/* Split keywords="" with semicolons.
* Commas are also frequently used, but will break
* entries like:
* keywords="Microscopy, Confocal"
* Returns BIBL_OK or BIBL_ERR_MEMERR
*/
static int
process_keywords( fields *info, newstr *d, int level )
{
int fstatus, status = BIBL_OK;
newstr keyword;
char *p;
if ( !d || d->len==0 ) return BIBL_OK;
p = d->data;
newstr_init( &keyword );
while ( *p ) {
p = newstr_cpytodelim( &keyword, skip_ws( p ), ";", 1 );
newstr_trimendingws( &keyword );
if ( newstr_memerr( &keyword ) ) {
status = BIBL_ERR_MEMERR;
goto out;
}
if ( keyword.len ) {
fstatus = fields_add( info, "KEYWORD", keyword.data, level );
if ( fstatus!=FIELDS_OK ) {
status = BIBL_ERR_MEMERR;
goto out;
}
}
}
out:
newstr_free( &keyword );
return status;
}
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;
}
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 );
}
static int
test_trimws( newstr *s )
{
char str1[] = " ksjadfk lajskfjds askdjflkj ";
char str2[] = " ";
int failed = 0;
newstr_empty( s );
newstr_trimstartingws( s );
if ( string_mismatch( s, 0, "" ) ) failed++;
newstr_trimendingws( s );
if ( string_mismatch( s, 0, "" ) ) failed++;
newstr_strcpy( s, str2 );
newstr_trimstartingws( s );
if ( string_mismatch( s, 0, "" ) ) failed++;
newstr_strcpy( s, str2 );
newstr_trimendingws( s );
if ( string_mismatch( s, 0, "" ) ) failed++;
newstr_strcpy( s, str1 );
newstr_trimstartingws( s );
if ( string_mismatch( s, strlen("ksjadfk lajskfjds askdjflkj "), "ksjadfk lajskfjds askdjflkj " ) ) failed++;
newstr_trimendingws( s );
if ( string_mismatch( s, strlen("ksjadfk lajskfjds askdjflkj"), "ksjadfk lajskfjds askdjflkj" ) ) failed++;
newstr_strcpy( s, str1 );
newstr_trimendingws( s );
if ( string_mismatch( s, strlen(" ksjadfk lajskfjds askdjflkj"), " ksjadfk lajskfjds askdjflkj" ) ) failed++;
newstr_trimstartingws( s );
if ( string_mismatch( s, strlen("ksjadfk lajskfjds askdjflkj"), "ksjadfk lajskfjds askdjflkj" ) ) failed++;
newstr_empty( s );
newstr_stripws( s );
if ( string_mismatch( s, 0, "" ) ) failed++;
newstr_strcpy( s, "0123456789" );
newstr_stripws( s );
if ( string_mismatch( s, 10, "0123456789" ) ) failed++;
newstr_strcpy( s, str1 );
newstr_stripws( s );
if ( string_mismatch( s, strlen("ksjadfklajskfjdsaskdjflkj"), "ksjadfklajskfjdsaskdjflkj" ) ) failed++;
return failed;
}
static char *
process_endline2( newstr *tag, newstr *data, char *p )
{
while ( *p==' ' || *p=='\t' ) p++;
while ( *p && *p!='\r' && *p!='\n' )
newstr_addchar( data, *p++ );
newstr_trimendingws( data );
while ( *p=='\r' || *p=='\n' ) p++;
return p;
}
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;
}
static char*
copacin_addtag2( char *p, newstr *tag, newstr *data )
{
int i;
i =0;
while ( i<3 && *p ) {
newstr_addchar( tag, *p++ );
i++;
}
while ( *p==' ' || *p=='\t' ) p++;
while ( *p && *p!='\r' && *p!='\n' ) {
newstr_addchar( data, *p );
