int main(int argc, char** argv){
char number[5]="123ff";
int inum=charconv(number[0]);
printf("value:%d\n",inum);
inum = inum*(charconv( number[1]));
printf("value:%d\n",inum);
inum = inum*(charconv(number[2]));
printf("value:%d\n",inum);
inum = inum*(charconv( number[3]));
printf("value:%d\n",inum);
inum = inum*(charconv( number[4]));
printf("value:%d\n",inum);
return 0;
}
void Lexer::string() {
char start = _stream->current();
// look for unescaped end same as start.
char c;
while (true) {
c = _stream->read();
if (c == '\\') { // escaped.
// read next char and continue.
_word[_wi++] = charconv(_stream->read());
continue;
} else if (c == 0)
break;
if (start == c)
break;
_word[_wi++] = c;
}
_word[_wi++] = '\0';
add(new (collect) Term(TString, dup_str(_word)));
_wi = 0;
}
LIBETPAN_EXPORT
int mailmime_encoded_phrase_parse(const char * default_fromcode,
const char * message, size_t length,
size_t * indx, const char * tocode,
char ** result)
{
MMAPString * gphrase;
struct mailmime_encoded_word * word;
int first;
size_t cur_token;
int r;
int res;
char * str;
char * wordutf8;
int type;
cur_token = * indx;
gphrase = mmap_string_new("");
if (gphrase == NULL) {
res = MAILIMF_ERROR_MEMORY;
goto err;
}
first = TRUE;
type = TYPE_ERROR; /* XXX - removes a gcc warning */
while (1) {
int has_fwd = 0;
word = NULL;
r = mailmime_encoded_word_parse(message, length, &cur_token, &word, &has_fwd);
if (r == MAILIMF_NO_ERROR) {
if ((!first) && has_fwd) {
if (type != TYPE_ENCODED_WORD) {
if (mmap_string_append_c(gphrase, ' ') == NULL) {
mailmime_encoded_word_free(word);
res = MAILIMF_ERROR_MEMORY;
goto free;
}
}
}
type = TYPE_ENCODED_WORD;
wordutf8 = NULL;
r = charconv(tocode, word->wd_charset, word->wd_text,
strlen(word->wd_text), &wordutf8);
switch (r) {
case MAIL_CHARCONV_ERROR_MEMORY:
mailmime_encoded_word_free(word);
res = MAILIMF_ERROR_MEMORY;
goto free;
case MAIL_CHARCONV_ERROR_UNKNOWN_CHARSET:
r = charconv(tocode, "iso-8859-1", word->wd_text,
strlen(word->wd_text), &wordutf8);
break;
case MAIL_CHARCONV_ERROR_CONV:
mailmime_encoded_word_free(word);
res = MAILIMF_ERROR_PARSE;
goto free;
}
switch (r) {
case MAIL_CHARCONV_ERROR_MEMORY:
mailmime_encoded_word_free(word);
res = MAILIMF_ERROR_MEMORY;
goto free;
case MAIL_CHARCONV_ERROR_CONV:
mailmime_encoded_word_free(word);
res = MAILIMF_ERROR_PARSE;
goto free;
}
if (wordutf8 != NULL) {
if (mmap_string_append(gphrase, wordutf8) == NULL) {
mailmime_encoded_word_free(word);
free(wordutf8);
res = MAILIMF_ERROR_MEMORY;
goto free;
}
free(wordutf8);
}
mailmime_encoded_word_free(word);
first = FALSE;
}
else if (r == MAILIMF_ERROR_PARSE) {
/* do nothing */
}
else {
res = r;
goto free;
}
if (r == MAILIMF_ERROR_PARSE) {
char * raw_word;
raw_word = NULL;
r = mailmime_non_encoded_word_parse(message, length,
&cur_token, &raw_word, &has_fwd);
if (r == MAILIMF_NO_ERROR) {
if ((!first) && has_fwd) {
if (mmap_string_append_c(gphrase, ' ') == NULL) {
free(raw_word);
res = MAILIMF_ERROR_MEMORY;
goto free;
}
}
type = TYPE_WORD;
wordutf8 = NULL;
r = charconv(tocode, default_fromcode, raw_word,
strlen(raw_word), &wordutf8);
switch (r) {
case MAIL_CHARCONV_ERROR_MEMORY:
free(raw_word);
res = MAILIMF_ERROR_MEMORY;
goto free;
case MAIL_CHARCONV_ERROR_UNKNOWN_CHARSET:
