void compare_per_run()
{
int i, len;
char utf8buf_expect[16];
char utf8buf_commit[16];
printf( "Expected: %s", expect_string_buf );
printf( "Committed: ");
tested_word_count += (len = ueStrLen( expect_string_buf ) - 1);
/* omit the suffix character */
for ( i = 0; i < len; i++ ) {
ueStrNCpy( utf8buf_expect,
ueStrSeek( expect_string_buf, i ),
1, STRNCPY_CLOSE );
ueStrNCpy( utf8buf_commit,
ueStrSeek( commit_string_buf, i ),
1, STRNCPY_CLOSE );
if ( ! strcmp( utf8buf_expect, utf8buf_commit ) )
printf( "%s", utf8buf_commit );
else {
printf( "\033[44;37m%s\033[m", utf8buf_commit );
failed_word_count++;
}
}
memset( commit_string_buf, 0, MAXLEN );
printf( "\n\n" );
}
/* kpchen said, record is the index array of interval */
static void OutputRecordStr(
ChewingData *pgdata,
char *out_buf, int out_buf_len,
int *record, int nRecord,
uint16_t phoneSeq[], int nPhoneSeq,
char selectStr[][ MAX_PHONE_SEQ_LEN * MAX_UTF8_SIZE + 1 ],
IntervalType selectInterval[],
int nSelect, TreeDataType *ptd )
{
PhraseIntervalType inter;
int i;
LoadChar( pgdata, out_buf, out_buf_len, phoneSeq, nPhoneSeq );
for ( i = 0; i < nRecord; i++ ) {
inter = ptd->interval[ record[ i ] ];
ueStrNCpy(
ueStrSeek( out_buf, inter.from ),
( inter.p_phr )->phrase,
( inter.to - inter.from ), -1);
}
for ( i = 0; i < nSelect; i++ ) {
inter.from = selectInterval[ i ].from;
inter.to = selectInterval[ i ].to ;
ueStrNCpy(
ueStrSeek( out_buf, inter.from ),
selectStr[ i ], ( inter.to - inter.from ), -1);
}
}
int main ()
{
char *u8string;
int u8len;
int len;
char u8string2[16];
u8string = "HelloWorld";
u8len = ueStrLen(u8string);
len = strlen(u8string);
ok (u8len == len, "ueStrLen");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8len = ueStrLen(u8string);
ok (u8len == 6, "ueStrLen");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
ueStrNCpy(u8string2, u8string, 3, STRNCPY_CLOSE);
ok (!strcmp(u8string2, "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88" /* 測試計 */ ), "ueStrNCpy");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8string = ueStrSeek(u8string, 3);
ok (!strcmp(u8string, "\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6" /* 算長度 */ ), "ueStrSeek");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8string = ueStrSeek(u8string, 0);
ok (!strcmp(u8string, "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6" /* 測試計算長度 */ ), "ueStrSeek");
return exit_status();
}
static void ChangeSelectIntervalAndBreakpoint(ChewingData *pgdata, int from, int to, const char *str)
{
int i;
int user_alloc;
IntervalType inte;
inte.from = from;
inte.to = to;
for (i = 0; i < pgdata->nSelect; i++) {
if (IsIntersect(inte, pgdata->selectInterval[i])) {
RemoveSelectElement(i, pgdata);
i--;
}
}
pgdata->selectInterval[pgdata->nSelect].from = from;
pgdata->selectInterval[pgdata->nSelect].to = to;
/* No available selection */
if ((user_alloc = (to - from)) == 0)
return;
ueStrNCpy(pgdata->selectStr[pgdata->nSelect], str, user_alloc, 1);
pgdata->nSelect++;
if (user_alloc > 1) {
memset(&pgdata->bUserArrBrkpt[from + 1], 0, sizeof(int) * (user_alloc - 1));
memset(&pgdata->bUserArrCnnct[from + 1], 0, sizeof(int) * (user_alloc - 1));
}
}
void test_utf8()
{
char *u8string;
int u8len;
int len;
char u8string2[16];
start_testcase(NULL, fd);
u8string = "HelloWorld";
u8len = ueStrLen(u8string);
len = strlen(u8string);
