[PATCH v4 00/12] Wildmatch v4

STALE3716d

17 messages, 3 authors, 2016-06-15 · open the first message on its own page

[PATCH v4 00/12] Wildmatch v4

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Really small updates. I did not want to resend it this soon but this
may fix the compile errors for Junio. Changes are

 - cleanup in wildmatch.c so #include "cache.h" is at the top of the
   file.

 - wildmatch() returns no match if it encounters "**" without
   surrounding slashes. It returns a special code so we can actually
   show a warning at higher level. I don't want to do that now
   because I want to mark the pattern "broken" in attr.c/dir.c so
   the pattern will never be used again, and the message shown only
   once. That needs nd/attr-match-optim-more, but I don't want to tie
   this series to that just yet.

Nguyễn Thái Ngọc Duy (12):
  ctype: make sane_ctype[] const array
  ctype: support iscntrl, ispunct, isxdigit and isprint
  Import wildmatch from rsync
  wildmatch: remove unnecessary functions
  Integrate wildmatch to git
  wildmatch: make wildmatch's return value compatible with fnmatch
  wildmatch: remove static variable force_lower_case
  wildmatch: fix case-insensitive matching
  wildmatch: adjust "**" behavior
  wildmatch: make /**/ match zero or more directories
  Support "**" wildcard in .gitignore and .gitattributes
  t3070: disable two fnmatch tests that have different results on
    different libc

 .gitignore                         |   1 +
 Documentation/gitattributes.txt    |   6 +
 Documentation/gitignore.txt        |  19 +++
 Makefile                           |   3 +
 attr.c                             |   4 +-
 ctype.c                            |  20 +++-
 dir.c                              |   4 +-
 git-compat-util.h                  |  15 ++-
 t/t0003-attributes.sh              |  38 ++++++
 t/t3001-ls-files-others-exclude.sh |  19 +++
 t/t3070-wildmatch.sh               | 187 +++++++++++++++++++++++++++++
 test-wildmatch.c                   |  14 +++
 wildmatch.c                        | 240 +++++++++++++++++++++++++++++++++++++
 wildmatch.h                        |   9 ++
 14 files changed, 575 insertions(+), 4 deletions(-)
 create mode 100755 t/t3070-wildmatch.sh
 create mode 100644 test-wildmatch.c
 create mode 100644 wildmatch.c
 create mode 100644 wildmatch.h

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 01/12] ctype: make sane_ctype[] const array

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 ctype.c           | 2 +-
 git-compat-util.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/ctype.c b/ctype.c
index 9353271..faeaf34 100644
--- a/ctype.c
+++ b/ctype.c

@@ -14,7 +14,7 @@ enum {
 	P = GIT_PATHSPEC_MAGIC  /* other non-alnum, except for ] and } */
 };
 
-unsigned char sane_ctype[256] = {
+const unsigned char sane_ctype[256] = {
 	0, 0, 0, 0, 0, 0, 0, 0, 0, S, S, 0, 0, S, 0, 0,		/*   0.. 15 */
 	0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,		/*  16.. 31 */
 	S, P, P, P, R, P, P, P, R, R, G, R, P, P, R, P,		/*  32.. 47 */

diff --git a/git-compat-util.h b/git-compat-util.h
index 2fbf1fd..f8b859c 100644
--- a/git-compat-util.h
+++ b/git-compat-util.h

@@ -510,7 +510,7 @@ extern const char tolower_trans_tbl[256];
 #undef isupper
 #undef tolower
 #undef toupper
-extern unsigned char sane_ctype[256];
+extern const unsigned char sane_ctype[256];
 #define GIT_SPACE 0x01
 #define GIT_DIGIT 0x02
 #define GIT_ALPHA 0x04

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 02/12] ctype: support iscntrl, ispunct, isxdigit and isprint

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

A new table sane_ctype2[] is added, otherwise we'll need to OR with
current bits in sane_ctype[] and that looks ugly.

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 ctype.c           | 18 ++++++++++++++++++
 git-compat-util.h | 13 +++++++++++++
 2 files changed, 31 insertions(+)

diff --git a/ctype.c b/ctype.c
index faeaf34..b4bf48a 100644
--- a/ctype.c
+++ b/ctype.c

@@ -26,6 +26,24 @@ const unsigned char sane_ctype[256] = {
 	/* Nothing in the 128.. range */
 };
 
+enum {
+	CN = GIT_CNTRL,
+	PU = GIT_PUNCT,
+	XD = GIT_XDIGIT,
+};
+
+const unsigned char sane_ctype2[256] = {
+	CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, /*    0..15 */
+	CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, CN, /*   16..31 */
+	0,  PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, PU, /*   32..47 */
+	XD, XD, XD, XD, XD, XD, XD, XD, XD, XD, PU, PU, PU, PU, PU, PU, /*   48..63 */
+	PU, 0,	XD, 0,	XD, 0,	XD, 0,	0,  0,	0,  0,	0,  0,	0,  0,	/*   64..79 */
+	0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  PU, PU, PU, PU, PU, /*   80..95 */
+	PU, 0,	XD, 0,	XD, 0,	XD, 0,	0,  0,	0,  0,	0,  0,	0,  0,	/*  96..111 */
+	0,  0,	0,  0,	0,  0,	0,  0,	0,  0,	0,  PU, PU, PU, PU, CN, /* 112..127 */
+	/* Nothing in the 128.. range */
+};
+
 /* For case-insensitive kwset */
 const char tolower_trans_tbl[256] = {
 	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,

diff --git a/git-compat-util.h b/git-compat-util.h
index f8b859c..ea11694 100644
--- a/git-compat-util.h
+++ b/git-compat-util.h

@@ -510,14 +510,23 @@ extern const char tolower_trans_tbl[256];
 #undef isupper
 #undef tolower
 #undef toupper
+#undef iscntrl
+#undef ispunct
+#undef isxdigit
+#undef isprint
 extern const unsigned char sane_ctype[256];
+extern const unsigned char sane_ctype2[256];
 #define GIT_SPACE 0x01
 #define GIT_DIGIT 0x02
 #define GIT_ALPHA 0x04
 #define GIT_GLOB_SPECIAL 0x08
 #define GIT_REGEX_SPECIAL 0x10
 #define GIT_PATHSPEC_MAGIC 0x20
+#define GIT_CNTRL 0x01
+#define GIT_PUNCT 0x02
+#define GIT_XDIGIT 0x04
 #define sane_istest(x,mask) ((sane_ctype[(unsigned char)(x)] & (mask)) != 0)
+#define sane_istest2(x,mask) ((sane_ctype2[(unsigned char)(x)] & (mask)) != 0)
 #define isascii(x) (((x) & ~0x7f) == 0)
 #define isspace(x) sane_istest(x,GIT_SPACE)
 #define isdigit(x) sane_istest(x,GIT_DIGIT)

@@ -527,6 +536,10 @@ extern const unsigned char sane_ctype[256];
 #define isupper(x) sane_iscase(x, 0)
 #define is_glob_special(x) sane_istest(x,GIT_GLOB_SPECIAL)
 #define is_regex_special(x) sane_istest(x,GIT_GLOB_SPECIAL | GIT_REGEX_SPECIAL)
+#define iscntrl(x) sane_istest2(x, GIT_CNTRL)
+#define ispunct(x) sane_istest2(x, GIT_PUNCT)
+#define isxdigit(x) sane_istest2(x, GIT_XDIGIT)
+#define isprint(x) (isalnum(x) || isspace(x) || ispunct(x))
 #define tolower(x) sane_case((unsigned char)(x), 0x20)
 #define toupper(x) sane_case((unsigned char)(x), 0)
 #define is_pathspec_magic(x) sane_istest(x,GIT_PATHSPEC_MAGIC)

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 03/12] Import wildmatch from rsync

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

These files are from rsync.git commit
f92f5b166e3019db42bc7fe1aa2f1a9178cd215d, which was the last commit
before rsync turned GPL-3. All files are imported as-is and
no-op. Adaptation is done in a separate patch.

rsync.git           ->  git.git
lib/wildmatch.[ch]      wildmatch.[ch]
wildtest.txt            t/t3070/wildtest.txt

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 t/t3070/wildtest.txt | 165 +++++++++++++++++++++++
 wildmatch.c          | 368 +++++++++++++++++++++++++++++++++++++++++++++++++++
 wildmatch.h          |   6 +
 3 files changed, 539 insertions(+)
 create mode 100644 t/t3070/wildtest.txt
 create mode 100644 wildmatch.c
 create mode 100644 wildmatch.h

diff --git a/t/t3070/wildtest.txt b/t/t3070/wildtest.txt
new file mode 100644
index 0000000..42c1678
--- /dev/null
+++ b/t/t3070/wildtest.txt

