mirror of
https://github.com/rhasspy/espeak-ng.git
synced 2026-11-21 03:24:48 +00:00
Replace the *2 wchar/wctype APIs with their standard versions. These will be fixed using compatibility headers.
This commit is contained in:
@@ -26,6 +26,7 @@
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <wctype.h>
|
||||
|
||||
#include <espeak-ng/espeak_ng.h>
|
||||
#include <espeak-ng/speak_lib.h>
|
||||
@@ -585,7 +586,7 @@ static int compile_line(char *linebuf, char *dict_line, int *hash)
|
||||
ix = utf8_in(&c2, p);
|
||||
if (c2 == 0)
|
||||
break;
|
||||
if (iswupper2(c2))
|
||||
if (iswupper(c2))
|
||||
utf8_out(towlower2(c2), p);
|
||||
else
|
||||
all_upper_case = 0;
|
||||
|
||||
@@ -24,6 +24,7 @@
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <wctype.h>
|
||||
#include <wchar.h>
|
||||
|
||||
#include <espeak-ng/espeak_ng.h>
|
||||
@@ -614,7 +615,7 @@ const char *GetTranslatedPhonemeString(int phoneme_mode)
|
||||
p += utf8_in(&c, p);
|
||||
if (use_tie != 0) {
|
||||
// look for non-inital alphabetic character, but not diacritic, superscript etc.
|
||||
if ((count > 0) && !(flags & (1 << (count-1))) && ((c < 0x2b0) || (c > 0x36f)) && iswalpha2(c))
|
||||
if ((count > 0) && !(flags & (1 << (count-1))) && ((c < 0x2b0) || (c > 0x36f)) && iswalpha(c))
|
||||
buf += utf8_out(use_tie, buf);
|
||||
}
|
||||
buf += utf8_out(c, buf);
|
||||
@@ -832,7 +833,7 @@ int Unpronouncable(Translator *tr, char *word, int posn)
|
||||
break;
|
||||
}
|
||||
|
||||
if ((c != '\'') && !iswalpha2(c))
|
||||
if ((c != '\'') && !iswalpha(c))
|
||||
return 0;
|
||||
}
|
||||
|
||||
@@ -1787,7 +1788,7 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
|
||||
failed = 1;
|
||||
break;
|
||||
case RULE_NONALPHA:
|
||||
if (!iswalpha2(letter_w)) {
|
||||
if (!iswalpha(letter_w)) {
|
||||
add_points = (21-distance_right);
|
||||
post_ptr += letter_xbytes;
|
||||
} else
|
||||
@@ -1996,7 +1997,7 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
|
||||
failed = 1;
|
||||
break;
|
||||
case RULE_NONALPHA:
|
||||
if (!iswalpha2(letter_w)) {
|
||||
if (!iswalpha(letter_w)) {
|
||||
add_points = (21-distance_right);
|
||||
pre_ptr -= letter_xbytes;
|
||||
} else
|
||||
@@ -2300,7 +2301,7 @@ int TranslateRules(Translator *tr, char *p_start, char *phonemes, int ph_size, c
|
||||
|
||||
if (tr->letter_bits_offset > 0) {
|
||||
// not a Latin alphabet, switch to the default Latin alphabet language
|
||||
if ((letter <= 0x241) && iswalpha2(letter)) {
|
||||
if ((letter <= 0x241) && iswalpha(letter)) {
|
||||
sprintf(phonemes, "%c%s", phonSWITCH, tr->langopts.ascii_language);
|
||||
return 0;
|
||||
}
|
||||
|
||||
@@ -701,7 +701,7 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
|
||||
|
||||
if (control & 2) {
|
||||
// include CAPITAL information
|
||||
if (iswupper2(letter))
|
||||
if (iswupper(letter))
|
||||
Lookup(tr, "_cap", capital);
|
||||
}
|
||||
letter = towlower2(letter);
|
||||
@@ -845,7 +845,7 @@ int TranslateLetter(Translator *tr, char *word, char *phonemes, int control)
|
||||
if (ph_buf[0] == 0) {
|
||||
speak_letter_number = 1;
|
||||
if (!(al_flags & AL_NO_SYMBOL)) {
|
||||
if (iswalpha2(letter))
|
||||
if (iswalpha(letter))
|
||||
Lookup(translator, "_?A", ph_buf);
|
||||
|
||||
if ((ph_buf[0] == 0) && !iswspace(letter))
|
||||
@@ -2055,7 +2055,7 @@ static int TranslateNumber_1(Translator *tr, char *word, char *ph_out, unsigned
|
||||
if ((tr->langopts.numbers & NUM_NOPAUSE) && (next_char == ' '))
|
||||
utf8_in(&next_char, p);
|
||||
|
||||
if (!iswalpha2(next_char) && (thousands_exact == 0))
|
||||
if (!iswalpha(next_char) && (thousands_exact == 0))
|
||||
strcat(ph_out, str_pause); // don't add pause for 100s, 6th, etc.
