remove the syllable-chain-based implementation
This commit is contained in:
parent
f67fb24333
commit
2981a65e25
10 changed files with 21 additions and 377 deletions
7
compat.h
Normal file
7
compat.h
Normal file
|
|
@ -0,0 +1,7 @@
|
|||
#ifndef __COMPAT_H
|
||||
#define __COMPAT_H
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#endif /* __COMPAT_H */
|
||||
174
iast-czech.c
174
iast-czech.c
|
|
@ -1,174 +0,0 @@
|
|||
/* SPDX-License-Identifier: GPL-2.0 */
|
||||
|
||||
/* https://cs.wikipedia.org/wiki/Wikipedie:Transkripce_hindštiny */
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <ctype.h>
|
||||
|
||||
#include "iast-czech.h"
|
||||
#include "transliteration.h"
|
||||
|
||||
|
||||
static const struct transliteration_letter table[] = {
|
||||
|
||||
/* Vowels */
|
||||
{0x0905, FLAG_REGULAR, "a"}, /* अ */ /* 1 */
|
||||
{0x0906, FLAG_REGULAR, "á"}, /* आ */ /* 2 */
|
||||
{0x0907, FLAG_REGULAR, "i"}, /* इ */ /* 3 */
|
||||
{0x0908, FLAG_REGULAR, "í"}, /* ई */ /* 4 */
|
||||
{0x0909, FLAG_REGULAR, "u"}, /* उ */ /* 5 */
|
||||
{0x090a, FLAG_REGULAR, "ú"}, /* ऊ */ /* 6 */
|
||||
{0x090b, FLAG_REGULAR, "r"}, /* ऋ */ /* 7 */
|
||||
{0x0960, FLAG_REGULAR, "r"}, /* ॠ */ /* 8 */
|
||||
{0x090c, FLAG_REGULAR, "l"}, /* ऌ */ /* 9 */
|
||||
{0x0961, FLAG_REGULAR, "l"}, /* ॡ */ /* 10 */
|
||||
{0x090f, FLAG_REGULAR, "é"}, /* ए */ /* 11 */
|
||||
{0x0910, FLAG_REGULAR, "ai"}, /* ऐ */ /* 12 */
|
||||
{0x0913, FLAG_REGULAR, "ó"}, /* ओ */ /* 13 */
|
||||
{0x0914, FLAG_REGULAR, "au"}, /* औ */ /* 14 */
|
||||
|
||||
/* Consonants */
|
||||
{0x0915, FLAG_REGULAR, "ka"}, /* क */ /* 1 */
|
||||
{0x0916, FLAG_REGULAR, "kha"}, /* ख */ /* 2 */
|
||||
{0x0917, FLAG_REGULAR, "ga"}, /* ग */ /* 3 */
|
||||
{0x0918, FLAG_REGULAR, "gha"}, /* घ */ /* 4 */
|
||||
{0x0919, FLAG_REGULAR, "na"}, /* ङ */ /* 5 */
|
||||
{0x0939, FLAG_REGULAR, "ha"}, /* ह */ /* 6 */
|
||||
{0x091a, FLAG_REGULAR, "ca"}, /* च */ /* 7 */
|
||||
{0x091b, FLAG_REGULAR, "cha"}, /* छ */ /* 8 */
|
||||
{0x091c, FLAG_REGULAR, "dža"}, /* ज */ /* 9 */
|
||||
{0x091d, FLAG_REGULAR, "džha"}, /* झ */ /* 10 */
|
||||
{0x091e, FLAG_REGULAR, "ňa"}, /* ञ */ /* 11 */
|
||||
{0x092f, FLAG_REGULAR, "ja"}, /* य */ /* 12 */
|
||||
{0x0936, FLAG_REGULAR, "ša"}, /* श */ /* 13 */
|
||||
{0x091F, FLAG_REGULAR, "ta"}, /* ट */ /* 14 */
|
||||
{0x0920, FLAG_REGULAR, "tha"}, /* ठ */ /* 15 */
|
||||
{0x0921, FLAG_REGULAR, "da"}, /* ड */ /* 16 */
|
||||
{0x0922, FLAG_REGULAR, "dha"}, /* ढ */ /* 17 */
|
||||
{0x0923, FLAG_REGULAR, "na"}, /* ण */ /* 18 */
|
||||
{0x0930, FLAG_REGULAR, "ra"}, /* र */ /* 19 */
|
||||
{0x0937, FLAG_REGULAR, "ša"}, /* श */ /* 20 */
|
||||
{0x0924, FLAG_REGULAR, "ta"}, /* त */ /* 21 */
|
||||
{0x0925, FLAG_REGULAR, "tha"}, /* थ */ /* 22 */
|
||||
{0x0926, FLAG_REGULAR, "da"}, /* द */ /* 23 */
|
||||
{0x0927, FLAG_REGULAR, "dha"}, /* ध */ /* 24 */
|
||||
{0x0928, FLAG_REGULAR, "na"}, /* न */ /* 25 */
|
||||
{0x0932, FLAG_REGULAR, "la"}, /* ल */ /* 26 */
|
||||
{0x0938, FLAG_REGULAR, "sa"}, /* स */ /* 27 */
|
||||
{0x092a, FLAG_REGULAR, "pa"}, /* प */ /* 28 */
|
||||
{0x092b, FLAG_REGULAR, "pha"}, /* फ */ /* 29 */