@@ -0,0 +1,165 @@
+# Input is in the following format (all items white-space separated):
+#
+# The first two items are 1 or 0 indicating if the wildmat call is expected to
+# succeed and if fnmatch works the same way as wildmat, respectively.  After
+# that is a text string for the match, and a pattern string.  Strings can be
+# quoted (if desired) in either double or single quotes, as well as backticks.
+#
+# MATCH FNMATCH_SAME "text to match" 'pattern to use'
+
+# Basic wildmat features
+1 1 foo			foo
+0 1 foo			bar
+1 1 ''			""
+1 1 foo			???
+0 1 foo			??
+1 1 foo			*
+1 1 foo			f*
+0 1 foo			*f
+1 1 foo			*foo*
+1 1 foobar		*ob*a*r*
+1 1 aaaaaaabababab	*ab
+1 1 foo*		foo\*
+0 1 foobar		foo\*bar
+1 1 f\oo		f\\oo
+1 1 ball		*[al]?
+0 1 ten			[ten]
+1 1 ten			**[!te]
+0 1 ten			**[!ten]
+1 1 ten			t[a-g]n
+0 1 ten			t[!a-g]n
+1 1 ton			t[!a-g]n
+1 1 ton			t[^a-g]n
+1 1 a]b			a[]]b
+1 1 a-b			a[]-]b
+1 1 a]b			a[]-]b
+0 1 aab			a[]-]b
+1 1 aab			a[]a-]b
+1 1 ]			]
+
+# Extended slash-matching features
+0 1 foo/baz/bar		foo*bar
+1 1 foo/baz/bar		foo**bar
+0 1 foo/bar		foo?bar
+0 1 foo/bar		foo[/]bar
+0 1 foo/bar		f[^eiu][^eiu][^eiu][^eiu][^eiu]r
+1 1 foo-bar		f[^eiu][^eiu][^eiu][^eiu][^eiu]r
+0 1 foo			**/foo
+1 1 /foo		**/foo
+1 1 bar/baz/foo		**/foo
+0 1 bar/baz/foo		*/foo
+0 0 foo/bar/baz		**/bar*
+1 1 deep/foo/bar/baz	**/bar/*
+0 1 deep/foo/bar/baz/	**/bar/*
+1 1 deep/foo/bar/baz/	**/bar/**
+0 1 deep/foo/bar	**/bar/*
+1 1 deep/foo/bar/	**/bar/**
+1 1 foo/bar/baz		**/bar**
+1 1 foo/bar/baz/x	*/bar/**
+0 0 deep/foo/bar/baz/x	*/bar/**
+1 1 deep/foo/bar/baz/x	**/bar/*/*
+
+# Various additional tests
+0 1 acrt		a[c-c]st
+1 1 acrt		a[c-c]rt
+0 1 ]			[!]-]
+1 1 a			[!]-]
+0 1 ''			\
+0 1 \			\
+0 1 /\			*/\
+1 1 /\			*/\\
+1 1 foo			foo
+1 1 @foo		@foo
+0 1 foo			@foo
+1 1 [ab]		\[ab]
+1 1 [ab]		[[]ab]
+1 1 [ab]		[[:]ab]
+0 1 [ab]		[[::]ab]
+1 1 [ab]		[[:digit]ab]
+1 1 [ab]		[\[:]ab]
+1 1 ?a?b		\??\?b
+1 1 abc			\a\b\c
+0 1 foo			''
+1 1 foo/bar/baz/to	**/t[o]
+
+# Character class tests
+1 1 a1B		[[:alpha:]][[:digit:]][[:upper:]]
+0 1 a		[[:digit:][:upper:][:space:]]
+1 1 A		[[:digit:][:upper:][:space:]]
+1 1 1		[[:digit:][:upper:][:space:]]
+0 1 1		[[:digit:][:upper:][:spaci:]]
+1 1 ' '		[[:digit:][:upper:][:space:]]
+0 1 .		[[:digit:][:upper:][:space:]]
+1 1 .		[[:digit:][:punct:][:space:]]
+1 1 5		[[:xdigit:]]
+1 1 f		[[:xdigit:]]
+1 1 D		[[:xdigit:]]
+1 1 _		[[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
+#1 1 �		[^[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
+1 1 		[^[:alnum:][:alpha:][:blank:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
+1 1 .		[^[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:lower:][:space:][:upper:][:xdigit:]]
+1 1 5		[a-c[:digit:]x-z]
+1 1 b		[a-c[:digit:]x-z]
+1 1 y		[a-c[:digit:]x-z]
+0 1 q		[a-c[:digit:]x-z]
+
+# Additional tests, including some malformed wildmats
+1 1 ]		[\\-^]
+0 1 [		[\\-^]
+1 1 -		[\-_]
+1 1 ]		[\]]
+0 1 \]		[\]]
+0 1 \		[\]]
+0 1 ab		a[]b
+0 1 a[]b	a[]b
+0 1 ab[		ab[
+0 1 ab		[!
+0 1 ab		[-
+1 1 -		[-]
+0 1 -		[a-
+0 1 -		[!a-
+1 1 -		[--A]
+1 1 5		[--A]
+1 1 ' '		'[ --]'
+1 1 $		'[ --]'
+1 1 -		'[ --]'
+0 1 0		'[ --]'
+1 1 -		[---]
+1 1 -		[------]
+0 1 j		[a-e-n]
+1 1 -		[a-e-n]
+1 1 a		[!------]
+0 1 [		[]-a]
+1 1 ^		[]-a]
+0 1 ^		[!]-a]
+1 1 [		[!]-a]
+1 1 ^		[a^bc]
+1 1 -b]		[a-]b]
+0 1 \		[\]
+1 1 \		[\\]
+0 1 \		[!\\]
+1 1 G		[A-\\]
+0 1 aaabbb	b*a
+0 1 aabcaa	*ba*
+1 1 ,		[,]
+1 1 ,		[\\,]
+1 1 \		[\\,]
+1 1 -		[,-.]
+0 1 +		[,-.]
+0 1 -.]		[,-.]
+1 1 2		[\1-\3]
+1 1 3		[\1-\3]
+0 1 4		[\1-\3]
+1 1 \		[[-\]]
+1 1 [		[[-\]]
+1 1 ]		[[-\]]
+0 1 -		[[-\]]
+
+# Test recursion and the abort code (use "wildtest -i" to see iteration counts)
+1 1 -adobe-courier-bold-o-normal--12-120-75-75-m-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
+0 1 -adobe-courier-bold-o-normal--12-120-75-75-X-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
+0 1 -adobe-courier-bold-o-normal--12-120-75-75-/-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
+1 1 /adobe/courier/bold/o/normal//12/120/75/75/m/70/iso8859/1	/*/*/*/*/*/*/12/*/*/*/m/*/*/*
+0 1 /adobe/courier/bold/o/normal//12/120/75/75/X/70/iso8859/1	/*/*/*/*/*/*/12/*/*/*/m/*/*/*
+1 1 abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txt		**/*a*b*g*n*t
+0 1 abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txtz		**/*a*b*g*n*t

diff --git a/wildmatch.c b/wildmatch.c
new file mode 100644
index 0000000..f3a1731
--- /dev/null
+++ b/wildmatch.c

