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;
}
static void
construct_url( char *prefix, newstr *id, newstr *id_url )
{
if ( !strncasecmp( id->data, "http:", 5 ) )
newstr_newstrcpy( id_url, id );
else {
newstr_strcpy( id_url, prefix );
if ( id->data[0]!='/' ) newstr_addchar( id_url, '/' );
newstr_newstrcat( id_url, id );
}
}
static int
string_concatenate( list *tokens, fields *bibin )
{
newstr *s, *t;
int i, status;
i = 0;
while ( i < tokens->n ) {
s = list_get( tokens, i );
if ( !strcmp( s->data, "#" ) ) {
if ( i==0 || i==tokens->n-1 ) {
fprintf( stderr, "%s: Warning: Stray string concatenation "
"('#' character) in reference\n", progname );
status = list_remove( tokens, i );
if ( status!=LIST_OK ) return BIBL_ERR_MEMERR;
continue;
}
s = list_get( tokens, i-1 );
if ( s->data[0]!='\"' && s->data[s->len-1]!='\"' )
fprintf( stderr, "%s: Warning: String concentation should "
"be used in context of quotations marks.\n", progname );
t = list_get( tokens, i+1 );
if ( t->data[0]!='\"' && t->data[s->len-1]!='\"' )
fprintf( stderr, "%s: Warning: String concentation should "
"be used in context of quotations marks.\n", progname );
if ( ( s->data[s->len-1]=='\"' && t->data[0]=='\"') || (s->data[s->len-1]=='}' && t->data[0]=='{') ) {
newstr_trimend( s, 1 );
newstr_trimbegin( t, 1 );
newstr_newstrcat( s, t );
} else {
newstr_newstrcat( s, t );
}
status = list_remove( tokens, i );
if ( status!=LIST_OK ) return BIBL_ERR_MEMERR;
status = list_remove( tokens, i );
if ( status!=LIST_OK ) return BIBL_ERR_MEMERR;
} else i++;
}
return BIBL_OK;
}
/* 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;
}
/* 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_addname( fields *info, char *tag, newstr *name, int level, list *asis,
list *corps )
{
char *usetag = tag, editor[]="EDITOR";
newstr usename, *s;
list tokens;
int comma = 0, i, ok;
if ( list_find( asis, name->data ) !=-1 ||
list_find( corps, name->data ) !=-1 ) {
ok = name_add( info, tag, name->data, level, asis, corps );
if ( ok ) return BIBL_OK;
else return BIBL_ERR_MEMERR;
}
list_init( &tokens );
newstr_init( &usename );
list_tokenize( &tokens, name, " ", 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 ) {
if ( i ) newstr_addchar( &usename, ' ' );
newstr_newstrcat( &usename, list_get( &tokens, i ) );
}
list_free( &tokens );
ok = name_add( info, usetag, usename.data, level, asis, corps );
if ( ok ) return BIBL_OK;
else return BIBL_ERR_MEMERR;
}
static int
test_newstrcat( newstr *s )
{
int numshort = 5, numstrings = 1000, i;
int failed = 0;
newstr t;
newstr_init( &t );
/* ...adding empty strings to an empty string shouldn't change length */
newstr_empty( s );
for ( i=0; i<numstrings; ++i )
newstr_newstrcat( s, &t );
if ( string_mismatch( s, 0, "" ) ) failed++;
/* ...adding empty strings to a defined string shouldn't change string */
newstr_strcpy( s, "1" );
for ( i=0; i<numstrings; ++i )
newstr_newstrcat( s, &t );
if ( string_mismatch( s, 1, "1" ) ) failed++;
/* ...build "1111" with newstr_strcat */
newstr_empty( s );
newstr_strcpy( &t, "1" );
for ( i=0; i<numshort; ++i )
newstr_newstrcat( s, &t );
if ( string_mismatch( s, numshort, "11111" ) ) failed++;
/* ...build "xoxoxoxoxo" with newstr_strcat */
newstr_empty( s );
newstr_strcpy( &t, "xo" );
for ( i=0; i<numshort; ++i )
newstr_newstrcat( s, &t );
if ( string_mismatch( s, numshort*2, "xoxoxoxoxo" ) ) failed++;
newstr_empty( s );
newstr_strcpy( &t, "1" );
for ( i=0; i<numstrings; ++i )
newstr_newstrcat( s, &t );
if ( inconsistent_len( s, numstrings ) ) failed++;
newstr_empty( s );
newstr_strcpy( &t, "XXOO" );
for ( i=0; i<numstrings; ++i )
newstr_newstrcat( s, &t );
if ( inconsistent_len( s, numstrings*4 ) ) failed++;
newstr_free( &t );
return failed;
}
static void
output_fileattach( FILE *fp, fields *info, int format_opts )
{
newstr data;
int i;
newstr_init( &data );
for ( i=0; i<info->nfields; ++i ) {
if ( strcasecmp( info->tag[i].data, "FILEATTACH" ) ) continue;
newstr_strcpy( &data, ":" );
newstr_newstrcat( &data, &(info->data[i]) );
if ( strsearch( info->data[i].data, ".pdf" ) )
newstr_strcat( &data, ":PDF" );
else if ( strsearch( info->data[i].data, ".html" ) )
newstr_strcat( &data, ":HTML" );
else newstr_strcat( &data, ":TYPE" );
output_element( fp, "file", data.data, format_opts );
fields_setused( info, i );
newstr_empty( &data );
}
newstr_free( &data );
}
/* return NULL on memory error */
static char *
process_bibtexline( char *p, newstr *tag, newstr *data, uchar stripquotes, fields *bibin )
{
int i, status;
list tokens;
newstr *s;
newstr_empty( data );
p = bibtex_tag( p, tag );
if ( p==NULL || tag->len==0 ) return p;
list_init( &tokens );
if ( *p=='=' ) {
p = bibtex_data( p+1, bibin, &tokens );
if ( p==NULL ) goto out;
}
replace_strings( &tokens, bibin );
status = string_concatenate( &tokens, bibin );
if ( status!=BIBL_OK ) {
p = NULL;
goto out;
}
for ( i=0; i<tokens.n; i++ ) {
s = list_get( &tokens, i );
if ( ( stripquotes && s->data[0]=='\"' && s->data[s->len-1]=='\"' ) ||
( s->data[0]=='{' && s->data[s->len-1]=='}' ) ) {
newstr_trimbegin( s, 1 );
newstr_trimend( s, 1 );
}
newstr_newstrcat( data, list_get( &tokens, i ) );
}
out:
list_free( &tokens );
return p;
}