|
||||
{0x092c, FLAG_REGULAR, "ba"}, /* ब */ /* 30 */
|
||||
{0x092d, FLAG_REGULAR, "bha"}, /* भ */ /* 31 */
|
||||
{0x092e, FLAG_REGULAR, "ma"}, /* म */ /* 32 */
|
||||
{0x0935, FLAG_REGULAR, "va"}, /* व */ /* 33 */
|
||||
|
||||
/* Codas */
|
||||
{0x0902, FLAG_REGULAR, "m"}, /* ं (anusvara) */
|
||||
{0x0903, FLAG_REGULAR, ""}, /* ः (visarga) */
|
||||
{0x093d, FLAG_REGULAR, "'"}, /* ऽ (avagrada) */
|
||||
|
||||
/* Special characters */
|
||||
{0x0950, FLAG_REGULAR, "óm"}, /* ॐ */
|
||||
|
||||
/* Numbers */
|
||||
{0x0966, FLAG_REGULAR, "0"},
|
||||
{0x0967, FLAG_REGULAR, "1"},
|
||||
{0x0968, FLAG_REGULAR, "2"},
|
||||
{0x0969, FLAG_REGULAR, "3"},
|
||||
{0x096a, FLAG_REGULAR, "4"},
|
||||
{0x096b, FLAG_REGULAR, "5"},
|
||||
{0x096c, FLAG_REGULAR, "6"},
|
||||
{0x096d, FLAG_REGULAR, "7"},
|
||||
{0x096e, FLAG_REGULAR, "8"},
|
||||
{0x096f, FLAG_REGULAR, "9"},
|
||||
|
||||
/* Diacritic modifiers */
|
||||
{0x093e, FLAG_MODIFIER, "á"}, /* ा */
|
||||
{0x093f, FLAG_MODIFIER, "i"}, /* ि */
|
||||
{0x0940, FLAG_MODIFIER, "í"}, /* ी */
|
||||
{0x0941, FLAG_MODIFIER, "u"}, /* ु */
|
||||
{0x0942, FLAG_MODIFIER, "ú"}, /* ू */
|
||||
{0x0943, FLAG_MODIFIER, "r"}, /* ृ */
|
||||
{0x0944, FLAG_MODIFIER, "r"}, /* ॄ */
|
||||
{0x0962, FLAG_MODIFIER, "l"}, /* ॢ */
|
||||
{0x0963, FLAG_MODIFIER, "l"}, /* ॣ */
|
||||
{0x0947, FLAG_MODIFIER, "é"}, /* े */
|
||||
{0x0948, FLAG_MODIFIER, "ai"}, /* ै */
|
||||
{0x094b, FLAG_MODIFIER, "ó"}, /* ो */
|
||||
{0x094c, FLAG_MODIFIER, "au"}, /* ौ */
|
||||
{0x094d, FLAG_MODIFIER, ""}, /* ् (virama) */
|
||||
|
||||
{0, 0, NULL}
|
||||
};
|
||||
|
||||
static void nasal_consonants_filter(struct syllable *chain)
|
||||
{
|
||||
struct syllable *syllable = chain;
|
||||
|
||||
while (syllable) {
|
||||
if (is_devanagari(syllable->code) && strcmp(syllable->data, "n") == 0) {
|
||||
if (syllable->next != NULL) {
|
||||
if (syllable->next->data[0] == 'p' ||
|
||||
syllable->next->data[0] == 'b' ||
|
||||
syllable->next->data[0] == 'm') {
|
||||
free(syllable->data);
|
||||
syllable->data = strdup("m");
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
if (is_devanagari(syllable->code) && strcmp(syllable->data, "m") == 0) {
|
||||
if (syllable->next != NULL) {
|
||||
if (syllable->next->data[0] != 'p' &&
|
||||
syllable->next->data[0] != 'b' &&
|
||||
syllable->next->data[0] != 'm') {
|
||||
free(syllable->data);
|
||||
syllable->data = strdup("n");
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
syllable = syllable->next;
|
||||
}
|
||||
}
|
||||
|
||||
static void end_of_word_filter(struct syllable *chain)
|
||||
{
|
||||
struct syllable *syllable = chain;
|
||||
unsigned int n;
|
||||
|
||||
while (syllable) {
|
||||
if (syllable->next == NULL || isspace(syllable->next->data[0])) {
|
||||
if (!strcmp(syllable->data, "m")) {
|
||||
free(syllable->data);
|
||||
syllable->data = strdup("");
|
||||
}
|
||||
|
||||
n = strlen(syllable->data);
|
||||
if (!strcmp(syllable->data + n - 2, "á")) {
|
||||
syllable->data[n - 2] = 'a';
|
||||
syllable->data[n - 1] = '\0';
|
||||
}
|
||||
}
|
||||
|
||||
syllable = syllable->next;
|