@@ -0,0 +1,368 @@
+/*
+**  Do shell-style pattern matching for ?, \, [], and * characters.
+**  It is 8bit clean.
+**
+**  Written by Rich $alz, mirror!rs, Wed Nov 26 19:03:17 EST 1986.
+**  Rich $alz is now <rsalz@bbn.com>.
+**
+**  Modified by Wayne Davison to special-case '/' matching, to make '**'
+**  work differently than '*', and to fix the character-class code.
+*/
+
+#include "rsync.h"
+
+/* What character marks an inverted character class? */
+#define NEGATE_CLASS	'!'
+#define NEGATE_CLASS2	'^'
+
+#define FALSE 0
+#define TRUE 1
+#define ABORT_ALL -1
+#define ABORT_TO_STARSTAR -2
+
+#define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \
+				    && *(class) == *(litmatch) \
+				    && strncmp((char*)class, litmatch, len) == 0)
+
+#if defined STDC_HEADERS || !defined isascii
+# define ISASCII(c) 1
+#else
+# define ISASCII(c) isascii(c)
+#endif
+
+#ifdef isblank
+# define ISBLANK(c) (ISASCII(c) && isblank(c))
+#else
+# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
+#endif
+
+#ifdef isgraph
+# define ISGRAPH(c) (ISASCII(c) && isgraph(c))
+#else
+# define ISGRAPH(c) (ISASCII(c) && isprint(c) && !isspace(c))
+#endif
+
+#define ISPRINT(c) (ISASCII(c) && isprint(c))
+#define ISDIGIT(c) (ISASCII(c) && isdigit(c))
+#define ISALNUM(c) (ISASCII(c) && isalnum(c))
+#define ISALPHA(c) (ISASCII(c) && isalpha(c))
+#define ISCNTRL(c) (ISASCII(c) && iscntrl(c))
+#define ISLOWER(c) (ISASCII(c) && islower(c))
+#define ISPUNCT(c) (ISASCII(c) && ispunct(c))
+#define ISSPACE(c) (ISASCII(c) && isspace(c))
+#define ISUPPER(c) (ISASCII(c) && isupper(c))
+#define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
+
+#ifdef WILD_TEST_ITERATIONS
+int wildmatch_iteration_count;
+#endif
+
+static int force_lower_case = 0;
+
+/* Match pattern "p" against the a virtually-joined string consisting
+ * of "text" and any strings in array "a". */
+static int dowild(const uchar *p, const uchar *text, const uchar*const *a)
+{
+    uchar p_ch;
+
+#ifdef WILD_TEST_ITERATIONS
+    wildmatch_iteration_count++;
+#endif
+
+    for ( ; (p_ch = *p) != '\0'; text++, p++) {
+	int matched, special;
+	uchar t_ch, prev_ch;
+	while ((t_ch = *text) == '\0') {
+	    if (*a == NULL) {
+		if (p_ch != '*')
+		    return ABORT_ALL;
+		break;
+	    }
+	    text = *a++;
+	}
+	if (force_lower_case && ISUPPER(t_ch))
+	    t_ch = tolower(t_ch);
+	switch (p_ch) {
+	  case '\\':
+	    /* Literal match with following character.  Note that the test
+	     * in "default" handles the p[1] == '\0' failure case. */
+	    p_ch = *++p;
+	    /* FALLTHROUGH */
+	  default:
+	    if (t_ch != p_ch)
+		return FALSE;
+	    continue;
+	  case '?':
+	    /* Match anything but '/'. */
+	    if (t_ch == '/')
+		return FALSE;
+	    continue;
+	  case '*':
+	    if (*++p == '*') {
+		while (*++p == '*') {}
+		special = TRUE;
+	    } else
+		special = FALSE;
+	    if (*p == '\0') {
+		/* Trailing "**" matches everything.  Trailing "*" matches
+		 * only if there are no more slash characters. */
+		if (!special) {
+		    do {
+			if (strchr((char*)text, '/') != NULL)
+			    return FALSE;
+		    } while ((text = *a++) != NULL);
+		}
+		return TRUE;
+	    }
+	    while (1) {
+		if (t_ch == '\0') {
+		    if ((text = *a++) == NULL)
+			break;
+		    t_ch = *text;
+		    continue;
+		}
+		if ((matched = dowild(p, text, a)) != FALSE) {
+		    if (!special || matched != ABORT_TO_STARSTAR)
+			return matched;
+		} else if (!special && t_ch == '/')
+		    return ABORT_TO_STARSTAR;
+		t_ch = *++text;
+	    }
+	    return ABORT_ALL;
+	  case '[':
+	    p_ch = *++p;
+#ifdef NEGATE_CLASS2
+	    if (p_ch == NEGATE_CLASS2)
+		p_ch = NEGATE_CLASS;
+#endif
+	    /* Assign literal TRUE/FALSE because of "matched" comparison. */
+	    special = p_ch == NEGATE_CLASS? TRUE : FALSE;
+	    if (special) {
+		/* Inverted character class. */
+		p_ch = *++p;
+	    }
+	    prev_ch = 0;
+	    matched = FALSE;
+	    do {
+		if (!p_ch)
+		    return ABORT_ALL;
+		if (p_ch == '\\') {
+		    p_ch = *++p;
+		    if (!p_ch)
+			return ABORT_ALL;
+		    if (t_ch == p_ch)
+			matched = TRUE;
+		} else if (p_ch == '-' && prev_ch && p[1] && p[1] != ']') {
+		    p_ch = *++p;
+		    if (p_ch == '\\') {
+			p_ch = *++p;
+			if (!p_ch)
+			    return ABORT_ALL;
+		    }
+		    if (t_ch <= p_ch && t_ch >= prev_ch)
+			matched = TRUE;
+		    p_ch = 0; /* This makes "prev_ch" get set to 0. */
+		} else if (p_ch == '[' && p[1] == ':') {
+		    const uchar *s;
+		    int i;
+		    for (s = p += 2; (p_ch = *p) && p_ch != ']'; p++) {} /*SHARED ITERATOR*/
+		    if (!p_ch)
+			return ABORT_ALL;
+		    i = p - s - 1;
+		    if (i < 0 || p[-1] != ':') {
+			/* Didn't find ":]", so treat like a normal set. */
+			p = s - 2;
+			p_ch = '[';
+			if (t_ch == p_ch)
+			    matched = TRUE;
+			continue;
+		    }
+		    if (CC_EQ(s,i, "alnum")) {
+			if (ISALNUM(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "alpha")) {
+			if (ISALPHA(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "blank")) {
+			if (ISBLANK(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "cntrl")) {
+			if (ISCNTRL(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "digit")) {
+			if (ISDIGIT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "graph")) {
+			if (ISGRAPH(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "lower")) {
+			if (ISLOWER(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "print")) {
+			if (ISPRINT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "punct")) {
+			if (ISPUNCT(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "space")) {
+			if (ISSPACE(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "upper")) {
+			if (ISUPPER(t_ch))
+			    matched = TRUE;
+		    } else if (CC_EQ(s,i, "xdigit")) {
+			if (ISXDIGIT(t_ch))
+			    matched = TRUE;
+		    } else /* malformed [:class:] string */
+			return ABORT_ALL;
+		    p_ch = 0; /* This makes "prev_ch" get set to 0. */
+		} else if (t_ch == p_ch)
+		    matched = TRUE;
+	    } while (prev_ch = p_ch, (p_ch = *++p) != ']');
+	    if (matched == special || t_ch == '/')
+		return FALSE;
+	    continue;
+	}
+    }
+
+    do {
+	if (*text)
+	    return FALSE;
+    } while ((text = *a++) != NULL);
+
+    return TRUE;
+}
+
+/* Match literal string "s" against the a virtually-joined string consisting
+ * of "text" and any strings in array "a". */
+static int doliteral(const uchar *s, const uchar *text, const uchar*const *a)
+{
+    for ( ; *s != '\0'; text++, s++) {
+	while (*text == '\0') {
+	    if ((text = *a++) == NULL)
+		return FALSE;
+	}
+	if (*text != *s)
+	    return FALSE;
+    }
+
+    do {
+	if (*text)
+	    return FALSE;
+    } while ((text = *a++) != NULL);
+
+    return TRUE;
+}
+
+/* Return the last "count" path elements from the concatenated string.
+ * We return a string pointer to the start of the string, and update the
+ * array pointer-pointer to point to any remaining string elements. */
+static const uchar *trailing_N_elements(const uchar*const **a_ptr, int count)
+{
+    const uchar*const *a = *a_ptr;
+    const uchar*const *first_a = a;
+
+    while (*a)
+	    a++;
+
+    while (a != first_a) {
+	const uchar *s = *--a;
+	s += strlen((char*)s);
+	while (--s >= *a) {
+	    if (*s == '/' && !--count) {
+		*a_ptr = a+1;
+		return s+1;
+	    }
+	}
+    }
+
+    if (count == 1) {
+	*a_ptr = a+1;
+	return *a;
+    }
+
+    return NULL;
+}
+
+/* Match the "pattern" against the "text" string. */
+int wildmatch(const char *pattern, const char *text)
+{
+    static const uchar *nomore[1]; /* A NULL pointer. */
+#ifdef WILD_TEST_ITERATIONS
+    wildmatch_iteration_count = 0;
+#endif
+    return dowild((const uchar*)pattern, (const uchar*)text, nomore) == TRUE;
+}
+
+/* Match the "pattern" against the forced-to-lower-case "text" string. */
+int iwildmatch(const char *pattern, const char *text)
+{
+    static const uchar *nomore[1]; /* A NULL pointer. */
+    int ret;
+#ifdef WILD_TEST_ITERATIONS
+    wildmatch_iteration_count = 0;
+#endif
+    force_lower_case = 1;
+    ret = dowild((const uchar*)pattern, (const uchar*)text, nomore) == TRUE;
+    force_lower_case = 0;
+    return ret;
+}
+
+/* Match pattern "p" against the a virtually-joined string consisting
+ * of all the pointers in array "texts" (which has a NULL pointer at the
+ * end).  The int "where" can be 0 (normal matching), > 0 (match only
+ * the trailing N slash-separated filename components of "texts"), or < 0
+ * (match the "pattern" at the start or after any slash in "texts"). */
+int wildmatch_array(const char *pattern, const char*const *texts, int where)
+{
+    const uchar *p = (const uchar*)pattern;
+    const uchar*const *a = (const uchar*const*)texts;
+    const uchar *text;
+    int matched;
+
+#ifdef WILD_TEST_ITERATIONS
+    wildmatch_iteration_count = 0;
+#endif
+
+    if (where > 0)
+	text = trailing_N_elements(&a, where);
+    else
+	text = *a++;
+    if (!text)
+	return FALSE;
+
+    if ((matched = dowild(p, text, a)) != TRUE && where < 0
+     && matched != ABORT_ALL) {
+	while (1) {
+	    if (*text == '\0') {
+		if ((text = (uchar*)*a++) == NULL)
+		    return FALSE;
+		continue;
+	    }
+	    if (*text++ == '/' && (matched = dowild(p, text, a)) != FALSE
+	     && matched != ABORT_TO_STARSTAR)
+		break;
+	}
+    }
+    return matched == TRUE;
+}
+
+/* Match literal string "s" against the a virtually-joined string consisting
+ * of all the pointers in array "texts" (which has a NULL pointer at the
+ * end).  The int "where" can be 0 (normal matching), or > 0 (match
+ * only the trailing N slash-separated filename components of "texts"). */
+int litmatch_array(const char *string, const char*const *texts, int where)
+{
+    const uchar *s = (const uchar*)string;
+    const uchar*const *a = (const uchar* const*)texts;
+    const uchar *text;
+
+    if (where > 0)
+	text = trailing_N_elements(&a, where);
+    else
+	text = *a++;
+    if (!text)
+	return FALSE;
+
+    return doliteral(s, text, a) == TRUE;
+}

diff --git a/wildmatch.h b/wildmatch.h
new file mode 100644
index 0000000..e7f1a35
--- /dev/null
+++ b/wildmatch.h

@@ -0,0 +1,6 @@
+/* wildmatch.h */
+
+int wildmatch(const char *pattern, const char *text);
+int iwildmatch(const char *pattern, const char *text);
+int wildmatch_array(const char *pattern, const char*const *texts, int where);
+int litmatch_array(const char *string, const char*const *texts, int where);

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 04/12] wildmatch: remove unnecessary functions

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 wildmatch.c | 161 ++++--------------------------------------------------------
 wildmatch.h |   2 -
 2 files changed, 9 insertions(+), 154 deletions(-)

diff --git a/wildmatch.c b/wildmatch.c
index f3a1731..71dba76 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -53,33 +53,19 @@
 #define ISUPPER(c) (ISASCII(c) && isupper(c))
 #define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
 
-#ifdef WILD_TEST_ITERATIONS
-int wildmatch_iteration_count;
-#endif
-
 static int force_lower_case = 0;
 
 /* Match pattern "p" against the a virtually-joined string consisting
  * of "text" and any strings in array "a". */
-static int dowild(const uchar *p, const uchar *text, const uchar*const *a)
+static int dowild(const uchar *p, const uchar *text)
 {
     uchar p_ch;
 
-#ifdef WILD_TEST_ITERATIONS
-    wildmatch_iteration_count++;
-#endif
-
     for ( ; (p_ch = *p) != '\0'; text++, p++) {
 	int matched, special;
 	uchar t_ch, prev_ch;
-	while ((t_ch = *text) == '\0') {
-	    if (*a == NULL) {
-		if (p_ch != '*')
-		    return ABORT_ALL;
-		break;
-	    }
-	    text = *a++;
-	}
+	if ((t_ch = *text) == '\0' && p_ch != '*')
+		return ABORT_ALL;
 	if (force_lower_case && ISUPPER(t_ch))
 	    t_ch = tolower(t_ch);
 	switch (p_ch) {

@@ -107,21 +93,15 @@ static int dowild(const uchar *p, const uchar *text, const uchar*const *a)
 		/* Trailing "**" matches everything.  Trailing "*" matches
 		 * only if there are no more slash characters. */
 		if (!special) {
-		    do {
 			if (strchr((char*)text, '/') != NULL)
 			    return FALSE;
-		    } while ((text = *a++) != NULL);
 		}
 		return TRUE;
 	    }
 	    while (1) {
-		if (t_ch == '\0') {
-		    if ((text = *a++) == NULL)
-			break;
-		    t_ch = *text;
-		    continue;
-		}
-		if ((matched = dowild(p, text, a)) != FALSE) {
+		if (t_ch == '\0')
+		    break;
+		if ((matched = dowild(p, text)) != FALSE) {
 		    if (!special || matched != ABORT_TO_STARSTAR)
 			return matched;
 		} else if (!special && t_ch == '/')

@@ -225,144 +205,21 @@ static int dowild(const uchar *p, const uchar *text, const uchar*const *a)
 	}
     }
 