ok(u8len == len, "ueStrLen");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8len = ueStrLen(u8string);
ok(u8len == 6, "ueStrLen");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
ueStrNCpy(u8string2, u8string, 3, STRNCPY_CLOSE);
ok(!strcmp(u8string2, "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88" /* 測試計 */ ), "ueStrNCpy");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8string = ueStrSeek(u8string, 3);
ok(!strcmp(u8string, "\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6" /* 算長度 */ ), "ueStrSeek");
u8string = "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6"; /* 測試計算長度 */
u8string = ueStrSeek(u8string, 0);
ok(!strcmp
(u8string, "\xE6\xB8\xAC\xE8\xA9\xA6\xE8\xA8\x88\xE7\xAE\x97\xE9\x95\xB7\xE5\xBA\xA6" /* 測試計算長度 */ ),
"ueStrSeek");
}
int main (int argc, char *argv[])
{
char *u8string;
int u8len;
u8string = "HelloWorld";
u8len = ueStrLen(u8string);
int len = strlen(u8string);
ok (u8len == len, "ueStrLen");
u8string = "測試計算長度";
u8len = ueStrLen(u8string);
ok (u8len == 6, "ueStrLen");
u8string = "測試計算長度";
char u8string2[16];
ueStrNCpy(u8string2, u8string, 3, STRNCPY_CLOSE);
ok (!strcmp(u8string2, "測試計"), "ueStrNCpy");
u8string = "測試計算長度";
u8string = ueStrSeek(u8string, 3);
ok (!strcmp(u8string, "算長度"), "ueStrSeek");
u8string = "測試計算長度";
u8string = ueStrSeek(u8string, 0);
ok (!strcmp(u8string, "測試計算長度"), "ueStrSeek");
return exit_status();
}
int faft_data_add_fixed(const char *s, ChewingData *data, ArrayCode ac)
{
if (s == NULL || s[0] == '\0')
return 0;
AddChi(ac, data);
--(data->chiSymbolCursor);
int i_p = PhoneSeqCursor(data);
++(data->chiSymbolCursor);
/* Add selected character */
ueStrNCpy(data->selectStr[data->nSelect], s, 1, STRNCPY_CLOSE);
/* Add one character interval */
data->selectInterval[data->nSelect].from = i_p;
data->selectInterval[data->nSelect].to = i_p+1;
++(data->nSelect);
return 1;
}
/* Routines */
int faft_data_add_symbol(const char *s, ChewingData *data)
{
if (s == NULL || s[0] == '\0')
return 0;
/* This part is copied from libchewing */
memmove(&(data->chiSymbolBuf[data->chiSymbolCursor + 1]),
&(data->chiSymbolBuf[data->chiSymbolCursor]),
sizeof(wch_t)
* (data->chiSymbolBufLen - data->chiSymbolCursor));
data->chiSymbolBuf[data->chiSymbolCursor].wch = (wchar_t) 0;
ueStrNCpy((char *) data->chiSymbolBuf[data->chiSymbolCursor].s,
s, 1, STRNCPY_CLOSE);
int key = FindSymbolKey(s);
data->symbolKeyBuf[data->chiSymbolCursor] = key ? key : '1';
data->bUserArrCnnct[ PhoneSeqCursor(data)] = 0;
++(data->chiSymbolBufLen);
++(data->chiSymbolCursor);
return 1;
}
static int CheckUserChoose(
ChewingData *pgdata,
uint16_t *new_phoneSeq, int from , int to,
Phrase **pp_phr,
char selectStr[][ MAX_PHONE_SEQ_LEN * MAX_UTF8_SIZE + 1 ],
IntervalType selectInterval[], int nSelect )
{
IntervalType inte, c;
int chno, len;
int user_alloc;
UserPhraseData *pUserPhraseData;
Phrase *p_phr = ALC( Phrase, 1 );
assert( p_phr );
inte.from = from;
inte.to = to;
*pp_phr = NULL;
/* pass 1
* if these exist one selected interval which is not contained by inte
* but has intersection with inte, then inte is an unacceptable interval
*/
for ( chno = 0; chno < nSelect; chno++ ) {
c = selectInterval[ chno ];
if ( IsIntersect( inte, c ) && ! IsContain( inte, c ) ) {
free( p_phr );
return 0;
}
}
/* pass 2
* if there exist one phrase satisfied all selectStr then return 1, else return 0.
