2009-06-29 Bruno Haible <bruno@clisp.org>
+ New module 'unicase/u8-prefix-context'.
+ * lib/unicase/u8-prefix-context.c: New file.
+ * lib/unicase/u-prefix-context.h: New file.
+ * lib/unicase/context.h: New file.
+ * modules/unicase/u8-prefix-context: New file.
+
New module 'unicase/empty-prefix-context'.
* lib/unicase/empty-prefix-context.c: New file.
* modules/unicase/empty-prefix-context: New file.
--- /dev/null
+/* Case-mapping contexts of UTF-8/UTF-16/UTF-32 substring.
+ Copyright (C) 2009 Free Software Foundation, Inc.
+ Written by Bruno Haible <bruno@clisp.org>, 2009.
+
+ This program is free software: you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published
+ by the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+
+/* The context of a prefix string combines the information of the "Before C"
+ conditions of the Unicode Standard,
+ <http://www.unicode.org/versions/Unicode5.0.0/ch03.pdf>, section 3.13,
+ table 3-14 "Context Specification for Casing".
+
+ casing_prefix_context_t contains the following fields:
+
+ // Helper for evaluating the FINAL_SIGMA condition:
+ // Last character that was not case-ignorable.
+ ucs4_t last_char_except_ignorable;
+
+ // Helper for evaluating the AFTER_SOFT_DOTTED and AFTER_I conditions:
+ // Last character that was of combining class 230 ("Above") or 0.
+ ucs4_t last_char_normal_or_above;
+
+ Three bits would be sufficient to carry the context information, but
+ that would require to invoke uc_is_cased and uc_is_property_soft_dotted
+ ahead of time, more often than actually needed. */
+
+
+/* The context of a suffix string combines the information of the "After C"
+ conditions of the Unicode Standard,
+ <http://www.unicode.org/versions/Unicode5.0.0/ch03.pdf>, section 3.13,
+ table 3-14 "Context Specification for Casing".
+
+ casing_suffix_context_t contains the following fields:
+
+ // For evaluating the FINAL_SIGMA condition:
+ // Bit 0 is set if the suffix starts with a sequence consisting of a
+ // case-ignorable sequence and then a cased letter.
+ //
+ // For evaluating the MORE_ABOVE condition:
+ // Bit 1 is set if the suffix contains a character of combining class
+ // 230 (Above) with no character of combining class 0 or 230 (Above)
+ // before it.
+ //
+ // For evaluating the BEFORE_DOT condition:
+ // Bit 2 is set if the suffix contains a COMBINING DOT ABOVE (U+0307)
+ // with no character of combining class 0 or 230 (Above) before it.
+ //
+ uint32_t bits;
+ */
+#define SCC_FINAL_SIGMA_MASK 1
+#define SCC_MORE_ABOVE_MASK 2
+#define SCC_BEFORE_DOT_MASK 4
--- /dev/null
+/* Case-mapping context of prefix UTF-8/UTF-16/UTF-32 string.
+ Copyright (C) 2009 Free Software Foundation, Inc.
+ Written by Bruno Haible <bruno@clisp.org>, 2009.
+
+ This program is free software: you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published
+ by the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+casing_prefix_context_t
+FUNC1 (const UNIT *s, size_t n)
+{
+ return FUNC2 (s, n, unicase_empty_prefix_context);
+}
+
+casing_prefix_context_t
+FUNC2 (const UNIT *s, size_t n, casing_prefix_context_t a_context)
+{
+#if 0
+ /* Forward iteration. Slow for long strings. */
+ casing_prefix_context_t context = a_context;
+ const UNIT *s_end = s + n;
+
+ while (s < s_end)
+ {
+ ucs4_t uc;
+ int count = U_MBTOUC_UNSAFE (&uc, s, s_end - s);
+
+ if (!uc_is_case_ignorable (uc))
+ context.last_char_except_ignorable = uc;
+
+ {
+ int ccc = uc_combining_class (uc);
+ if (ccc == UC_CCC_A || ccc == UC_CCC_NR)
+ context.last_char_normal_or_above = uc;
+ }
+
+ s += count;
+ }
+
+ return context;
+#else
+ /* Iterate backwards, only as far as needed. */
+ casing_prefix_context_t context;
+ ucs4_t last_char_except_ignorable = (ucs4_t)(-1);
+ ucs4_t last_char_normal_or_above = (ucs4_t)(-1);
+ const UNIT *p = s + n;
+
+ for (;;)
+ {
+ ucs4_t uc;
+ p = U_PREV (&uc, p, s);
+ if (p == NULL)
+ break;
+
+ if (last_char_except_ignorable == (ucs4_t)(-1))
+ {
+ if (!uc_is_case_ignorable (uc))
+ last_char_except_ignorable = uc;
+ }
+
+ if (last_char_normal_or_above == (ucs4_t)(-1))
+ {
+ int ccc = uc_combining_class (uc);
+ if (ccc == UC_CCC_A || ccc == UC_CCC_NR)
+ last_char_normal_or_above = uc;
+ }
+
+ if (last_char_except_ignorable != (ucs4_t)(-1)
+ && last_char_normal_or_above != (ucs4_t)(-1))
+ break;
+ }
+ context.last_char_except_ignorable =
+ (last_char_except_ignorable != (ucs4_t)(-1)
+ ? last_char_except_ignorable
+ : a_context.last_char_except_ignorable);
+ context.last_char_normal_or_above =
+ (last_char_normal_or_above != (ucs4_t)(-1)
+ ? last_char_normal_or_above
+ : a_context.last_char_normal_or_above);
+
+ return context;
+#endif
+}
--- /dev/null
+/* Case-mapping context of prefix UTF-8 string.
+ Copyright (C) 2009 Free Software Foundation, Inc.
+ Written by Bruno Haible <bruno@clisp.org>, 2009.
+
+ This program is free software: you can redistribute it and/or modify it
+ under the terms of the GNU Lesser General Public License as published
+ by the Free Software Foundation; either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+#include <config.h>
+
+/* Specification. */
+#include "unicase.h"
+
+#include "unictype.h"
+#include "unistr.h"
+#include "caseprop.h"
+#include "context.h"
+
+#define FUNC1 u8_casing_prefix_context
+#define FUNC2 u8_casing_prefixes_context
+#define UNIT uint8_t
+#define U_MBTOUC_UNSAFE u8_mbtouc_unsafe
+#define U_PREV u8_prev
+#include "u-prefix-context.h"
--- /dev/null
+Description:
+Case-mapping context of prefix UTF-8 string.
+
+Files:
+lib/unicase/u8-prefix-context.c
+lib/unicase/u-prefix-context.h
+lib/unicase/context.h
+
+Depends-on:
+unicase/base
+unicase/empty-prefix-context
+unicase/ignorable
+unictype/combining-class
+unistr/u8-mbtouc-unsafe
+unistr/u8-prev
+
+configure.ac:
+
+Makefile.am:
+lib_SOURCES += unicase/u8-prefix-context.c
+
+Include:
+"unicase.h"
+
+License:
+LGPL
+
+Maintainer:
+Bruno Haible
+
projects / pspp / commitdiff