-    do {
-	if (*text)
-	    return FALSE;
-    } while ((text = *a++) != NULL);
-
-    return TRUE;
-}
-
-/* Match literal string "s" against the a virtually-joined string consisting
- * of "text" and any strings in array "a". */
-static int doliteral(const uchar *s, const uchar *text, const uchar*const *a)
-{
-    for ( ; *s != '\0'; text++, s++) {
-	while (*text == '\0') {
-	    if ((text = *a++) == NULL)
-		return FALSE;
-	}
-	if (*text != *s)
-	    return FALSE;
-    }
-
-    do {
-	if (*text)
-	    return FALSE;
-    } while ((text = *a++) != NULL);
-
-    return TRUE;
-}
-
-/* Return the last "count" path elements from the concatenated string.
- * We return a string pointer to the start of the string, and update the
- * array pointer-pointer to point to any remaining string elements. */
-static const uchar *trailing_N_elements(const uchar*const **a_ptr, int count)
-{
-    const uchar*const *a = *a_ptr;
-    const uchar*const *first_a = a;
-
-    while (*a)
-	    a++;
-
-    while (a != first_a) {
-	const uchar *s = *--a;
-	s += strlen((char*)s);
-	while (--s >= *a) {
-	    if (*s == '/' && !--count) {
-		*a_ptr = a+1;
-		return s+1;
-	    }
-	}
-    }
-
-    if (count == 1) {
-	*a_ptr = a+1;
-	return *a;
-    }
-
-    return NULL;
+    return *text ? FALSE : TRUE;
 }
 
 /* Match the "pattern" against the "text" string. */
 int wildmatch(const char *pattern, const char *text)
 {
-    static const uchar *nomore[1]; /* A NULL pointer. */
-#ifdef WILD_TEST_ITERATIONS
-    wildmatch_iteration_count = 0;
-#endif
-    return dowild((const uchar*)pattern, (const uchar*)text, nomore) == TRUE;
+    return dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
 }
 
 /* Match the "pattern" against the forced-to-lower-case "text" string. */
 int iwildmatch(const char *pattern, const char *text)
 {
-    static const uchar *nomore[1]; /* A NULL pointer. */
     int ret;
-#ifdef WILD_TEST_ITERATIONS
-    wildmatch_iteration_count = 0;
-#endif
     force_lower_case = 1;
-    ret = dowild((const uchar*)pattern, (const uchar*)text, nomore) == TRUE;
+    ret = dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
     force_lower_case = 0;
     return ret;
 }
-
-/* Match pattern "p" against the a virtually-joined string consisting
- * of all the pointers in array "texts" (which has a NULL pointer at the
- * end).  The int "where" can be 0 (normal matching), > 0 (match only
- * the trailing N slash-separated filename components of "texts"), or < 0
- * (match the "pattern" at the start or after any slash in "texts"). */
-int wildmatch_array(const char *pattern, const char*const *texts, int where)
-{
-    const uchar *p = (const uchar*)pattern;
-    const uchar*const *a = (const uchar*const*)texts;
-    const uchar *text;
-    int matched;
-
-#ifdef WILD_TEST_ITERATIONS
-    wildmatch_iteration_count = 0;
-#endif
-
-    if (where > 0)
-	text = trailing_N_elements(&a, where);
-    else
-	text = *a++;
-    if (!text)
-	return FALSE;
-
-    if ((matched = dowild(p, text, a)) != TRUE && where < 0
-     && matched != ABORT_ALL) {
-	while (1) {
-	    if (*text == '\0') {
-		if ((text = (uchar*)*a++) == NULL)
-		    return FALSE;
-		continue;
-	    }
-	    if (*text++ == '/' && (matched = dowild(p, text, a)) != FALSE
-	     && matched != ABORT_TO_STARSTAR)
-		break;
-	}
-    }
-    return matched == TRUE;
-}
-
-/* Match literal string "s" against the a virtually-joined string consisting
- * of all the pointers in array "texts" (which has a NULL pointer at the
- * end).  The int "where" can be 0 (normal matching), or > 0 (match
- * only the trailing N slash-separated filename components of "texts"). */
-int litmatch_array(const char *string, const char*const *texts, int where)
-{
-    const uchar *s = (const uchar*)string;
-    const uchar*const *a = (const uchar* const*)texts;
-    const uchar *text;
-
-    if (where > 0)
-	text = trailing_N_elements(&a, where);
-    else
-	text = *a++;
-    if (!text)
-	return FALSE;
-
-    return doliteral(s, text, a) == TRUE;
-}

diff --git a/wildmatch.h b/wildmatch.h
index e7f1a35..562faa3 100644
--- a/wildmatch.h
+++ b/wildmatch.h

@@ -2,5 +2,3 @@
 
 int wildmatch(const char *pattern, const char *text);
 int iwildmatch(const char *pattern, const char *text);
-int wildmatch_array(const char *pattern, const char*const *texts, int where);
-int litmatch_array(const char *string, const char*const *texts, int where);

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 05/12] Integrate wildmatch to git

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 .gitignore           |   1 +
 Makefile             |   3 +
 t/t3070-wildmatch.sh | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++
 t/t3070/wildtest.txt | 165 -----------------------------------------------
 test-wildmatch.c     |  14 ++++
 wildmatch.c          |   5 +-
 6 files changed, 200 insertions(+), 166 deletions(-)
 create mode 100755 t/t3070-wildmatch.sh
 delete mode 100644 t/t3070/wildtest.txt
 create mode 100644 test-wildmatch.c

diff --git a/.gitignore b/.gitignore
index a188a82..37c3507 100644
--- a/.gitignore
+++ b/.gitignore

@@ -197,6 +197,7 @@
 /test-string-list
 /test-subprocess
 /test-svn-fe
+/test-wildmatch
 /common-cmds.h
 *.tar.gz
 *.dsc

diff --git a/Makefile b/Makefile
index 8413606..9a97379 100644
--- a/Makefile
+++ b/Makefile

@@ -523,6 +523,7 @@ TEST_PROGRAMS_NEED_X += test-sigchain
 TEST_PROGRAMS_NEED_X += test-string-list
 TEST_PROGRAMS_NEED_X += test-subprocess
 TEST_PROGRAMS_NEED_X += test-svn-fe
+TEST_PROGRAMS_NEED_X += test-wildmatch
 
 TEST_PROGRAMS = $(patsubst %,%$X,$(TEST_PROGRAMS_NEED_X))

@@ -695,6 +696,7 @@ LIB_H += userdiff.h
 LIB_H += utf8.h
 LIB_H += varint.h
 LIB_H += walker.h
+LIB_H += wildmatch.h
 LIB_H += wt-status.h
 LIB_H += xdiff-interface.h
 LIB_H += xdiff/xdiff.h

@@ -826,6 +828,7 @@ LIB_OBJS += utf8.o
 LIB_OBJS += varint.o
 LIB_OBJS += version.o
 LIB_OBJS += walker.o
+LIB_OBJS += wildmatch.o
 LIB_OBJS += wrapper.o
 LIB_OBJS += write_or_die.o
 LIB_OBJS += ws.o

diff --git a/t/t3070-wildmatch.sh b/t/t3070-wildmatch.sh
new file mode 100755
index 0000000..bb92f8d
--- /dev/null
+++ b/t/t3070-wildmatch.sh

