static char *
bibtex_item( char *p, newstr *s )
{
int nquotes = 0;
int nbrackets = 0;
while ( *p ) {
if ( !nquotes && !nbrackets ) {
if ( *p==',' || *p=='=' || *p=='}' || *p==')' )
goto out;
}
if ( *p=='\"' && *(p-1)!='\\') {
nquotes = ( nquotes==0 );
newstr_addchar( s, *p );
} else if ( *p=='{' ) {
nbrackets++;
/*if ( s->len!=0 )*/ newstr_addchar( s, *p );
} else if ( *p=='}' ) {
nbrackets--;
/*if ( nbrackets>0 )*/ newstr_addchar( s, *p );
} else {
if ( s->len!=0 || ( s->len==0 && !is_ws( *p ) ) )
newstr_addchar( s, *p );
}
p++;
}
out:
newstr_trimendingws( s );
return p;
}
int
test_addchar( newstr *s )
{
int failed = 0;
int numshort = 5, numchars = 1000, i;
/* ...appending '\0' characters won't increase length */
newstr_empty( s );
for ( i=0; i<numshort; ++i )
newstr_addchar( s, '\0' );
if ( test_consistency( s, 0, __FUNCTION__ ) || test_identity( s, "" ) )
failed++;
/* ...build "11111" with newstr_addchar */
newstr_empty( s );
for ( i=0; i<numshort; ++i )
newstr_addchar( s, '1' );
if ( test_consistency( s, 5, __FUNCTION__ ) || test_identity( s, "11111" ) )
failed++;
newstr_empty( s );
for ( i=0; i<numchars; ++i ) {
newstr_addchar( s, ( i % 64 ) + 64);
}
if ( test_consistency( s, numchars, __FUNCTION__ ) )
failed++;
return failed;
}
static void
bibtex_split( list *tokens, newstr *s )
{
newstr currtok;
int nquotes = 0, nbrackets = 0;
int i, n = s->len;
newstr_init( &currtok );
for ( i=0; i<n; ++i ) {
if ( s->data[i]=='\"' ) {
if ( nquotes ) nquotes = 0;
else nquotes = 1;
newstr_addchar( &currtok, '\"' );
} else if ( s->data[i]=='{' ) {
nbrackets++;
newstr_addchar( &currtok, '{' );
} else if ( s->data[i]=='}' ) {
nbrackets--;
newstr_addchar( &currtok, '}' );
} else if ( s->data[i]=='#' && !nquotes && !nbrackets ) {
if ( currtok.len ) list_add( tokens, currtok.data );
newstr_empty( &currtok );
} else if ( !is_ws( s->data[i] ) || nquotes || nbrackets ) {
newstr_addchar( &currtok, s->data[i] );
}
}
if ( currtok.len ) list_add( tokens, currtok.data );
for ( i=0; i<tokens->n; ++i ) {
newstr_trimendingws( &(tokens->str[i]) );
}
newstr_free( &currtok );
}
/* get reference name */
static char*
process_bibtexid( char *p, newstr *data )
{
newstr tmp;
char *start_p = p;
newstr_init( &tmp );
newstr_empty( data );
while ( *p && *p!=',' ) newstr_addchar( &tmp, *p++ );
if ( *p==',' ) p++;
p = skip_ws( p ); /* skip ending newline/carriage return */
if ( tmp.len ) {
if ( strchr( tmp.data, '=' ) ) {
/* Endnote writes bibtex files w/o fields, try to
* distinguish via presence of an equal sign.... if
* it's there, assume that it's a tag/data pair instead
* and roll back.
