[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[no subject]
|
From: |
Patrice Dumas |
|
Date: |
Sat, 13 Jan 2024 14:34:22 -0500 (EST) |
branch: master
commit 40fd859eb21c9be2d0f5b677ece1c240796037ac
Author: Patrice Dumas <pertusus@free.fr>
AuthorDate: Fri Jan 12 12:02:54 2024 +0100
* tp/Texinfo/XS/main/unicode.h (DIACRITIC_UNICODE),
tp/maintain/setup_converters_code_tables.pl (unicode_diacritics): add
hexadecimal representation of diacritics to the unicode_diacritics
table.
* tp/Texinfo/XS/convert/convert_html.c
(set_case_if_only_word_characters)
(html_accent_entities_html_accent_internal)
(after_escaped_characters, css_string_accent)
(css_string_convert_accent_command, html_converter_initialize):
implement css_string_convert_accent_command in C. Create
set_case_if_only_word_characters based on
html_accent_entities_html_accent_internal tu use common code.
---
ChangeLog | 16 ++
tp/Texinfo/XS/convert/convert_html.c | 246 +++++++++++++++++++++++++++-
tp/Texinfo/XS/main/unicode.h | 3 +-
tp/maintain/setup_converters_code_tables.pl | 5 +-
4 files changed, 261 insertions(+), 9 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index ed0d64da34..7a4f06fa93 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,19 @@
+2024-01-12 Patrice Dumas <pertusus@free.fr>
+
+ * tp/Texinfo/XS/main/unicode.h (DIACRITIC_UNICODE),
+ tp/maintain/setup_converters_code_tables.pl (unicode_diacritics): add
+ hexadecimal representation of diacritics to the unicode_diacritics
+ table.
+
+ * tp/Texinfo/XS/convert/convert_html.c
+ (set_case_if_only_word_characters)
+ (html_accent_entities_html_accent_internal)
+ (after_escaped_characters, css_string_accent)
+ (css_string_convert_accent_command, html_converter_initialize):
+ implement css_string_convert_accent_command in C. Create
+ set_case_if_only_word_characters based on
+ html_accent_entities_html_accent_internal tu use common code.
+
2024-01-11 Patrice Dumas <pertusus@free.fr>
* tp/Texinfo/XS/main/convert_to_text.c: make two functions static.
diff --git a/tp/Texinfo/XS/convert/convert_html.c
b/tp/Texinfo/XS/convert/convert_html.c
index c1be81ef30..1ea789534a 100644
--- a/tp/Texinfo/XS/convert/convert_html.c
+++ b/tp/Texinfo/XS/convert/convert_html.c
@@ -9220,10 +9220,8 @@ convert_math_command (CONVERTER *self, const enum
command_id cmd,
free (attribute_class);
}
-char *
-html_accent_entities_html_accent_internal (CONVERTER *self, const char *text,
- const ELEMENT *element, int set_case,
- int use_numeric_entities)
+static char *
+set_case_if_only_word_characters (const char *text, int set_case)
{
char *text_set;
@@ -9244,6 +9242,16 @@ html_accent_entities_html_accent_internal (CONVERTER
*self, const char *text,
else
text_set = strdup (text);
+ return text_set;
+}
+
+char *
+html_accent_entities_html_accent_internal (CONVERTER *self, const char *text,
+ const ELEMENT *element, int set_case,
+ int use_numeric_entities)
+{
+ char *text_set = set_case_if_only_word_characters (text, set_case);
+
/* do not return a dotless i or j as such if it is further composed
with an accented letter, return the letter as is */
if (element->cmd == CM_dotless
@@ -9322,6 +9330,7 @@ convert_accent_command (CONVERTER *self, const enum
command_id cmd,
const HTML_ARGS_FORMATTED *args_formatted,
const char *content, TEXT *result)
{
+ char *accent_text;
char *(*format_accents)(CONVERTER *self, const char *text,
const ELEMENT *element, int set_case);
@@ -9332,7 +9341,225 @@ convert_accent_command (CONVERTER *self, const enum
command_id cmd,
else
format_accents = &html_accent_entities_html_accent;
- char *accent_text = convert_accents (self, element, &html_convert_tree,
+ accent_text = convert_accents (self, element, &html_convert_tree,
+ format_accents, output_encoded_characters,
+ html_in_upper_case (self));
+
+ text_append (result, accent_text);
+ free (accent_text);
+}
+
+/* same as matching the regex /^\\[a-zA-Z0-9]+ /
+ */
+static char *
+after_escaped_characters (char *text)
+{
+ char *p = text;
+ if (*p != '\\')
+ return 0;
+
+ p++;
+
+ if (!isascii_alnum (*p))
+ return 0;
+
+ while (isascii_alnum (*p))
+ p++;
+
+ if (*p == ' ')
+ return p+1;
+
+ return 0;
+}
+
+char *
+css_string_accent (CONVERTER *self, const char *text,
+ const ELEMENT *element, int set_case)
+{
+ char *text_set = set_case_if_only_word_characters (text, set_case);
+
+ if (element->cmd == CM_dotless)
+ {
+ /* corresponds in perl, and for dotless, to
+ Texinfo::Convert::Unicode::unicode_accented_letters{$accent}->{$text} */