|
||||
}
|
||||
|
||||
|
||||
@@ -289,88 +289,13 @@ static const short wchar_toupper[] = {
|
||||
0, 0
|
||||
};
|
||||
|
||||
// use internal data for iswalpha up to U+024F
|
||||
// iswalpha() on Windows is unreliable (U+AA, U+BA).
|
||||
int iswalpha2(int c)
|
||||
{
|
||||
if (c < 0x80)
|
||||
return isalpha(c);
|
||||
if ((c > 0x3040) && (c <= 0xa700))
|
||||
return 1; // japanese, chinese characters
|
||||
if (c > MAX_WALPHA)
|
||||
return iswalpha(c);
|
||||
return walpha_tab[c-0x80];
|
||||
}
|
||||
|
||||
int iswlower2(int c)
|
||||
{
|
||||
if (c < 0x80)
|
||||
return islower(c);
|
||||
if (c > MAX_WALPHA)
|
||||
return iswlower(c);
|
||||
if (walpha_tab[c-0x80] == 0xff)
|
||||
return 1;
|
||||
return 0;
|
||||
}
|
||||
|
||||
int iswupper2(int c)
|
||||
{
|
||||
int x;
|
||||
if (c < 0x80)
|
||||
return isupper(c);
|
||||
if (c > MAX_WALPHA)
|
||||
return iswupper(c);
|
||||
if (((x = walpha_tab[c-0x80]) > 0) && (x < 0xfe))
|
||||
return 1;
|
||||
return 0;
|
||||
}
|
||||
|
||||
int towlower2(unsigned int c)
|
||||
{
|
||||
int x;
|
||||
int ix;
|
||||
|
||||
// check for non-standard upper to lower case conversions
|
||||
if (c == 'I') {
|
||||
if (translator->langopts.dotless_i)
|
||||
c = 0x131; // I -> ı
|
||||
}
|
||||
if (c == 'I' && translator->langopts.dotless_i)
|
||||
return 0x131; // I -> ı
|
||||
|
||||
if (c < 0x80)
|
||||
return tolower(c);
|
||||
|
||||
if (c > MAX_WALPHA)
|
||||
return towlower(c);
|
||||
|
||||
if ((x = walpha_tab[c-0x80]) >= 0xfe)
|
||||
return c; // this is not an upper case letter
|
||||
|
||||
if (x == 0xfd) {
|
||||
// special cases, lookup translation table
|
||||
for (ix = 0; wchar_tolower[ix] != 0; ix += 2) {
|
||||
if (wchar_tolower[ix] == (int)c)
|
||||
return wchar_tolower[ix+1];
|
||||
}
|
||||
}
|
||||
return c + x; // convert to lower case
|
||||
}
|
||||
|
||||
int towupper2(unsigned int c)
|
||||
{
|
||||
int ix;
|
||||
if (c > MAX_WALPHA)
|
||||
return towupper(c);
|
||||
|
||||
// check whether a previous character code is the upper-case equivalent of this character
|
||||
if (towlower2(c-32) == (int)c)
|
||||
return c-32; // yes, use it
|
||||
if (towlower2(c-1) == (int)c)
|
||||
return c-1;
|
||||
for (ix = 0; wchar_toupper[ix] != 0; ix += 2) {
|
||||
if (wchar_toupper[ix] == (int)c)
|
||||
return wchar_toupper[ix+1];
|
||||
}
|
||||
return c; // no
|
||||
return towlower(c);
|
||||
}
|
||||
|
||||
static int IsRomanU(unsigned int c)
|
||||
@@ -2015,7 +1940,7 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
while (!Eof() && (c1 != '>'))
|
||||
c1 = GetC();
|
||||
c2 = ' ';
|
||||
} else if ((c2 == '/') || iswalpha2(c2)) {
|
||||