* also store the phrase with highest freq
*/
pUserPhraseData = UserGetPhraseFirst( pgdata, new_phoneSeq );
p_phr->freq = -1;
do {
for ( chno = 0; chno < nSelect; chno++ ) {
c = selectInterval[ chno ];
if ( IsContain( inte, c ) ) {
/*
* find a phrase of ph_id where the text contains
* 'selectStr[chno]' test if not ok then return 0,
* if ok then continue to test. */
len = c.to - c.from;
if ( memcmp(
ueStrSeek( pUserPhraseData->wordSeq, c.from - from ),
selectStr[ chno ],
ueStrNBytes( selectStr[ chno ], len ) ) )
break;
}
}
if ( chno == nSelect ) {
/* save phrase data to "pp_phr" */
if ( pUserPhraseData->userfreq > p_phr->freq ) {
if ( ( user_alloc = ( to - from ) ) > 0 ) {
ueStrNCpy( p_phr->phrase,
pUserPhraseData->wordSeq,
user_alloc, 1);
}
p_phr->freq = pUserPhraseData->userfreq;
*pp_phr = p_phr;
}
}
} while ( ( pUserPhraseData = UserGetPhraseNext( pgdata, new_phoneSeq ) ) != NULL );
if ( p_phr->freq != -1 )
return 1;
free( p_phr );
return 0;
}
/** @brief Loading all possible phrases of certain length.
*
* Loading all possible phrases of certain length into ChoiceInfo structure from static
* and dynamic dictionaries,\n
* including number of total pages and the number of current page.\n
*/
static void SetChoiceInfo(
ChoiceInfo *pci,AvailInfo *pai, uint16 *phoneSeq, int cursor,
int candPerPage )
{
Word tempWord;
Phrase tempPhrase;
int len;
UserPhraseData *pUserPhraseData;
uint16 userPhoneSeq[ MAX_PHONE_SEQ_LEN ];
/* Clears previous candidates. */
memset( pci->totalChoiceStr, '\0', sizeof(char) * MAX_CHOICE * MAX_PHRASE_LEN * MAX_UTF8_SIZE + 1);
pci->nTotalChoice = 0;
len = pai->avail[ pai->currentAvail ].len;
assert(len);
/* secondly, read tree phrase */
if ( len == 1 ) { /* single character */
GetCharFirst( &tempWord, phoneSeq[ cursor ] );
do {
if ( ChoiceTheSame( pci, tempWord.word, ueBytesFromChar( tempWord.word[0] ) * sizeof( char ) ) )
continue;
memcpy(
pci->totalChoiceStr[ pci->nTotalChoice ],
tempWord.word, ueBytesFromChar( tempWord.word[0] ) * sizeof( char ) );
assert(pci->nTotalChoice <= MAX_CHOICE);
pci->totalChoiceStr[ pci->nTotalChoice ][ ueBytesFromChar( tempWord.word[0] ) ] = '\0';
pci->nTotalChoice++;
} while( GetCharNext( &tempWord ) );
}
/* phrase */
else {
if ( pai->avail[ pai->currentAvail ].id != -1 ) {
GetPhraseFirst( &tempPhrase, pai->avail[ pai->currentAvail ].id );
do {
if ( ChoiceTheSame(
pci,
tempPhrase.phrase,
len * ueBytesFromChar( tempPhrase.phrase[0] ) * sizeof( char ) ) ) {
continue;
}
ueStrNCpy( pci->totalChoiceStr[ pci->nTotalChoice ],
tempPhrase.phrase, len, 1);
pci->nTotalChoice++;
} while( GetPhraseNext( &tempPhrase ) );
}
memcpy( userPhoneSeq, &phoneSeq[ cursor ], sizeof( uint16 ) * len );
userPhoneSeq[ len ] = 0;
pUserPhraseData = UserGetPhraseFirst( userPhoneSeq );
if ( pUserPhraseData ) {
do {
/* check if the phrase is already in the choice list */
if ( ChoiceTheSame(
pci,
pUserPhraseData->wordSeq,
len * ueBytesFromChar( pUserPhraseData->wordSeq[0] ) * sizeof( char ) ) )
continue;
/* otherwise store it */
ueStrNCpy(
pci->totalChoiceStr[ pci->nTotalChoice ],
pUserPhraseData->wordSeq,
len, 1);
pci->nTotalChoice++;
} while( ( pUserPhraseData =
UserGetPhraseNext( userPhoneSeq ) ) != NULL );
}
}
/* magic number */
pci->nChoicePerPage = candPerPage;
if ( pci->nChoicePerPage > MAX_SELKEY )
pci->nChoicePerPage = MAX_SELKEY;
pci->nPage = CEIL_DIV( pci->nTotalChoice, pci->nChoicePerPage );
pci->pageNo = 0;
}
/** @brief Loading all possible phrases of certain length.
