Mercurial > hg > octave-lojdl > gnulib-hg

changeset 11662:8e561c7cff21

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
New functions for case mapping of substrings.
author Bruno Haible <bruno@clisp.org>
date Mon, 29 Jun 2009 23:21:51 +0200
parents 06bbe9b0bb80
children 328bb9352b98
files ChangeLog lib/unicase.h
diffstat 2 files changed, 171 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,22 @@
+2009-06-29  Bruno Haible  <bruno@clisp.org>
+
+	New functions for case mapping of substrings.
+	* lib/unicase.h (casing_prefix_context_t): New type.
+	(unicase_empty_prefix_context): New variable.
+	(u8_casing_prefix_context, u16_casing_prefix_context,
+	u32_casing_prefix_context, u8_casing_prefixes_context,
+	u16_casing_prefixes_context, u32_casing_prefixes_context): New
+	declarations.
+	(casing_suffix_context_t): New type.
+	(unicase_empty_suffix_context): New variable.
+	(u8_casing_suffix_context, u16_casing_suffix_context,
+	u32_casing_suffix_context, u8_casing_suffixes_context,
+	u16_casing_suffixes_context, u32_casing_suffixes_context,
+	u8_ct_toupper, u16_ct_toupper, u32_ct_toupper, u8_ct_tolower,
+	u16_ct_tolower, u32_ct_tolower, u8_ct_totitle, u16_ct_totitle,
+	u32_ct_totitle, u8_ct_casefold, u16_ct_casefold, u32_ct_casefold): New
+	declarations.
+
 2009-06-28  Jim Meyering  <meyering@redhat.com>
 
 	boostrap: indent only with spaces
--- a/lib/unicase.h
+++ b/lib/unicase.h
@@ -134,6 +134,136 @@
 		    uninorm_t nf,
 		    uint32_t *resultbuf, size_t *lengthp);
 