} else if ((c2 == '/') || iswalpha(c2)) {
|
||||
// check for space in the output buffer for embedded commands produced by the SSML tag
|
||||
if (ix > (n_buf - 20)) {
|
||||
// Perhaps not enough room, end the clause before the SSML tag
|
||||
@@ -2167,9 +2092,9 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
}
|
||||
}
|
||||
|
||||
if (iswupper2(c1)) {
|
||||
if (iswupper(c1)) {
|
||||
tr->clause_upper_count++;
|
||||
if ((option_capitals == 2) && (sayas_mode == 0) && !iswupper2(cprev)) {
|
||||
if ((option_capitals == 2) && (sayas_mode == 0) && !iswupper(cprev)) {
|
||||
char text_buf[40];
|
||||
char text_buf2[30];
|
||||
if (LookupSpecial(tr, "_cap", text_buf2) != NULL) {
|
||||
@@ -2181,7 +2106,7 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
}
|
||||
}
|
||||
}
|
||||
} else if (iswalpha2(c1))
|
||||
} else if (iswalpha(c1))
|
||||
tr->clause_lower_count++;
|
||||
|
||||
if (option_phoneme_input) {
|
||||
@@ -2238,7 +2163,7 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
// i.e. is dot followed by an upper-case letter?
|
||||
|
||||
if (!iswspace(c1)) {
|
||||
if (!IsAlpha(c1) || !iswlower2(c1)) {
|
||||
if (!IsAlpha(c1) || !iswlower(c1)) {
|
||||
UngetC(c2);
|
||||
ungot_char2 = c1;
|
||||
buf[end_clause_index] = ' '; // delete the end-clause punctuation
|
||||
@@ -2320,7 +2245,7 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
punct_data |= CLAUSE_DOT;
|
||||
|
||||
if (nl_count == 0) {
|
||||
if ((c1 == ',') && (cprev == '.') && (tr->translator_name == L('h', 'u')) && iswdigit(cprev2) && (iswdigit(c_next) || (iswlower2(c_next)))) {
|
||||
if ((c1 == ',') && (cprev == '.') && (tr->translator_name == L('h', 'u')) && iswdigit(cprev2) && (iswdigit(c_next) || (iswlower(c_next)))) {
|
||||
// lang=hu, fix for ordinal numbers, eg: "december 2., szerda", ignore ',' after ordinal number
|
||||
c1 = CHAR_COMMA_BREAK;
|
||||
is_end_clause = 0;
|
||||
@@ -2332,11 +2257,11 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
|
||||
// dot after a number indicates an ordinal number
|
||||
if (!iswdigit(cprev))
|
||||
is_end_clause = 0; // Roman number followed by dot
|
||||
else if (iswlower2(c_next) || (c_next == '-')) // hyphen is needed for lang-hu (eg. 2.-kal)
|
||||
else if (iswlower(c_next) || (c_next == '-')) // hyphen is needed for lang-hu (eg. 2.-kal)
|
||||
is_end_clause = 0; // only if followed by lower-case, (or if there is a XML tag)
|
||||
} else if (c_next == '\'')
|
||||
is_end_clause = 0; // eg. u.s.a.'s
|
||||
if (iswlower2(c_next)) {
|
||||
if (iswlower(c_next)) {
|
||||
// next word has no capital letter, this dot is probably from an abbreviation
|
||||
is_end_clause = 0;
|
||||
}
|
||||
|