case MAIL_CHARCONV_ERROR_CONV:
free(raw_word);
res = MAILIMF_ERROR_PARSE;
goto free;
}
if (mmap_string_append(gphrase, wordutf8) == NULL) {
free(wordutf8);
free(raw_word);
res = MAILIMF_ERROR_MEMORY;
goto free;
}
free(wordutf8);
free(raw_word);
first = FALSE;
}
else if (r == MAILIMF_ERROR_PARSE) {
r = mailimf_fws_parse(message, length, &cur_token);
if (r != MAILIMF_NO_ERROR) {
break;
}
if (mmap_string_append_c(gphrase, ' ') == NULL) {
res = MAILIMF_ERROR_MEMORY;
goto free;
}
first = FALSE;
break;
}
else {
res = r;
goto free;
}
}
}
if (first) {
if (cur_token != length) {
res = MAILIMF_ERROR_PARSE;
goto free;
}
}
str = strdup(gphrase->str);
if (str == NULL) {
res = MAILIMF_ERROR_MEMORY;
goto free;
}
mmap_string_free(gphrase);
* result = str;
* indx = cur_token;
return MAILIMF_NO_ERROR;
free:
mmap_string_free(gphrase);
err:
return res;
}
int
main(int argc, char *argv[])
{
FILE *fp;
char header[512];
time_t now;
char *binfile, *lrfile, *rlfile, *outfile;
int i;
char *from_code, *to_code, *buf;
boolean charconv_enabled = FALSE;
boolean force_swap = FALSE;
WORD_ID w;
binfile = lrfile = rlfile = outfile = NULL;
from_code = to_code = NULL;
if (argc <= 1) {
usage(argv[0]);
return -1;
}
for(i=1;i<argc;i++) {
if (argv[i][0] == '-') {
if (argv[i][1] == 'd') {
if (++i >= argc) {
printf("Error: no argument for option \"%s\"\n", argv[i]);
usage(argv[0]);
return -1;
}
binfile = argv[i];
} else if (argv[i][1] == 'n') {
switch(argv[i][2]) {
case 'l':
if (++i >= argc) {
printf("Error: no argument for option \"%s\"\n", argv[i]);
usage(argv[0]);
return -1;
}
lrfile = argv[i];
break;
case 'r':
if (++i >= argc) {
printf("Error: no argument for option \"%s\"\n", argv[i]);
usage(argv[0]);
return -1;
}
rlfile = argv[i];
break;
default:
printf("Error: no such option \"%s\"\n", argv[i]);
usage(argv[0]);
return -1;
}
} else if (argv[i][1] == 'c') {
if (++i >= argc) {
printf("Error: no argument for option \"%s\"\n", argv[i]);
usage(argv[0]);
return -1;
}
from_code = strcpy((char*)mymalloc(strlen(argv[i])+1), argv[i]);
if (++i >= argc) {
printf("Error: no argument for option \"%s\"\n", argv[i]);
usage(argv[0]);
free(from_code);
return -1;
}
to_code = strcpy((char*)mymalloc(strlen(argv[i])+1),argv[i]);
charconv_enabled = TRUE;
} else if (argv[i][1] == 's') {
force_swap = TRUE;
}
} else {
if (outfile == NULL) {
outfile = argv[i];
} else {
printf("Error: more than one output file\n");
usage(argv[0]);
return -1;
}
}
}
if (!outfile) {
printf("Error: no output file specified\n");
usage(argv[0]);
return -1;
}
if (binfile) {
if (lrfile || rlfile) {
printf("Error: both binary file and ARPA file are specified\n");
usage(argv[0]);
return -1;
}
printf("bingram: %s\n", binfile);
} else {
if (rlfile) {
printf("backward n-gram: %s\n", rlfile);
if (lrfile) {
printf("additional forward 2-gram for 1st pass: %s\n", lrfile);
}
} else if (lrfile) {
printf("forward n-gram: %s\n", lrfile);
} else {
printf("Error: no input N-gram file specified\n");
usage(argv[0]);
return -1;
}
}
printf("\nSTART LOADING\n\n");
/* make header string */
now = time(NULL);
if (binfile) {
sprintf(header, "converted at %s\nfrom bingram = %s\n", ctime(&now), binfile);
} else {
if (rlfile && lrfile) {