*/
p = start_p;
} else {
/* add '{' and '}' to protect from string expansion */
newstr_addchar( data, '{' );
newstr_strcat( data, tmp.data );
newstr_addchar( data, '}' );
}
}
newstr_free( &tmp );
return p;
}
/* extract_range()
*
* Handle input strings like:
*
* "1-15"
* " 1 - 15 "
* " 1000--- 1500"
* " 1 <<em-dash>> 10"
* " 107 111"
*/
static void
extract_range( newstr *input, newstr *begin, newstr *end )
{
/* -30 is the first character of a UTF8 em-dash and en-dash */
const char terminators[] = { ' ', '-', '\t', '\r', '\n', -30, '\0' };
char *p;
newstr_empty( begin );
newstr_empty( end );
if ( input->len==0 ) return;
p = skip_ws( input->data );
while ( *p && !strchr( terminators, *p ) )
newstr_addchar( begin, *p++ );
p = skip_ws( p );
while ( *p=='-' ) p++;
while ( utf8_is_emdash( p ) ) p+=3;
while ( utf8_is_endash( p ) ) p+=3;
p = skip_ws( p );
while ( *p && !strchr( terminators, *p ) )
newstr_addchar( end, *p++ );
}
static void
output_person( FILE *fp, char *name )
{
newstr family, given, suffix;
char *p = name;
newstrs_init( &family, &given, &suffix, NULL );
while ( *p && *p!='|' )
newstr_addchar( &family, *p++ );
while ( *p=='|' && *(p+1)!='|' ) {
p++;
if ( *p!='|' ) newstr_addchar( &given, *p++ );
while ( *p && *p!='|' ) p++;
}
if ( *p=='|' && *(p+1)=='|' ) {
p += 2;
while ( *p && *p!='|' ) newstr_addchar( &suffix, *p++ );
}
if ( family.len ) fprintf( fp, "%s", family.data );
if ( suffix.len ) {
if ( family.len ) fprintf( fp, " %s", suffix.data );
else fprintf( fp, "%s", suffix.data );
}
if ( given.len ) fprintf( fp, ", %s", given.data );
newstrs_free( &family, &given, &suffix, NULL );
}
/* <AuthorList CompleteYN="Y">
* <Author>
* <LastName>Barondeau</LastName>
* <ForeName>David P</ForeName>
* ( or <FirstName>David P</FirstName> )
* <Initials>DP</Initials>
* </Author>
* <Author>
* <CollectiveName>Organization</CollectiveName>
* </Author>
* </AuthorList>
*/
static int
medin_author( xml *node, newstr *name )
{
char *p;
if ( xml_tagexact( node, "LastName" ) ) {
if ( name->len ) {
newstr_prepend( name, "|" );
newstr_prepend( name, xml_data( node ) );
}
else newstr_strcat( name, xml_data( node ) );
} else if ( xml_tagexact( node, "ForeName" ) ||
xml_tagexact( node, "FirstName" ) ) {
p = xml_data( node );
while ( p && *p ) {
if ( name->len ) newstr_addchar( name, '|' );
while ( *p && *p==' ' ) p++;
while ( *p && *p!=' ' ) newstr_addchar( name, *p++ );
}
} else if ( xml_tagexact( node, "Initials" ) && !strchr( name->data, '|' )) {
p = xml_data( node );
while ( p && *p ) {
if ( name->len ) newstr_addchar( name, '|' );
if ( !is_ws(*p) ) newstr_addchar( name, *p++ );
}
}
if ( node->next ) medin_author( node->next, name );
return BIBL_OK;
}
static int
test_addchar( newstr *s )
{
int failed = 0;
int numshort = 5, numchars = 1000, i;
/* ...appending '\0' characters won't increase length */
newstr_empty( s );
for ( i=0; i<numshort; ++i )
newstr_addchar( s, '\0' );
if ( string_mismatch( s, 0, "" ) ) failed++;
/* ...build "11111" with newstr_addchar */
newstr_empty( s );
for ( i=0; i<numshort; ++i )
newstr_addchar( s, '1' );
if ( string_mismatch( s, 5, "11111" ) ) failed++;
/* ...build a bunch of random characters */
newstr_empty( s );
for ( i=0; i<numchars; ++i ) {
