You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@subversion.apache.org by br...@apache.org on 2017/09/20 20:55:40 UTC
svn commit: r1809081 [3/5] - in /subversion/upstream/utf8proc: ./ bench/
data/ pgsql/ ruby/ test/
Modified: subversion/upstream/utf8proc/utf8proc.c
URL: http://svn.apache.org/viewvc/subversion/upstream/utf8proc/utf8proc.c?rev=1809081&r1=1809080&r2=1809081&view=diff
==============================================================================
--- subversion/upstream/utf8proc/utf8proc.c (original)
+++ subversion/upstream/utf8proc/utf8proc.c Wed Sep 20 20:55:39 2017
@@ -1,4 +1,6 @@
+/* -*- mode: c; c-basic-offset: 2; tab-width: 2; indent-tabs-mode: nil -*- */
/*
+ * Copyright (c) 2015 Steven G. Johnson, Jiahao Chen, Peter Colberg, Tony Kelman, Scott P. Jones, and other contributors.
* Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
*
* Permission is hereby granted, free of charge, to any person obtaining a
@@ -43,7 +45,7 @@
#include "utf8proc_data.c"
-const int8_t utf8proc_utf8class[256] = {
+UTF8PROC_DLLEXPORT const utf8proc_int8_t utf8proc_utf8class[256] = {
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
@@ -81,25 +83,16 @@ const int8_t utf8proc_utf8class[256] = {
#define UTF8PROC_HANGUL_S_START 0xAC00
#define UTF8PROC_HANGUL_S_END 0xD7A4
-
-#define UTF8PROC_BOUNDCLASS_START 0
-#define UTF8PROC_BOUNDCLASS_OTHER 1
-#define UTF8PROC_BOUNDCLASS_CR 2
-#define UTF8PROC_BOUNDCLASS_LF 3
-#define UTF8PROC_BOUNDCLASS_CONTROL 4
-#define UTF8PROC_BOUNDCLASS_EXTEND 5
-#define UTF8PROC_BOUNDCLASS_L 6
-#define UTF8PROC_BOUNDCLASS_V 7
-#define UTF8PROC_BOUNDCLASS_T 8
-#define UTF8PROC_BOUNDCLASS_LV 9
-#define UTF8PROC_BOUNDCLASS_LVT 10
-
-
-const char *utf8proc_version(void) {
- return "1.1.5";
+/* Should follow semantic-versioning rules (semver.org) based on API
+ compatibility. (Note that the shared-library version number will
+ be different, being based on ABI compatibility.): */
+#define STRINGIZEx(x) #x
+#define STRINGIZE(x) STRINGIZEx(x)
+UTF8PROC_DLLEXPORT const char *utf8proc_version(void) {
+ return STRINGIZE(UTF8PROC_VERSION_MAJOR) "." STRINGIZE(UTF8PROC_VERSION_MINOR) "." STRINGIZE(UTF8PROC_VERSION_PATCH) "";
}
-const char *utf8proc_errmsg(ssize_t errcode) {
+UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode) {
switch (errcode) {
case UTF8PROC_ERROR_NOMEM:
return "Memory for processing UTF-8 data could not be allocated.";
@@ -112,88 +105,122 @@ const char *utf8proc_errmsg(ssize_t errc
case UTF8PROC_ERROR_INVALIDOPTS:
return "Invalid options for UTF-8 processing chosen.";
default:
- return "An unknown error occured while processing UTF-8 data.";
+ return "An unknown error occurred while processing UTF-8 data.";
}
}
-ssize_t utf8proc_iterate(
- const uint8_t *str, ssize_t strlen, int32_t *dst
+#define utf_cont(ch) (((ch) & 0xc0) == 0x80)
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *dst
) {
- int length;
- int i;
- int32_t uc = -1;
+ utf8proc_uint32_t uc;
+ const utf8proc_uint8_t *end;
+
*dst = -1;
if (!strlen) return 0;
- length = utf8proc_utf8class[str[0]];
- if (!length) return UTF8PROC_ERROR_INVALIDUTF8;
- if (strlen >= 0 && length > strlen) return UTF8PROC_ERROR_INVALIDUTF8;
- for (i=1; i<length; i++) {
- if ((str[i] & 0xC0) != 0x80) return UTF8PROC_ERROR_INVALIDUTF8;
- }
- switch (length) {
- case 1:
- uc = str[0];
- break;
- case 2:
- uc = ((str[0] & 0x1F) << 6) + (str[1] & 0x3F);
- if (uc < 0x80) uc = -1;
- break;
- case 3:
- uc = ((str[0] & 0x0F) << 12) + ((str[1] & 0x3F) << 6)
- + (str[2] & 0x3F);
- if (uc < 0x800 || (uc >= 0xD800 && uc < 0xE000) ||
- (uc >= 0xFDD0 && uc < 0xFDF0)) uc = -1;
- break;
- case 4:
- uc = ((str[0] & 0x07) << 18) + ((str[1] & 0x3F) << 12)
- + ((str[2] & 0x3F) << 6) + (str[3] & 0x3F);
- if (uc < 0x10000 || uc >= 0x110000) uc = -1;
- break;
- }
- if (uc < 0 || ((uc & 0xFFFF) >= 0xFFFE))
- return UTF8PROC_ERROR_INVALIDUTF8;
- *dst = uc;
- return length;
+ end = str + ((strlen < 0) ? 4 : strlen);
+ uc = *str++;
+ if (uc < 0x80) {
+ *dst = uc;
+ return 1;
+ }
+ // Must be between 0xc2 and 0xf4 inclusive to be valid
+ if ((uc - 0xc2) > (0xf4-0xc2)) return UTF8PROC_ERROR_INVALIDUTF8;
+ if (uc < 0xe0) { // 2-byte sequence
+ // Must have valid continuation character
+ if (str >= end || !utf_cont(*str)) return UTF8PROC_ERROR_INVALIDUTF8;
+ *dst = ((uc & 0x1f)<<6) | (*str & 0x3f);
+ return 2;
+ }
+ if (uc < 0xf0) { // 3-byte sequence
+ if ((str + 1 >= end) || !utf_cont(*str) || !utf_cont(str[1]))
+ return UTF8PROC_ERROR_INVALIDUTF8;
+ // Check for surrogate chars
+ if (uc == 0xed && *str > 0x9f)
+ return UTF8PROC_ERROR_INVALIDUTF8;
+ uc = ((uc & 0xf)<<12) | ((*str & 0x3f)<<6) | (str[1] & 0x3f);
+ if (uc < 0x800)
+ return UTF8PROC_ERROR_INVALIDUTF8;
+ *dst = uc;
+ return 3;
+ }
+ // 4-byte sequence
+ // Must have 3 valid continuation characters
+ if ((str + 2 >= end) || !utf_cont(*str) || !utf_cont(str[1]) || !utf_cont(str[2]))
+ return UTF8PROC_ERROR_INVALIDUTF8;
+ // Make sure in correct range (0x10000 - 0x10ffff)
+ if (uc == 0xf0) {
+ if (*str < 0x90) return UTF8PROC_ERROR_INVALIDUTF8;
+ } else if (uc == 0xf4) {
+ if (*str > 0x8f) return UTF8PROC_ERROR_INVALIDUTF8;
+ }
+ *dst = ((uc & 7)<<18) | ((*str & 0x3f)<<12) | ((str[1] & 0x3f)<<6) | (str[2] & 0x3f);
+ return 4;
}
-bool utf8proc_codepoint_valid(int32_t uc) {
- if (uc < 0 || uc >= 0x110000 ||
- ((uc & 0xFFFF) >= 0xFFFE) || (uc >= 0xD800 && uc < 0xE000) ||
- (uc >= 0xFDD0 && uc < 0xFDF0)) return false;
- else return true;
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t uc) {
+ return (((utf8proc_uint32_t)uc)-0xd800 > 0x07ff) && ((utf8proc_uint32_t)uc < 0x110000);
}
-ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst) {
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t uc, utf8proc_uint8_t *dst) {
if (uc < 0x00) {
return 0;
} else if (uc < 0x80) {
- dst[0] = uc;
+ dst[0] = (utf8proc_uint8_t) uc;
return 1;
} else if (uc < 0x800) {
- dst[0] = 0xC0 + (uc >> 6);
- dst[1] = 0x80 + (uc & 0x3F);
+ dst[0] = (utf8proc_uint8_t)(0xC0 + (uc >> 6));
+ dst[1] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 2;
- } else if (uc == 0xFFFF) {
- dst[0] = 0xFF;
- return 1;
- } else if (uc == 0xFFFE) {
- dst[0] = 0xFE;
- return 1;
+ // Note: we allow encoding 0xd800-0xdfff here, so as not to change
+ // the API, however, these are actually invalid in UTF-8
} else if (uc < 0x10000) {
- dst[0] = 0xE0 + (uc >> 12);
- dst[1] = 0x80 + ((uc >> 6) & 0x3F);
- dst[2] = 0x80 + (uc & 0x3F);
+ dst[0] = (utf8proc_uint8_t)(0xE0 + (uc >> 12));
+ dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
+ dst[2] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 3;
} else if (uc < 0x110000) {
- dst[0] = 0xF0 + (uc >> 18);
- dst[1] = 0x80 + ((uc >> 12) & 0x3F);
- dst[2] = 0x80 + ((uc >> 6) & 0x3F);
- dst[3] = 0x80 + (uc & 0x3F);
+ dst[0] = (utf8proc_uint8_t)(0xF0 + (uc >> 18));
+ dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 12) & 0x3F));
+ dst[2] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
+ dst[3] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
return 4;
} else return 0;
}
-const utf8proc_property_t *utf8proc_get_property(int32_t uc) {
+/* internal "unsafe" version that does not check whether uc is in range */
+static utf8proc_ssize_t unsafe_encode_char(utf8proc_int32_t uc, utf8proc_uint8_t *dst) {
+ if (uc < 0x00) {
+ return 0;
+ } else if (uc < 0x80) {
+ dst[0] = (utf8proc_uint8_t)uc;
+ return 1;
+ } else if (uc < 0x800) {
+ dst[0] = (utf8proc_uint8_t)(0xC0 + (uc >> 6));
+ dst[1] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
+ return 2;
+ } else if (uc == 0xFFFF) {
+ dst[0] = (utf8proc_uint8_t)0xFF;
+ return 1;
+ } else if (uc == 0xFFFE) {
+ dst[0] = (utf8proc_uint8_t)0xFE;
+ return 1;
+ } else if (uc < 0x10000) {
+ dst[0] = (utf8proc_uint8_t)(0xE0 + (uc >> 12));
+ dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
+ dst[2] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
+ return 3;
+ } else if (uc < 0x110000) {
+ dst[0] = (utf8proc_uint8_t)(0xF0 + (uc >> 18));
+ dst[1] = (utf8proc_uint8_t)(0x80 + ((uc >> 12) & 0x3F));
+ dst[2] = (utf8proc_uint8_t)(0x80 + ((uc >> 6) & 0x3F));
+ dst[3] = (utf8proc_uint8_t)(0x80 + (uc & 0x3F));
+ return 4;
+ } else return 0;
+}
+
+/* internal "unsafe" version that does not check whether uc is in range */
+static const utf8proc_property_t *unsafe_get_property(utf8proc_int32_t uc) {
/* ASSERT: uc >= 0 && uc < 0x110000 */
return utf8proc_properties + (
utf8proc_stage2table[
@@ -202,22 +229,183 @@ const utf8proc_property_t *utf8proc_get_
);
}
+UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t uc) {
+ return uc < 0 || uc >= 0x110000 ? utf8proc_properties : unsafe_get_property(uc);
+}
+
+/* return whether there is a grapheme break between boundclasses lbc and tbc
+ (according to the definition of extended grapheme clusters)
+
+ Rule numbering refers to TR29 Version 29 (Unicode 9.0.0):
+ http://www.unicode.org/reports/tr29/tr29-29.html
+
+ CAVEATS:
+ Please note that evaluation of GB10 (grapheme breaks between emoji zwj sequences)
+ and GB 12/13 (regional indicator code points) require knowledge of previous characters
+ and are thus not handled by this function. This may result in an incorrect break before
+ an E_Modifier class codepoint and an incorrectly missing break between two
+ REGIONAL_INDICATOR class code points if such support does not exist in the caller.