p++;
}
newstr_trimendingws( data );
while ( *p=='\n' || *p=='\r' ) p++;
return p;
}
static char *
process_isiline( newstr *tag, newstr *data, char *p )
{
int i;
/* collect tag and skip past it */
i = 0;
while ( i<2 && *p && *p!='\r' && *p!='\n') {
newstr_addchar( tag, *p++ );
i++;
}
while ( *p==' ' || *p=='\t' ) p++;
while ( *p && *p!='\r' && *p!='\n' )
newstr_addchar( data, *p++ );
newstr_trimendingws( data );
while ( *p=='\r' || *p=='\n' ) p++;
return p;
}
static int
bibtex_split( list *tokens, newstr *s )
{
int i, n = s->len, nbrackets = 0, status = BIBL_OK;
newstr tok, *t;
newstr_init( &tok );
for ( i=0; i<n; ++i ) {
if ( s->data[i]=='{' && ( i==0 || s->data[i-1]!='\\' ) ) {
nbrackets++;
newstr_addchar( &tok, '{' );
} else if ( s->data[i]=='}' && ( i==0 || s->data[i-1]!='\\' ) ) {
nbrackets--;
newstr_addchar( &tok, '}' );
} else if ( !is_ws( s->data[i] ) || nbrackets ) {
newstr_addchar( &tok, s->data[i] );
} else if ( is_ws( s->data[i] ) ) {
if ( tok.len ) {
t = list_add( tokens, &tok );
if ( !t ) {
status = BIBL_ERR_MEMERR;
goto out;
}
}
newstr_empty( &tok );
}
}
if ( tok.len ) {
t = list_add( tokens, &tok );
if ( !t ) {
status = BIBL_ERR_MEMERR;
goto out;
}
}
for ( i=0; i<tokens->n; ++i ) {
newstr_trimstartingws( list_get( tokens, i ) );
newstr_trimendingws( list_get( tokens, i ) );
}
out:
newstr_free( &tok );
return status;
}
/*
* sentelink = {file://localhost/full/path/to/file.pdf,Sente,PDF}
*/
static int
process_sente( fields *info, newstr *d, int level )
{
int fstatus, status = BIBL_OK;
newstr link;
newstr_init( &link );
newstr_cpytodelim( &link, skip_ws( d->data ), ",", 0 );
newstr_trimendingws( &link );
if ( newstr_memerr( &link ) ) status = BIBL_ERR_MEMERR;
if ( status==BIBL_OK && link.len ) {
fstatus = fields_add( info, "FILEATTACH", link.data, level );
if ( fstatus!=FIELDS_OK ) status = BIBL_ERR_MEMERR;
}
newstr_free( &link );
return status;
}
/*
* file={Description:/full/path/to/file.pdf:PDF}
*/
static int
process_file( fields *info, newstr *d, int level )
{
int fstatus, status = BIBL_OK;
char *p = d->data;
int i, n, n1, n2;
newstr link;
n = count_colons( p );
if ( n > 1 ) {
/* A DOS file can contain a colon ":C:/....pdf:PDF" */
/* Extract after 1st and up to last colons */
n1 = first_colon( p ) + 1;
n2 = last_colon( p );
newstr_init( &link );
for ( i=n1; i<n2; ++i ) {
newstr_addchar( &link, p[i] );
}
newstr_trimstartingws( &link );
newstr_trimendingws( &link );
if ( newstr_memerr( &link ) ) {
status = BIBL_ERR_MEMERR;
goto out;
}
if ( link.len ) {
fstatus = fields_add( info, "FILEATTACH", link.data, level );
if ( fstatus!=FIELDS_OK ) status = BIBL_ERR_MEMERR;
}
out:
newstr_free( &link );
} else {
/* This field isn't formatted properly, so just copy directly */
fstatus = fields_add( info, "FILEATTACH", p, level );
if ( fstatus!=FIELDS_OK ) status = BIBL_ERR_MEMERR;
}
return status;
}