newstr_addchar( s, ( i % 64 ) + 64);
}
if ( inconsistent_len( s, numchars ) ) failed++;
return failed;
}
static int
generate_citekey( fields *info, int nref )
{
newstr citekey;
int n1, n2;
char *p, buf[100];
newstr_init( &citekey );
n1 = fields_find( info, "AUTHOR", 0 );
if ( n1==-1 ) n1 = fields_find( info, "AUTHOR", -1 );
n2 = fields_find( info, "YEAR", 0 );
if ( n2==-1 ) n2 = fields_find( info, "YEAR", -1 );
if ( n2==-1 ) n2 = fields_find( info, "PARTYEAR", 0 );
if ( n2==-1 ) n2 = fields_find( info, "PARTYEAR", -1 );
if ( n1!=-1 && n2!=-1 ) {
p = info->data[n1].data;
while ( p && *p && *p!='|' ) {
if ( !is_ws( *p ) ) newstr_addchar( &citekey, *p );
p++;
}
p = info->data[n2].data;
while ( p && *p ) {
if ( !is_ws( *p ) ) newstr_addchar( &citekey, *p );
p++;
}
fields_add( info, "REFNUM", citekey.data, 0 );
} else {
sprintf( buf, "ref%d\n", nref );
newstr_strcpy( &citekey, buf );
}
newstr_free( &citekey );
return fields_find( info, "REFNUM", -1 );
}
static void
resolve_citekeys( bibl *b, list *citekeys, int *dup )
{
char abc[]="abcdefghijklmnopqrstuvwxyz";
newstr tmp;
int nsame, ntmp, n, i, j;
newstr_init( &tmp );
for ( i=0; i<citekeys->n; ++i ) {
if ( dup[i]==-1 ) continue;
nsame = 0;
for ( j=i; j<citekeys->n; ++j ) {
if ( dup[j]!=i ) continue;
newstr_newstrcpy( &tmp, &(citekeys->str[j]) );
ntmp = nsame;
while ( ntmp >= 26 ) {
newstr_addchar( &tmp, 'a' );
ntmp -= 26;
}
if ( ntmp<26 && ntmp>=0 )
newstr_addchar( &tmp, abc[ntmp] );
nsame++;
dup[j] = -1;
n = fields_find( b->ref[j], "REFNUM", -1 );
if ( n!=-1 )
newstr_newstrcpy(&((b->ref[j])->data[n]),&tmp);
}
}
newstr_free( &tmp );
}
static int
build_refnum( fields *info, long nrefs )
{
newstr refnum;
char *p, num[512];
int y, a;
newstr_init( &refnum );
y = fields_find( info, "YEAR", -1 );
if ( y==-1 ) y = fields_find( info, "PARTYEAR", -1 );
a = fields_find( info, "AUTHOR", -1 );
if ( a==-1 ) a = fields_find( info, "AUTHOR:CORP", -1 );
if ( a!=-1 && y!=-1 ) {
p = info->data[a].data;
while ( p && *p && *p!='|' )
newstr_addchar( &refnum, *p++ );
p = info->data[y].data;
while ( p && *p && *p!=' ' && *p!='\t' )
newstr_addchar( &refnum, *p++ );
} else {
sprintf( num, "%ld", nrefs );
newstr_strcpy( &refnum, "ref" );
newstr_strcat( &refnum, num );
}
fields_add( info, "REFNUM", refnum.data, 0 );
newstr_free( &refnum );
return fields_find( info, "REFNUM", 0 );
}
char *
xml_findend( char *buffer, char *tag )
{
newstr endtag;
char *p;
newstr_init( &endtag );
newstr_strcpy( &endtag, "</" );
if ( xml_pns ) {
newstr_strcat( &endtag, xml_pns );
newstr_addchar( &endtag, ':' );
}
newstr_strcat( &endtag, tag );
newstr_addchar( &endtag, '>' );
p = strsearch( buffer, endtag.data );
if ( p && *p ) {
if ( *p ) p++; /* skip <random_tag></end> combo */
while ( *p && *(p-1)!='>' ) p++;
}
newstr_free( &endtag );
return p;
}
/*
* xml_processtag
*
* XML_COMMENT <!-- .... -->
* XML_DESCRIPTOR <?.....>
* XML_OPEN <A>
* XML_CLOSE </A>
* XML_OPENCLOSE <A/>
*/
static char *
xml_processtag( char *p, newstr *tag, xml_attrib **attrib, int *type )
{
*attrib = NULL;
if ( *p=='<' ) p++;
if ( *p=='!' ) {
while ( *p && *p!='>' ) newstr_addchar( tag, *p++ );
*type = XML_COMMENT;
} else if ( *p=='?' ) {
*type = XML_DESCRIPTOR;
p++; /* skip '?' */
