void IdnaConfTest::Test(void){
if (!ReadAndConvertFile())return;
UnicodeString s;
UnicodeString key;
UnicodeString value;
// skip everything before the first "=====" and "=====" itself
do {
if (!ReadOneLine(s)) {
errln("End of file prematurely found");
break;
}
}
while (s.compare(C_TAG, -1) != 0); //"====="
while(ReadOneLine(s)){
s.trim();
key.remove();
value.remove();
if (s.compare(C_TAG, -1) == 0){ //"====="
Call();
} else {
// explain key:value
int p = s.indexOf((UChar)0x3A); // :
key.setTo(s,0,p).trim();
value.setTo(s,p+1).trim();
if (key.compare(C_TYPE, -1) == 0){
if (value.compare(C_TOASCII, -1) == 0) {
type = 0;
} else if (value.compare(C_TOUNICODE, -1) == 0){
type = 1;
}
} else if (key.compare(C_PASSFAIL, -1) == 0){
if (value.compare(C_PASS, -1) == 0){
passfail = 0;
} else if (value.compare(C_FAIL, -1) == 0){
passfail = 1;
}
} else if (key.compare(C_DESC, -1) == 0){
if (value.indexOf(C_USESTD3ASCIIRULES, u_strlen(C_USESTD3ASCIIRULES), 0) == -1){
option = 1; // not found
} else {
option = 0;
}
id.setTo(value, 0, value.indexOf((UChar)0x20)); // space
} else if (key.compare(C_NAMEZONE, -1) == 0){
ExplainCodePointTag(value);
namezone.setTo(value);
} else if (key.compare(C_NAMEBASE, -1) == 0){
ExplainCodePointTag(value);
namebase.setTo(value);
}
// just skip other lines
}
}
Call(); // for last record
}
/* ReadOneLine: */
static Boolean ReadOneLine(FILE * fp, char *line)
{
if (feof(fp))
return FALSE;
line[0] = 0;
fgets(line, 256, fp);
if (line[0] == '#' || line[0] == 0x0a || line[0] == 0x0d)
return ReadOneLine(fp, line);
return TRUE;
}
/* LoadScaleRate: Load scale rate file for parameter updating */
static void LoadScaleRate(char *scalefile)
{
char *line;
FILE *fp;
int i, p;
float fScale;
line = (char *) malloc(256 * sizeof(char));
fp = fopen(scalefile, "rt");
/* read the scale */
ReadOneLine(fp, line);
fScale = (float) atof(line);
for (p = 1; p <= genInfo->nPdfStream[0]; p++) {
for (i = 1; i <= genInfo->pst[p].order; i++) {
if (!ReadOneLine(fp, line))
break;
mtInfo->SRMean[p][i] = (float) atof(line) * fScale;
}
}
/* read the scale */
ReadOneLine(fp, line);
fScale = (float) atof(line);
for (p = 1; p <= genInfo->nPdfStream[0]; p++) {
for (i = 1; i <= genInfo->pst[p].order; i++) {
if (!ReadOneLine(fp, line))
break;
mtInfo->SRVar[p][i] = (float) atof(line) * fScale;
}
}
/* read the scale */
ReadOneLine(fp, line);
fScale = (float) atof(line);
for (p = 1; p <= genInfo->nPdfStream[0]; p++) {
for (i = 1; i <= genInfo->pst[p].order; i++) {
if (!ReadOneLine(fp, line))
break;
mtInfo->DWght[p][i] = (float) atof(line) * fScale;
}
}
free(line);
fclose(fp);
}
/* Function: SeqfileFormat()
*
* Purpose: Determine format of seqfile, and return it
* through ret_format. From Gilbert's seqFileFormat().
*
* If filename is "-", we will read from stdin and
* assume that the stream is coming in FASTA format --
* either unaligned or aligned.
*
* Args: filename - name of sequence file
* ret_format - RETURN: format code for file, see squid.h
* for codes.
* env - name of environment variable containing
* a directory path that filename might also be
* found in. "BLASTDB", for example. Can be NULL.