||||
@@ -363,7 +363,7 @@ int IsAlpha(unsigned int c)
|
||||
0
|
||||
};
|
||||
|
||||
if (iswalpha2(c))
|
||||
if (iswalpha(c))
|
||||
return 1;
|
||||
|
||||
if (c < 0x300)
|
||||
@@ -610,7 +610,7 @@ int IsAllUpper(const char *word)
|
||||
int c;
|
||||
while ((*word != 0) && !isspace2(*word)) {
|
||||
word += utf8_in(&c, word);
|
||||
if (!iswupper2(c))
|
||||
if (!iswupper(c))
|
||||
return 0;
|
||||
}
|
||||
return 1;
|
||||
@@ -904,7 +904,7 @@ int TranslateWord(Translator *tr, char *word_start, WORD_TAB *wtab, char *word_o
|
||||
}
|
||||
}
|
||||
|
||||
if ((wflags & FLAG_ALL_UPPER) && (word_length > 1) && iswalpha2(first_char)) {
|
||||
if ((wflags & FLAG_ALL_UPPER) && (word_length > 1) && iswalpha(first_char)) {
|
||||
if ((option_tone_flags & OPTION_EMPHASIZE_ALLCAPS) && !(dictionary_flags[0] & FLAG_ABBREV)) {
|
||||
// emphasize words which are in capitals
|
||||
emphasize_allcaps = FLAG_EMPHASIZED;
|
||||
@@ -1331,7 +1331,7 @@ int TranslateWord(Translator *tr, char *word_start, WORD_TAB *wtab, char *word_o
|
||||
tr->expect_past--;
|
||||
}
|
||||
|
||||
if ((word_length == 1) && (tr->translator_name == L('e', 'n')) && iswalpha2(first_char) && (first_char != 'i')) {
|
||||
if ((word_length == 1) && (tr->translator_name == L('e', 'n')) && iswalpha(first_char) && (first_char != 'i')) {
|
||||
// English Specific !!!!
|
||||
// any single letter before a dot is an abbreviation, except 'I'
|
||||
dictionary_flags[0] |= FLAG_ALLOW_DOT;
|
||||
@@ -1587,7 +1587,7 @@ static int TranslateWord2(Translator *tr, char *word, WORD_TAB *wtab, int pre_pa
|
||||
while (*p2 != ' ') p2++;
|
||||
|
||||
utf8_in(&c_word2, p2+1); // first character of the next word;
|
||||
if (!iswalpha2(c_word2))
|
||||
if (!iswalpha(c_word2))
|
||||
ok = 0;
|
||||
|
||||
if (ok != 0) {
|
||||
@@ -1957,7 +1957,7 @@ static int SubstituteChar(Translator *tr, unsigned int c, unsigned int next_in,
|
||||
|
||||
// there is a list of character codes to be substituted with alternative codes
|
||||
|
||||
if (iswupper2(c_lower = c)) {
|
||||
if (iswupper(c_lower = c)) {
|
||||
c_lower = towlower2(c);
|
||||
upper_case = 1;
|
||||
}
|
||||
@@ -1984,14 +1984,14 @@ static int SubstituteChar(Translator *tr, unsigned int c, unsigned int next_in,
|
||||
// there is a second character to be inserted
|
||||
// don't convert the case of the second character unless the next letter is also upper case
|
||||
c2 = new_c >> 16;
|
||||
if (upper_case && iswupper2(next_in))
|
||||
c2 = towupper2(c2);
|
||||
if (upper_case && iswupper(next_in))
|
||||
c2 = towupper(c2);
|
||||
*insert = c2;
|
||||
new_c &= 0xffff;
|
||||
}
|
||||
|
||||
if (upper_case)
|
||||