while ( *p && !strchr( " \t", *p ) && !xml_terminator(p,type) )
newstr_addchar( tag, *p++ );
if ( *p==' ' || *p=='\t' )
p = xml_processattrib( p, attrib, type );
} else if ( *p=='/' ) {
while ( *p && !strchr( " \t", *p ) && !xml_terminator(p,type) )
newstr_addchar( tag, *p++ );
*type = XML_CLOSE;
if ( *p==' ' || *p=='\t' )
p = xml_processattrib( p, attrib, type );
} else {
*type = XML_OPEN;
while ( *p && !strchr( " \t", *p ) && !xml_terminator(p,type) )
newstr_addchar( tag, *p++ );
if ( *p==' ' || *p=='\t' )
p = xml_processattrib( p, attrib, type );
}
while ( *p && *p!='>' ) p++;
if ( *p=='>' ) p++;
return p;
}
int
copacin_readf( FILE *fp, char *buf, int bufsize, int *bufpos, newstr *line, newstr *reference, int *fcharset )
{
int haveref = 0, inref=0;
char *p;
while ( !haveref && readmore( fp, buf, bufsize, bufpos, line ) ) {
/* blank line separates */
if ( line->data==NULL ) continue;
if ( inref && line->len==0 ) haveref=1;
p = &(line->data[0]);
if ( copacin_istag( p ) ) {
if ( inref ) newstr_addchar( reference, '\n' );
newstr_strcat( reference, p );
newstr_empty( line );
inref = 1;
} else if ( inref ) {
if ( p ) {
/* copac puts tag only on 1st line */
newstr_addchar( reference, ' ' );
if ( *p ) p++;
if ( *p ) p++;
if ( *p ) p++;
newstr_strcat( reference, p );
}
newstr_empty( line );
} else {
newstr_empty( line );
}
}
*fcharset = CHARSET_UNKNOWN;
return haveref;
}
/* name_nocomma()
*
* names in the format "H. F. Author"
*/
void
name_nocomma( char *start, newstr *outname )
{
char *p, *last, *end;
int uplast, lowlast, upfirst, lowfirst, splitfirst;
/* move to end */
p = start;
while ( *p && *(p+1) ) p++;
/* point to last name */
end = p;
while ( p>start && !is_ws( *p ) ) p--;
if ( !strcasecmp( p, "Jr." ) || !strcasecmp( p, "III" ) ) {
while ( p>start && is_ws( *p ) ) p--;
while ( p>start && !is_ws( *p ) ) p--;
}
last = p;
p = skip_ws( p );
/* look for upper and lower case in last name */
check_case( p, end+1, &uplast, &lowlast );
/* copy last name */
while ( p<=end )
newstr_addchar( outname, *p++ );
if ( start==last ) return; /*Only last name */
/* Given names */
newstr_addchar( outname, '|' );
/* look for upper and lower case in given name(s) */
check_case( start, last, &upfirst, &lowfirst );
splitfirst = should_split( uplast, lowlast, upfirst, lowfirst );
/* copy given name(s), splitfirst to identify cases of "HF Author" */
p = start;
while ( p!=last ) {
if ( *p!=' ' && *p!='\t' ) {
if ( !(splitfirst && ( *p=='.' || *p=='-' ) ) ) {
p = name_addmultibytechar( outname, p, last );
if ( splitfirst )
newstr_addchar(outname,'|');
} else p++;
} else {
while ( p!=last && ( *p==' ' || *p=='\t' ) )
p++;
if ( p!=last && !splitfirst )
newstr_addchar( outname, '|' );
}
}
}
static int
risin_readf( FILE *fp, char *buf, int bufsize, int *bufpos, newstr *line,
newstr *reference, int *fcharset )
{
int haveref = 0, inref = 0, readtoofar = 0;
unsigned char *up;
char *p;
*fcharset = CHARSET_UNKNOWN;
while ( !haveref && readmore( fp, buf, bufsize, bufpos, line ) ) {
if ( !line->data || line->len==0 ) continue;
p = &( line->data[0] );
/* Recognize UTF8 BOM */
up = (unsigned char * ) p;
if ( line->len > 2 &&
up[0]==0xEF && up[1]==0xBB && up[2]==0xBF ) {
*fcharset = CHARSET_UNICODE;
p += 3;
}