||||
}
|
||||
}
|
||||
|
||||
static const transliteration_filter_t filters[] = {
|
||||
nasal_consonants_filter,
|
||||
end_of_word_filter,
|
||||
NULL
|
||||
};
|
||||
|
||||
static const struct transliteration_context context = {
|
||||
.table = table,
|
||||
.filters = filters
|
||||
};
|
||||
|
||||
const struct transliteration_context *get_iast_czech_transliteration_context()
|
||||
{
|
||||
return &context;
|
||||
}
|
||||
10
iast-czech.h
10
iast-czech.h
|
|
@ -1,10 +0,0 @@
|
|||
/* SPDX-License-Identifier: GPL-2.0 */
|
||||
|
||||
#ifndef __IAST_CZECH_H
|
||||
#define __IAST_CZECH_H
|
||||
|
||||
#include "transliteration.h"
|
||||
|
||||
const struct transliteration_context *get_iast_czech_transliteration_context();
|
||||
|
||||
#endif /* __IAST_CZECH_H */
|
||||
15
iast.c
15
iast.c
|
|
@ -4,7 +4,7 @@
|
|||
#include "compat.h"
|
||||
#include "iast.h"
|
||||
|
||||
static const struct translit_letter table[] = {
|
||||
static struct translit_letter table[] = {
|
||||
|
||||
/* Vowels */
|
||||
{0x0910, VOWEL, "ai"}, /* 01 */
|
||||
|
|
@ -96,16 +96,7 @@ static const struct translit_letter table[] = {
|
|||
{0, 0, NULL}
|
||||
};
|
||||
|
||||
static const transliteration_filter_t filters[] = {
|
||||
NULL
|
||||
};
|
||||
|
||||
static struct translit_context context = {
|
||||
.table = table,
|
||||
.filters = filters
|
||||
};
|
||||
|
||||
struct translit_context *get_iast_transliteration_context()
|
||||
struct translit_letter *get_iast_transliteration_table()
|
||||
{
|
||||
return &context;
|
||||
return table;
|
||||
}
|
||||
|
|
|
|||
2
iast.h
2
iast.h
|
|
@ -5,6 +5,6 @@
|
|||
|
||||
#include "transliteration.h"
|
||||
|
||||
struct translit_context *get_iast_transliteration_context();
|
||||
struct translit_letter *get_iast_transliteration_table();
|
||||
|
||||
#endif /* __IAST_H */
|
||||
|
|
|
|||
10
main.c
10
main.c
|
|
@ -7,7 +7,6 @@
|
|||
|
||||
#include "transliteration.h"
|
||||
#include "iast.h"
|
||||
#include "iast-czech.h"
|
||||
#include "encoder.h"
|
||||
|
||||
#define PROGNAME "iast"
|
||||
|
|
@ -88,7 +87,6 @@ int main(int argc, const char **argv)
|
|||
const char *arg;
|
||||
const char *queue[argc];
|
||||
char *input, *output;
|
||||
const struct transliteration_context *context;
|
||||
|
||||
for (i = 1; i < argc; i++) {
|
||||
arg = argv[i];
|
||||
|
|
@ -119,10 +117,6 @@ int main(int argc, const char **argv)
|
|||
}
|
||||
}
|
||||
|
||||
context = (flags & FLAG_CZECH)
|
||||
? get_iast_czech_transliteration_context()
|
||||
: get_iast_transliteration_context();
|
||||
|
||||
if (flags & FLAG_STDIN) {
|
||||
input = stdin_read();
|
||||
if (input == NULL) {
|
||||
|
|
@ -133,7 +127,7 @@ int main(int argc, const char **argv)
|
|||
if (flags & FLAG_ENCODE) {
|
||||
output = encode_iast_punctation(input);
|
||||
} else {
|
||||
output = transliterate_devanagari_to_latin(input, context);
|
||||
output = transliterate_devanagari_to_latin(input);
|
||||
}
|
||||
|
||||
fprintf(stdout, "%s\n", output);
|
||||
|
|
@ -145,7 +139,7 @@ int main(int argc, const char **argv)
|
|||
if (flags & FLAG_ENCODE) {
|
||||