+
+ See the special support in grapheme_break_extended, for required bookkeeping by the caller.
+*/
+static utf8proc_bool grapheme_break_simple(int lbc, int tbc) {
+ return
+ (lbc == UTF8PROC_BOUNDCLASS_START) ? true : // GB1
+ (lbc == UTF8PROC_BOUNDCLASS_CR && // GB3
+ tbc == UTF8PROC_BOUNDCLASS_LF) ? false : // ---
+ (lbc >= UTF8PROC_BOUNDCLASS_CR && lbc <= UTF8PROC_BOUNDCLASS_CONTROL) ? true : // GB4
+ (tbc >= UTF8PROC_BOUNDCLASS_CR && tbc <= UTF8PROC_BOUNDCLASS_CONTROL) ? true : // GB5
+ (lbc == UTF8PROC_BOUNDCLASS_L && // GB6
+ (tbc == UTF8PROC_BOUNDCLASS_L || // ---
+ tbc == UTF8PROC_BOUNDCLASS_V || // ---
+ tbc == UTF8PROC_BOUNDCLASS_LV || // ---
+ tbc == UTF8PROC_BOUNDCLASS_LVT)) ? false : // ---
+ ((lbc == UTF8PROC_BOUNDCLASS_LV || // GB7
+ lbc == UTF8PROC_BOUNDCLASS_V) && // ---
+ (tbc == UTF8PROC_BOUNDCLASS_V || // ---
+ tbc == UTF8PROC_BOUNDCLASS_T)) ? false : // ---
+ ((lbc == UTF8PROC_BOUNDCLASS_LVT || // GB8
+ lbc == UTF8PROC_BOUNDCLASS_T) && // ---
+ tbc == UTF8PROC_BOUNDCLASS_T) ? false : // ---
+ (tbc == UTF8PROC_BOUNDCLASS_EXTEND || // GB9
+ tbc == UTF8PROC_BOUNDCLASS_ZWJ || // ---
+ tbc == UTF8PROC_BOUNDCLASS_SPACINGMARK || // GB9a
+ lbc == UTF8PROC_BOUNDCLASS_PREPEND) ? false : // GB9b
+ ((lbc == UTF8PROC_BOUNDCLASS_E_BASE || // GB10 (requires additional handling below)
+ lbc == UTF8PROC_BOUNDCLASS_E_BASE_GAZ) && // ----
+ tbc == UTF8PROC_BOUNDCLASS_E_MODIFIER) ? false : // ----
+ (lbc == UTF8PROC_BOUNDCLASS_ZWJ && // GB11
+ (tbc == UTF8PROC_BOUNDCLASS_GLUE_AFTER_ZWJ || // ----
+ tbc == UTF8PROC_BOUNDCLASS_E_BASE_GAZ)) ? false : // ----
+ (lbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR && // GB12/13 (requires additional handling below)
+ tbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR) ? false : // ----
+ true; // GB999
+}
+
+static utf8proc_bool grapheme_break_extended(int lbc, int tbc, utf8proc_int32_t *state)
+{
+ int lbc_override = lbc;
+ if (state && *state != UTF8PROC_BOUNDCLASS_START)
+ lbc_override = *state;
+ utf8proc_bool break_permitted = grapheme_break_simple(lbc_override, tbc);
+ if (state) {
+ // Special support for GB 12/13 made possible by GB999. After two RI
+ // class codepoints we want to force a break. Do this by resetting the
+ // second RI's bound class to UTF8PROC_BOUNDCLASS_OTHER, to force a break
+ // after that character according to GB999 (unless of course such a break is
+ // forbidden by a different rule such as GB9).
+ if (*state == tbc && tbc == UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR)
+ *state = UTF8PROC_BOUNDCLASS_OTHER;
+ // Special support for GB10. Fold any EXTEND codepoints into the previous
+ // boundclass if we're dealing with an emoji base boundclass.
+ else if ((*state == UTF8PROC_BOUNDCLASS_E_BASE ||
+ *state == UTF8PROC_BOUNDCLASS_E_BASE_GAZ) &&
+ tbc == UTF8PROC_BOUNDCLASS_EXTEND)
+ *state = UTF8PROC_BOUNDCLASS_E_BASE;
+ else
+ *state = tbc;
+ }
+ return break_permitted;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful(
+ utf8proc_int32_t c1, utf8proc_int32_t c2, utf8proc_int32_t *state) {
+
+ return grapheme_break_extended(utf8proc_get_property(c1)->boundclass,
+ utf8proc_get_property(c2)->boundclass,
+ state);
+}
+
+
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break(
+ utf8proc_int32_t c1, utf8proc_int32_t c2) {
+ return utf8proc_grapheme_break_stateful(c1, c2, NULL);
+}
+
+static utf8proc_int32_t seqindex_decode_entry(const utf8proc_uint16_t **entry)
+{
+ utf8proc_int32_t entry_cp = **entry;
+ if ((entry_cp & 0xF800) == 0xD800) {
+ *entry = *entry + 1;
+ entry_cp = ((entry_cp & 0x03FF) << 10) | (**entry & 0x03FF);
+ entry_cp += 0x10000;
+ }
+ return entry_cp;
+}
+
+static utf8proc_int32_t seqindex_decode_index(const utf8proc_uint32_t seqindex)
+{
+ const utf8proc_uint16_t *entry = &utf8proc_sequences[seqindex];
+ return seqindex_decode_entry(&entry);
+}
+
+static utf8proc_ssize_t seqindex_write_char_decomposed(utf8proc_uint16_t seqindex, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, utf8proc_option_t options, int *last_boundclass) {
+ utf8proc_ssize_t written = 0;
+ const utf8proc_uint16_t *entry = &utf8proc_sequences[seqindex & 0x1FFF];
+ int len = seqindex >> 13;
+ if (len >= 7) {
+ len = *entry;
+ entry++;
+ }
+ for (; len >= 0; entry++, len--) {
+ utf8proc_int32_t entry_cp = seqindex_decode_entry(&entry);
+
+ written += utf8proc_decompose_char(entry_cp, dst+written,
+ (bufsize > written) ? (bufsize - written) : 0, options,
+ last_boundclass);
+ if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
+ }
+ return written;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c)
+{
+ utf8proc_int32_t cl = utf8proc_get_property(c)->lowercase_seqindex;
+ return cl != UINT16_MAX ? seqindex_decode_index(cl) : c;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c)
+{
+ utf8proc_int32_t cu = utf8proc_get_property(c)->uppercase_seqindex;
+ return cu != UINT16_MAX ? seqindex_decode_index(cu) : c;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c)
+{
+ utf8proc_int32_t cu = utf8proc_get_property(c)->titlecase_seqindex;
+ return cu != UINT16_MAX ? seqindex_decode_index(cu) : c;
+}
+
+/* return a character width analogous to wcwidth (except portable and
+ hopefully less buggy than most system wcwidth functions). */
+UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t c) {
+ return utf8proc_get_property(c)->charwidth;
+}
+
+UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t c) {
+ return utf8proc_get_property(c)->category;
+}
+
+UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t c) {
+ static const char s[][3] = {"Cn","Lu","Ll","Lt","Lm","Lo","Mn","Mc","Me","Nd","Nl","No","Pc","Pd","Ps","Pe","Pi","Pf","Po","Sm","Sc","Sk","So","Zs","Zl","Zp","Cc","Cf","Cs","Co"};
+ return s[utf8proc_category(c)];
+}
+
#define utf8proc_decompose_lump(replacement_uc) \
return utf8proc_decompose_char((replacement_uc), dst, bufsize, \
options & ~UTF8PROC_LUMP, last_boundclass)
-ssize_t utf8proc_decompose_char(int32_t uc, int32_t *dst, ssize_t bufsize,
- int options, int *last_boundclass) {
- /* ASSERT: uc >= 0 && uc < 0x110000 */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char(utf8proc_int32_t uc, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize, utf8proc_option_t options, int *last_boundclass) {
const utf8proc_property_t *property;
utf8proc_propval_t category;
- int32_t hangul_sindex;
- property = utf8proc_get_property(uc);
+ utf8proc_int32_t hangul_sindex;
+ if (uc < 0 || uc >= 0x110000) return UTF8PROC_ERROR_NOTASSIGNED;
+ property = unsafe_get_property(uc);
category = property->category;
hangul_sindex = uc - UTF8PROC_HANGUL_SBASE;
if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT) {
- int32_t hangul_tindex;
+ utf8proc_int32_t hangul_tindex;
if (bufsize >= 1) {
dst[0] = UTF8PROC_HANGUL_LBASE +
hangul_sindex / UTF8PROC_HANGUL_NCOUNT;
@@ -268,79 +456,20 @@ ssize_t utf8proc_decompose_char(int32_t
category == UTF8PROC_CATEGORY_ME) return 0;
}
if (options & UTF8PROC_CASEFOLD) {
- if (property->casefold_mapping) {
- const int32_t *casefold_entry;
- ssize_t written = 0;
- for (casefold_entry = property->casefold_mapping;
- *casefold_entry >= 0; casefold_entry++) {
- written += utf8proc_decompose_char(*casefold_entry, dst+written,
- (bufsize > written) ? (bufsize - written) : 0, options,
- last_boundclass);
- if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
- }
- return written;
+ if (property->casefold_seqindex != UINT16_MAX) {
+ return seqindex_write_char_decomposed(property->casefold_seqindex, dst, bufsize, options, last_boundclass);
}
}
if (options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) {
- if (property->decomp_mapping &&
+ if (property->decomp_seqindex != UINT16_MAX &&
(!property->decomp_type || (options & UTF8PROC_COMPAT))) {
- const int32_t *decomp_entry;
- ssize_t written = 0;
- for (decomp_entry = property->decomp_mapping;
- *decomp_entry >= 0; decomp_entry++) {
- written += utf8proc_decompose_char(*decomp_entry, dst+written,
- (bufsize > written) ? (bufsize - written) : 0, options,
- last_boundclass);
- if (written < 0) return UTF8PROC_ERROR_OVERFLOW;
- }
- return written;
+ return seqindex_write_char_decomposed(property->decomp_seqindex, dst, bufsize, options, last_boundclass);
}
}
if (options & UTF8PROC_CHARBOUND) {
- bool boundary;
- int tbc, lbc;
- tbc =
- (uc == 0x000D) ? UTF8PROC_BOUNDCLASS_CR :
- (uc == 0x000A) ? UTF8PROC_BOUNDCLASS_LF :
- ((category == UTF8PROC_CATEGORY_ZL ||
- category == UTF8PROC_CATEGORY_ZP ||
- category == UTF8PROC_CATEGORY_CC ||
- category == UTF8PROC_CATEGORY_CF) &&
- !(uc == 0x200C || uc == 0x200D)) ? UTF8PROC_BOUNDCLASS_CONTROL :
- property->extend ? UTF8PROC_BOUNDCLASS_EXTEND :
- ((uc >= UTF8PROC_HANGUL_L_START && uc < UTF8PROC_HANGUL_L_END) ||
- uc == UTF8PROC_HANGUL_L_FILLER) ? UTF8PROC_BOUNDCLASS_L :
- (uc >= UTF8PROC_HANGUL_V_START && uc < UTF8PROC_HANGUL_V_END) ?