/* Each reference starts with 'TY - ' &&
* ends with 'ER - ' */
if ( strncmp(p,"TY - ",6)==0 ) {
if ( !inref ) {
inref = 1;
} else {
/* we've read too far.... */
readtoofar = 1;
inref = 0;
}
}
if ( risin_istag( p ) ) {
if ( !inref ) {
fprintf(stderr,"Warning. Tagged line not "
"in properly started reference.\n");
fprintf(stderr,"Ignored: '%s'\n", p );
} else if ( !strncmp(p,"ER -",5) ) {
inref = 0;
} else {
newstr_addchar( reference, '\n' );
newstr_strcat( reference, p );
}
}
/* not a tag, but we'll append to last values ...*/
else if ( inref && strncmp(p,"ER -",5)) {
newstr_addchar( reference, '\n' );
newstr_strcat( reference, p );
}
if ( !inref && reference->len ) haveref = 1;
if ( !readtoofar ) newstr_empty( line );
}
if ( inref ) haveref = 1;
return haveref;
}
static void
copacin_adddate( fields *info, char *tag, char *newtag, char *p, int level )
{
char *months[12]={ "January", "February", "March", "April",
"May", "June", "July", "August", "September",
"October", "November", "December" };
char month[10];
int found,i,part;
newstr date;
newstr_init( &date );
part = (!strncasecmp(newtag,"PART",4));
if ( !strcasecmp( tag, "%D" ) ) {
while ( *p ) newstr_addchar( &date, *p++ );
if ( date.len>0 ) {
if ( part )
fields_add(info, "PARTYEAR", date.data, level);
else
fields_add( info, "YEAR", date.data, level );
}
} else if ( !strcasecmp( tag, "%8" ) ) {
while ( *p && *p!=' ' && *p!=',' ) newstr_addchar( &date, *p++ );
if ( date.len>0 ) {
found = -1;
for ( i=0; i<12 && found==-1; ++i )
if ( !strncasecmp( date.data, months[i], 3 ) )
found = i;
if ( found!=-1 ) {
if (found>8) sprintf( month, "%d", found+1 );
else sprintf( month, "0%d", found+1 );
if ( part )
fields_add( info, "PARTMONTH", month, level );
else fields_add( info, "MONTH", month, level );
} else {
if ( part )
fields_add( info, "PARTMONTH", date.data, level );
else
fields_add( info, "MONTH", date.data, level );
}
}
newstr_empty( &date );
p = skip_ws( p );
while ( *p && *p!='\n' && *p!=',' )
newstr_addchar( &date, *p++ );
if ( date.len>0 && date.len<3 ) {
if ( part )
fields_add( info, "PARTDAY", date.data, level );
else
fields_add( info, "DAY", date.data, level );
}
}
newstr_free( &date );
}
/* copac names appear to always start with last name first, but don't
* always seem to have a comma after the name
*
* editors seem to be stuck in as authors with the tag "[Editor]" in it
*/
static int
copacin_person( fields *bibin, newstr *intag, newstr *invalue, int level, param *pm, char *outtag, fields *bibout )
{
char *usetag = outtag, editor[]="EDITOR";
newstr usename, *s;
list tokens;
int comma = 0, i, ok;
if ( list_find( &(pm->asis), invalue->data ) !=-1 ||
list_find( &(pm->corps), invalue->data ) !=-1 ) {
ok = name_add( bibout, outtag, invalue->data, level, &(pm->asis), &(pm->corps) );
if ( ok ) return BIBL_OK;
else return BIBL_ERR_MEMERR;
}
list_init( &tokens );
newstr_init( &usename );
list_tokenize( &tokens, invalue, " ", 1 );
for ( i=0; i<tokens.n; ++i ) {
s = list_get( &tokens, i );
if ( !strcmp( s->data, "[Editor]" ) ) {
usetag = editor;
newstr_strcpy( s, "" );
} else if ( s->len && s->data[s->len-1]==',' ) {
comma++;
}
}
if ( comma==0 && tokens.n ) {
s = list_get( &tokens, 0 );
newstr_addchar( s, ',' );
}
for ( i=0; i<tokens.n; ++i ) {
