regex: Add BSR_ANYCRLF and BSR_ANY match options

These flags override the compile option at match time. They use PCRE_BSR_ANYCRLF
and PCRE_BSR_UNICODE, resp., which make \R match only CR, LF and CRLF, or any
Unicode newline character or character sequences, resp.

diff --git a/glib/gregex.c b/glib/gregex.c
index a315411..a0aee80 100644 (file)
--- a/glib/gregex.c
+++ b/glib/gregex.c
@@ -126,7 +126,9 @@
                             G_REGEX_MATCH_NEWLINE_LF       | \
                             G_REGEX_MATCH_NEWLINE_CRLF     | \
                             G_REGEX_MATCH_NEWLINE_ANY      | \
-                            G_REGEX_MATCH_NEWLINE_ANYCRLF)
+                            G_REGEX_MATCH_NEWLINE_ANYCRLF  | \
+                            G_REGEX_MATCH_BSR_ANYCRLF      | \
+                            G_REGEX_MATCH_BSR_ANY)
 
 /* we rely on these flags having the same values */
 G_STATIC_ASSERT (G_REGEX_CASELESS        == PCRE_CASELESS);
@@ -154,6 +156,8 @@ G_STATIC_ASSERT (G_REGEX_MATCH_NEWLINE_LF      == PCRE_NEWLINE_LF);
 G_STATIC_ASSERT (G_REGEX_MATCH_NEWLINE_CRLF    == PCRE_NEWLINE_CRLF);
 G_STATIC_ASSERT (G_REGEX_MATCH_NEWLINE_ANY     == PCRE_NEWLINE_ANY);
 G_STATIC_ASSERT (G_REGEX_MATCH_NEWLINE_ANYCRLF == PCRE_NEWLINE_ANYCRLF);
+G_STATIC_ASSERT (G_REGEX_MATCH_BSR_ANYCRLF     == PCRE_BSR_ANYCRLF);
+G_STATIC_ASSERT (G_REGEX_MATCH_BSR_ANY         == PCRE_BSR_UNICODE);
 
 /* if the string is in UTF-8 use g_utf8_ functions, else use
  * use just +/- 1. */

diff --git a/glib/gregex.h b/glib/gregex.h
index 9045376..8705230 100644 (file)
--- a/glib/gregex.h
+++ b/glib/gregex.h
@@ -346,6 +346,15 @@ typedef enum
  * @G_REGEX_MATCH_NEWLINE_ANYCRLF: Overrides the newline definition set when
  *     creating a new #GRegex; any '\r', '\n', or '\r\n' character sequence
  *     is recognized as a newline. Since: 2.34
+ * @G_REGEX_MATCH_BSR_ANYCRLF: Overrides the newline definition for "\R" set when
+ *     creating a new #GRegex; only '\r', '\n', or '\r\n' character sequences
+ *     are recognized as a newline by "\R". Since: 2.34
+ * @G_REGEX_MATCH_BSR_ANY: Overrides the newline definition for "\R" set when
+ *     creating a new #GRegex; any Unicode newline character or character sequence
+ *     are recognized as a newline by "\R". These are '\r', '\n' and '\rn', and the
+ *     single characters U+000B LINE TABULATION, U+000C FORM FEED (FF),
+ *     U+0085 NEXT LINE (NEL), U+2028 LINE SEPARATOR and
+ *     U+2029 PARAGRAPH SEPARATOR. Since: 2.34
  *
  * Flags specifying match-time options.
  *
@@ -364,7 +373,9 @@ typedef enum
   G_REGEX_MATCH_NEWLINE_LF      = 1 << 21,
   G_REGEX_MATCH_NEWLINE_CRLF    = G_REGEX_MATCH_NEWLINE_CR | G_REGEX_MATCH_NEWLINE_LF,
   G_REGEX_MATCH_NEWLINE_ANY     = 1 << 22,
-  G_REGEX_MATCH_NEWLINE_ANYCRLF = G_REGEX_MATCH_NEWLINE_CR | G_REGEX_MATCH_NEWLINE_ANY
+  G_REGEX_MATCH_NEWLINE_ANYCRLF = G_REGEX_MATCH_NEWLINE_CR | G_REGEX_MATCH_NEWLINE_ANY,
+  G_REGEX_MATCH_BSR_ANYCRLF     = 1 << 23,
+  G_REGEX_MATCH_BSR_ANY         = 1 << 24
 } GRegexMatchFlags;
 
 /**