- UTF8PROC_BOUNDCLASS_V :
- (uc >= UTF8PROC_HANGUL_T_START && uc < UTF8PROC_HANGUL_T_END) ?
- UTF8PROC_BOUNDCLASS_T :
- (uc >= UTF8PROC_HANGUL_S_START && uc < UTF8PROC_HANGUL_S_END) ? (
- ((uc-UTF8PROC_HANGUL_SBASE) % UTF8PROC_HANGUL_TCOUNT == 0) ?
- UTF8PROC_BOUNDCLASS_LV : UTF8PROC_BOUNDCLASS_LVT
- ) :
- UTF8PROC_BOUNDCLASS_OTHER;
- lbc = *last_boundclass;
- boundary =
- (tbc == UTF8PROC_BOUNDCLASS_EXTEND) ? false :
- (lbc == UTF8PROC_BOUNDCLASS_START) ? true :
- (lbc == UTF8PROC_BOUNDCLASS_CR &&
- tbc == UTF8PROC_BOUNDCLASS_LF) ? false :
- (lbc == UTF8PROC_BOUNDCLASS_CONTROL) ? true :
- (tbc == UTF8PROC_BOUNDCLASS_CONTROL) ? true :
- (lbc == UTF8PROC_BOUNDCLASS_L &&
- (tbc == UTF8PROC_BOUNDCLASS_L ||
- tbc == UTF8PROC_BOUNDCLASS_V ||
- tbc == UTF8PROC_BOUNDCLASS_LV ||
- tbc == UTF8PROC_BOUNDCLASS_LVT)) ? false :
- ((lbc == UTF8PROC_BOUNDCLASS_LV ||
- lbc == UTF8PROC_BOUNDCLASS_V) &&
- (tbc == UTF8PROC_BOUNDCLASS_V ||
- tbc == UTF8PROC_BOUNDCLASS_T)) ? false :
- ((lbc == UTF8PROC_BOUNDCLASS_LVT ||
- lbc == UTF8PROC_BOUNDCLASS_T) &&
- tbc == UTF8PROC_BOUNDCLASS_T) ? false :
- true;
- *last_boundclass = tbc;
+ utf8proc_bool boundary;
+ int tbc = property->boundclass;
+ boundary = grapheme_break_extended(*last_boundclass, tbc, last_boundclass);
if (boundary) {
if (bufsize >= 1) dst[0] = 0xFFFF;
if (bufsize >= 2) dst[1] = uc;
@@ -351,26 +480,34 @@ ssize_t utf8proc_decompose_char(int32_t
return 1;
}
-ssize_t utf8proc_decompose(
- const uint8_t *str, ssize_t strlen,
- int32_t *buffer, ssize_t bufsize, int options
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
+ utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options
+) {
+ return utf8proc_decompose_custom(str, strlen, buffer, bufsize, options, NULL, NULL);
+}
+
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
+ utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options,
+ utf8proc_custom_func custom_func, void *custom_data
) {
/* strlen will be ignored, if UTF8PROC_NULLTERM is set in options */
- ssize_t wpos = 0;
+ utf8proc_ssize_t wpos = 0;
if ((options & UTF8PROC_COMPOSE) && (options & UTF8PROC_DECOMPOSE))
return UTF8PROC_ERROR_INVALIDOPTS;
if ((options & UTF8PROC_STRIPMARK) &&
!(options & UTF8PROC_COMPOSE) && !(options & UTF8PROC_DECOMPOSE))
return UTF8PROC_ERROR_INVALIDOPTS;
{
- int32_t uc;
- ssize_t rpos = 0;
- ssize_t decomp_result;
+ utf8proc_int32_t uc;
+ utf8proc_ssize_t rpos = 0;
+ utf8proc_ssize_t decomp_result;
int boundclass = UTF8PROC_BOUNDCLASS_START;
while (1) {
if (options & UTF8PROC_NULLTERM) {
rpos += utf8proc_iterate(str + rpos, -1, &uc);
- /* checking of return value is not neccessary,
+ /* checking of return value is not necessary,
as 'uc' is < 0 in case of error */
if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
if (rpos < 0) return UTF8PROC_ERROR_OVERFLOW;
@@ -380,6 +517,9 @@ ssize_t utf8proc_decompose(
rpos += utf8proc_iterate(str + rpos, strlen - rpos, &uc);
if (uc < 0) return UTF8PROC_ERROR_INVALIDUTF8;
}
+ if (custom_func != NULL) {
+ uc = custom_func(uc, custom_data); /* user-specified custom mapping */
+ }
decomp_result = utf8proc_decompose_char(
uc, buffer + wpos, (bufsize > wpos) ? (bufsize - wpos) : 0, options,
&boundclass
@@ -387,19 +527,20 @@ ssize_t utf8proc_decompose(
if (decomp_result < 0) return decomp_result;
wpos += decomp_result;
/* prohibiting integer overflows due to too long strings: */
- if (wpos < 0 || wpos > SSIZE_MAX/sizeof(int32_t)/2)
+ if (wpos < 0 ||
+ wpos > (utf8proc_ssize_t)(SSIZE_MAX/sizeof(utf8proc_int32_t)/2))
return UTF8PROC_ERROR_OVERFLOW;
}
}
if ((options & (UTF8PROC_COMPOSE|UTF8PROC_DECOMPOSE)) && bufsize >= wpos) {
- ssize_t pos = 0;
+ utf8proc_ssize_t pos = 0;
while (pos < wpos-1) {
- int32_t uc1, uc2;
+ utf8proc_int32_t uc1, uc2;
const utf8proc_property_t *property1, *property2;
uc1 = buffer[pos];
uc2 = buffer[pos+1];
- property1 = utf8proc_get_property(uc1);
- property2 = utf8proc_get_property(uc2);
+ property1 = unsafe_get_property(uc1);
+ property2 = unsafe_get_property(uc2);
if (property1->combining_class > property2->combining_class &&
property2->combining_class > 0) {
buffer[pos] = uc2;
@@ -413,13 +554,12 @@ ssize_t utf8proc_decompose(
return wpos;
}
-ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int 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)) {
- ssize_t rpos;
- ssize_t wpos = 0;
- int32_t uc;
+ utf8proc_ssize_t rpos;
+ utf8proc_ssize_t wpos = 0;
+ utf8proc_int32_t uc;
for (rpos = 0; rpos < length; rpos++) {
uc = buffer[rpos];
if (uc == 0x000D && rpos < length-1 && buffer[rpos+1] == 0x000A) rpos++;
@@ -448,23 +588,23 @@ ssize_t utf8proc_reencode(int32_t *buffe
length = wpos;
}
if (options & UTF8PROC_COMPOSE) {
- int32_t *starter = NULL;
- int32_t current_char;
+ utf8proc_int32_t *starter = NULL;
+ utf8proc_int32_t current_char;
const utf8proc_property_t *starter_property = NULL, *current_property;
utf8proc_propval_t max_combining_class = -1;
- ssize_t rpos;
- ssize_t wpos = 0;
- int32_t composition;
+ utf8proc_ssize_t rpos;
+ utf8proc_ssize_t wpos = 0;
+ utf8proc_int32_t composition;
for (rpos = 0; rpos < length; rpos++) {
current_char = buffer[rpos];
- current_property = utf8proc_get_property(current_char);
+ current_property = unsafe_get_property(current_char);
if (starter && current_property->combining_class > max_combining_class) {
/* combination perhaps possible */
- int32_t hangul_lindex;
- int32_t hangul_sindex;
+ utf8proc_int32_t hangul_lindex;
+ utf8proc_int32_t hangul_sindex;
hangul_lindex = *starter - UTF8PROC_HANGUL_LBASE;
if (hangul_lindex >= 0 && hangul_lindex < UTF8PROC_HANGUL_LCOUNT) {
- int32_t hangul_vindex;
+ utf8proc_int32_t hangul_vindex;
hangul_vindex = current_char - UTF8PROC_HANGUL_VBASE;
if (hangul_vindex >= 0 && hangul_vindex < UTF8PROC_HANGUL_VCOUNT) {
*starter = UTF8PROC_HANGUL_SBASE +
@@ -477,7 +617,7 @@ ssize_t utf8proc_reencode(int32_t *buffe
hangul_sindex = *starter - UTF8PROC_HANGUL_SBASE;
if (hangul_sindex >= 0 && hangul_sindex < UTF8PROC_HANGUL_SCOUNT &&
(hangul_sindex % UTF8PROC_HANGUL_TCOUNT) == 0) {
- int32_t hangul_tindex;
+ utf8proc_int32_t hangul_tindex;
hangul_tindex = current_char - UTF8PROC_HANGUL_TBASE;
if (hangul_tindex >= 0 && hangul_tindex < UTF8PROC_HANGUL_TCOUNT) {
*starter += hangul_tindex;
@@ -486,19 +626,26 @@ ssize_t utf8proc_reencode(int32_t *buffe
}
}
if (!starter_property) {
- starter_property = utf8proc_get_property(*starter);
+ starter_property = unsafe_get_property(*starter);
}
- if (starter_property->comb1st_index >= 0 &&
- current_property->comb2nd_index >= 0) {
- composition = utf8proc_combinations[
- starter_property->comb1st_index +
- current_property->comb2nd_index
- ];
- if (composition >= 0 && (!(options & UTF8PROC_STABLE) ||
- !(utf8proc_get_property(composition)->comp_exclusion))) {
- *starter = composition;
- starter_property = NULL;
- continue;
+ if (starter_property->comb_index < 0x8000 &&
+ current_property->comb_index != UINT16_MAX &&
+ current_property->comb_index >= 0x8000) {
+ int sidx = starter_property->comb_index;
+ int idx = (current_property->comb_index & 0x3FFF) - utf8proc_combinations[sidx];
+ if (idx >= 0 && idx <= utf8proc_combinations[sidx + 1] ) {
+ idx += sidx + 2;
+ if (current_property->comb_index & 0x4000) {
+ composition = (utf8proc_combinations[idx] << 16) | utf8proc_combinations[idx+1];