s = list_get( &tokens, i );
if ( s->len==0 ) continue;
if ( i ) newstr_addchar( &usename, ' ' );
newstr_newstrcat( &usename, s );
}
list_free( &tokens );
ok = name_add( bibout, usetag, usename.data, level, &(pm->asis), &(pm->corps) );
newstr_free( &usename );
if ( ok ) return BIBL_OK;
else return BIBL_ERR_MEMERR;
}
static int
wordin_pages( xml *node, fields *info )
{
int i, status, ret = BIBL_OK;
newstr sp, ep;
char *p;
newstrs_init( &sp, &ep, NULL );
p = xml_data( node );
while ( *p && *p!='-' )
newstr_addchar( &sp, *p++ );
if ( newstr_memerr( &sp ) ) {
ret = BIBL_ERR_MEMERR;
goto out;
}
if ( *p=='-' ) p++;
while ( *p )
newstr_addchar( &ep, *p++ );
if ( newstr_memerr( &ep ) ) {
ret = BIBL_ERR_MEMERR;
goto out;
}
if ( sp.len ) {
status = fields_add( info, "PAGES:START", sp.data, 1 );
if ( status!=FIELDS_OK ) {
ret = BIBL_ERR_MEMERR;
goto out;
}
}
if ( ep.len ) {
if ( sp.len > ep.len ) {
for ( i=sp.len-ep.len; i<sp.len; ++i )
sp.data[i] = ep.data[i-sp.len+ep.len];
status = fields_add( info, "PAGES:STOP", sp.data, 1 );
} else
status = fields_add( info, "PAGES:STOP", ep.data, 1 );
if ( status!=FIELDS_OK ) {
ret = BIBL_ERR_MEMERR;
goto out;
}
}
out:
newstrs_free( &sp, &ep, NULL );
return ret;
}
/* name_addmultibytechar
*
* Add character to newstring s starting at pointer p.
*
* Handles the case for multibyte Unicode chars (with high bits
* set). Do not progress past the lastp barrier.
*
* Since we can progress more than one byte in the string,
* return the properly updated pointer p.
*/
static char *
name_addmultibytechar( newstr *s, char *p, char *lastp )
{
if ( ! ((*p) & 128) ) {
newstr_addchar( s, *p );
p++;
} else {
while ( p!=lastp && ((*p) & 128) ) {
newstr_addchar( s, *p );
p++;
}
}
return p;
}
static void
output_name( FILE *outptr, char *p, int level )
{
newstr family, part, suffix;
int n=0;
newstrs_init( &family, &part, &suffix, NULL );
while ( *p && *p!='|' ) newstr_addchar( &family, *p++ );
if ( *p=='|' ) p++;
while ( *p ) {
while ( *p && *p!='|' ) newstr_addchar( &part, *p++ );
/* truncate periods from "A. B. Jones" names */
if ( part.len ) {
if ( part.len==2 && part.data[1]=='.' ) {
part.len=1;
part.data[1]='\0';
}
if ( n==0 )
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, "type", "personal", NULL );
output_tag( outptr, lvl2indent(incr_level(level,1)), "namePart", part.data, TAG_OPENCLOSE, TAG_NEWLINE, "type", "given", NULL );
n++;
}
if ( *p=='|' ) {
p++;
if ( *p=='|' ) {
p++;
while ( *p && *p!='|' ) newstr_addchar( &suffix, *p++ );
}
newstr_empty( &part );
}
}
if ( family.len ) {
if ( n==0 )
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, "type", "personal", NULL );
output_tag( outptr, lvl2indent(incr_level(level,1)), "namePart", family.data, TAG_OPENCLOSE, TAG_NEWLINE, "type", "family", NULL );
n++;
}
if ( suffix.len ) {
if ( n==0 )
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, "type", "personal", NULL );
output_tag( outptr, lvl2indent(incr_level(level,1)), "namePart", suffix.data, TAG_OPENCLOSE, TAG_NEWLINE, "type", "suffix", NULL );
}
newstrs_free( &part, &family, &suffix, NULL );
}
/* newstr_fget()
* returns 0 if we're done, 1 if we're not done
* extracts line by line (regardless of end characters)
* and feeds from buf....