@@ -0,0 +1,178 @@
+#!/bin/sh
+
+test_description='wildmatch tests'
+
+. ./test-lib.sh
+
+match() {
+    test_expect_success "wildmatch $*" "
+	if [ $1 = 1 ]; then
+	    test-wildmatch wildmatch '$3' '$4'
+	else
+	    ! test-wildmatch wildmatch '$3' '$4'
+	fi &&
+	if [ $2 = 1 ]; then
+	    test-wildmatch fnmatch '$3' '$4'
+	else
+	    ! test-wildmatch fnmatch '$3' '$4'
+	fi
+    "
+}
+
+# Basic wildmat features
+match 1 1 foo foo
+match 0 0 foo bar
+match 1 1 '' ""
+match 1 1 foo '???'
+match 0 0 foo '??'
+match 1 1 foo '*'
+match 1 1 foo 'f*'
+match 0 0 foo '*f'
+match 1 1 foo '*foo*'
+match 1 1 foobar '*ob*a*r*'
+match 1 1 aaaaaaabababab '*ab'
+match 1 1 'foo*' 'foo\*'
+match 0 0 foobar 'foo\*bar'
+match 1 1 'f\oo' 'f\\oo'
+match 1 1 ball '*[al]?'
+match 0 0 ten '[ten]'
+match 1 1 ten '**[!te]'
+match 0 0 ten '**[!ten]'
+match 1 1 ten 't[a-g]n'
+match 0 0 ten 't[!a-g]n'
+match 1 1 ton 't[!a-g]n'
+match 1 1 ton 't[^a-g]n'
+match 1 1 'a]b' 'a[]]b'
+match 1 1 a-b 'a[]-]b'
+match 1 1 'a]b' 'a[]-]b'
+match 0 0 aab 'a[]-]b'
+match 1 1 aab 'a[]a-]b'
+match 1 1 ']' ']'
+
+# Extended slash-matching features
+match 0 0 'foo/baz/bar' 'foo*bar'
+match 1 0 'foo/baz/bar' 'foo**bar'
+match 0 0 'foo/bar' 'foo?bar'
+match 0 0 'foo/bar' 'foo[/]bar'
+match 0 0 'foo/bar' 'f[^eiu][^eiu][^eiu][^eiu][^eiu]r'
+match 1 1 'foo-bar' 'f[^eiu][^eiu][^eiu][^eiu][^eiu]r'
+match 0 0 'foo' '**/foo'
+match 1 1 '/foo' '**/foo'
+match 1 0 'bar/baz/foo' '**/foo'
+match 0 0 'bar/baz/foo' '*/foo'
+match 0 0 'foo/bar/baz' '**/bar*'
+match 1 0 'deep/foo/bar/baz' '**/bar/*'
+match 0 0 'deep/foo/bar/baz/' '**/bar/*'
+match 1 0 'deep/foo/bar/baz/' '**/bar/**'
+match 0 0 'deep/foo/bar' '**/bar/*'
+match 1 0 'deep/foo/bar/' '**/bar/**'
+match 1 0 'foo/bar/baz' '**/bar**'
+match 1 0 'foo/bar/baz/x' '*/bar/**'
+match 0 0 'deep/foo/bar/baz/x' '*/bar/**'
+match 1 0 'deep/foo/bar/baz/x' '**/bar/*/*'
+
+# Various additional tests
+match 0 0 'acrt' 'a[c-c]st'
+match 1 1 'acrt' 'a[c-c]rt'
+match 0 0 ']' '[!]-]'
+match 1 1 'a' '[!]-]'
+match 0 0 '' '\'
+match 0 0 '\' '\'
+match 0 0 '/\' '*/\'
+match 1 1 '/\' '*/\\'
+match 1 1 'foo' 'foo'
+match 1 1 '@foo' '@foo'
+match 0 0 'foo' '@foo'
+match 1 1 '[ab]' '\[ab]'
+match 1 1 '[ab]' '[[]ab]'
+match 1 1 '[ab]' '[[:]ab]'
+match 0 0 '[ab]' '[[::]ab]'
+match 1 1 '[ab]' '[[:digit]ab]'
+match 1 1 '[ab]' '[\[:]ab]'
+match 1 1 '?a?b' '\??\?b'
+match 1 1 'abc' '\a\b\c'
+match 0 0 'foo' ''
+match 1 0 'foo/bar/baz/to' '**/t[o]'
+
+# Character class tests
+match 1 1 'a1B' '[[:alpha:]][[:digit:]][[:upper:]]'
+match 0 0 'a' '[[:digit:][:upper:][:space:]]'
+match 1 1 'A' '[[:digit:][:upper:][:space:]]'
+match 1 0 '1' '[[:digit:][:upper:][:space:]]'
+match 0 0 '1' '[[:digit:][:upper:][:spaci:]]'
+match 1 1 ' ' '[[:digit:][:upper:][:space:]]'
+match 0 0 '.' '[[:digit:][:upper:][:space:]]'
+match 1 1 '.' '[[:digit:][:punct:][:space:]]'
+match 1 1 '5' '[[:xdigit:]]'
+match 1 1 'f' '[[:xdigit:]]'
+match 1 1 'D' '[[:xdigit:]]'
+match 1 0 '_' '[[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]'
+match 1 0 '_' '[[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]'
+match 1 1 '.' '[^[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:lower:][:space:][:upper:][:xdigit:]]'
+match 1 1 '5' '[a-c[:digit:]x-z]'
+match 1 1 'b' '[a-c[:digit:]x-z]'
+match 1 1 'y' '[a-c[:digit:]x-z]'
+match 0 0 'q' '[a-c[:digit:]x-z]'
+
+# Additional tests, including some malformed wildmats
+match 1 1 ']' '[\\-^]'
+match 0 0 '[' '[\\-^]'
+match 1 1 '-' '[\-_]'
+match 1 1 ']' '[\]]'
+match 0 0 '\]' '[\]]'
+match 0 0 '\' '[\]]'
+match 0 0 'ab' 'a[]b'
+match 0 1 'a[]b' 'a[]b'
+match 0 1 'ab[' 'ab['
+match 0 0 'ab' '[!'
+match 0 0 'ab' '[-'
+match 1 1 '-' '[-]'
+match 0 0 '-' '[a-'
+match 0 0 '-' '[!a-'
+match 1 1 '-' '[--A]'
+match 1 1 '5' '[--A]'
+match 1 1 ' ' '[ --]'
+match 1 1 '$' '[ --]'
+match 1 1 '-' '[ --]'
+match 0 0 '0' '[ --]'
+match 1 1 '-' '[---]'
+match 1 1 '-' '[------]'
+match 0 0 'j' '[a-e-n]'
+match 1 1 '-' '[a-e-n]'
+match 1 1 'a' '[!------]'
+match 0 0 '[' '[]-a]'
+match 1 1 '^' '[]-a]'
+match 0 0 '^' '[!]-a]'
+match 1 1 '[' '[!]-a]'
+match 1 1 '^' '[a^bc]'
+match 1 1 '-b]' '[a-]b]'
+match 0 0 '\' '[\]'
+match 1 1 '\' '[\\]'
+match 0 0 '\' '[!\\]'
+match 1 1 'G' '[A-\\]'
+match 0 0 'aaabbb' 'b*a'
+match 0 0 'aabcaa' '*ba*'
+match 1 1 ',' '[,]'
+match 1 1 ',' '[\\,]'
+match 1 1 '\' '[\\,]'
+match 1 1 '-' '[,-.]'
+match 0 0 '+' '[,-.]'
+match 0 0 '-.]' '[,-.]'
+match 1 1 '2' '[\1-\3]'
+match 1 1 '3' '[\1-\3]'
+match 0 0 '4' '[\1-\3]'
+match 1 1 '\' '[[-\]]'
+match 1 1 '[' '[[-\]]'
+match 1 1 ']' '[[-\]]'
+match 0 0 '-' '[[-\]]'
+
+# Test recursion and the abort code (use "wildtest -i" to see iteration counts)
+match 1 1 '-adobe-courier-bold-o-normal--12-120-75-75-m-70-iso8859-1' '-*-*-*-*-*-*-12-*-*-*-m-*-*-*'
+match 0 0 '-adobe-courier-bold-o-normal--12-120-75-75-X-70-iso8859-1' '-*-*-*-*-*-*-12-*-*-*-m-*-*-*'
+match 0 0 '-adobe-courier-bold-o-normal--12-120-75-75-/-70-iso8859-1' '-*-*-*-*-*-*-12-*-*-*-m-*-*-*'
+match 1 1 '/adobe/courier/bold/o/normal//12/120/75/75/m/70/iso8859/1' '/*/*/*/*/*/*/12/*/*/*/m/*/*/*'
+match 0 0 '/adobe/courier/bold/o/normal//12/120/75/75/X/70/iso8859/1' '/*/*/*/*/*/*/12/*/*/*/m/*/*/*'
+match 1 0 'abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txt' '**/*a*b*g*n*t'
+match 0 0 'abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txtz' '**/*a*b*g*n*t'
+
+test_done

diff --git a/t/t3070/wildtest.txt b/t/t3070/wildtest.txt
deleted file mode 100644
index 42c1678..0000000
--- a/t/t3070/wildtest.txt
+++ /dev/null

@@ -1,165 +0,0 @@
-# Input is in the following format (all items white-space separated):
-#
-# The first two items are 1 or 0 indicating if the wildmat call is expected to
-# succeed and if fnmatch works the same way as wildmat, respectively.  After
-# that is a text string for the match, and a pattern string.  Strings can be
-# quoted (if desired) in either double or single quotes, as well as backticks.
-#
-# MATCH FNMATCH_SAME "text to match" 'pattern to use'
-
-# Basic wildmat features
-1 1 foo			foo
-0 1 foo			bar
-1 1 ''			""
-1 1 foo			???
-0 1 foo			??
-1 1 foo			*
-1 1 foo			f*
-0 1 foo			*f
-1 1 foo			*foo*
-1 1 foobar		*ob*a*r*
-1 1 aaaaaaabababab	*ab
-1 1 foo*		foo\*
-0 1 foobar		foo\*bar
-1 1 f\oo		f\\oo
-1 1 ball		*[al]?
-0 1 ten			[ten]
-1 1 ten			**[!te]
-0 1 ten			**[!ten]
-1 1 ten			t[a-g]n
-0 1 ten			t[!a-g]n
-1 1 ton			t[!a-g]n
-1 1 ton			t[^a-g]n
-1 1 a]b			a[]]b
-1 1 a-b			a[]-]b
-1 1 a]b			a[]-]b
-0 1 aab			a[]-]b
-1 1 aab			a[]a-]b
-1 1 ]			]
-
-# Extended slash-matching features
-0 1 foo/baz/bar		foo*bar
-1 1 foo/baz/bar		foo**bar
-0 1 foo/bar		foo?bar
-0 1 foo/bar		foo[/]bar
-0 1 foo/bar		f[^eiu][^eiu][^eiu][^eiu][^eiu]r
-1 1 foo-bar		f[^eiu][^eiu][^eiu][^eiu][^eiu]r
-0 1 foo			**/foo
-1 1 /foo		**/foo
-1 1 bar/baz/foo		**/foo
-0 1 bar/baz/foo		*/foo
-0 0 foo/bar/baz		**/bar*
-1 1 deep/foo/bar/baz	**/bar/*
-0 1 deep/foo/bar/baz/	**/bar/*
-1 1 deep/foo/bar/baz/	**/bar/**
-0 1 deep/foo/bar	**/bar/*
-1 1 deep/foo/bar/	**/bar/**
-1 1 foo/bar/baz		**/bar**
-1 1 foo/bar/baz/x	*/bar/**
-0 0 deep/foo/bar/baz/x	*/bar/**
-1 1 deep/foo/bar/baz/x	**/bar/*/*
-
-# Various additional tests
-0 1 acrt		a[c-c]st
-1 1 acrt		a[c-c]rt
-0 1 ]			[!]-]
-1 1 a			[!]-]
-0 1 ''			\
-0 1 \			\
-0 1 /\			*/\
-1 1 /\			*/\\
-1 1 foo			foo
-1 1 @foo		@foo
-0 1 foo			@foo
-1 1 [ab]		\[ab]
-1 1 [ab]		[[]ab]
-1 1 [ab]		[[:]ab]
-0 1 [ab]		[[::]ab]
-1 1 [ab]		[[:digit]ab]
-1 1 [ab]		[\[:]ab]
-1 1 ?a?b		\??\?b
-1 1 abc			\a\b\c
-0 1 foo			''
-1 1 foo/bar/baz/to	**/t[o]
-
-# Character class tests
-1 1 a1B		[[:alpha:]][[:digit:]][[:upper:]]
-0 1 a		[[:digit:][:upper:][:space:]]
-1 1 A		[[:digit:][:upper:][:space:]]
-1 1 1		[[:digit:][:upper:][:space:]]
-0 1 1		[[:digit:][:upper:][:spaci:]]
-1 1 ' '		[[:digit:][:upper:][:space:]]
-0 1 .		[[:digit:][:upper:][:space:]]
-1 1 .		[[:digit:][:punct:][:space:]]
-1 1 5		[[:xdigit:]]
-1 1 f		[[:xdigit:]]
-1 1 D		[[:xdigit:]]
-1 1 _		[[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
-#1 1 �		[^[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
-1 1 		[^[:alnum:][:alpha:][:blank:][:digit:][:graph:][:lower:][:print:][:punct:][:space:][:upper:][:xdigit:]]
-1 1 .		[^[:alnum:][:alpha:][:blank:][:cntrl:][:digit:][:lower:][:space:][:upper:][:xdigit:]]
-1 1 5		[a-c[:digit:]x-z]
-1 1 b		[a-c[:digit:]x-z]
-1 1 y		[a-c[:digit:]x-z]
-0 1 q		[a-c[:digit:]x-z]
-
-# Additional tests, including some malformed wildmats
-1 1 ]		[\\-^]
-0 1 [		[\\-^]
-1 1 -		[\-_]
-1 1 ]		[\]]
-0 1 \]		[\]]
-0 1 \		[\]]
-0 1 ab		a[]b
-0 1 a[]b	a[]b
-0 1 ab[		ab[
-0 1 ab		[!
-0 1 ab		[-
-1 1 -		[-]
-0 1 -		[a-
-0 1 -		[!a-
-1 1 -		[--A]
-1 1 5		[--A]
-1 1 ' '		'[ --]'
-1 1 $		'[ --]'
-1 1 -		'[ --]'
-0 1 0		'[ --]'
-1 1 -		[---]
-1 1 -		[------]
-0 1 j		[a-e-n]
-1 1 -		[a-e-n]
-1 1 a		[!------]
-0 1 [		[]-a]
-1 1 ^		[]-a]
-0 1 ^		[!]-a]
-1 1 [		[!]-a]
-1 1 ^		[a^bc]
-1 1 -b]		[a-]b]
-0 1 \		[\]
-1 1 \		[\\]
-0 1 \		[!\\]
-1 1 G		[A-\\]
-0 1 aaabbb	b*a
-0 1 aabcaa	*ba*
-1 1 ,		[,]
-1 1 ,		[\\,]
-1 1 \		[\\,]
-1 1 -		[,-.]
-0 1 +		[,-.]
-0 1 -.]		[,-.]
-1 1 2		[\1-\3]
-1 1 3		[\1-\3]
-0 1 4		[\1-\3]
-1 1 \		[[-\]]
-1 1 [		[[-\]]
-1 1 ]		[[-\]]
-0 1 -		[[-\]]
-
-# Test recursion and the abort code (use "wildtest -i" to see iteration counts)
-1 1 -adobe-courier-bold-o-normal--12-120-75-75-m-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
-0 1 -adobe-courier-bold-o-normal--12-120-75-75-X-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
-0 1 -adobe-courier-bold-o-normal--12-120-75-75-/-70-iso8859-1	-*-*-*-*-*-*-12-*-*-*-m-*-*-*
-1 1 /adobe/courier/bold/o/normal//12/120/75/75/m/70/iso8859/1	/*/*/*/*/*/*/12/*/*/*/m/*/*/*
-0 1 /adobe/courier/bold/o/normal//12/120/75/75/X/70/iso8859/1	/*/*/*/*/*/*/12/*/*/*/m/*/*/*
-1 1 abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txt		**/*a*b*g*n*t
-0 1 abcd/abcdefg/abcdefghijk/abcdefghijklmnop.txtz		**/*a*b*g*n*t