+ } else
+ composition = utf8proc_combinations[idx];
+
+ if (composition > 0 && (!(options & UTF8PROC_STABLE) ||
+ !(unsafe_get_property(composition)->comp_exclusion))) {
+ *starter = composition;
+ starter_property = NULL;
+ continue;
+ }
}
}
}
@@ -516,29 +663,51 @@ ssize_t utf8proc_reencode(int32_t *buffe
}
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;
{
- ssize_t rpos, wpos = 0;
- int32_t uc;
- for (rpos = 0; rpos < length; rpos++) {
- uc = buffer[rpos];
- wpos += utf8proc_encode_char(uc, ((uint8_t *)buffer) + wpos);
+ utf8proc_ssize_t rpos, wpos = 0;
+ utf8proc_int32_t uc;
+ if (options & UTF8PROC_CHARBOUND) {
+ for (rpos = 0; rpos < length; rpos++) {
+ uc = buffer[rpos];
+ wpos += unsafe_encode_char(uc, ((utf8proc_uint8_t *)buffer) + wpos);
+ }
+ } else {
+ for (rpos = 0; rpos < length; rpos++) {
+ uc = buffer[rpos];
+ wpos += utf8proc_encode_char(uc, ((utf8proc_uint8_t *)buffer) + wpos);
+ }
}
- ((uint8_t *)buffer)[wpos] = 0;
+ ((utf8proc_uint8_t *)buffer)[wpos] = 0;
return wpos;
}
}
-ssize_t utf8proc_map(
- const uint8_t *str, ssize_t strlen, uint8_t **dstptr, int options
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options
) {
- int32_t *buffer;
- ssize_t result;
+ return utf8proc_map_custom(str, strlen, dstptr, options, NULL, NULL);
+}
+
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options,
+ utf8proc_custom_func custom_func, void *custom_data
+) {
+ utf8proc_int32_t *buffer;
+ utf8proc_ssize_t result;
*dstptr = NULL;
- result = utf8proc_decompose(str, strlen, NULL, 0, options);
+ result = utf8proc_decompose_custom(str, strlen, NULL, 0, options, custom_func, custom_data);
if (result < 0) return result;
- buffer = malloc(result * sizeof(int32_t) + 1);
+ buffer = (utf8proc_int32_t *) malloc(result * sizeof(utf8proc_int32_t) + 1);
if (!buffer) return UTF8PROC_ERROR_NOMEM;
- result = utf8proc_decompose(str, strlen, buffer, result, options);
+ result = utf8proc_decompose_custom(str, strlen, buffer, result, options, custom_func, custom_data);
if (result < 0) {
free(buffer);
return result;
@@ -549,39 +718,38 @@ ssize_t utf8proc_map(
return result;
}
{
- int32_t *newptr;
- newptr = realloc(buffer, (size_t)result+1);
+ utf8proc_int32_t *newptr;
+ newptr = (utf8proc_int32_t *) realloc(buffer, (size_t)result+1);
if (newptr) buffer = newptr;
}
- *dstptr = (uint8_t *)buffer;
+ *dstptr = (utf8proc_uint8_t *)buffer;
return result;
}
-uint8_t *utf8proc_NFD(const uint8_t *str) {
- uint8_t *retval;
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str) {
+ utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_DECOMPOSE);
return retval;
}
-uint8_t *utf8proc_NFC(const uint8_t *str) {
- uint8_t *retval;
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str) {
+ utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_COMPOSE);
return retval;
}
-uint8_t *utf8proc_NFKD(const uint8_t *str) {
- uint8_t *retval;
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str) {
+ utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_DECOMPOSE | UTF8PROC_COMPAT);
return retval;
}
-uint8_t *utf8proc_NFKC(const uint8_t *str) {
- uint8_t *retval;
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str) {
+ utf8proc_uint8_t *retval;
utf8proc_map(str, 0, &retval, UTF8PROC_NULLTERM | UTF8PROC_STABLE |
UTF8PROC_COMPOSE | UTF8PROC_COMPAT);
return retval;
}
-
Modified: subversion/upstream/utf8proc/utf8proc.h
URL: http://svn.apache.org/viewvc/subversion/upstream/utf8proc/utf8proc.h?rev=1809081&r1=1809080&r2=1809081&view=diff
==============================================================================
--- subversion/upstream/utf8proc/utf8proc.h (original)
+++ subversion/upstream/utf8proc/utf8proc.h Wed Sep 20 20:55:39 2017
@@ -1,77 +1,136 @@
/*
- * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
+ * Copyright (c) 2015 Steven G. Johnson, Jiahao Chen, Peter Colberg, Tony Kelman, Scott P. Jones, and other contributors.
+ * Copyright (c) 2009 Public Software Group e. V., Berlin, Germany
*
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
- * DEALINGS IN THE SOFTWARE.
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
*/
-/*
- * File name: utf8proc.h
+/**
+ * @mainpage
+ *
+ * utf8proc is a free/open-source (MIT/expat licensed) C library
+ * providing Unicode normalization, case-folding, and other operations
+ * for strings in the UTF-8 encoding, supporting Unicode version
+ * 8.0.0. See the utf8proc home page (http://julialang.org/utf8proc/)
+ * for downloads and other information, or the source code on github
+ * (https://github.com/JuliaLang/utf8proc).
+ *
+ * For the utf8proc API documentation, see: @ref utf8proc.h
*
- * Description:
- * Header files for libutf8proc, which is a mapping tool for UTF-8 strings
- * with following features:
- * - decomposing and composing of strings
- * - replacing compatibility characters with their equivalents
- * - stripping of "default ignorable characters"
- * like SOFT-HYPHEN or ZERO-WIDTH-SPACE
- * - folding of certain characters for string comparison
- * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-")
- * (see "LUMP" option)
- * - optional rejection of strings containing non-assigned code points
- * - stripping of control characters
- * - stripping of character marks (accents, etc.)
- * - transformation of LF, CRLF, CR and NEL to line-feed (LF)
- * or to the unicode chararacters for paragraph separation (PS)
- * or line separation (LS).
- * - unicode case folding (for case insensitive string comparisons)
- * - rejection of illegal UTF-8 data
- * (i.e. UTF-8 encoded UTF-16 surrogates)
- * - support for korean hangul characters
- * Unicode Version 5.0.0 is supported.
+ * The features of utf8proc include:
+ *
+ * - Transformation of strings (@ref utf8proc_map) to:
+ * - decompose (@ref UTF8PROC_DECOMPOSE) or compose (@ref UTF8PROC_COMPOSE) Unicode combining characters (http://en.wikipedia.org/wiki/Combining_character)
+ * - canonicalize Unicode compatibility characters (@ref UTF8PROC_COMPAT)
+ * - strip "ignorable" (@ref UTF8PROC_IGNORE) characters, control characters (@ref UTF8PROC_STRIPCC), or combining characters such as accents (@ref UTF8PROC_STRIPMARK)
+ * - case-folding (@ref UTF8PROC_CASEFOLD)
+ * - Unicode normalization: @ref utf8proc_NFD, @ref utf8proc_NFC, @ref utf8proc_NFKD, @ref utf8proc_NFKC
+ * - Detecting grapheme boundaries (@ref utf8proc_grapheme_break and @ref UTF8PROC_CHARBOUND)
+ * - Character-width computation: @ref utf8proc_charwidth
+ * - Classification of characters by Unicode category: @ref utf8proc_category and @ref utf8proc_category_string
+ * - Encode (@ref utf8proc_encode_char) and decode (@ref utf8proc_iterate) Unicode codepoints to/from UTF-8.
*/
+/** @file */
#ifndef UTF8PROC_H
#define UTF8PROC_H
+/** @name API version
+ *
+ * The utf8proc API version MAJOR.MINOR.PATCH, following
+ * semantic-versioning rules (http://semver.org) based on API
+ * compatibility.
+ *
+ * This is also returned at runtime by @ref utf8proc_version; however, the
+ * runtime version may append a string like "-dev" to the version number
+ * for prerelease versions.
+ *
+ * @note The shared-library version number in the Makefile
+ * (and CMakeLists.txt, and MANIFEST) may be different,
+ * being based on ABI compatibility rather than API compatibility.
