utf8proc

commit 70bbed8626e902d8c1e2b8277b0c61efb8a460bb
parent caef918abd0a9425b3942df3859c7bea7b8986e0
Author: Michael Drake <mike@smoothartist.com>
Date:   Mon, 21 Nov 2016 14:22:39 +0000

Tlsa/ucs4 normalize (#88)

* Split codepoint sequence normalisation out into separate function.

This creates utf8proc_normalize_utf32() which takes and returns
a UTF-32 string, applying the following options:

- UTF8PROC_NLF2LS
- UTF8PROC_NLF2PS
- UTF8PROC_NLF2LF
- UTF8PROC_STRIPCC
- UTF8PROC_COMPOSE
- UTF8PROC_STABLE

The utf8proc_reencode() function has been updated to call the
new utf8proc_normalize_utf32().

* Update code documentation: utf8proc_reencode handles UTF8PROC_CHARBOUND.

Diffstat:
M
utf8proc.c
 | 
13
++++++++++---
M
utf8proc.h
 | 
32
++++++++++++++++++++++++++++++--

2 files changed, 40 insertions(+), 5 deletions(-)
diff --git a/utf8proc.c b/utf8proc.c
@@ -545,9 +545,8 @@ UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
   return wpos;
 }
 
-UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options) {
-  /* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored
-     ASSERT: 'buffer' has one spare byte of free space at the end! */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options) {
+  /* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored */
   if (options & (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS | UTF8PROC_STRIPCC)) {
     utf8proc_ssize_t rpos;
     utf8proc_ssize_t wpos = 0;
@@ -655,6 +654,14 @@ UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, 
     }
     length = wpos;
   }
+  return length;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options) {
+  /* UTF8PROC_NULLTERM option will be ignored, 'length' is never ignored
+     ASSERT: 'buffer' has one spare byte of free space at the end! */
+  length = utf8proc_normalize_utf32(buffer, length, options);
+  if (length < 0) return length;
   {
     utf8proc_ssize_t rpos, wpos = 0;
     utf8proc_int32_t uc;
diff --git a/utf8proc.h b/utf8proc.h
@@ -491,8 +491,34 @@ UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
 );
 
 /**
+ * Normalizes the sequence of `length` codepoints pointed to by `buffer`
+ * in-place (i.e., the result is also stored in `buffer`).
+ *
+ * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode.
+ * @param length the length (in codepoints) of the buffer.
+ * @param options a bitwise or (`|`) of one or more of the following flags:
+ * - @ref UTF8PROC_NLF2LS  - convert LF, CRLF, CR and NEL into LS
+ * - @ref UTF8PROC_NLF2PS  - convert LF, CRLF, CR and NEL into PS
+ * - @ref UTF8PROC_NLF2LF  - convert LF, CRLF, CR and NEL into LF
+ * - @ref UTF8PROC_STRIPCC - strip or convert all non-affected control characters
+ * - @ref UTF8PROC_COMPOSE - try to combine decomposed codepoints into composite
+ *                           codepoints
+ * - @ref UTF8PROC_STABLE  - prohibit combining characters that would violate
+ *                           the unicode versioning stability
+ *
+ * @return
+ * In case of success, the length (in codepoints) of the normalized UTF-32 string is
+ * returned; otherwise, a negative error code is returned (@ref utf8proc_errmsg).
+ *
+ * @warning The entries of the array pointed to by `str` have to be in the
+ *          range `0x0000` to `0x10FFFF`. Otherwise, the program might crash!
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_normalize_utf32(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options);
+
+/**
  * Reencodes the sequence of `length` codepoints pointed to by `buffer`
  * UTF-8 data in-place (i.e., the result is also stored in `buffer`).
+ * Can optionally normalize the UTF-32 sequence prior to UTF-8 conversion.
  *
  * @param buffer the (native-endian UTF-32) unicode codepoints to re-encode.
  * @param length the length (in codepoints) of the buffer.
@@ -505,10 +531,12 @@ UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
  *                           codepoints
  * - @ref UTF8PROC_STABLE  - prohibit combining characters that would violate
  *                           the unicode versioning stability
+ * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster
  *
  * @return
- * In case of success, the length (in bytes) of the resulting UTF-8 string is
- * returned; otherwise, a negative error code is returned (@ref utf8proc_errmsg).
+ * In case of success, the length (in bytes) of the resulting nul-terminated
+ * UTF-8 string is returned; otherwise, a negative error code is returned
+ * (@ref utf8proc_errmsg).
  *
  * @warning The amount of free space pointed to by `buffer` must
  *          exceed the amount of the input data by one byte, and the