+ if (!strcmp (text_set, "i"))
+ {
+ free (text_set);
+ return strdup ("\\" "0131" " ");
+ }
+ else if (!strcmp (text_set, "j"))
+ {
+ free (text_set);
+ return strdup ("\\" "0237" " ");
+ }
+ }
+
+ if (unicode_diacritics[element->cmd].text)
+ {
+ char *accent_and_diacritic;
+ char *normalized_accent_text;
+ static TEXT accented_text;
+ if (element->cmd == CM_tieaccent)
+ {
+ /* tieaccent diacritic is naturally and correctly composed
+ between two characters */
+ /* we consider that letters are either characters or
+ escaped characters as they appear in CSS strings */
+ /* p non NUL corresponds to escaped characters */
+ char *p = after_escaped_characters (text_set);
+ char *next_text = 0;
+ ucs4_t first_char;
+ const uint8_t *next = 0;
+ uint8_t *encoded_u8 = 0;
+
+ if (!p)
+ {
+ /* check if a character matches */
+ encoded_u8 = u8_strconv_from_encoding (text, "UTF-8",
+ iconveh_question_mark);
+ next = u8_next (&first_char, encoded_u8);
+ if (next && (uc_is_general_category (first_char, UC_CATEGORY_L)
+ /* ASCII digits */
+ || (first_char >= 0x0030 && first_char <= 0x0039)))
+ {
+ next_text = u8_strconv_to_encoding (next, "UTF-8",
+ iconveh_question_mark);
+ }
+ }
+ else
+ {
+ next_text = p;
+ }
+
+ if (next_text)
+ {
+ ucs4_t second_char;
+ const char *q = after_escaped_characters (next_text);
+
+ if (!q)
+ {
+ const uint8_t *remaining;
+ if (!next)
+ {
+ encoded_u8 = u8_strconv_from_encoding (text, "UTF-8",
+ iconveh_question_mark);
+ next = encoded_u8;
+ }
+ remaining = u8_next (&second_char, next);
+ if (remaining
+ && (uc_is_general_category (second_char, UC_CATEGORY_L)
+ /* ASCII digits */
+ || (second_char >= 0x0030 && second_char <= 0x0039)))
+ {
+ /* next_text remains as the text to add after
+ the diacritic */
+ }
+ else
+ next_text = 0;
+ }
+
+ if (next_text)
+ {
+ /* add the first character or escaped text */
+ text_init (&accented_text);
+ if (!p)
+ {
+ char *first_char_text;
+ uint8_t *first_char_u8 = malloc (7 * sizeof(uint8_t));
+ int first_char_len
+ = u8_uctomb (first_char_u8, first_char, 6);
+ if (first_char_len < 0)
+ fatal ("u8_uctomb returns negative value");
+ first_char_u8[first_char_len] = 0;
+ first_char_text = u8_strconv_to_encoding (first_char_u8,
+ "UTF-8", iconveh_question_mark);
+ free (first_char_u8);
+ text_append (&accented_text, first_char_text);
+ free (first_char_text);
+ }
+ else
+ text_append_n (&accented_text, p, p - text_set);
+
+ /* add the tie accent */
+ text_printf (&accented_text, "\\%s ",
+ unicode_diacritics[element->cmd].hex_codepoint);
+ /* add the remaining, second character or escaped text
+ and everything else after (which is in general invalid
+ but we do not care) */
+ text_append (&accented_text, next_text);
+ if (!p)
+ free (next_text);
+ }
+ }
+ free (encoded_u8);
+ if (next_text)
+ return accented_text.text;
+ }
+
+ /* case of text and diacritic (including fallback for invalid tie
+ accent) */
+ text_init (&accented_text);
+ /* check if the normalization leads to merging text and diacritic,
+ if yes use the merged character, if not output text and diacitic
+ to be set up for composition */
+ xasprintf (&accent_and_diacritic, "%s%s",
+ text, unicode_diacritics[element->cmd].text);
+ normalized_accent_text = normalize_NFC (accent_and_diacritic);
+ free (accent_and_diacritic);
+ /* check if the normalization led to merging text and diacritic
+ as one character. If not, the leading text remains, this
+ is what the comparison checks */
+ if (!strncmp (normalized_accent_text, text, strlen (text)))
+ {
+ /* no normalization as one character, output text and diacritic
+ such that they could be composed */
+ text_append (&accented_text, text);
+ text_printf (&accented_text, "\\%s ",
+ unicode_diacritics[element->cmd].hex_codepoint);
+ }
+ else
+ {
+ /* determine the hexadecimal unicode point of the normalized
+ character to output in the format expected in CSS strings */
+ char *next_text;
+ uint8_t *encoded_u8 = u8_strconv_from_encoding (
+ normalized_accent_text, "UTF-8",
+ iconveh_question_mark);
+ ucs4_t first_char;
+ const uint8_t *next = u8_next (&first_char, encoded_u8);
+ text_printf (&accented_text, "\\%04lX ", first_char);
+ next_text = u8_strconv_to_encoding (next, "UTF-8",
+ iconveh_question_mark);
+ free (encoded_u8);
+ text_append (&accented_text, next_text);
+ free (next_text);
+ }
+ free (normalized_accent_text);
+ free (text_set);
+ return accented_text.text;
+ }
+
+ /* There are diacritics for every accent command except for dotless.