sprintf(header, "converted at %s\nfrom n-gram = %s, LR 2-gram = %s\n", ctime(&now), rlfile, lrfile);
} else if (rlfile) {
sprintf(header, "converted at %s\nfrom n-gram = %s\n", ctime(&now), rlfile);
} else {
sprintf(header, "converted at %s\nfrom n-gram = %s\n", ctime(&now), lrfile);
}
}
ngram = ngram_info_new();
if (binfile) {
/* read in bingram */
if (init_ngram_bin(ngram, binfile) == FALSE) return -1;
} else {
/* read in ARPA n-gram */
if (force_swap) {
ngram->bos_eos_swap = TRUE;
}
if (rlfile) {
if (init_ngram_arpa(ngram, rlfile, DIR_RL) == FALSE) return -1;
if (lrfile) {
if (init_ngram_arpa_additional(ngram, lrfile) == FALSE) return -1;
}
} else if (lrfile) {
if (init_ngram_arpa(ngram, lrfile, DIR_LR) == FALSE) return -1;
}
}
print_ngram_info(stdout, ngram);
if (charconv_enabled == TRUE) {
/* do character conversion */
if (charconv_setup(from_code, to_code) == -1) {
fprintf(stderr, "failed to setup character convertsion\n");
return -1;
}
buf = (char *)mymalloc(4096);
for (w = 0; w < ngram->max_word_num; w++) {
charconv(ngram->wname[w], buf, 4096);
ngram->wname[w] = mybmalloc2(strlen(buf)+1, &(ngram->mroot));
strcpy(ngram->wname[w], buf);
}
free(buf);
}
/* write in JULIUS binary format */
if ((fp = fopen_writefile(outfile)) == NULL) {
fprintf(stderr, "failed to open \"%s\"\n", outfile);
return -1;
}
printf("\nWriting in v5 format to \"%s\"...\n", outfile);
if (ngram_write_bin(fp, ngram, header) == FALSE){/* failed */
fprintf(stderr, "failed to write \"%s\"\n",outfile);
return -1;
}
fclose_writefile(fp);
printf("completed\n");
return 0;
}
LIBETPAN_EXPORT
int mailmime_encoded_phrase_parse2(const char * default_fromcode,
const char * message,
const char * tocode,
char ** result)
{
long res;
char * charset = NULL;
char * encoding = NULL;
char * text = NULL;
size_t current = 0;
size_t firstQuotes,secondQuotes;
if (message == NULL) {
return 0;
}
while(message[current] != '\0'){
if(message[current] == '\'') {
break;
}
current++;
}
if (message[current] == '\0' || message[current+1] == '\0') {
goto free;
}
firstQuotes = current;
current++;
while(message[current] != '\0'){
if(message[current] == '\'') {
break;
}
current++;
}
if (message[current] == '\0' || message[current+1] == '\0') {
goto free;
}
secondQuotes = current;
current++;
charset = (char*)malloc(firstQuotes + 1);
strncpy(charset, message, firstQuotes);
charset[firstQuotes] = '\0';
encoding = (char*)malloc(secondQuotes - firstQuotes);
strncpy(encoding, message+firstQuotes+1, secondQuotes - firstQuotes -1);
encoding[secondQuotes-firstQuotes] = '\0';
size_t length = strlen(message);
text = (char*)malloc(length - secondQuotes);
memset(text, 0, length - secondQuotes);
res = urldecode(text, message+current, 0);
if (res > 0 && strcasecmp(tocode,charset) != 0) {
char * wordutf8 = NULL;
if (strcasecmp(charset, "unknown") == 0) {
charconv(tocode, default_fromcode, text, strlen(text), &wordutf8);
}else{
charconv(tocode, charset, text, strlen(text), &wordutf8);
}
if (wordutf8 != NULL) {
* result = wordutf8;
} else {
* result = text;
text = NULL;
}
} else {
* result = text;
text = NULL;
}
free:
if (charset != NULL) {
free(charset);
}
if (encoding != NULL) {
free(encoding);
}
if (text != NULL) {
free(text);
}
return 0;
}