+ */
+/** @{ */
+/** The MAJOR version number (increased when backwards API compatibility is broken). */
+#define UTF8PROC_VERSION_MAJOR 2
+/** The MINOR version number (increased when new functionality is added in a backwards-compatible manner). */
+#define UTF8PROC_VERSION_MINOR 1
+/** The PATCH version (increased for fixes that do not change the API). */
+#define UTF8PROC_VERSION_PATCH 0
+/** @} */
#include <stdlib.h>
#include <sys/types.h>
-#ifdef _MSC_VER
-typedef signed char int8_t;
-typedef unsigned char uint8_t;
-typedef short int16_t;
-typedef unsigned short uint16_t;
-typedef int int32_t;
-#ifdef _WIN64
-#define ssize_t __int64
+#if defined(_MSC_VER) && _MSC_VER < 1800
+// MSVC prior to 2013 lacked stdbool.h and inttypes.h
+typedef signed char utf8proc_int8_t;
+typedef unsigned char utf8proc_uint8_t;
+typedef short utf8proc_int16_t;
+typedef unsigned short utf8proc_uint16_t;
+typedef int utf8proc_int32_t;
+typedef unsigned int utf8proc_uint32_t;
+# ifdef _WIN64
+typedef __int64 utf8proc_ssize_t;
+typedef unsigned __int64 utf8proc_size_t;
+# else
+typedef int utf8proc_ssize_t;
+typedef unsigned int utf8proc_size_t;
+# endif
+# ifndef __cplusplus
+// emulate C99 bool
+typedef unsigned char utf8proc_bool;
+# ifndef __bool_true_false_are_defined
+# define false 0
+# define true 1
+# define __bool_true_false_are_defined 1
+# endif
+# else
+typedef bool utf8proc_bool;
+# endif
#else
-#define ssize_t int
+# include <stddef.h>
+# include <stdbool.h>
+# include <inttypes.h>
+typedef int8_t utf8proc_int8_t;
+typedef uint8_t utf8proc_uint8_t;
+typedef int16_t utf8proc_int16_t;
+typedef uint16_t utf8proc_uint16_t;
+typedef int32_t utf8proc_int32_t;
+typedef uint32_t utf8proc_uint32_t;
+typedef size_t utf8proc_size_t;
+typedef ptrdiff_t utf8proc_ssize_t;
+typedef bool utf8proc_bool;
#endif
-typedef unsigned char bool;
-enum {false, true};
+#include <limits.h>
+
+#ifdef _WIN32
+# ifdef UTF8PROC_EXPORTS
+# define UTF8PROC_DLLEXPORT __declspec(dllexport)
+# else
+# define UTF8PROC_DLLEXPORT __declspec(dllimport)
+# endif
+#elif __GNUC__ >= 4
+# define UTF8PROC_DLLEXPORT __attribute__ ((visibility("default")))
#else
-#include <stdbool.h>
-#include <inttypes.h>
+# define UTF8PROC_DLLEXPORT
#endif
-#include <limits.h>
#ifdef __cplusplus
extern "C" {
@@ -81,305 +140,560 @@ extern "C" {
#define SSIZE_MAX ((size_t)SIZE_MAX/2)
#endif
-#define UTF8PROC_NULLTERM (1<<0)
-#define UTF8PROC_STABLE (1<<1)
-#define UTF8PROC_COMPAT (1<<2)
-#define UTF8PROC_COMPOSE (1<<3)
-#define UTF8PROC_DECOMPOSE (1<<4)
-#define UTF8PROC_IGNORE (1<<5)
-#define UTF8PROC_REJECTNA (1<<6)
-#define UTF8PROC_NLF2LS (1<<7)
-#define UTF8PROC_NLF2PS (1<<8)
-#define UTF8PROC_NLF2LF (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS)
-#define UTF8PROC_STRIPCC (1<<9)
-#define UTF8PROC_CASEFOLD (1<<10)
-#define UTF8PROC_CHARBOUND (1<<11)
-#define UTF8PROC_LUMP (1<<12)
-#define UTF8PROC_STRIPMARK (1<<13)
-/*
- * Flags being regarded by several functions in the library:
- * NULLTERM: The given UTF-8 input is NULL terminated.
- * STABLE: Unicode Versioning Stability has to be respected.
- * COMPAT: Compatiblity decomposition
- * (i.e. formatting information is lost)
- * COMPOSE: Return a result with composed characters.
- * DECOMPOSE: Return a result with decomposed characters.
- * IGNORE: Strip "default ignorable characters"
- * REJECTNA: Return an error, if the input contains unassigned
- * code points.
- * NLF2LS: Indicating that NLF-sequences (LF, CRLF, CR, NEL) are
- * representing a line break, and should be converted to the
- * unicode character for line separation (LS).
- * NLF2PS: Indicating that NLF-sequences are representing a paragraph
- * break, and should be converted to the unicode character for
- * paragraph separation (PS).
- * NLF2LF: Indicating that the meaning of NLF-sequences is unknown.
- * STRIPCC: Strips and/or convers control characters.
- * NLF-sequences are transformed into space, except if one of
- * the NLF2LS/PS/LF options is given.
- * HorizontalTab (HT) and FormFeed (FF) are treated as a
- * NLF-sequence in this case.
- * All other control characters are simply removed.
- * CASEFOLD: Performs unicode case folding, to be able to do a
- * case-insensitive string comparison.
- * CHARBOUND: Inserts 0xFF bytes at the beginning of each sequence which
- * is representing a single grapheme cluster (see UAX#29).
- * LUMP: Lumps certain characters together
- * (e.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-").
- * (See lump.txt for details.)
- * If NLF2LF is set, this includes a transformation of
- * paragraph and line separators to ASCII line-feed (LF).
- * STRIPMARK: Strips all character markings
- * (non-spacing, spacing and enclosing) (i.e. accents)
- * NOTE: this option works only with COMPOSE or DECOMPOSE
+#ifndef UINT16_MAX
+# define UINT16_MAX 65535U
+#endif
+
+/**
+ * Option flags used by several functions in the library.
*/
+typedef enum {
+ /** The given UTF-8 input is NULL terminated. */
+ UTF8PROC_NULLTERM = (1<<0),
+ /** Unicode Versioning Stability has to be respected. */
+ UTF8PROC_STABLE = (1<<1),
+ /** Compatibility decomposition (i.e. formatting information is lost). */
+ UTF8PROC_COMPAT = (1<<2),
+ /** Return a result with decomposed characters. */
+ UTF8PROC_COMPOSE = (1<<3),
+ /** Return a result with decomposed characters. */
+ UTF8PROC_DECOMPOSE = (1<<4),
+ /** Strip "default ignorable characters" such as SOFT-HYPHEN or ZERO-WIDTH-SPACE. */
+ UTF8PROC_IGNORE = (1<<5),
+ /** Return an error, if the input contains unassigned codepoints. */
+ UTF8PROC_REJECTNA = (1<<6),
+ /**
+ * Indicating that NLF-sequences (LF, CRLF, CR, NEL) are representing a
+ * line break, and should be converted to the codepoint for line
+ * separation (LS).
+ */
+ UTF8PROC_NLF2LS = (1<<7),
+ /**
+ * Indicating that NLF-sequences are representing a paragraph break, and
+ * should be converted to the codepoint for paragraph separation
+ * (PS).
+ */
+ UTF8PROC_NLF2PS = (1<<8),
+ /** Indicating that the meaning of NLF-sequences is unknown. */
+ UTF8PROC_NLF2LF = (UTF8PROC_NLF2LS | UTF8PROC_NLF2PS),
+ /** Strips and/or convers control characters.
+ *
+ * NLF-sequences are transformed into space, except if one of the
+ * NLF2LS/PS/LF options is given. HorizontalTab (HT) and FormFeed (FF)
+ * are treated as a NLF-sequence in this case. All other control
+ * characters are simply removed.
+ */
+ UTF8PROC_STRIPCC = (1<<9),
+ /**
+ * Performs unicode case folding, to be able to do a case-insensitive
+ * string comparison.
+ */
+ UTF8PROC_CASEFOLD = (1<<10),
+ /**
+ * Inserts 0xFF bytes at the beginning of each sequence which is
+ * representing a single grapheme cluster (see UAX#29).
+ */
+ UTF8PROC_CHARBOUND = (1<<11),
+ /** Lumps certain characters together.
+ *
+ * E.g. HYPHEN U+2010 and MINUS U+2212 to ASCII "-". See lump.md for details.
+ *
+ * If NLF2LF is set, this includes a transformation of paragraph and
+ * line separators to ASCII line-feed (LF).
+ */
+ UTF8PROC_LUMP = (1<<12),
+ /** Strips all character markings.
+ *
+ * This includes non-spacing, spacing and enclosing (i.e. accents).
+ * @note This option works only with @ref UTF8PROC_COMPOSE or
+ * @ref UTF8PROC_DECOMPOSE
+ */
+ UTF8PROC_STRIPMARK = (1<<13),
+} utf8proc_option_t;
+/** @name Error codes
+ * Error codes being returned by almost all functions.
+ */
+/** @{ */
+/** Memory could not be allocated. */
#define UTF8PROC_ERROR_NOMEM -1
+/** The given string is too long to be processed. */
#define UTF8PROC_ERROR_OVERFLOW -2
+/** The given string is not a legal UTF-8 string. */
#define UTF8PROC_ERROR_INVALIDUTF8 -3
+/** The @ref UTF8PROC_REJECTNA flag was set and an unassigned codepoint was found. */
#define UTF8PROC_ERROR_NOTASSIGNED -4
+/** Invalid options have been used. */
#define UTF8PROC_ERROR_INVALIDOPTS -5
-/*
- * Error codes being returned by almost all functions:
- * ERROR_NOMEM: Memory could not be allocated.
- * ERROR_OVERFLOW: The given string is too long to be processed.
- * ERROR_INVALIDUTF8: The given string is not a legal UTF-8 string.
- * ERROR_NOTASSIGNED: The REJECTNA flag was set,
- * and an unassigned code point was found.
- * ERROR_INVALIDOPTS: Invalid options have been used.
- */
+/** @} */
-typedef int16_t utf8proc_propval_t;
+/* @name Types */
+
+/** Holds the value of a property. */
+typedef utf8proc_int16_t utf8proc_propval_t;
+
+/** Struct containing information about a codepoint. */
typedef struct utf8proc_property_struct {
+ /**
+ * Unicode category.
+ * @see utf8proc_category_t.
+ */
utf8proc_propval_t category;
utf8proc_propval_t combining_class;
+ /**
+ * Bidirectional class.
+ * @see utf8proc_bidi_class_t.
+ */
utf8proc_propval_t bidi_class;
+ /**
+ * @anchor Decomposition type.
+ * @see utf8proc_decomp_type_t.
