mjohnson/espeak-ng
1
0
Fork 0
mirror of https://github.com/rhasspy/espeak-ng.git synced 2026-11-12 08:50:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

tokenizer.c: Support colon tokens.

Browse Source
This commit is contained in:
Reece H. Dunn
2017-05-01 19:22:48 +01:00
parent 7560070dcd
commit af7e8fc5a3
3 changed files with 21 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/libespeak-ng/tokenizer.c
+8 -1
View File
@@ -96,9 +96,10 @@ typedef enum {
ESPEAKNG_CTYPE_QUESTION_MARK,
ESPEAKNG_CTYPE_EXCLAMATION_MARK,
ESPEAKNG_CTYPE_COMMA,
ESPEAKNG_CTYPE_COLON,
} espeakng_CTYPE;
#define ESPEAKNG_CTYPE_PROPERTY_MASK 0xF00000000000C001ull
#define ESPEAKNG_CTYPE_PROPERTY_MASK 0xF80000000000C001ull
// Reference: http://www.unicode.org/reports/tr14/tr14-32.html -- Unicode Line Breaking Algorithm
static espeakng_CTYPE codepoint_type(uint32_t c)
@@ -145,6 +146,8 @@ static espeakng_CTYPE codepoint_type(uint32_t c)
return ESPEAKNG_CTYPE_EXCLAMATION_MARK;
case ESPEAKNG_PROPERTY_COMMA:
return ESPEAKNG_CTYPE_COMMA;
case ESPEAKNG_PROPERTY_COLON:
return ESPEAKNG_CTYPE_COLON;
}
// 4. Classify the remaining codepoints.
@@ -292,6 +295,10 @@ tokenizer_state_default(espeak_ng_TOKENIZER *tokenizer)
current += utf8_out(c, current);
*current = '\0';
return ESPEAKNG_TOKEN_COMMA;
case ESPEAKNG_CTYPE_COLON:
current += utf8_out(c, current);
*current = '\0';
return ESPEAKNG_TOKEN_COLON;
default:
current += utf8_out(c, current);
*current = '\0';
src/libespeak-ng/tokenizer.h
+1
View File
@@ -55,6 +55,7 @@ typedef enum
ESPEAKNG_TOKEN_QUESTION_MARK,
ESPEAKNG_TOKEN_EXCLAMATION_MARK,
ESPEAKNG_TOKEN_COMMA,
ESPEAKNG_TOKEN_COLON,
} espeak_ng_TOKEN_TYPE;
espeak_ng_TOKEN_TYPE
tests/tokenizer.c
+12 -1
View File
@@ -465,7 +465,7 @@ test_Latn_punctuation_tokens()
espeak_ng_TOKENIZER *tokenizer = create_tokenizer();
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, ". ? ! ,", -1, ESPEAKNG_ENCODING_US_ASCII) == ENS_OK);
assert(text_decoder_decode_string(decoder, ". ? ! , :", -1, ESPEAKNG_ENCODING_US_ASCII) == ENS_OK);
assert(tokenizer_reset(tokenizer, decoder, ESPEAKNG_TOKENIZER_OPTION_TEXT) == 1);
assert(tokenizer_read_next_token(tokenizer) == ESPEAKNG_TOKEN_FULL_STOP);
@@ -496,6 +496,14 @@ test_Latn_punctuation_tokens()
assert(tokenizer_get_token_text(tokenizer) != NULL);
assert(strcmp(tokenizer_get_token_text(tokenizer), ",") == 0);
assert(tokenizer_read_next_token(tokenizer) == ESPEAKNG_TOKEN_WHITESPACE);
assert(tokenizer_get_token_text(tokenizer) != NULL);
assert(strcmp(tokenizer_get_token_text(tokenizer), " ") == 0);
assert(tokenizer_read_next_token(tokenizer) == ESPEAKNG_TOKEN_COLON);
assert(tokenizer_get_token_text(tokenizer) != NULL);
assert(strcmp(tokenizer_get_token_text(tokenizer), ":") == 0);
assert(tokenizer_read_next_token(tokenizer) == ESPEAKNG_TOKEN_END_OF_BUFFER);
assert(tokenizer_get_token_text(tokenizer) != NULL);
assert(*tokenizer_get_token_text(tokenizer) == '\0');
@@ -597,6 +605,9 @@ print_tokens(espeak_ng_TEXT_DECODER *decoder)
case ESPEAKNG_TOKEN_COMMA:
printf("comma : %s\n", tokenizer_get_token_text(tokenizer));
break;
case ESPEAKNG_TOKEN_COLON:
printf("colon : %s\n", tokenizer_get_token_text(tokenizer));
break;
}
}