diff --git a/test-wildmatch.c b/test-wildmatch.c
new file mode 100644
index 0000000..08962d5
--- /dev/null
+++ b/test-wildmatch.c

@@ -0,0 +1,14 @@
+#include "cache.h"
+#include "wildmatch.h"
+
+int main(int argc, char **argv)
+{
+	if (!strcmp(argv[1], "wildmatch"))
+		return wildmatch(argv[3], argv[2]) ? 0 : 1;
+	else if (!strcmp(argv[1], "iwildmatch"))
+		return iwildmatch(argv[3], argv[2]) ? 0 : 1;
+	else if (!strcmp(argv[1], "fnmatch"))
+		return fnmatch(argv[3], argv[2], FNM_PATHNAME);
+	else
+		return 1;
+}

diff --git a/wildmatch.c b/wildmatch.c
index 71dba76..6092bde 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -9,7 +9,10 @@
 **  work differently than '*', and to fix the character-class code.
 */
 
-#include "rsync.h"
+#include "cache.h"
+#include "wildmatch.h"
+
+typedef unsigned char uchar;
 
 /* What character marks an inverted character class? */
 #define NEGATE_CLASS	'!'

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 06/12] wildmatch: make wildmatch's return value compatible with fnmatch

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

wildmatch returns non-zero if matched, zero otherwise. This patch
makes it return zero if matches, non-zero otherwise, like fnmatch().

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 test-wildmatch.c |  4 ++--
 wildmatch.c      | 21 ++++++++++++---------
 2 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/test-wildmatch.c b/test-wildmatch.c
index 08962d5..596029c 100644
--- a/test-wildmatch.c
+++ b/test-wildmatch.c

@@ -4,9 +4,9 @@
 int main(int argc, char **argv)
 {
 	if (!strcmp(argv[1], "wildmatch"))
-		return wildmatch(argv[3], argv[2]) ? 0 : 1;
+		return wildmatch(argv[3], argv[2]);
 	else if (!strcmp(argv[1], "iwildmatch"))
-		return iwildmatch(argv[3], argv[2]) ? 0 : 1;
+		return iwildmatch(argv[3], argv[2]);
 	else if (!strcmp(argv[1], "fnmatch"))
 		return fnmatch(argv[3], argv[2], FNM_PATHNAME);
 	else

diff --git a/wildmatch.c b/wildmatch.c
index 6092bde..7c180e3 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -20,6 +20,9 @@ typedef unsigned char uchar;
 
 #define FALSE 0
 #define TRUE 1
+
+#define NOMATCH 1
+#define MATCH 0
 #define ABORT_ALL -1
 #define ABORT_TO_STARSTAR -2

@@ -79,12 +82,12 @@ static int dowild(const uchar *p, const uchar *text)
 	    /* FALLTHROUGH */
 	  default:
 	    if (t_ch != p_ch)
-		return FALSE;
+		return NOMATCH;
 	    continue;
 	  case '?':
 	    /* Match anything but '/'. */
 	    if (t_ch == '/')
-		return FALSE;
+		return NOMATCH;
 	    continue;
 	  case '*':
 	    if (*++p == '*') {

@@ -97,14 +100,14 @@ static int dowild(const uchar *p, const uchar *text)
 		 * only if there are no more slash characters. */
 		if (!special) {
 			if (strchr((char*)text, '/') != NULL)
-			    return FALSE;
+			    return NOMATCH;
 		}
-		return TRUE;
+		return MATCH;
 	    }
 	    while (1) {
 		if (t_ch == '\0')
 		    break;
-		if ((matched = dowild(p, text)) != FALSE) {
+		if ((matched = dowild(p, text)) != NOMATCH) {
 		    if (!special || matched != ABORT_TO_STARSTAR)
 			return matched;
 		} else if (!special && t_ch == '/')

@@ -203,18 +206,18 @@ static int dowild(const uchar *p, const uchar *text)
 		    matched = TRUE;
 	    } while (prev_ch = p_ch, (p_ch = *++p) != ']');
 	    if (matched == special || t_ch == '/')
-		return FALSE;
+		return NOMATCH;
 	    continue;
 	}
     }
 
-    return *text ? FALSE : TRUE;
+    return *text ? NOMATCH : MATCH;
 }
 
 /* Match the "pattern" against the "text" string. */
 int wildmatch(const char *pattern, const char *text)
 {
-    return dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
+    return dowild((const uchar*)pattern, (const uchar*)text);
 }
 
 /* Match the "pattern" against the forced-to-lower-case "text" string. */

@@ -222,7 +225,7 @@ int iwildmatch(const char *pattern, const char *text)
 {
     int ret;
     force_lower_case = 1;
-    ret = dowild((const uchar*)pattern, (const uchar*)text) == TRUE;
+    ret = dowild((const uchar*)pattern, (const uchar*)text);
     force_lower_case = 0;
     return ret;
 }

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 07/12] wildmatch: remove static variable force_lower_case

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

One place less to worry about thread safety. Also combine wildmatch
and iwildmatch into one.

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 test-wildmatch.c |  4 ++--
 wildmatch.c      | 21 +++++----------------
 wildmatch.h      |  3 +--
 3 files changed, 8 insertions(+), 20 deletions(-)

diff --git a/test-wildmatch.c b/test-wildmatch.c
index 596029c..d716852 100644
--- a/test-wildmatch.c
+++ b/test-wildmatch.c

@@ -4,9 +4,9 @@
 int main(int argc, char **argv)
 {
 	if (!strcmp(argv[1], "wildmatch"))
-		return wildmatch(argv[3], argv[2]);
+		return wildmatch(argv[3], argv[2], 0);
 	else if (!strcmp(argv[1], "iwildmatch"))
-		return iwildmatch(argv[3], argv[2]);
+		return wildmatch(argv[3], argv[2], FNM_CASEFOLD);
 	else if (!strcmp(argv[1], "fnmatch"))
 		return fnmatch(argv[3], argv[2], FNM_PATHNAME);
 	else

diff --git a/wildmatch.c b/wildmatch.c
index 7c180e3..d18d721 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -59,11 +59,9 @@ typedef unsigned char uchar;
 #define ISUPPER(c) (ISASCII(c) && isupper(c))
 #define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
 
-static int force_lower_case = 0;
-
 /* Match pattern "p" against the a virtually-joined string consisting
  * of "text" and any strings in array "a". */
-static int dowild(const uchar *p, const uchar *text)
+static int dowild(const uchar *p, const uchar *text, int force_lower_case)
 {
     uchar p_ch;

@@ -107,7 +105,7 @@ static int dowild(const uchar *p, const uchar *text)
 	    while (1) {
 		if (t_ch == '\0')
 		    break;
-		if ((matched = dowild(p, text)) != NOMATCH) {
+		if ((matched = dowild(p, text, force_lower_case)) != NOMATCH) {
 		    if (!special || matched != ABORT_TO_STARSTAR)
 			return matched;
 		} else if (!special && t_ch == '/')

@@ -215,17 +213,8 @@ static int dowild(const uchar *p, const uchar *text)
 }
 
 /* Match the "pattern" against the "text" string. */
-int wildmatch(const char *pattern, const char *text)
-{
-    return dowild((const uchar*)pattern, (const uchar*)text);
-}
-
-/* Match the "pattern" against the forced-to-lower-case "text" string. */
-int iwildmatch(const char *pattern, const char *text)
+int wildmatch(const char *pattern, const char *text, int flags)
 {
-    int ret;
-    force_lower_case = 1;
-    ret = dowild((const uchar*)pattern, (const uchar*)text);
-    force_lower_case = 0;
-    return ret;
+    return dowild((const uchar*)pattern, (const uchar*)text,
+		  flags & FNM_CASEFOLD ? 1 : 0);
 }

diff --git a/wildmatch.h b/wildmatch.h
index 562faa3..e974f9a 100644
--- a/wildmatch.h
+++ b/wildmatch.h

@@ -1,4 +1,3 @@
 /* wildmatch.h */
 
-int wildmatch(const char *pattern, const char *text);
-int iwildmatch(const char *pattern, const char *text);
+int wildmatch(const char *pattern, const char *text, int flags);

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 08/12] wildmatch: fix case-insensitive matching

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

dowild() does case insensitive matching by lower-casing the text. That
means lower case letters in patterns imply case-insensitive matching,
but upper case means exact matching.