+ */
utf8proc_propval_t decomp_type;
- const int32_t *decomp_mapping;
+ utf8proc_uint16_t decomp_seqindex;
+ utf8proc_uint16_t casefold_seqindex;
+ utf8proc_uint16_t uppercase_seqindex;
+ utf8proc_uint16_t lowercase_seqindex;
+ utf8proc_uint16_t titlecase_seqindex;
+ utf8proc_uint16_t comb_index;
unsigned bidi_mirrored:1;
- int32_t uppercase_mapping;
- int32_t lowercase_mapping;
- int32_t titlecase_mapping;
- int32_t comb1st_index;
- int32_t comb2nd_index;
unsigned comp_exclusion:1;
+ /**
+ * Can this codepoint be ignored?
+ *
+ * Used by @ref utf8proc_decompose_char when @ref UTF8PROC_IGNORE is
+ * passed as an option.
+ */
unsigned ignorable:1;
unsigned control_boundary:1;
- unsigned extend:1;
- const int32_t *casefold_mapping;
+ /** The width of the codepoint. */
+ unsigned charwidth:2;
+ unsigned pad:2;
+ /**
+ * Boundclass.
+ * @see utf8proc_boundclass_t.
+ */
+ unsigned boundclass:8;
} utf8proc_property_t;
-#define UTF8PROC_CATEGORY_LU 1
-#define UTF8PROC_CATEGORY_LL 2
-#define UTF8PROC_CATEGORY_LT 3
-#define UTF8PROC_CATEGORY_LM 4
-#define UTF8PROC_CATEGORY_LO 5
-#define UTF8PROC_CATEGORY_MN 6
-#define UTF8PROC_CATEGORY_MC 7
-#define UTF8PROC_CATEGORY_ME 8
-#define UTF8PROC_CATEGORY_ND 9
-#define UTF8PROC_CATEGORY_NL 10
-#define UTF8PROC_CATEGORY_NO 11
-#define UTF8PROC_CATEGORY_PC 12
-#define UTF8PROC_CATEGORY_PD 13
-#define UTF8PROC_CATEGORY_PS 14
-#define UTF8PROC_CATEGORY_PE 15
-#define UTF8PROC_CATEGORY_PI 16
-#define UTF8PROC_CATEGORY_PF 17
-#define UTF8PROC_CATEGORY_PO 18
-#define UTF8PROC_CATEGORY_SM 19
-#define UTF8PROC_CATEGORY_SC 20
-#define UTF8PROC_CATEGORY_SK 21
-#define UTF8PROC_CATEGORY_SO 22
-#define UTF8PROC_CATEGORY_ZS 23
-#define UTF8PROC_CATEGORY_ZL 24
-#define UTF8PROC_CATEGORY_ZP 25
-#define UTF8PROC_CATEGORY_CC 26
-#define UTF8PROC_CATEGORY_CF 27
-#define UTF8PROC_CATEGORY_CS 28
-#define UTF8PROC_CATEGORY_CO 29
-#define UTF8PROC_CATEGORY_CN 30
-#define UTF8PROC_BIDI_CLASS_L 1
-#define UTF8PROC_BIDI_CLASS_LRE 2
-#define UTF8PROC_BIDI_CLASS_LRO 3
-#define UTF8PROC_BIDI_CLASS_R 4
-#define UTF8PROC_BIDI_CLASS_AL 5
-#define UTF8PROC_BIDI_CLASS_RLE 6
-#define UTF8PROC_BIDI_CLASS_RLO 7
-#define UTF8PROC_BIDI_CLASS_PDF 8
-#define UTF8PROC_BIDI_CLASS_EN 9
-#define UTF8PROC_BIDI_CLASS_ES 10
-#define UTF8PROC_BIDI_CLASS_ET 11
-#define UTF8PROC_BIDI_CLASS_AN 12
-#define UTF8PROC_BIDI_CLASS_CS 13
-#define UTF8PROC_BIDI_CLASS_NSM 14
-#define UTF8PROC_BIDI_CLASS_BN 15
-#define UTF8PROC_BIDI_CLASS_B 16
-#define UTF8PROC_BIDI_CLASS_S 17
-#define UTF8PROC_BIDI_CLASS_WS 18
-#define UTF8PROC_BIDI_CLASS_ON 19
-#define UTF8PROC_DECOMP_TYPE_FONT 1
-#define UTF8PROC_DECOMP_TYPE_NOBREAK 2
-#define UTF8PROC_DECOMP_TYPE_INITIAL 3
-#define UTF8PROC_DECOMP_TYPE_MEDIAL 4
-#define UTF8PROC_DECOMP_TYPE_FINAL 5
-#define UTF8PROC_DECOMP_TYPE_ISOLATED 6
-#define UTF8PROC_DECOMP_TYPE_CIRCLE 7
-#define UTF8PROC_DECOMP_TYPE_SUPER 8
-#define UTF8PROC_DECOMP_TYPE_SUB 9
-#define UTF8PROC_DECOMP_TYPE_VERTICAL 10
-#define UTF8PROC_DECOMP_TYPE_WIDE 11
-#define UTF8PROC_DECOMP_TYPE_NARROW 12
-#define UTF8PROC_DECOMP_TYPE_SMALL 13
-#define UTF8PROC_DECOMP_TYPE_SQUARE 14
-#define UTF8PROC_DECOMP_TYPE_FRACTION 15
-#define UTF8PROC_DECOMP_TYPE_COMPAT 16
+/** Unicode categories. */
+typedef enum {
+ UTF8PROC_CATEGORY_CN = 0, /**< Other, not assigned */
+ UTF8PROC_CATEGORY_LU = 1, /**< Letter, uppercase */
+ UTF8PROC_CATEGORY_LL = 2, /**< Letter, lowercase */
+ UTF8PROC_CATEGORY_LT = 3, /**< Letter, titlecase */
+ UTF8PROC_CATEGORY_LM = 4, /**< Letter, modifier */
+ UTF8PROC_CATEGORY_LO = 5, /**< Letter, other */
+ UTF8PROC_CATEGORY_MN = 6, /**< Mark, nonspacing */
+ UTF8PROC_CATEGORY_MC = 7, /**< Mark, spacing combining */
+ UTF8PROC_CATEGORY_ME = 8, /**< Mark, enclosing */
+ UTF8PROC_CATEGORY_ND = 9, /**< Number, decimal digit */
+ UTF8PROC_CATEGORY_NL = 10, /**< Number, letter */
+ UTF8PROC_CATEGORY_NO = 11, /**< Number, other */
+ UTF8PROC_CATEGORY_PC = 12, /**< Punctuation, connector */
+ UTF8PROC_CATEGORY_PD = 13, /**< Punctuation, dash */
+ UTF8PROC_CATEGORY_PS = 14, /**< Punctuation, open */
+ UTF8PROC_CATEGORY_PE = 15, /**< Punctuation, close */
+ UTF8PROC_CATEGORY_PI = 16, /**< Punctuation, initial quote */
+ UTF8PROC_CATEGORY_PF = 17, /**< Punctuation, final quote */
+ UTF8PROC_CATEGORY_PO = 18, /**< Punctuation, other */
+ UTF8PROC_CATEGORY_SM = 19, /**< Symbol, math */
+ UTF8PROC_CATEGORY_SC = 20, /**< Symbol, currency */
+ UTF8PROC_CATEGORY_SK = 21, /**< Symbol, modifier */
+ UTF8PROC_CATEGORY_SO = 22, /**< Symbol, other */
+ UTF8PROC_CATEGORY_ZS = 23, /**< Separator, space */
+ UTF8PROC_CATEGORY_ZL = 24, /**< Separator, line */
+ UTF8PROC_CATEGORY_ZP = 25, /**< Separator, paragraph */
+ UTF8PROC_CATEGORY_CC = 26, /**< Other, control */
+ UTF8PROC_CATEGORY_CF = 27, /**< Other, format */
+ UTF8PROC_CATEGORY_CS = 28, /**< Other, surrogate */
+ UTF8PROC_CATEGORY_CO = 29, /**< Other, private use */
+} utf8proc_category_t;
+
+/** Bidirectional character classes. */
+typedef enum {
+ UTF8PROC_BIDI_CLASS_L = 1, /**< Left-to-Right */
+ UTF8PROC_BIDI_CLASS_LRE = 2, /**< Left-to-Right Embedding */
+ UTF8PROC_BIDI_CLASS_LRO = 3, /**< Left-to-Right Override */
+ UTF8PROC_BIDI_CLASS_R = 4, /**< Right-to-Left */
+ UTF8PROC_BIDI_CLASS_AL = 5, /**< Right-to-Left Arabic */
+ UTF8PROC_BIDI_CLASS_RLE = 6, /**< Right-to-Left Embedding */
+ UTF8PROC_BIDI_CLASS_RLO = 7, /**< Right-to-Left Override */
+ UTF8PROC_BIDI_CLASS_PDF = 8, /**< Pop Directional Format */
+ UTF8PROC_BIDI_CLASS_EN = 9, /**< European Number */
+ UTF8PROC_BIDI_CLASS_ES = 10, /**< European Separator */
+ UTF8PROC_BIDI_CLASS_ET = 11, /**< European Number Terminator */
+ UTF8PROC_BIDI_CLASS_AN = 12, /**< Arabic Number */
+ UTF8PROC_BIDI_CLASS_CS = 13, /**< Common Number Separator */
+ UTF8PROC_BIDI_CLASS_NSM = 14, /**< Nonspacing Mark */
+ UTF8PROC_BIDI_CLASS_BN = 15, /**< Boundary Neutral */
+ UTF8PROC_BIDI_CLASS_B = 16, /**< Paragraph Separator */
+ UTF8PROC_BIDI_CLASS_S = 17, /**< Segment Separator */
+ UTF8PROC_BIDI_CLASS_WS = 18, /**< Whitespace */
+ UTF8PROC_BIDI_CLASS_ON = 19, /**< Other Neutrals */
+ UTF8PROC_BIDI_CLASS_LRI = 20, /**< Left-to-Right Isolate */
+ UTF8PROC_BIDI_CLASS_RLI = 21, /**< Right-to-Left Isolate */
+ UTF8PROC_BIDI_CLASS_FSI = 22, /**< First Strong Isolate */
+ UTF8PROC_BIDI_CLASS_PDI = 23, /**< Pop Directional Isolate */
+} utf8proc_bidi_class_t;
+
+/** Decomposition type. */
+typedef enum {
+ UTF8PROC_DECOMP_TYPE_FONT = 1, /**< Font */
+ UTF8PROC_DECOMP_TYPE_NOBREAK = 2, /**< Nobreak */
+ UTF8PROC_DECOMP_TYPE_INITIAL = 3, /**< Initial */
+ UTF8PROC_DECOMP_TYPE_MEDIAL = 4, /**< Medial */
+ UTF8PROC_DECOMP_TYPE_FINAL = 5, /**< Final */
+ UTF8PROC_DECOMP_TYPE_ISOLATED = 6, /**< Isolated */
+ UTF8PROC_DECOMP_TYPE_CIRCLE = 7, /**< Circle */
+ UTF8PROC_DECOMP_TYPE_SUPER = 8, /**< Super */
+ UTF8PROC_DECOMP_TYPE_SUB = 9, /**< Sub */
+ UTF8PROC_DECOMP_TYPE_VERTICAL = 10, /**< Vertical */
+ UTF8PROC_DECOMP_TYPE_WIDE = 11, /**< Wide */
+ UTF8PROC_DECOMP_TYPE_NARROW = 12, /**< Narrow */
+ UTF8PROC_DECOMP_TYPE_SMALL = 13, /**< Small */
+ UTF8PROC_DECOMP_TYPE_SQUARE = 14, /**< Square */
+ UTF8PROC_DECOMP_TYPE_FRACTION = 15, /**< Fraction */
+ UTF8PROC_DECOMP_TYPE_COMPAT = 16, /**< Compat */
+} utf8proc_decomp_type_t;
+
+/** Boundclass property. (TR29) */
+typedef enum {
+ UTF8PROC_BOUNDCLASS_START = 0, /**< Start */
+ UTF8PROC_BOUNDCLASS_OTHER = 1, /**< Other */
+ UTF8PROC_BOUNDCLASS_CR = 2, /**< Cr */
+ UTF8PROC_BOUNDCLASS_LF = 3, /**< Lf */
+ UTF8PROC_BOUNDCLASS_CONTROL = 4, /**< Control */
+ UTF8PROC_BOUNDCLASS_EXTEND = 5, /**< Extend */
+ UTF8PROC_BOUNDCLASS_L = 6, /**< L */
+ UTF8PROC_BOUNDCLASS_V = 7, /**< V */
+ UTF8PROC_BOUNDCLASS_T = 8, /**< T */
+ UTF8PROC_BOUNDCLASS_LV = 9, /**< Lv */
+ UTF8PROC_BOUNDCLASS_LVT = 10, /**< Lvt */
+ UTF8PROC_BOUNDCLASS_REGIONAL_INDICATOR = 11, /**< Regional indicator */
+ UTF8PROC_BOUNDCLASS_SPACINGMARK = 12, /**< Spacingmark */
+ UTF8PROC_BOUNDCLASS_PREPEND = 13, /**< Prepend */