+/* The case-mapping context given by a prefix string.  */
+typedef struct casing_prefix_context
+	{
+	  /* These fields are private, undocumented.  */
+	  uint32_t last_char_except_ignorable;
+	  uint32_t last_char_normal_or_above;
+	}
+	casing_prefix_context_t;
+/* The case-mapping context of the empty prefix string.  */
+extern const casing_prefix_context_t unicase_empty_prefix_context;
+/* Return the case-mapping context of a given prefix string.  */
+extern casing_prefix_context_t
+       u8_casing_prefix_context (const uint8_t *s, size_t n);
+extern casing_prefix_context_t
+       u16_casing_prefix_context (const uint16_t *s, size_t n);
+extern casing_prefix_context_t
+       u32_casing_prefix_context (const uint32_t *s, size_t n);
+/* Return the case-mapping context of the prefix concat(A, S), given the
+   case-mapping context of the prefix A.  */
+extern casing_prefix_context_t
+       u8_casing_prefixes_context (const uint8_t *s, size_t n,
+				   casing_prefix_context_t a_context);
+extern casing_prefix_context_t
+       u16_casing_prefixes_context (const uint16_t *s, size_t n,
+				    casing_prefix_context_t a_context);
+extern casing_prefix_context_t
+       u32_casing_prefixes_context (const uint32_t *s, size_t n,
+				    casing_prefix_context_t a_context);
+
+/* The case-mapping context given by a suffix string.  */
+typedef struct casing_suffix_context
+	{
+	  /* These fields are private, undocumented.  */
+	  uint32_t bits;
+	  uint32_t unused_bits;
+	}
+	casing_suffix_context_t;
+/* The case-mapping context of the empty suffix string.  */
+extern const casing_suffix_context_t unicase_empty_suffix_context;
+/* Return the case-mapping context of a given suffix string.  */
+extern casing_suffix_context_t
+       u8_casing_suffix_context (const uint8_t *s, size_t n);
+extern casing_suffix_context_t
+       u16_casing_suffix_context (const uint16_t *s, size_t n);
+extern casing_suffix_context_t
+       u32_casing_suffix_context (const uint32_t *s, size_t n);
+/* Return the case-mapping context of the suffix concat(S, A), given the
+   case-mapping context of the suffix A.  */
+extern casing_suffix_context_t
+       u8_casing_suffixes_context (const uint8_t *s, size_t n,
+				   casing_suffix_context_t a_context);
+extern casing_suffix_context_t
+       u16_casing_suffixes_context (const uint16_t *s, size_t n,
+				    casing_suffix_context_t a_context);
+extern casing_suffix_context_t
+       u32_casing_suffixes_context (const uint32_t *s, size_t n,
+				    casing_suffix_context_t a_context);
+
+/* Return the uppercase mapping of a string that is surrounded by a prefix
+   and a suffix.  */
+extern uint8_t *
+       u8_ct_toupper (const uint8_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint8_t *resultbuf, size_t *lengthp);
+extern uint16_t *
+       u16_ct_toupper (const uint16_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint16_t *resultbuf, size_t *lengthp);
+extern uint32_t *
+       u32_ct_toupper (const uint32_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint32_t *resultbuf, size_t *lengthp);
+
+/* Return the lowercase mapping of a string that is surrounded by a prefix
+   and a suffix.  */
+extern uint8_t *
+       u8_ct_tolower (const uint8_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint8_t *resultbuf, size_t *lengthp);
+extern uint16_t *
+       u16_ct_tolower (const uint16_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint16_t *resultbuf, size_t *lengthp);
+extern uint32_t *
+       u32_ct_tolower (const uint32_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint32_t *resultbuf, size_t *lengthp);
+
+/* Return the titlecase mapping of a string that is surrounded by a prefix
+   and a suffix.  */
+extern uint8_t *
+       u8_ct_totitle (const uint8_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint8_t *resultbuf, size_t *lengthp);
+extern uint16_t *
+       u16_ct_totitle (const uint16_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint16_t *resultbuf, size_t *lengthp);
+extern uint32_t *
+       u32_ct_totitle (const uint32_t *s, size_t n,
+		      casing_prefix_context_t prefix_context,
+		      casing_suffix_context_t suffix_context,
+		      const char *iso639_language,
+		      uninorm_t nf,
+		      uint32_t *resultbuf, size_t *lengthp);
+
 /* Return the case folded string.
    Comparing uN_casefold (S1) and uN_casefold (S2) with uN_cmp2() is equivalent
    to comparing S1 and S2 with uN_casecmp().
@@ -151,6 +281,28 @@
        u32_casefold (const uint32_t *s, size_t n, const char *iso639_language,
 		     uninorm_t nf,
 		     uint32_t *resultbuf, size_t *lengthp);
+/* Likewise, for a string that is surrounded by a prefix and a suffix.  */
+extern uint8_t *
+       u8_ct_casefold (const uint8_t *s, size_t n,
+		       casing_prefix_context_t prefix_context,
+		       casing_suffix_context_t suffix_context,
+		       const char *iso639_language,
+		       uninorm_t nf,
+		       uint8_t *resultbuf, size_t *lengthp);
+extern uint16_t *
+       u16_ct_casefold (const uint16_t *s, size_t n,
+			casing_prefix_context_t prefix_context,
+			casing_suffix_context_t suffix_context,
+			const char *iso639_language,
+			uninorm_t nf,
+			uint16_t *resultbuf, size_t *lengthp);
+extern uint32_t *
+       u32_ct_casefold (const uint32_t *s, size_t n,
+			casing_prefix_context_t prefix_context,
+			casing_suffix_context_t suffix_context,
+			const char *iso639_language,
+			uninorm_t nf,
+			uint32_t *resultbuf, size_t *lengthp);
 
 /* Compare S1 and S2, ignoring differences in case and normalization.
    The nf argument identifies the normalization form to apply after the