We do not want that subtlety. Lower case pattern too so iwildmatch()
always does what we expect it to do.

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 wildmatch.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/wildmatch.c b/wildmatch.c
index d18d721..b700631 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -72,6 +72,8 @@ static int dowild(const uchar *p, const uchar *text, int force_lower_case)
 		return ABORT_ALL;
 	if (force_lower_case && ISUPPER(t_ch))
 	    t_ch = tolower(t_ch);
+	if (force_lower_case && ISUPPER(p_ch))
+	    p_ch = tolower(p_ch);
 	switch (p_ch) {
 	  case '\\':
 	    /* Literal match with following character.  Note that the test

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 09/12] wildmatch: adjust "**" behavior

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Standard wildmatch() sees consecutive asterisks as "*" that can also
match slashes. But that may be hard to explain to users as
"abc/**/def" can match "abcdef", "abcxyzdef", "abc/def", "abc/x/def",
"abc/x/y/def"...

This patch changes wildmatch so that users can do

- "**/def" -> all paths ending with file/directory 'def'
- "abc/**" - equivalent to "/abc/"
- "abc/**/def" -> "abc/x/def", "abc/x/y/def"...
- otherwise consider the pattern malformed if "**" is found

Basically the magic of "**" only remains if it's wrapped around by
slashes.

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 t/t3070-wildmatch.sh |  5 +++--
 wildmatch.c          | 13 +++++++------
 wildmatch.h          |  6 ++++++
 3 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/t/t3070-wildmatch.sh b/t/t3070-wildmatch.sh
index bb92f8d..f6c64be 100755
--- a/t/t3070-wildmatch.sh
+++ b/t/t3070-wildmatch.sh

@@ -36,7 +36,7 @@ match 0 0 foobar 'foo\*bar'
 match 1 1 'f\oo' 'f\\oo'
 match 1 1 ball '*[al]?'
 match 0 0 ten '[ten]'
-match 1 1 ten '**[!te]'
+match 0 1 ten '**[!te]'
 match 0 0 ten '**[!ten]'
 match 1 1 ten 't[a-g]n'
 match 0 0 ten 't[!a-g]n'

@@ -51,7 +51,8 @@ match 1 1 ']' ']'
 
 # Extended slash-matching features
 match 0 0 'foo/baz/bar' 'foo*bar'
-match 1 0 'foo/baz/bar' 'foo**bar'
+match 0 0 'foo/baz/bar' 'foo**bar'
+match 0 1 'foobazbar' 'foo**bar'
 match 0 0 'foo/bar' 'foo?bar'
 match 0 0 'foo/bar' 'foo[/]bar'
 match 0 0 'foo/bar' 'f[^eiu][^eiu][^eiu][^eiu][^eiu]r'

diff --git a/wildmatch.c b/wildmatch.c
index b700631..e09a459 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -21,11 +21,6 @@ typedef unsigned char uchar;
 #define FALSE 0
 #define TRUE 1
 
-#define NOMATCH 1
-#define MATCH 0
-#define ABORT_ALL -1
-#define ABORT_TO_STARSTAR -2
-
 #define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \
 				    && *(class) == *(litmatch) \
 				    && strncmp((char*)class, litmatch, len) == 0)

@@ -91,8 +86,14 @@ static int dowild(const uchar *p, const uchar *text, int force_lower_case)
 	    continue;
 	  case '*':
 	    if (*++p == '*') {
+		const uchar *prev_p = p - 2;
 		while (*++p == '*') {}
-		special = TRUE;
+		if ((prev_p == text || *prev_p == '/') ||
+		    (*p == '\0' || *p == '/' ||
+		     (p[0] == '\\' && p[1] == '/'))) {
+		    special = TRUE;
+		} else
+		    return ABORT_MALFORMED;
 	    } else
 		special = FALSE;
 	    if (*p == '\0') {

diff --git a/wildmatch.h b/wildmatch.h
index e974f9a..984a38c 100644
--- a/wildmatch.h
+++ b/wildmatch.h

@@ -1,3 +1,9 @@
 /* wildmatch.h */
 
+#define ABORT_MALFORMED 2
+#define NOMATCH 1
+#define MATCH 0
+#define ABORT_ALL -1
+#define ABORT_TO_STARSTAR -2
+
 int wildmatch(const char *pattern, const char *text, int flags);

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 10/12] wildmatch: make /**/ match zero or more directories

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

"foo/**/bar" matches "foo/x/bar", "foo/x/y/bar"... but not
"foo/bar". We make a special case, when foo/**/ is detected (and
"foo/" part is already matched), try matching "bar" with the rest of
the string.

"Match one or more directories" semantics can be easily achieved using
"foo/*/**/bar".

This also makes "**/foo" match "foo" in addition to "x/foo",
"x/y/foo"..

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 t/t3070-wildmatch.sh |  8 +++++++-
 wildmatch.c          | 17 +++++++++++++++++
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/t/t3070-wildmatch.sh b/t/t3070-wildmatch.sh
index f6c64be..f21da6c 100755
--- a/t/t3070-wildmatch.sh
+++ b/t/t3070-wildmatch.sh

@@ -53,11 +53,17 @@ match 1 1 ']' ']'
 match 0 0 'foo/baz/bar' 'foo*bar'
 match 0 0 'foo/baz/bar' 'foo**bar'
 match 0 1 'foobazbar' 'foo**bar'
+match 1 1 'foo/baz/bar' 'foo/**/bar'
+match 1 0 'foo/baz/bar' 'foo/**/**/bar'
+match 1 0 'foo/b/a/z/bar' 'foo/**/bar'
+match 1 0 'foo/b/a/z/bar' 'foo/**/**/bar'
+match 1 0 'foo/bar' 'foo/**/bar'
+match 1 0 'foo/bar' 'foo/**/**/bar'
 match 0 0 'foo/bar' 'foo?bar'
 match 0 0 'foo/bar' 'foo[/]bar'
 match 0 0 'foo/bar' 'f[^eiu][^eiu][^eiu][^eiu][^eiu]r'
 match 1 1 'foo-bar' 'f[^eiu][^eiu][^eiu][^eiu][^eiu]r'
-match 0 0 'foo' '**/foo'
+match 1 0 'foo' '**/foo'
 match 1 1 '/foo' '**/foo'
 match 1 0 'bar/baz/foo' '**/foo'
 match 0 0 'bar/baz/foo' '*/foo'

diff --git a/wildmatch.c b/wildmatch.c
index e09a459..f060bb0 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -91,6 +91,23 @@ static int dowild(const uchar *p, const uchar *text, int force_lower_case)
 		if ((prev_p == text || *prev_p == '/') ||
 		    (*p == '\0' || *p == '/' ||
 		     (p[0] == '\\' && p[1] == '/'))) {
+			/*
+			 * Assuming we already match 'foo/' and are at
+			 * <star star slash>, just assume it matches
+			 * nothing and go ahead match the rest of the
+			 * pattern with the remaining string. This
+			 * helps make foo/<*><*>/bar (<> because
+			 * otherwise it breaks C comment syntax) match
+			 * both foo/bar and foo/a/bar.
+			 *
+			 * Crazy patterns like /<*><*>/<*><*>/ are
+			 * treated like /<*><*>/. But undefined
+			 * behavior is even appropriate for people
+			 * writing such a pattern.
+			 */
+			if (p[0] == '/' &&
+			    dowild(p + 1, text, force_lower_case) == MATCH)
+				return MATCH;
 		    special = TRUE;
 		} else
 		    return ABORT_MALFORMED;

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 11/12] Support "**" wildcard in .gitignore and .gitattributes

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 Documentation/gitattributes.txt    |  6 ++++++
 Documentation/gitignore.txt        | 19 +++++++++++++++++++
 attr.c                             |  4 +++-
 dir.c                              |  4 +++-
 t/t0003-attributes.sh              | 38 ++++++++++++++++++++++++++++++++++++++
 t/t3001-ls-files-others-exclude.sh | 19 +++++++++++++++++++
 6 files changed, 88 insertions(+), 2 deletions(-)

diff --git a/Documentation/gitattributes.txt b/Documentation/gitattributes.txt
index 99ed04d..93c542b 100644
--- a/Documentation/gitattributes.txt
+++ b/Documentation/gitattributes.txt

@@ -57,6 +57,12 @@ overrides an earlier line.  This overriding is done per
 attribute.  The rules how the pattern matches paths are the
 same as in `.gitignore` files; see linkgit:gitignore[5].
 
+Note that if a .gitignore rule matches a directory, the directory
+is ignored, which may be seen as "ignore" attribute propagated to
+to all files and directories inside. If a .gitattributes rule
+matches a directory, it manipulates attributes on that directory
+only, not files and directories inside.
+
 When deciding what attributes are assigned to a path, git
 consults `$GIT_DIR/info/attributes` file (which has the highest
 precedence), `.gitattributes` file in the same directory as the

diff --git a/Documentation/gitignore.txt b/Documentation/gitignore.txt
index 96639e0..b564b4b 100644
--- a/Documentation/gitignore.txt
+++ b/Documentation/gitignore.txt

@@ -104,6 +104,25 @@ PATTERN FORMAT
    For example, "/{asterisk}.c" matches "cat-file.c" but not
    "mozilla-sha1/sha1.c".
 