+ UTF8PROC_BOUNDCLASS_ZWJ = 14, /**< Zero Width Joiner */
+ UTF8PROC_BOUNDCLASS_E_BASE = 15, /**< Emoji Base */
+ UTF8PROC_BOUNDCLASS_E_MODIFIER = 16, /**< Emoji Modifier */
+ UTF8PROC_BOUNDCLASS_GLUE_AFTER_ZWJ = 17, /**< Glue_After_ZWJ */
+ UTF8PROC_BOUNDCLASS_E_BASE_GAZ = 18, /**< E_BASE + GLUE_AFTER_ZJW */
+} utf8proc_boundclass_t;
+
+/**
+ * Function pointer type passed to @ref utf8proc_map_custom and
+ * @ref utf8proc_decompose_custom, which is used to specify a user-defined
+ * mapping of codepoints to be applied in conjunction with other mappings.
+ */
+typedef utf8proc_int32_t (*utf8proc_custom_func)(utf8proc_int32_t codepoint, void *data);
+
+/**
+ * Array containing the byte lengths of a UTF-8 encoded codepoint based
+ * on the first byte.
+ */
+UTF8PROC_DLLEXPORT extern const utf8proc_int8_t utf8proc_utf8class[256];
+
+/**
+ * Returns the utf8proc API version as a string MAJOR.MINOR.PATCH
+ * (http://semver.org format), possibly with a "-dev" suffix for
+ * development versions.
+ */
+UTF8PROC_DLLEXPORT const char *utf8proc_version(void);
+
+/**
+ * Returns an informative error string for the given utf8proc error code
+ * (e.g. the error codes returned by @ref utf8proc_map).
+ */
+UTF8PROC_DLLEXPORT const char *utf8proc_errmsg(utf8proc_ssize_t errcode);
+
+/**
+ * Reads a single codepoint from the UTF-8 sequence being pointed to by `str`.
+ * The maximum number of bytes read is `strlen`, unless `strlen` is
+ * negative (in which case up to 4 bytes are read).
+ *
+ * If a valid codepoint could be read, it is stored in the variable
+ * pointed to by `codepoint_ref`, otherwise that variable will be set to -1.
+ * In case of success, the number of bytes read is returned; otherwise, a
+ * negative error code is returned.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_iterate(const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_int32_t *codepoint_ref);
+
+/**
+ * Check if a codepoint is valid (regardless of whether it has been
+ * assigned a value by the current Unicode standard).
+ *
+ * @return 1 if the given `codepoint` is valid and otherwise return 0.
+ */
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_codepoint_valid(utf8proc_int32_t codepoint);
-extern const int8_t utf8proc_utf8class[256];
+/**
+ * Encodes the codepoint as an UTF-8 string in the byte array pointed
+ * to by `dst`. This array must be at least 4 bytes long.
+ *
+ * In case of success the number of bytes written is returned, and
+ * otherwise 0 is returned.
+ *
+ * This function does not check whether `codepoint` is valid Unicode.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_encode_char(utf8proc_int32_t codepoint, utf8proc_uint8_t *dst);
-const char *utf8proc_version(void);
+/**
+ * Look up the properties for a given codepoint.
+ *
+ * @param codepoint The Unicode codepoint.
+ *
+ * @returns
+ * A pointer to a (constant) struct containing information about
+ * the codepoint.
+ * @par
+ * If the codepoint is unassigned or invalid, a pointer to a special struct is
+ * returned in which `category` is 0 (@ref UTF8PROC_CATEGORY_CN).
+ */
+UTF8PROC_DLLEXPORT const utf8proc_property_t *utf8proc_get_property(utf8proc_int32_t codepoint);
-const char *utf8proc_errmsg(ssize_t errcode);
-/*
- * Returns a static error string for the given error code.
+/** Decompose a codepoint into an array of codepoints.
+ *
+ * @param codepoint the codepoint.
+ * @param dst the destination buffer.
+ * @param bufsize the size of the destination buffer.
+ * @param options one or more of the following flags:
+ * - @ref UTF8PROC_REJECTNA - return an error `codepoint` is unassigned
+ * - @ref UTF8PROC_IGNORE - strip "default ignorable" codepoints
+ * - @ref UTF8PROC_CASEFOLD - apply Unicode casefolding
+ * - @ref UTF8PROC_COMPAT - replace certain codepoints with their
+ * compatibility decomposition
+ * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster
+ * - @ref UTF8PROC_LUMP - lump certain different codepoints together
+ * - @ref UTF8PROC_STRIPMARK - remove all character marks
+ * @param last_boundclass
+ * Pointer to an integer variable containing
+ * the previous codepoint's boundary class if the @ref UTF8PROC_CHARBOUND
+ * option is used. Otherwise, this parameter is ignored.
+ *
+ * @return
+ * In case of success, the number of codepoints written is returned; in case
+ * of an error, a negative error code is returned (@ref utf8proc_errmsg).
+ * @par
+ * If the number of written codepoints would be bigger than `bufsize`, the
+ * required buffer size is returned, while the buffer will be overwritten with
+ * undefined data.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_char(
+ utf8proc_int32_t codepoint, utf8proc_int32_t *dst, utf8proc_ssize_t bufsize,
+ utf8proc_option_t options, int *last_boundclass
+);
+
+/**
+ * The same as @ref utf8proc_decompose_char, but acts on a whole UTF-8
+ * string and orders the decomposed sequences correctly.
+ *
+ * If the @ref UTF8PROC_NULLTERM flag in `options` is set, processing
+ * will be stopped, when a NULL byte is encounted, otherwise `strlen`
+ * bytes are processed. The result (in the form of 32-bit unicode
+ * codepoints) is written into the buffer being pointed to by
+ * `buffer` (which must contain at least `bufsize` entries). In case of
+ * success, the number of codepoints written is returned; in case of an
+ * error, a negative error code is returned (@ref utf8proc_errmsg).
+ * See @ref utf8proc_decompose_custom to supply additional transformations.
+ *
+ * If the number of written codepoints would be bigger than `bufsize`, the
+ * required buffer size is returned, while the buffer will be overwritten with
+ * undefined data.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
+ utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options
+);
+
+/**
+ * The same as @ref utf8proc_decompose, but also takes a `custom_func` mapping function
+ * that is called on each codepoint in `str` before any other transformations
+ * (along with a `custom_data` pointer that is passed through to `custom_func`).
+ * The `custom_func` argument is ignored if it is `NULL`. See also @ref utf8proc_map_custom.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_decompose_custom(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen,
+ utf8proc_int32_t *buffer, utf8proc_ssize_t bufsize, utf8proc_option_t options,
+ utf8proc_custom_func custom_func, void *custom_data
+);
+
+/**
+ * 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);
-ssize_t utf8proc_iterate(const uint8_t *str, ssize_t strlen, int32_t *dst);
-/*
- * Reads a single char from the UTF-8 sequence being pointed to by 'str'.
- * The maximum number of bytes read is 'strlen', unless 'strlen' is
- * negative.
- * If a valid unicode char could be read, it is stored in the variable
- * being pointed to by 'dst', otherwise that variable will be set to -1.
- * In case of success the number of bytes read is returned, otherwise a
- * negative error code is returned.
+/**
+ * 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.
+ * @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
+ * - @ref UTF8PROC_CHARBOUND - insert 0xFF bytes before each grapheme cluster
+ *
+ * @return
+ * 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
+ * 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_reencode(utf8proc_int32_t *buffer, utf8proc_ssize_t length, utf8proc_option_t options);
+
+/**
+ * Given a pair of consecutive codepoints, return whether a grapheme break is
+ * permitted between them (as defined by the extended grapheme clusters in UAX#29).
+ *
+ * @param state Beginning with Version 29 (Unicode 9.0.0), this algorithm requires
+ * state to break graphemes. This state can be passed in as a pointer
+ * in the `state` argument and should initially be set to 0. If the
+ * state is not passed in (i.e. a null pointer is passed), UAX#29 rules
+ * GB10/12/13 which require this state will not be applied, essentially
+ * matching the rules in Unicode 8.0.0.