*
* Loading all possible phrases of certain length into ChoiceInfo structure
* from static and dynamic dictionaries, including number of total pages and
* the number of current page.
*/
static void SetChoiceInfo(ChewingData *pgdata)
{
Phrase tempPhrase;
int len;
UserPhraseData *pUserPhraseData;
uint16_t userPhoneSeq[MAX_PHONE_SEQ_LEN];
ChoiceInfo *pci = &(pgdata->choiceInfo);
AvailInfo *pai = &(pgdata->availInfo);
uint16_t *phoneSeq = pgdata->phoneSeq;
uint16_t *phoneSeqAlt = pgdata->phoneSeqAlt;
int cursor = PhoneSeqCursor(pgdata);
int candPerPage = pgdata->config.candPerPage;
/* Clears previous candidates. */
memset(pci->totalChoiceStr, '\0', MAX_CHOICE * MAX_PHRASE_LEN * MAX_UTF8_SIZE + 1);
pci->nTotalChoice = 0;
len = pai->avail[pai->currentAvail].len;
assert(len);
/* secondly, read tree phrase */
if (len == 1) { /* single character */
ChoiceInfoAppendChi(pgdata, pci, phoneSeq[cursor]);
if (phoneSeq[cursor] != phoneSeqAlt[cursor]) {
ChoiceInfoAppendChi(pgdata, pci, phoneSeqAlt[cursor]);
}
if (pgdata->bopomofoData.kbtype == KB_HSU || pgdata->bopomofoData.kbtype == KB_DVORAK_HSU) {
switch (phoneSeq[cursor]) {
case 0x2800: /* 'ㄘ' */
ChoiceInfoAppendChi(pgdata, pci, 0x30); /* 'ㄟ' */
break;
case 0x80: /* 'ㄧ' */
ChoiceInfoAppendChi(pgdata, pci, 0x20); /* 'ㄝ' */
break;
case 0x2A00: /* 'ㄙ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1); /* '˙' */
break;
case 0xA00: /* 'ㄉ' */
ChoiceInfoAppendChi(pgdata, pci, 0x2); /* 'ˊ' */
break;
case 0x800: /* 'ㄈ' */
ChoiceInfoAppendChi(pgdata, pci, 0x3); /* 'ˇ' */
break;
case 0x18: /* 'ㄜ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1200); /* 'ㄍ' */
break;
case 0x10: /* 'ㄛ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1600); /* 'ㄏ' */
break;
case 0x1E00: /* 'ㄓ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1800); /* 'ㄐ' */
ChoiceInfoAppendChi(pgdata, pci, 0x4); /* 'ˋ' */
break;
case 0x58: /* 'ㄤ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1400); /* 'ㄎ' */
break;
case 0x68: /* 'ㄦ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1000); /* 'ㄌ' */
ChoiceInfoAppendChi(pgdata, pci, 0x60); /* 'ㄥ' */
break;
case 0x2200: /* 'ㄕ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1C00); /* 'ㄒ' */
break;
case 0x2000: /* 'ㄔ' */
ChoiceInfoAppendChi(pgdata, pci, 0x1A00); /* 'ㄑ' */
break;
case 0x50: /* 'ㄣ' */
ChoiceInfoAppendChi(pgdata, pci, 0xE00); /* 'ㄋ' */
break;
case 0x48: /* 'ㄢ' */
ChoiceInfoAppendChi(pgdata, pci, 0x600); /* 'ㄇ' */
break;
default:
break;
}
}
}
/* phrase */
else {
if (pai->avail[pai->currentAvail].id) {
GetPhraseFirst(pgdata, &tempPhrase, pai->avail[pai->currentAvail].id);
do {
if (ChoiceTheSame(pci, tempPhrase.phrase, len * ueBytesFromChar(tempPhrase.phrase[0]))) {
continue;
}
ueStrNCpy(pci->totalChoiceStr[pci->nTotalChoice], tempPhrase.phrase, len, 1);
pci->nTotalChoice++;
} while (GetVocabNext(pgdata, &tempPhrase));
}
memcpy(userPhoneSeq, &phoneSeq[cursor], sizeof(uint16_t) * len);
userPhoneSeq[len] = 0;
pUserPhraseData = UserGetPhraseFirst(pgdata, userPhoneSeq);
