convert.c: simplify text_stat
Simplify the statistics: lonecr counts the CR which is not followed by a LF, lonelf counts the LF which is not preceded by a CR, crlf counts CRLF combinations. This simplifies the evaluation of the statistics. Signed-off-by: Torsten Bögershausen <tboegi@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Torsten Bögershausen committed
Feb 10, 2016 at 17:24 UTC
6e336a530b0dd98d1c751440ca5666a6999ab020
1 file changed
+22
-25
convert.c
+22
-25
@@ -31,7 +31,7 @@ enum crlf_action {
31
32
struct text_stat {
33
/* NUL, CR, LF and CRLF counts */
34
- unsigned nul, cr, lf, crlf;
34
+ unsigned nul, lonecr, lonelf, crlf;
35
36
/* These are just approximations! */
37
unsigned printable, nonprintable;
@@ -46,13 +46,15 @@ static void gather_stats(const char *buf, unsigned long size, struct text_stat *
46
for (i = 0; i < size; i++) {
47
unsigned char c = buf[i];
48
if (c == '\r') {
49
- stats->cr++;
50
- if (i+1 < size && buf[i+1] == '\n')
49
+ if (i+1 < size && buf[i+1] == '\n') {
50
stats->crlf++;
51
+ i++;
52
+ } else
53
+ stats->lonecr++;
54
continue;
55
}
56
if (c == '\n') {
55
- stats->lf++;
57
+ stats->lonelf++;
58
continue;
59
}
60
if (c == 127)
@@ -86,7 +88,7 @@ static void gather_stats(const char *buf, unsigned long size, struct text_stat *
88
*/
89
static int convert_is_binary(unsigned long size, const struct text_stat *stats)
90
{
89
- if (stats->cr != stats->crlf)
91
+ if (stats->lonecr)
92
return 1;
93
if (stats->nul)
94
return 1;
@@ -98,19 +100,18 @@ static int convert_is_binary(unsigned long size, const struct text_stat *stats)
100
static unsigned int gather_convert_stats(const char *data, unsigned long size)
101
{
102
struct text_stat stats;
103
+ int ret = 0;
104
if (!data || !size)
105
return 0;
106
gather_stats(data, size, &stats);
107
if (convert_is_binary(size, &stats))
105
- return CONVERT_STAT_BITS_BIN;
106
- else if (stats.crlf && stats.crlf == stats.lf)
107
- return CONVERT_STAT_BITS_TXT_CRLF;
108
- else if (stats.crlf && stats.lf)
109
- return CONVERT_STAT_BITS_TXT_CRLF | CONVERT_STAT_BITS_TXT_LF;
110
- else if (stats.lf)
111
- return CONVERT_STAT_BITS_TXT_LF;
112
- else
113
- return 0;
108
+ ret |= CONVERT_STAT_BITS_BIN;
109
+ if (stats.crlf)
110
+ ret |= CONVERT_STAT_BITS_TXT_CRLF;
111
+ if (stats.lonelf)
112
+ ret |= CONVERT_STAT_BITS_TXT_LF;
113
+
114
+ return ret;
115
}
116
117
static const char *gather_convert_stats_ascii(const char *data, unsigned long size)
@@ -207,7 +208,7 @@ static void check_safe_crlf(const char *path, enum crlf_action crlf_action,
208
* CRLFs would be added by checkout:
209
* check if we have "naked" LFs
210
*/
210
- if (stats->lf != stats->crlf) {
211
+ if (stats->lonelf) {
212
if (checksafe == SAFE_CRLF_WARN)
213
warning("LF will be replaced by CRLF in %s.\nThe file will have its original line endings in your working directory.", path);
214
else /* i.e. SAFE_CRLF_FAIL */
@@ -266,8 +267,8 @@ static int crlf_to_git(const char *path, const char *src, size_t len,
267
268
check_safe_crlf(path, crlf_action, &stats, checksafe);
269
269
- /* Optimization: No CR? Nothing to convert, regardless. */
270
- if (!stats.cr)
270
+ /* Optimization: No CRLF? Nothing to convert, regardless. */
271
+ if (!stats.crlf)
272
return 0;
273
274
/*
@@ -314,19 +315,15 @@ static int crlf_to_worktree(const char *path, const char *src, size_t len,
315
316
gather_stats(src, len, &stats);
317
317
- /* No LF? Nothing to convert, regardless. */
318
- if (!stats.lf)
319
- return 0;
320
-
321
- /* Was it already in CRLF format? */
322
- if (stats.lf == stats.crlf)
318
+ /* No "naked" LF? Nothing to convert, regardless. */
319
+ if (!stats.lonelf)
320
return 0;
321
322
if (crlf_action == CRLF_AUTO || crlf_action == CRLF_AUTO_INPUT || crlf_action == CRLF_AUTO_CRLF) {
323
if (crlf_action == CRLF_AUTO_INPUT || crlf_action == CRLF_AUTO_CRLF) {
324
/* If we have any CR or CRLF line endings, we do not touch it */
325
/* This is the new safer autocrlf-handling */
329
- if (stats.cr > 0 || stats.crlf > 0)
326
+ if (stats.lonecr || stats.crlf )
327
return 0;
328
}
329
@@ -338,7 +335,7 @@ static int crlf_to_worktree(const char *path, const char *src, size_t len,
335
if (src == buf->buf)
336
to_free = strbuf_detach(buf, NULL);
337
341
- strbuf_grow(buf, len + stats.lf - stats.crlf);
338
+ strbuf_grow(buf, len + stats.lonelf);
339
for (;;) {
340
const char *nl = memchr(src, '\n', len);
341
if (!nl)