+ *
+ * @warning If the state parameter is used, `utf8proc_grapheme_break_stateful` must
+ * be called IN ORDER on ALL potential breaks in a string.
*/
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break_stateful(
+ utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2, utf8proc_int32_t *state);
-bool utf8proc_codepoint_valid(int32_t uc);
-/*
- * Returns 1, if the given unicode code-point is valid, otherwise 0.
+/**
+ * Same as @ref utf8proc_grapheme_break_stateful, except without support for the
+ * Unicode 9 additions to the algorithm. Supported for legacy reasons.
*/
+UTF8PROC_DLLEXPORT utf8proc_bool utf8proc_grapheme_break(
+ utf8proc_int32_t codepoint1, utf8proc_int32_t codepoint2);
-ssize_t utf8proc_encode_char(int32_t uc, uint8_t *dst);
-/*
- * Encodes the unicode char with the code point 'uc' as an UTF-8 string in
- * the byte array being pointed to by 'dst'. This array has to be at least
- * 4 bytes long.
- * In case of success the number of bytes written is returned,
- * otherwise 0.
- * This function does not check if 'uc' is a valid unicode code point.
+
+/**
+ * Given a codepoint `c`, return the codepoint of the corresponding
+ * lower-case character, if any; otherwise (if there is no lower-case
+ * variant, or if `c` is not a valid codepoint) return `c`.
*/
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_tolower(utf8proc_int32_t c);
-const utf8proc_property_t *utf8proc_get_property(int32_t uc);
-/*
- * Returns a pointer to a (constant) struct containing information about
- * the unicode char with the given code point 'uc'.
- * If the character is not existent a pointer to a special struct is
- * returned, where 'category' is a NULL pointer.
- * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
- * 0x10FFFF, otherwise the program might crash!
+/**
+ * Given a codepoint `c`, return the codepoint of the corresponding
+ * upper-case character, if any; otherwise (if there is no upper-case
+ * variant, or if `c` is not a valid codepoint) return `c`.
*/
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_toupper(utf8proc_int32_t c);
-ssize_t utf8proc_decompose_char(
- int32_t uc, int32_t *dst, ssize_t bufsize,
- int options, int *last_boundclass
-);
-/*
- * Writes a decomposition of the unicode char 'uc' into the array being
- * pointed to by 'dst'.
- * Following flags in the 'options' field are regarded:
- * REJECTNA: an unassigned unicode code point leads to an error
- * IGNORE: "default ignorable" chars are stripped
- * CASEFOLD: unicode casefolding is applied
- * COMPAT: replace certain characters with their
- * compatibility decomposition
- * CHARBOUND: Inserts 0xFF bytes before each grapheme cluster
- * LUMP: lumps certain different characters together
- * STRIPMARK: removes all character marks
- * The pointer 'last_boundclass' has to point to an integer variable which
- * is storing the last character boundary class, if the CHARBOUND option
- * is used.
- * In case of success the number of chars written is returned,
- * in case of an error, a negative error code is returned.
- * If the number of written chars would be bigger than 'bufsize',
- * the buffer (up to 'bufsize') has inpredictable data, and the needed
- * buffer size is returned.
- * WARNING: The parameter 'uc' has to be in the range of 0x0000 to
- * 0x10FFFF, otherwise the program might crash!
- */
-
-ssize_t utf8proc_decompose(
- const uint8_t *str, ssize_t strlen,
- int32_t *buffer, ssize_t bufsize, int options
-);
-/*
- * Does the same as 'utf8proc_decompose_char', but acts on a whole UTF-8
- * string, and orders the decomposed sequences correctly.
- * If the NULLTERM flag in 'options' is set, processing will be stopped,
- * when a NULL byte is encounted, otherwise 'strlen' bytes are processed.
- * The result in form of unicode code points is written into the buffer
- * being pointed to by 'buffer', having the length of 'bufsize' entries.
- * In case of success the number of chars written is returned,
- * in case of an error, a negative error code is returned.
- * If the number of written chars would be bigger than 'bufsize',
- * the buffer (up to 'bufsize') has inpredictable data, and the needed
- * buffer size is returned.
+/**
+ * Given a codepoint `c`, return the codepoint of the corresponding
+ * title-case character, if any; otherwise (if there is no title-case
+ * variant, or if `c` is not a valid codepoint) return `c`.
*/
+UTF8PROC_DLLEXPORT utf8proc_int32_t utf8proc_totitle(utf8proc_int32_t c);
-ssize_t utf8proc_reencode(int32_t *buffer, ssize_t length, int options);
-/*
- * Reencodes the sequence of unicode characters given by the pointer
- * 'buffer' and 'length' as UTF-8.
- * The result is stored in the same memory area where the data is read.
- * Following flags in the 'options' field are regarded:
- * NLF2LS: converts LF, CRLF, CR and NEL into LS
- * NLF2PS: converts LF, CRLF, CR and NEL into PS
- * NLF2LF: converts LF, CRLF, CR and NEL into LF
- * STRIPCC: strips or converts all non-affected control characters
- * COMPOSE: tries to combine decomposed characters into composite
- * characters
- * STABLE: prohibits combining characters which would violate
- * the unicode versioning stability
- * In case of success the length of the resulting UTF-8 string is
- * returned, otherwise a negative error code is returned.
- * WARNING: The amount of free space being pointed to by 'buffer', has to
- * exceed the amount of the input data by one byte, and the
- * entries of the array pointed to by 'str' have to be in the
- * range of 0x0000 to 0x10FFFF, otherwise the program might
- * crash!
+/**
+ * Given a codepoint, return a character width analogous to `wcwidth(codepoint)`,
+ * except that a width of 0 is returned for non-printable codepoints
+ * instead of -1 as in `wcwidth`.
+ *
+ * @note
+ * If you want to check for particular types of non-printable characters,
+ * (analogous to `isprint` or `iscntrl`), use @ref utf8proc_category. */
+UTF8PROC_DLLEXPORT int utf8proc_charwidth(utf8proc_int32_t codepoint);
+
+/**
+ * Return the Unicode category for the codepoint (one of the
+ * @ref utf8proc_category_t constants.)
+ */
+UTF8PROC_DLLEXPORT utf8proc_category_t utf8proc_category(utf8proc_int32_t codepoint);
+
+/**
+ * Return the two-letter (nul-terminated) Unicode category string for
+ * the codepoint (e.g. `"Lu"` or `"Co"`).
+ */
+UTF8PROC_DLLEXPORT const char *utf8proc_category_string(utf8proc_int32_t codepoint);
+
+/**
+ * Maps the given UTF-8 string pointed to by `str` to a new UTF-8
+ * string, allocated dynamically by `malloc` and returned via `dstptr`.
+ *
+ * If the @ref UTF8PROC_NULLTERM flag in the `options` field is set,
+ * the length is determined by a NULL terminator, otherwise the
+ * parameter `strlen` is evaluated to determine the string length, but
+ * in any case the result will be NULL terminated (though it might
+ * contain NULL characters with the string if `str` contained NULL
+ * characters). Other flags in the `options` field are passed to the
+ * functions defined above, and regarded as described. See also
+ * @ref utfproc_map_custom to supply a custom codepoint transformation.
+ *
+ * In case of success the length of the new string is returned,
+ * otherwise a negative error code is returned.
+ *
+ * @note The memory of the new UTF-8 string will have been allocated
+ * with `malloc`, and should therefore be deallocated with `free`.
*/
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options
+);
-ssize_t utf8proc_map(
- const uint8_t *str, ssize_t strlen, uint8_t **dstptr, int options
+/**
+ * Like @ref utf8proc_map, but also takes a `custom_func` mapping function
+ * that is called on each codepoint in `str` before any other transformations
+ * (along with a `custom_data` pointer that is passed through to `custom_func`).
+ * The `custom_func` argument is ignored if it is `NULL`.
+ */
+UTF8PROC_DLLEXPORT utf8proc_ssize_t utf8proc_map_custom(
+ const utf8proc_uint8_t *str, utf8proc_ssize_t strlen, utf8proc_uint8_t **dstptr, utf8proc_option_t options,
+ utf8proc_custom_func custom_func, void *custom_data
);
-/*
- * Maps the given UTF-8 string being pointed to by 'str' to a new UTF-8
- * string, which is allocated dynamically, and afterwards pointed to by
- * the pointer being pointed to by 'dstptr'.
- * If the NULLTERM flag in the 'options' field is set, the length is
- * determined by a NULL terminator, otherwise the parameter 'strlen' is
- * evaluated to determine the string length, but in any case the result
- * will be NULL terminated (though it might contain NULL characters
- * before). Other flags in the 'options' field are passed to the functions
- * defined above, and regarded as described.
- * In case of success the length of the new string is returned,
- * otherwise a negative error code is returned.
- * NOTICE: The memory of the new UTF-8 string will have been allocated with
- * 'malloc', and has theirfore to be freed with 'free'.
- */
-
-uint8_t *utf8proc_NFD(const uint8_t *str);
-uint8_t *utf8proc_NFC(const uint8_t *str);
-uint8_t *utf8proc_NFKD(const uint8_t *str);
-uint8_t *utf8proc_NFKC(const uint8_t *str);
-/*
- * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC
- * normalized version of the null-terminated string 'str'.
- */
+
+/** @name Unicode normalization
+ *
+ * Returns a pointer to newly allocated memory of a NFD, NFC, NFKD or NFKC
+ * normalized version of the null-terminated string `str`. These
+ * are shortcuts to calling @ref utf8proc_map with @ref UTF8PROC_NULLTERM
+ * combined with @ref UTF8PROC_STABLE and flags indicating the normalization.
+ */
+/** @{ */
+/** NFD normalization (@ref UTF8PROC_DECOMPOSE). */
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFD(const utf8proc_uint8_t *str);
+/** NFC normalization (@ref UTF8PROC_COMPOSE). */
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFC(const utf8proc_uint8_t *str);
+/** NFKD normalization (@ref UTF8PROC_DECOMPOSE and @ref UTF8PROC_COMPAT). */
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKD(const utf8proc_uint8_t *str);
+/** NFKC normalization (@ref UTF8PROC_COMPOSE and @ref UTF8PROC_COMPAT). */
+UTF8PROC_DLLEXPORT utf8proc_uint8_t *utf8proc_NFKC(const utf8proc_uint8_t *str);
+/** @} */
#ifdef __cplusplus
}
#endif
#endif
-