata)
                free (ecbdata->diff_words->minus.orig);
                free (ecbdata->diff_words->plus.text.ptr);
                free (ecbdata->diff_words->plus.orig);
+               free(ecbdata->diff_words->word_regex);
                free(ecbdata->diff_words);
                ecbdata->diff_words = NULL;
        }
@@ -1506,6 +1540,14 @@ static void builtin_diff(const char *name_a,
                        ecbdata.diff_words =
                                xcalloc(1, sizeof(struct diff_words_data));
                        ecbdata.diff_words->file = o->file;
+                       if (o->word_regex) {
+                               ecbdata.diff_words->word_regex = (regex_t *)
+                                       xmalloc(sizeof(regex_t));
+                               if (regcomp(ecbdata.diff_words->word_regex,
+                                               o->word_regex, REG_EXTENDED))
+                                       die ("Invalid regular expression: %s",
+                                                       o->word_regex);
+                       }
                }
                xdi_diff_outf(&mf1, &mf2, fn_out_consume, &ecbdata,
                              &xpp, &xecfg, &ecb);
@@ -2517,6 +2559,10 @@ int diff_opt_parse(struct diff_options *options, const char **av, int ac)
                DIFF_OPT_CLR(options, COLOR_DIFF);
        else if (!strcmp(arg, "--color-words"))
                options->flags |= DIFF_OPT_COLOR_DIFF | DIFF_OPT_COLOR_DIFF_WORDS;
+       else if (!prefixcmp(arg, "--color-words=")) {
+               options->flags |= DIFF_OPT_COLOR_DIFF | DIFF_OPT_COLOR_DIFF_WORDS;
+               options->word_regex = arg + 14;
+       }
        else if (!strcmp(arg, "--exit-code"))
                DIFF_OPT_SET(options, EXIT_WITH_STATUS);
        else if (!strcmp(arg, "--quiet"))
diff --git a/diff.h b/diff.h
index 4d5a32781da81295d5aa1b4dd33dd2765be3ff89..23cd90c2e64cf8be44999be812a0765cbe36c9f8 100644 (file)
--- a/diff.h
+++ b/diff.h
@@ -98,6 +98,7 @@ struct diff_options {
 
        int stat_width;
        int stat_name_width;
+       const char *word_regex;
 
        /* this is set by diffcore for DIFF_FORMAT_PATCH */
        int found_changes;
index b22195f8bbbe369349b24ba1a61734087b476b47..48734863013734002deac08b20b1442322cbfbd3 100755 (executable)
@@ -63,4 +63,61 @@ test_expect_success 'word diff with runs of whitespace' '
 
 '
 
+cat > expect <<\EOF
+<WHITE>diff --git a/pre b/post<RESET>
+<WHITE>index 330b04f..5ed8eff 100644<RESET>
+<WHITE>--- a/pre<RESET>
+<WHITE>+++ b/post<RESET>
+<BROWN>@@ -1,3 +1,7 @@<RESET>
+h(4),<GREEN>hh<RESET>[44]
+<RESET>
+a = b + c<RESET>
+
+<GREEN>aa = a<RESET>
+
+<GREEN>aeff = aeff * ( aaa<RESET> )
+EOF
+
+test_expect_success 'word diff with a regular expression' '
+
+       word_diff --color-words="[a-z]+"
+
+'
+
+echo 'aaa (aaa)' > pre
+echo 'aaa (aaa) aaa' > post
+
+cat > expect <<\EOF
+<WHITE>diff --git a/pre b/post<RESET>
+<WHITE>index c29453b..be22f37 100644<RESET>
+<WHITE>--- a/pre<RESET>
+<WHITE>+++ b/post<RESET>
+<BROWN>@@ -1 +1 @@<RESET>
+aaa (aaa) <GREEN>aaa<RESET>
+EOF
+
+test_expect_success 'test parsing words for newline' '
+
+       word_diff --color-words="a+"
+
+'
+
+echo '(:' > pre
+echo '(' > post
+
+cat > expect <<\EOF
+<WHITE>diff --git a/pre b/post<RESET>
+<WHITE>index 289cb9d..2d06f37 100644<RESET>
+<WHITE>--- a/pre<RESET>
+<WHITE>+++ b/post<RESET>
+<BROWN>@@ -1 +1 @@<RESET>
+(<RED>:<RESET>
+EOF
+
+test_expect_success 'test when words are only removed at the end' '
+
+       word_diff --color-words=.
+
+'
+
 test_done