*/
int
newstr_fget( FILE *fp, char *buf, int bufsize, int *pbufpos, newstr *outs )
{
int bufpos = *pbufpos, done = 0;
char *ok;
newstr_empty( outs );
while ( !done ) {
while ( buf[bufpos] && buf[bufpos]!='\r' && buf[bufpos]!='\n' )
newstr_addchar( outs, buf[bufpos++] );
if ( buf[bufpos]=='\0' ) {
ok = fgets( buf, bufsize, fp );
bufpos=*pbufpos=0;
if ( !ok && feof(fp) ) { /* end-of-file */
buf[bufpos] = 0;
if ( outs->len==0 ) return 0; /*nothing in out*/
else return 1; /*one last out */
}
} else if ( buf[bufpos]=='\r' || buf[bufpos]=='\n' ) done=1;
}
if ( ( buf[bufpos]=='\n' && buf[bufpos+1]=='\r') ||
( buf[bufpos]=='\r' && buf[bufpos+1]=='\n') ) bufpos+=2;
else if ( buf[bufpos]=='\n' || buf[bufpos]=='\r' ) bufpos+=1;
*pbufpos = bufpos;
return 1;
}
static char*
process_line( newstr *tag, newstr *data, char *p )
{
int i = 0;
while ( i<6 && *p ) {
if ( i<2 ) newstr_addchar( tag, *p );
p++;
i++;
}
while ( *p==' ' || *p=='\t' ) p++;
while ( *p && *p!='\r' && *p!='\n' )
newstr_addchar( data, *p++ );
newstr_trimendingws( data );
while ( *p=='\n' || *p=='\r' ) p++;
return p;
}
int
risin_processf( fields *risin, char *p, char *filename, long nref )
{
newstr tag, data;
newstr_init( &tag );
newstr_init( &data );
while ( *p ) {
if ( risin_istag( p ) ) {
p = process_line( &tag, &data, p );
/* no anonymous fields allowed */
/* if ( tag.len && data.len )*/
if ( tag.len )
fields_add( risin, tag.data, data.data, 0 );
} else {
p = process_line2( &tag, &data, p );
if ( data.len && risin->nfields>0 ) {
newstr *od;
od = &(risin->data[risin->nfields-1] );
newstr_addchar( od, ' ' );
newstr_strcat( od, data.data );
}
}
newstr_empty( &tag );
newstr_empty( &data );
}
newstr_free( &tag );
newstr_free( &data );
return 1;
}
/*
* readf()
*
* returns zero if cannot get reference and hit end of-file
* returns 1 if last reference in file, 2 if reference within file
*/
int
bibtexin_readf( FILE *fp, char *buf, int bufsize, int *bufpos, newstr *line, newstr *reference, int *fcharset )
{
int haveref = 0;
char *p;
*fcharset = CHARSET_UNKNOWN;
while ( haveref!=2 && readmore( fp, buf, bufsize, bufpos, line ) ) {
if ( line->len == 0 ) continue; /* blank line */
p = &(line->data[0]);
/* Recognize UTF8 BOM */
if ( line->len > 2 &&
(unsigned char)(p[0])==0xEF &&
(unsigned char)(p[1])==0xBB &&
(unsigned char)(p[2])==0xBF ) {
*fcharset = CHARSET_UNICODE;
p += 3;
}
p = skip_ws( p );
if ( *p == '%' ) { /* commented out line */
newstr_empty( line );
continue;
}
if ( *p == '@' ) haveref++;
if ( haveref && haveref<2 ) {
newstr_strcat( reference, p );
newstr_addchar( reference, '\n' );
newstr_empty( line );
} else if ( !haveref ) newstr_empty( line );
}
return haveref;
}
static int
bibtex_cleandata( newstr *tag, newstr *s, fields *info, param *p )
{
int i, status;
list tokens;
newstr *tok;
if ( !s->len ) return BIBL_OK;
/* protect url from undergoing any parsing */
if ( is_url_tag( tag ) ) return BIBL_OK;
list_init( &tokens );
status = bibtex_split( &tokens, s );
if ( status!=BIBL_OK ) goto out;
for ( i=0; i<tokens.n; ++i ) {
tok = list_get( &tokens, i );
if ( bibtex_protected( tok ) ) {
if (!strncasecmp(tok->data,"\\href{", 6)) {