+Two consecutive asterisks ("`**`") in patterns matched against
+full pathname may have special meaning:
+
+ - A leading "`**`" followed by a slash means match in all
+   directories. For example, "`**/foo`" matches file or directory
+   "`foo`" anywhere, the same as pattern "`foo`". "**/foo/bar"
+   matches file or directory "`bar`" anywhere that is directly
+   under directory "`foo`".
+
+ - A trailing "/**" matches everything inside. For example,
+   "abc/**" matches all files inside directory "abc", relative
+   to the location of the `.gitignore` file, with infinite depth.
+
+ - A slash followed by two consecutive asterisks then a slash
+   matches zero or more directories. For example, "`a/**/b`"
+   matches "`a/b`", "`a/x/b`", "`a/x/y/b`" and so on.
+
+ - Other consecutive asterisks are considered invalid.
+
 NOTES
 -----

diff --git a/attr.c b/attr.c
index 887a9ae..8010429 100644
--- a/attr.c
+++ b/attr.c

@@ -12,6 +12,7 @@
 #include "exec_cmd.h"
 #include "attr.h"
 #include "dir.h"
+#include "wildmatch.h"
 
 const char git_attr__true[] = "(builtin)true";
 const char git_attr__false[] = "\0(builtin)false";

@@ -666,7 +667,8 @@ static int path_matches(const char *pathname, int pathlen,
 		return 0;
 	if (baselen != 0)
 		baselen++;
-	return fnmatch_icase(pattern, pathname + baselen, FNM_PATHNAME) == 0;
+	return wildmatch(pattern, pathname + baselen,
+			 ignore_case ? FNM_CASEFOLD : 0) == 0;
 }
 
 static int macroexpand_one(int attr_nr, int rem);

diff --git a/dir.c b/dir.c
index 4868339..442db1c 100644
--- a/dir.c
+++ b/dir.c

@@ -8,6 +8,7 @@
 #include "cache.h"
 #include "dir.h"
 #include "refs.h"
+#include "wildmatch.h"
 
 struct path_simplify {
 	int len;

@@ -575,7 +576,8 @@ int excluded_from_list(const char *pathname,
 			namelen -= prefix;
 		}
 
-		if (!namelen || !fnmatch_icase(exclude, name, FNM_PATHNAME))
+		if (!namelen ||
+		    wildmatch(exclude, name, ignore_case ? FNM_CASEFOLD : 0) == 0)
 			return to_exclude;
 	}
 	return -1; /* undecided */

diff --git a/t/t0003-attributes.sh b/t/t0003-attributes.sh
index febc45c..67a5694 100755
--- a/t/t0003-attributes.sh
+++ b/t/t0003-attributes.sh

@@ -232,4 +232,42 @@ test_expect_success 'bare repository: test info/attributes' '
 	attr_check subdir/a/i unspecified
 '
 
+test_expect_success '"**" test' '
+	cd .. &&
+	echo "**/f foo=bar" >.gitattributes &&
+	cat <<\EOF >expect &&
+f: foo: bar
+a/f: foo: bar
+a/b/f: foo: bar
+a/b/c/f: foo: bar
+EOF
+	git check-attr foo -- "f" >actual 2>err &&
+	git check-attr foo -- "a/f" >>actual 2>>err &&
+	git check-attr foo -- "a/b/f" >>actual 2>>err &&
+	git check-attr foo -- "a/b/c/f" >>actual 2>>err &&
+	test_cmp expect actual &&
+	test_line_count = 0 err
+'
+
+test_expect_success '"**" with no slashes test' '
+	echo "a**f foo=bar" >.gitattributes &&
+	git check-attr foo -- "f" >actual &&
+	cat <<\EOF >expect &&
+f: foo: unspecified
+af: foo: bar
+axf: foo: bar
+a/f: foo: unspecified
+a/b/f: foo: unspecified
+a/b/c/f: foo: unspecified
+EOF
+	git check-attr foo -- "f" >actual 2>err &&
+	git check-attr foo -- "af" >>actual 2>err &&
+	git check-attr foo -- "axf" >>actual 2>err &&
+	git check-attr foo -- "a/f" >>actual 2>>err &&
+	git check-attr foo -- "a/b/f" >>actual 2>>err &&
+	git check-attr foo -- "a/b/c/f" >>actual 2>>err &&
+	test_cmp expect actual &&
+	test_line_count = 0 err
+'
+
 test_done

diff --git a/t/t3001-ls-files-others-exclude.sh b/t/t3001-ls-files-others-exclude.sh
index c8fe978..278315d 100755
--- a/t/t3001-ls-files-others-exclude.sh
+++ b/t/t3001-ls-files-others-exclude.sh

@@ -214,4 +214,23 @@ test_expect_success 'subdirectory ignore (l1)' '
 	test_cmp expect actual
 '
 
+
+test_expect_success 'ls-files with "**" patterns' '
+	cat <<\EOF >expect &&
+a.1
+one/a.1
+one/two/a.1
+three/a.1
+EOF
+	git ls-files -o -i --exclude "**/a.1" >actual
+	test_cmp expect actual
+'
+
+
+test_expect_success 'ls-files with "**" patterns and no slashes' '
+	: >expect &&
+	git ls-files -o -i --exclude "one**a.1" >actual &&
+	test_cmp expect actual
+'
+
 test_done

-- 
1.7.12.1.406.g6ab07c4

[PATCH v4 12/12] t3070: disable two fnmatch tests that have different results on different libc

From: Nguyễn Thái Ngọc Duy <hidden>
Date: 2016-06-15 22:54:59

fnmatch on glibc-2.12.1 returns no match. glibc-2.15 returns ok.

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 t/t3070-wildmatch.sh | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/t/t3070-wildmatch.sh b/t/t3070-wildmatch.sh
index f21da6c..38042eb 100755
--- a/t/t3070-wildmatch.sh
+++ b/t/t3070-wildmatch.sh

@@ -13,6 +13,8 @@ match() {
 	fi &&
 	if [ $2 = 1 ]; then
 	    test-wildmatch fnmatch '$3' '$4'
+	elif [ $2 = x ]; then
+	    true
 	else
 	    ! test-wildmatch fnmatch '$3' '$4'
 	fi

@@ -129,8 +131,8 @@ match 1 1 ']' '[\]]'
 match 0 0 '\]' '[\]]'
 match 0 0 '\' '[\]]'
 match 0 0 'ab' 'a[]b'
-match 0 1 'a[]b' 'a[]b'
-match 0 1 'ab[' 'ab['
+match 0 x 'a[]b' 'a[]b'
+match 0 x 'ab[' 'ab['
 match 0 0 'ab' '[!'
 match 0 0 'ab' '[-'
 match 1 1 '-' '[-]'

-- 
1.7.12.1.406.g6ab07c4

Re: [PATCH v4 04/12] wildmatch: remove unnecessary functions

From: Michael Haggerty <hidden>
Date: 2016-06-15 22:54:59

On 10/10/2012 12:40 PM, Nguyễn Thái Ngọc Duy wrote:

quoted hunk

Signed-off-by: Nguyễn Thái Ngọc Duy <redacted>
---
 wildmatch.c | 161 ++++--------------------------------------------------------
 wildmatch.h |   2 -
 2 files changed, 9 insertions(+), 154 deletions(-)

diff --git a/wildmatch.c b/wildmatch.c
index f3a1731..71dba76 100644
--- a/wildmatch.c
+++ b/wildmatch.c

@@ -53,33 +53,19 @@
 #define ISUPPER(c) (ISASCII(c) && isupper(c))
 #define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
 
-#ifdef WILD_TEST_ITERATIONS
-int wildmatch_iteration_count;
-#endif
-
 static int force_lower_case = 0;
 
 /* Match pattern "p" against the a virtually-joined string consisting
  * of "text" and any strings in array "a". */
-static int dowild(const uchar *p, const uchar *text, const uchar*const *a)
+static int dowild(const uchar *p, const uchar *text)

The comment still refers to array "a".

Michael

-- 
Michael Haggerty
mhagger@alum.mit.edu
http://softwareswirl.blogspot.com/

Re: [PATCH v4 12/12] t3070: disable two fnmatch tests that have different results on different libc

From: Johannes Sixt <hidden>
Date: 2016-06-15 22:55:00

Am 10/10/2012 12:40, schrieb Nguyễn Thái Ngọc Duy:

fnmatch on glibc-2.12.1 returns no match. glibc-2.15 returns ok.

There are many more cases that fail with the fnmatch() that we ship in
compat/fnmatch. To test this on Linux, you have to remove the "#if defined
_LIBC || !defined __GNU_LIBRARY__" brackets from compat/fnmatch/fnmatch.c
and build with NO_FNMATCH.

-- Hannes

Re: [PATCH v4 05/12] Integrate wildmatch to git

From: Johannes Sixt <hidden>
Date: 2016-06-15 22:55:00

--- >8 ---

From: Johannes Sixt <redacted>
Subject: [PATCH] test-wildmatch: avoid exit code -1

Our bash on Windows does not recognize -1 as failure.

Signed-off-by: Johannes Sixt <redacted>
---
 Please squash this in, in the next round.

 test-wildmatch.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/test-wildmatch.c b/test-wildmatch.c
index d716852..74c0864 100644
--- a/test-wildmatch.c
+++ b/test-wildmatch.c

@@ -4,11 +4,11 @@
 int main(int argc, char **argv)
 {
 	if (!strcmp(argv[1], "wildmatch"))
-		return wildmatch(argv[3], argv[2], 0);
+		return !!wildmatch(argv[3], argv[2], 0);
 	else if (!strcmp(argv[1], "iwildmatch"))
-		return wildmatch(argv[3], argv[2], FNM_CASEFOLD);
+		return !!wildmatch(argv[3], argv[2], FNM_CASEFOLD);
 	else if (!strcmp(argv[1], "fnmatch"))
-		return fnmatch(argv[3], argv[2], FNM_PATHNAME);
+		return !!fnmatch(argv[3], argv[2], FNM_PATHNAME);
 	else
 		return 1;
 }

-- 
1.8.0.rc2.1237.g5522246

Re: [PATCH v4 12/12] t3070: disable two fnmatch tests that have different results on different libc

From: Nguyen Thai Ngoc Duy <hidden>
Date: 2016-06-15 22:55:00

On Fri, Oct 12, 2012 at 2:22 PM, Johannes Sixt [off-list ref] wrote:

Am 10/10/2012 12:40, schrieb Nguyễn Thái Ngọc Duy:

quoted

fnmatch on glibc-2.12.1 returns no match. glibc-2.15 returns ok.

There are many more cases that fail with the fnmatch() that we ship in
compat/fnmatch. To test this on Linux, you have to remove the "#if defined
_LIBC || !defined __GNU_LIBRARY__" brackets from compat/fnmatch/fnmatch.c
and build with NO_FNMATCH.

Thanks. The point of checking against fnmatch is to make sure that
wildmatch does not divert in behavior from fnmatch. For some corner
cases that behavior is undefined, I think it is ok for different
fnmatch versions to behave differently.

But some of failed tests make me worry about fnmatch. compat/fnmatch
for example does not match '**/foo' with '/foo' (it does '*/foo' with
'/foo'). 'A' is matched with '[[:digit:][:space:][:upper:]]' but not
'[[:digit:][:upper:][:space:]]'. Perhaps we better off convert git to
use wildmatch only to keep matching behavior more reliable. wildmatch
does not support non-FNM_PATHNAME mode, but that should be easy to
fix. The only downside I see is fnmatch may support locale while
wildmatch does not.
-- 
Duy

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help