*
* Return: 1 on success, 0 on failure.
*/
int
SeqfileFormat(char *filename, int *ret_format, char *env)
{
int foundIG = 0;
int foundStrider = 0;
int foundGB = 0;
int foundEMBL = 0;
int foundPearson = 0;
int foundZuker = 0;
int gotGCGdata = 0;
int gotPIR = 0;
int gotSquid = 0;
int gotuw = 0;
int gotMSF = 0;
int gotClustal = 0;
int done = 0;
int format = kUnknown;
int nlines= 0, dnalines= 0;
int splen = 0;
char sp[LINEBUFLEN];
FILE *fseq;
/* First check if filename is "-": special case indicating
* a FASTA pipe.
*/
if (strcmp(filename, "-") == 0)
{ *ret_format = kPearson; return 1; }
#define ReadOneLine(sp) \
{ done |= (feof(fseq)); \
readline( fseq, sp); \
if (!done) { splen = (int) strlen(sp); ++nlines; } }
if ((fseq = fopen(filename, "r")) == NULL &&
(fseq = EnvFileOpen(filename, env)) == NULL)
{ squid_errno = SQERR_NOFILE; return 0; }
/* Look at a line at a time
*/
while ( !done ) {
ReadOneLine(sp);
if (sp==NULL || *sp=='\0')
/*EMPTY*/ ;
/* high probability identities: */
else if (strstr(sp, " MSF:") != NULL &&
strstr(sp, " Type:") != NULL &&
strstr(sp, " Check:") != NULL)
gotMSF = 1;
else if (strncmp(sp, "CLUSTAL ", 8) == 0 &&
strstr( sp, "multiple sequence alignment"))
gotClustal = 1;
else if (strstr(sp," Check: ") != NULL)
gotuw= 1;
else if (strncmp(sp, "///", 3) == 0 || strncmp(sp, "ENTRY ", 6) == 0)
gotPIR = 1;
else if (strncmp(sp, "++", 2) == 0 || strncmp(sp, "NAM ", 4) == 0)
gotSquid = 1;
else if (strncmp(sp, ">>>>", 4) == 0 && strstr(sp, "Len: "))
gotGCGdata = 1;
/* uncertain identities: */
else if (*sp ==';') {
if (strstr(sp,"Strider") !=NULL) foundStrider= 1;
else foundIG= 1;
}
else if (strncmp(sp,"LOCUS",5) == 0 || strncmp(sp,"ORIGIN",5) == 0)
foundGB= 1;
else if (*sp == '>') {
foundPearson = 1;
}
else if (strstr(sp,"ID ") == sp || strstr(sp,"SQ ") == sp)
foundEMBL= 1;
else if (*sp == '(')
foundZuker= 1;
else {
switch (Seqtype( sp )) {
case kDNA:
case kRNA: if (splen>20) dnalines++; break;
default: break;
}
}
if (gotMSF) {format = kMSF; done = 1; }
else if (gotClustal) {format = kClustal; done = 1; }
else if (gotSquid) {format = kSquid; done = 1; }
else if (gotPIR) {format = kPIR; done = 1; }
else if (gotGCGdata) {format = kGCGdata; done = 1; }
else if (gotuw)
{
if (foundIG) format= kIG; /* a TOIG file from GCG for certain */
else format= kGCG;
done= 1;
}
else if ((dnalines > 1) || done || (nlines > 500)) {
/* decide on most likely format */
/* multichar idents: */
if (foundStrider) format= kStrider;
else if (foundGB) format= kGenBank;
else if (foundEMBL) format= kEMBL;
/* single char idents: */
else if (foundIG) format= kIG;
else if (foundPearson) format= kPearson;
else if (foundZuker) format= kZuker;
/* spacing ident: */
else if (IsSELEXFormat(filename)) format= kSelex;
/* no format chars: */
else
{
squid_errno = SQERR_FORMAT;
return 0;
}
done= 1;
}
}
if (fseq!=NULL) fclose(fseq);
*ret_format = format;
return 1;
#undef ReadOneLine
}