bibtex_addtitleurl( info, tok );
}
}
if ( p->latexin && !is_name_tag( tag ) && !is_url_tag( tag ) )
bibtex_cleantoken( tok );
}
newstr_empty( s );
for ( i=0; i<tokens.n; ++i ) {
tok = list_get( &tokens, i );
if ( i>0 ) newstr_addchar( s, ' ' );
newstr_newstrcat( s, tok );
}
out:
list_free( &tokens );
return status;
}
int
endin_processf( fields *endin, char *p, char *filename, long nref )
{
newstr tag, data;
int n;
newstr_init( &tag );
newstr_init( &data );
while ( *p ) {
if ( endin_istag( p ) ) {
p = process_endline( &tag, &data, p );
/* no empty fields allowed */
if ( data.len ) {
fields_add( endin, tag.data, data.data, 0 );
}
} else {
p = process_endline2( &tag, &data, p );
/* endnote puts %K only on 1st line of keywords */
n = endin->nfields;
if ( n>0 && data.len ) {
if ( !strncmp( endin->tag[n-1].data, "%K", 2 ) ) {
fields_add( endin, "%K", data.data, 0 );
} else {
newstr_addchar( &(endin->data[n-1]), ' ' );
newstr_strcat( &(endin->data[n-1]), data.data );
}
}
}
newstr_empty( &tag );
newstr_empty( &data );
}
newstr_free( &tag );
newstr_free( &data );
return 1;
}
void
name_add( fields *info, char *tag, char *q, int level, list *asis, list *corps )
{
newstr inname;
char *p, *start, *end;
if ( !q ) return;
newstr_init( &inname );
while ( *q ) {
start = q = skip_ws( q );
/* strip tailing whitespace and commas */
while ( *q && *q!='|' ) q++;
end = q;
while ( is_ws( *end ) || *end==',' || *end=='|' || *end=='\0' )
end--;
for ( p=start; p<=end; p++ )
newstr_addchar( &inname, *p );
/* keep "names" like " , " from coredumping program */
if ( inname.len ) {
name_process( info, tag, level, &inname, asis, corps );
newstr_empty( &inname );
}
if ( *q=='|' ) q++;
}
newstr_free( &inname );
}
static void
addpage( fields *info, char *p, int level )
{
newstr page;
newstr_init( &page );
p = skip_ws( p );
while ( *p && !is_ws(*p) && *p!='-' && *p!='\r' && *p!='\n' )
newstr_addchar( &page, *p++ );
if ( page.len>0 ) fields_add( info, "PAGESTART", page.data, level );
newstr_empty( &page );
while ( *p && (is_ws(*p) || *p=='-' ) ) p++;
while ( *p && !is_ws(*p) && *p!='-' && *p!='\r' && *p!='\n' )
newstr_addchar( &page, *p++ );
if ( page.len>0 ) fields_add( info, "PAGEEND", page.data, level );
newstr_free( &page );
}
static void
output_title( FILE *fp, fields *info, unsigned long refnum, char *bibtag, int level, int format_opts )
{
newstr title;
int n1 = -1, n2 = -1;
/* Option is for short titles of journals */
if ( ( format_opts & BIBOUT_SHORTTITLE ) && level==1 ) {
n1 = fields_find( info, "SHORTTITLE", level );
n2 = fields_find( info, "SHORTSUBTITLE", level );
}
if ( n1==-1 ) {
n1 = fields_find( info, "TITLE", level );
n2 = fields_find( info, "SUBTITLE", level );
}
if ( n1!=-1 ) {
newstr_init( &title );
newstr_newstrcpy( &title, &(info->data[n1]) );
fields_setused( info, n1 );
if ( n2!=-1 ) {
if ( info->data[n1].data[info->data[n1].len]!='?' )
newstr_strcat( &title, ": " );
else newstr_addchar( &title, ' ' );
newstr_strcat( &title, info->data[n2].data );
fields_setused( info, n2 );
}
output_element( fp, bibtag, title.data, format_opts );
newstr_free( &title );
}
}