+ We should only get there with dotless if the argument is not recognized.
+ */
+ return text_set;
+}
+
+void
+css_string_convert_accent_command (CONVERTER *self, const enum command_id cmd,
+ const ELEMENT *element,
+ const HTML_ARGS_FORMATTED *args_formatted,
+ const char *content, TEXT *result)
+{
+ char *accent_text;
+ char *(*format_accents)(CONVERTER *self, const char *text,
+ const ELEMENT *element, int set_case);
+
+ int output_encoded_characters = (self->conf->OUTPUT_CHARACTERS.integer > 0);
+
+ format_accents = &css_string_accent;
+
+ accent_text = convert_accents (self, element, &html_convert_tree,
format_accents, output_encoded_characters,
html_in_upper_case (self));
@@ -9340,6 +9567,7 @@ convert_accent_command (CONVERTER *self, const enum
command_id cmd,
free (accent_text);
}
+
void
convert_indicateurl_command (CONVERTER *self, const enum command_id cmd,
const ELEMENT *element,
@@ -15969,7 +16197,7 @@ html_converter_initialize (CONVERTER *self)
}
}
- /* accents commands implemented in C, but not css strings accents */
+ /* accents commands implemented in C */
if (self->accent_cmd.number)
{
for (i = 0; i < self->accent_cmd.number; i++)
@@ -15977,6 +16205,8 @@ html_converter_initialize (CONVERTER *self)
enum command_id cmd = self->accent_cmd.list[i];
COMMAND_CONVERSION_FUNCTION *command_conversion
= &self->command_conversion_function[cmd];
+ COMMAND_CONVERSION_FUNCTION *css_string_command_conversion
+ = &self->css_string_command_conversion_function[cmd];
if (command_conversion->status == FRS_status_default_set)
{
command_conversion->formatting_reference = 0;
@@ -15984,6 +16214,10 @@ html_converter_initialize (CONVERTER *self)
command_conversion->command_conversion
= &convert_accent_command;
}
+ css_string_command_conversion->formatting_reference = 0;
+ css_string_command_conversion->status = FRS_status_internal;
+ css_string_command_conversion->command_conversion
+ = &css_string_convert_accent_command;
}
}
diff --git a/tp/Texinfo/XS/main/unicode.h b/tp/Texinfo/XS/main/unicode.h
index 15b3320a2f..2bd700b664 100644
--- a/tp/Texinfo/XS/main/unicode.h
+++ b/tp/Texinfo/XS/main/unicode.h
@@ -20,7 +20,8 @@ typedef struct COMMAND_UNICODE {
typedef struct DIACRITIC_UNICODE {
char *text; /* UTF-8 encoded */
- char *codepoint;
+ char *codepoint; /* numeric */
+ char *hex_codepoint; /* hexadecimal */
} DIACRITIC_UNICODE;
/* can be inlined in text parsing codes */
diff --git a/tp/maintain/setup_converters_code_tables.pl
b/tp/maintain/setup_converters_code_tables.pl
index bac7566369..30b5f2eabc 100755
--- a/tp/maintain/setup_converters_code_tables.pl
+++ b/tp/maintain/setup_converters_code_tables.pl
@@ -157,9 +157,10 @@ foreach my $command_name (@commands_order) {
my $numeric_codepoint = hex($unicode_diacritics{$command_name});
my $result = chr($numeric_codepoint);
my $protected = join ('', map {_protect_char($_)} split ('', $result));
- print UNIC "{\"$protected\", \"$numeric_codepoint\"}, /* $command */\n";
+ print UNIC "{\"$protected\", \"$numeric_codepoint\","
+ ." \"$unicode_diacritics{$command_name}\"}, /* $command */\n";
} else {
- print UNIC "{0, 0},\n";
+ print UNIC "{0, 0, 0},\n";
}
}
print UNIC "};\n\n";