new_c = towupper2(new_c);
|
||||
new_c = towupper(new_c);
|
||||
|
||||
*wordflags |= FLAG_CHAR_REPLACED;
|
||||
return new_c;
|
||||
@@ -2046,7 +2046,7 @@ static int TranslateChar(Translator *tr, char *ptr, int prev_in, unsigned int c,
|
||||
case L('n', 'l'):
|
||||
// look for 'n and replace by a special character (unicode: schwa)
|
||||
|
||||
if (!iswalpha2(prev_in)) {
|
||||
if (!iswalpha(prev_in)) {
|
||||
utf8_in(&next2, &ptr[1]);
|
||||
|
||||
if ((c == '\'') && IsSpace(next2)) {
|
||||
@@ -2400,7 +2400,7 @@ void *TranslateClause(Translator *tr, FILE *f_text, const void *vp_input, int *t
|
||||
if (!IsBracket(prev_out)) // ?? perhaps only set FLAG_NOSPACE for . - / (hyphenated words, URLs, etc)
|
||||
next_word_flags |= FLAG_NOSPACE;
|
||||
} else {
|
||||
if (iswupper2(c))
|
||||
if (iswupper(c))
|
||||
word_flags |= FLAG_FIRST_UPPER;
|
||||
|
||||
if ((prev_out == ' ') && iswdigit(sbuf[ix-2]) && !iswdigit(prev_in)) {
|
||||
@@ -2430,7 +2430,7 @@ void *TranslateClause(Translator *tr, FILE *f_text, const void *vp_input, int *t
|
||||
}
|
||||
}
|
||||
|
||||
if (iswupper2(c)) {
|
||||
if (iswupper(c)) {
|
||||
c = towlower2(c);
|
||||
|
||||
if ((j = tr->langopts.param[LOPT_CAPS_IN_WORD]) > 0) {
|
||||
@@ -2440,7 +2440,7 @@ void *TranslateClause(Translator *tr, FILE *f_text, const void *vp_input, int *t
|
||||
syllable_marked = 1;
|
||||
}
|
||||
} else {
|
||||
if (iswlower2(prev_in)) {
|
||||
if (iswlower(prev_in)) {
|
||||
// lower case followed by upper case in a word
|
||||
if (UpperCaseInWord(tr, &sbuf[ix], c) == 1) {
|
||||
// convert to lower case and continue
|
||||
@@ -2450,7 +2450,7 @@ void *TranslateClause(Translator *tr, FILE *f_text, const void *vp_input, int *t
|
||||
space_inserted = 1;
|
||||
prev_in_save = c;
|
||||
}
|
||||
} else if ((c != ' ') && iswupper2(prev_in) && iswlower2(next_in)) {
|
||||
} else if ((c != ' ') && iswupper(prev_in) && iswlower(next_in)) {
|
||||
int next2_in;
|
||||
utf8_in(&next2_in, &source[source_index + next_in_nbytes]);
|
||||
|
||||
|
||||
@@ -724,12 +724,8 @@ int IsDigit09(unsigned int c);
|
||||
int IsAlpha(unsigned int c);
|
||||
int IsVowel(Translator *tr, int c);
|
||||
int IsSuperscript(int letter);
|
||||
int iswalpha2(int c);
|
||||
int isspace2(unsigned int c);
|
||||
int iswlower2(int c);
|
||||
int iswupper2(int c);
|
||||
int towlower2(unsigned int c);
|
||||
int towupper2(unsigned int c);
|
||||
int towlower2(unsigned int c); // Supports Turkish I
|
||||
const char *GetTranslatedPhonemeString(int phoneme_mode);
|
||||
const char *WordToString2(unsigned int word);
|
||||
ALPHABET *AlphabetFromChar(int c);
|
||||
|
||||
Reference in New Issue
Block a user