if (pUserPhraseData) {
do {
/* check if the phrase is already in the choice list */
if (ChoiceTheSame(pci, pUserPhraseData->wordSeq, len * ueBytesFromChar(pUserPhraseData->wordSeq[0])))
continue;
/* otherwise store it */
ueStrNCpy(pci->totalChoiceStr[pci->nTotalChoice], pUserPhraseData->wordSeq, len, 1);
pci->nTotalChoice++;
} while ((pUserPhraseData = UserGetPhraseNext(pgdata, userPhoneSeq)) != NULL);
}
UserGetPhraseEnd(pgdata, userPhoneSeq);
}
/* magic number */
pci->nChoicePerPage = candPerPage;
assert(pci->nTotalChoice > 0);
pci->nPage = CEIL_DIV(pci->nTotalChoice, pci->nChoicePerPage);
pci->pageNo = 0;
pci->isSymbol = WORD_CHOICE;
}
CHEWING_API int chewing_handle_CtrlNum( ChewingContext *ctx, int key )
{
ChewingData *pgdata = ctx->data;
ChewingOutput *pgo = ctx->output;
int keystrokeRtn = KEYSTROKE_ABSORB;
int newPhraseLen;
int i;
uint16 addPhoneSeq[ MAX_PHONE_SEQ_LEN ];
char addWordSeq[ MAX_PHONE_SEQ_LEN * MAX_UTF8_SIZE + 1 ];
int phraseState;
int cursor;
CheckAndResetRange( pgdata );
if ( pgdata->bSelect )
return 0;
CallPhrasing( pgdata );
newPhraseLen = key - '0';
if ( key == '0' || key == '1' ) {
pgdata->bSelect = 1;
pgdata->choiceInfo.oldChiSymbolCursor = pgdata->chiSymbolCursor;
HaninSymbolInput(
&( pgdata->choiceInfo ),
&( pgdata->availInfo ),
pgdata->phoneSeq,
pgdata->config.candPerPage );
CallPhrasing( pgdata );
MakeOutputWithRtn( pgo, pgdata, keystrokeRtn );
return 0;
}
cursor = PhoneSeqCursor( pgdata );
if ( ! pgdata->config.bAddPhraseForward ) {
if (
newPhraseLen >= 1 &&
cursor + newPhraseLen - 1 <= pgdata->nPhoneSeq ) {
if ( NoSymbolBetween(
pgdata,
cursor,
cursor + newPhraseLen - 1 ) ) {
/* Manually add phrase to the user phrase database. */
memcpy( addPhoneSeq,
&pgdata->phoneSeq[ cursor ],
sizeof( uint16 ) * newPhraseLen );
addPhoneSeq[ newPhraseLen ] = 0;
ueStrNCpy( addWordSeq,
ueStrSeek( (char *) &pgdata->phrOut.chiBuf,
cursor ),
newPhraseLen, 1);
phraseState = UserUpdatePhrase( addPhoneSeq, addWordSeq );
SetUpdatePhraseMsg(
pgdata,
addWordSeq,
newPhraseLen,
phraseState );
/* Clear the breakpoint between the New Phrase */
for ( i = 1; i < newPhraseLen; i++ )
pgdata->bUserArrBrkpt[ cursor + i ] = 0;
}
}
}
else {
if (
newPhraseLen >= 1 &&
cursor - newPhraseLen >= 0 ) {
if ( NoSymbolBetween( pgdata,
cursor,
cursor - newPhraseLen ) ) {
/* Manually add phrase to the user phrase database. */
memcpy( addPhoneSeq,
&pgdata->phoneSeq[ cursor - newPhraseLen ],
sizeof( uint16 ) * newPhraseLen );
addPhoneSeq[ newPhraseLen ] = 0;
ueStrNCpy( addWordSeq,
ueStrSeek( (char *) &pgdata->phrOut.chiBuf,
cursor - newPhraseLen ),
newPhraseLen, 1);
phraseState = UserUpdatePhrase( addPhoneSeq, addWordSeq );
SetUpdatePhraseMsg(
pgdata,
addWordSeq,
newPhraseLen,
phraseState );
/* Clear the breakpoint between the New Phrase */
for ( i = 1; i < newPhraseLen; i++ )
pgdata->bUserArrBrkpt[ cursor - newPhraseLen + i ] = 0;
}
}
}
CallPhrasing( pgdata );
MakeOutputWithRtn( pgo, pgdata, keystrokeRtn );
MakeOutputAddMsgAndCleanInterval( pgo, pgdata );
return 0;
}