output = encode_iast_punctation(queue[i]);
|
||||
} else {
|
||||
output = transliterate_devanagari_to_latin(queue[i], context);
|
||||
output = transliterate_devanagari_to_latin(queue[i]);
|
||||
}
|
||||
|
||||
fprintf(stdout, "%s\n", output);
|
||||
|
|
|
|||
72
syllable.c
72
syllable.c
|
|
@ -1,72 +0,0 @@
|
|||
/* SPDX-License-Identifier: GPL-2.0 */
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include "syllable.h"
|
||||
#include "utf8.h"
|
||||
|
||||
|
||||
struct syllable *syllable_alloc(const char *data, unsigned int code)
|
||||
{
|
||||
struct syllable *ptr = malloc(sizeof(*ptr));
|
||||
|
||||
if (ptr == NULL)
|
||||
goto out;
|
||||
|
||||
ptr->data = data != NULL ? strdup(data) : utf8_code_to_string(code);
|
||||
ptr->code = code;
|
||||
ptr->prev = NULL;
|
||||
ptr->next = NULL;
|
||||
|
||||
out:
|
||||
return ptr;
|
||||
}
|
||||
|
||||
void syllable_drop(struct syllable *ptr)
|
||||
{
|
||||
if (ptr == NULL)
|
||||
return;
|
||||
|
||||
free(ptr->data);
|
||||
free(ptr);
|
||||
}
|
||||
|
||||
unsigned int syllable_chain_length(struct syllable *head)
|
||||
{
|
||||
struct syllable *walk = head;
|
||||
unsigned int length = 0;
|
||||
|
||||
while (walk) {
|
||||
length += strlen(walk->data);
|
||||
walk = walk->next;
|
||||
}
|
||||
|
||||
return length;
|
||||
}
|
||||
|
||||
char *syllable_chain_to_string(struct syllable *head)
|
||||
{
|
||||
struct syllable *walk = head;
|
||||
unsigned int length = syllable_chain_length(head);
|
||||
char *buffer = malloc(length + 1);
|
||||
char *ptr = buffer;
|
||||
|
||||
while (walk) {
|
||||
strcpy(ptr, walk->data);
|
||||
ptr += strlen(walk->data);
|
||||
walk = walk->next;
|
||||
}
|
||||
|
||||
return buffer;
|
||||
}
|
||||
|
||||
void syllable_chain_drop(struct syllable *head)
|
||||
{
|
||||
struct syllable *walk = head, *next;
|
||||
|
||||
while (walk) {
|
||||
next = walk->next;
|
||||
syllable_drop(walk);
|
||||
walk = next;
|
||||
}
|
||||
}
|
||||
20
syllable.h
20
syllable.h
|
|
@ -1,20 +0,0 @@
|
|||
/* SPDX-License-Identifier: GPL-2.0 */
|
||||
|
||||
#ifndef __SYLLABE_H
|
||||
#define __SYLLABE_H
|
||||
|
||||
struct syllable {
|
||||
char *data;
|
||||
unsigned int code;
|
||||
struct syllable *prev;
|
||||
struct syllable *next;
|
||||
};
|
||||
|
||||
struct syllable *syllable_alloc(const char *data, unsigned int code);
|
||||
void syllable_drop(struct syllable *syllable);
|
||||
|
||||
unsigned int syllable_chain_length(struct syllable *head);
|
||||
char *syllable_chain_to_string(struct syllable *head);
|
||||
void syllable_chain_drop(struct syllable *head);
|
||||
|
||||
#endif /* __SYLLABE_H */
|
||||
|
|
@ -1,84 +1,18 @@
|
|||
/* SPDX-License-Identifier: GPL-2.0 */
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "compat.h"
|
||||
#include "transliteration.h"
|
||||
#include "syllable.h"
|
||||
#include "iast.h"
|
||||
#include "utf8.h"
|
||||
|
||||
static const struct transliteration_letter *find_letter_by_code(unsigned int c,
|
||||
const struct transliteration_letter *table)
|
||||
char *transliterate_devanagari_to_latin(const char *devanagari)
|
||||
{
|
||||
