From 6fbb1463429748bed7ab6593779430ee1c0664f5 Mon Sep 17 00:00:00 2001
From: =?utf8?q?S=C3=A9bastien=20Wilmet?= <swilmet@gnome.org>
Date: Mon, 15 Jul 2013 13:52:14 +0200
Subject: [PATCH] GRegex: add g_regex_get_max_lookbehind()

It is useful for multi-segment regex matching.

A unit test is included.

https://bugzilla.gnome.org/show_bug.cgi?id=689794
---
 docs/reference/glib/glib-sections.txt |  1 +
 glib/gregex.c                         | 23 +++++++++++++++++++++++
 glib/gregex.h                         |  2 ++
 glib/tests/regex.c                    | 19 +++++++++++++++++++
 4 files changed, 45 insertions(+)

diff --git a/docs/reference/glib/glib-sections.txt b/docs/reference/glib/glib-sections.txt
index 54c7b6d..9c2a7f8 100644
--- a/docs/reference/glib/glib-sections.txt
+++ b/docs/reference/glib/glib-sections.txt
@@ -1018,6 +1018,7 @@ g_regex_get_pattern
 g_regex_get_max_backref
 g_regex_get_capture_count
 g_regex_get_has_cr_or_lf
+g_regex_get_max_lookbehind
 g_regex_get_string_number
 g_regex_get_compile_flags
 g_regex_get_match_flags
diff --git a/glib/gregex.c b/glib/gregex.c
index fa3a18f..6dbb9bf 100644
--- a/glib/gregex.c
+++ b/glib/gregex.c
@@ -1520,6 +1520,29 @@ g_regex_get_has_cr_or_lf (const GRegex *regex)
 }
 
 /**
+ * g_regex_get_max_lookbehind:
+ * @regex: a #GRegex structure
+ *
+ * Gets the number of characters in the longest lookbehind assertion in the
+ * pattern. This information is useful when doing multi-segment matching using
+ * the partial matching facilities.
+ *
+ * Returns: the number of characters in the longest lookbehind assertion.
+ *
+ * Since: 2.38
+ */
+gint
+g_regex_get_max_lookbehind (const GRegex *regex)
+{
+  gint max_lookbehind;
+
+  pcre_fullinfo (regex->pcre_re, regex->extra,
+                 PCRE_INFO_MAXLOOKBEHIND, &max_lookbehind);
+
+  return max_lookbehind;
+}
+
+/**
  * g_regex_get_compile_flags:
  * @regex: a #GRegex
  *
diff --git a/glib/gregex.h b/glib/gregex.h
index 57fcb67..77d9d04 100644
--- a/glib/gregex.h
+++ b/glib/gregex.h
@@ -454,6 +454,8 @@ GLIB_AVAILABLE_IN_ALL
 gint		  g_regex_get_capture_count	(const GRegex        *regex);
 GLIB_AVAILABLE_IN_ALL
 gboolean          g_regex_get_has_cr_or_lf      (const GRegex        *regex);
+GLIB_AVAILABLE_IN_2_38
+gint              g_regex_get_max_lookbehind    (const GRegex        *regex);
 GLIB_AVAILABLE_IN_ALL
 gint		  g_regex_get_string_number	(const GRegex        *regex, 
 						 const gchar         *name);
diff --git a/glib/tests/regex.c b/glib/tests/regex.c
index f205bf2..ea7385f 100644
--- a/glib/tests/regex.c
+++ b/glib/tests/regex.c
@@ -2084,6 +2084,24 @@ test_explicit_crlf (void)
   g_regex_unref (regex);
 }
 
+static void
+test_max_lookbehind (void)
+{
+  GRegex *regex;
+
+  regex = g_regex_new ("abc", 0, 0, NULL);
+  g_assert_cmpint (g_regex_get_max_lookbehind (regex), ==, 0);
+  g_regex_unref (regex);
+
+  regex = g_regex_new ("\\babc", 0, 0, NULL);
+  g_assert_cmpint (g_regex_get_max_lookbehind (regex), ==, 1);
+  g_regex_unref (regex);
+
+  regex = g_regex_new ("(?<=123)abc", 0, 0, NULL);
+  g_assert_cmpint (g_regex_get_max_lookbehind (regex), ==, 3);
+  g_regex_unref (regex);
+}
+
 int
 main (int argc, char *argv[])
 {
@@ -2102,6 +2120,7 @@ main (int argc, char *argv[])
   g_test_add_func ("/regex/recursion", test_recursion);
   g_test_add_func ("/regex/multiline", test_multiline);
   g_test_add_func ("/regex/explicit-crlf", test_explicit_crlf);
+  g_test_add_func ("/regex/max-lookbehind", test_max_lookbehind);
 
   /* TEST_NEW(pattern, compile_opts, match_opts) */
   TEST_NEW("[A-Z]+", G_REGEX_CASELESS | G_REGEX_EXTENDED | G_REGEX_OPTIMIZE, G_REGEX_MATCH_NOTBOL | G_REGEX_MATCH_PARTIAL);
-- 
2.7.4