const struct transliteration_letter *walk = table;
|
||||
|
||||
if (c == 0)
|
||||
return NULL;
|
||||
|
||||
while (walk->code != 0) {
|
||||
if (c == walk->code)
|
||||
return walk;
|
||||
walk++;
|
||||
}
|
||||
|
||||
return NULL;
|
||||
}
|
||||
|
||||
static void syllable_modify(struct syllable *syllable, const char *data)
|
||||
{
|
||||
char buffer[10];
|
||||
|
||||
if (syllable == NULL)
|
||||
return;
|
||||
|
||||
strcpy(buffer, syllable->data);
|
||||
buffer[strlen(buffer) - 1] = 0;
|
||||
strcat(buffer, data);
|
||||
|
||||
free(syllable->data);
|
||||
syllable->data = strdup(buffer);
|
||||
}
|
||||
|
||||
static void apply_transliteration_filters(struct syllable *head,
|
||||
const transliteration_filter_t *filters)
|
||||
{
|
||||
const transliteration_filter_t *filter = filters;
|
||||
|
||||
while (*filter) {
|
||||
(*filter)(head);
|
||||
filter++;
|
||||
}
|
||||
}
|
||||
|
||||
char *transliterate_devanagari_to_latin(const char *devanagari,
|
||||
const struct transliteration_context *context)
|
||||
{
|
||||
struct syllable *head = NULL, *tail = NULL, **indirect = &head;
|
||||
const struct transliteration_letter *letter;
|
||||
const char *ptr = devanagari;
|
||||
const char *end = ptr + strlen(devanagari);
|
||||
unsigned int code;
|
||||
struct translit_letter *table;
|
||||
char *retval;
|
||||
|
||||
while (ptr < end) {
|
||||
code = utf8_unpack_char(ptr);
|
||||
ptr += utf8_char_length(code);
|
||||
table = get_iast_transliteration_table();
|
||||
|
||||
letter = find_letter_by_code(code, context->table);
|
||||
if (letter && letter->flags & FLAG_MODIFIER) {
|
||||
syllable_modify(tail, letter->data);
|
||||
continue;
|
||||
}
|
||||
|
||||
*indirect = syllable_alloc(letter ? letter->data : NULL, code);
|
||||
tail = *indirect;
|
||||
indirect = &(*indirect)->next;
|
||||
}
|
||||
|
||||
apply_transliteration_filters(head, context->filters);
|
||||
|
||||
retval = syllable_chain_to_string(head);
|
||||
syllable_chain_drop(head);
|
||||
retval = strdup(""); /* FIXME */
|
||||
|
||||
return retval;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -3,8 +3,6 @@
|
|||
#ifndef __TRANSLITERATION_H
|
||||
#define __TRANSLITERATION_H
|
||||
|
||||
#include "syllable.h"
|
||||
|
||||
enum translit_letter_type {
|
||||
VOWEL,
|
||||
CONSONANT,
|
||||
|
|
@ -14,8 +12,6 @@ enum translit_letter_type {
|
|||
VOWEL_SIGN
|
||||
};
|
||||
|
||||
typedef void (*transliteration_filter_t)(struct syllable *syllable_chain);
|
||||
|
||||
struct translit_letter {
|
||||
unsigned int code;
|
||||
enum translit_letter_type type;
|
||||
|
|
@ -23,12 +19,10 @@ struct translit_letter {
|
|||
};
|
||||
|
||||
struct translit_context {
|
||||
const struct translit_letter *table;
|
||||
const transliteration_filter_t *filters;
|
||||
struct translit_letter *table;
|
||||
};
|
||||
|
||||
char *transliterate_devanagari_to_latin(const char *text,
|
||||
const struct transliteration_context *context);
|
||||
char *transliterate_devanagari_to_latin(const char *text);
|
||||
|
||||
static inline int is_devanagari(unsigned int code)
|
||||
{
|
||||
|
|
|
|||
Loading…
Reference in a new issue