static int
bibtex_addtitleurl( fields *info, newstr *in )
{
int fstatus, status = BIBL_OK;
newstr s;
char *p;
newstr_init( &s );
/* ...skip past "\href{" and copy to "}" */
p = newstr_cpytodelim( &s, in->data + 6, "}", 1 );
if ( newstr_memerr( &s ) ) { status = BIBL_ERR_MEMERR; goto out; }
/* ...add to URL */
fstatus = fields_add( info, "URL", s.data, 0 );
if ( fstatus!=FIELDS_OK ) { status = BIBL_ERR_MEMERR; goto out; }
/* ...return deleted fragment to newstr in */
p = newstr_cpytodelim( &s, p, "", 0 );
if ( newstr_memerr( &s ) ) { status = BIBL_ERR_MEMERR; goto out; }
newstr_swapstrings( &s, in );
out:
newstr_free( &s );
return status;
}
/* <MedlineDate>2003 Jan-Feb</MedlineDate> */
static int
medin_medlinedate( fields *info, char *p, int level )
{
int fstatus;
newstr tmp;
newstr_init( &tmp );
p = newstr_cpytodelim( &tmp, skip_ws( p ), " \t\n\r", 0 );
if ( newstr_memerr( &tmp ) ) return BIBL_ERR_MEMERR;
if ( tmp.len > 0 ) {
fstatus = fields_add( info, "PARTYEAR", tmp.data, level );
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
p = newstr_cpytodelim( &tmp, skip_ws( p ), " \t\n\r", 0 );
if ( newstr_memerr( &tmp ) ) return BIBL_ERR_MEMERR;
if ( tmp.len > 0 ) {
newstr_findreplace( &tmp, "-", "/" );
fstatus = fields_add( info, "PARTMONTH", tmp.data, level );
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
p = newstr_cpytodelim( &tmp, skip_ws( p ), " \t\n\r", 0 );
if ( newstr_memerr( &tmp ) ) return BIBL_ERR_MEMERR;
if ( tmp.len > 0 ) {
fstatus = fields_add( info, "PARTDAY", tmp.data, level );
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
newstr_free( &tmp );
return BIBL_OK;
}
int
pages_add( fields *bibout, char *outtag, newstr *invalue, int level )
{
int fstatus, status = 1;
newstr start, stop;
newstr_init( &start );
newstr_init( &stop );
extract_range( invalue, &start, &stop );
if ( newstr_memerr( &start ) || newstr_memerr( &stop ) ) {
status = 0;
goto out;
}
if ( start.len>0 ) {
fstatus = fields_add( bibout, "PAGES:START", start.data, level );
if ( fstatus!=FIELDS_OK ) {
status = 0;
goto out;
}
}
if ( stop.len>0 ) {
fstatus = fields_add( bibout, "PAGES:STOP", stop.data, level );
if ( fstatus!=FIELDS_OK ) status = 0;
}
out:
newstr_free( &start );
newstr_free( &stop );
return status;
}
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;
}
static int
medin_authorlist( xml *node, fields *info )
{
int fstatus, status;
newstr name;
char *tag;
newstr_init( &name );
node = node->down;
while ( node ) {
if ( xml_tagexact( node, "Author" ) && node->down ) {
status = medin_author( node->down, &name );
tag = "AUTHOR";
if ( !name.len ) {
status = medin_corpauthor( node->down, &name );
tag = "AUTHOR:CORP";
}
if ( newstr_memerr( &name ) || status!=BIBL_OK ) return BIBL_ERR_MEMERR;
if ( name.len ) {
fstatus = fields_add(info,tag,name.data,0);
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
newstr_empty( &name );
}
node = node->next;
}
newstr_free( &name );
return BIBL_OK;
}
static char *
bibtex_tag( char *p, newstr *tag )
{
p = newstr_cpytodelim( tag, skip_ws( p ), "= \t\r\n", 0 );
if ( newstr_memerr( tag ) ) return NULL;
return skip_ws( p );
}
/* 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;
}
/* wordin_person_last()
*
* From an xml list, extract the value from the first entry
* of <b:Last>xxxx</b:Last> and copy into name
*
* Additional <b:Last>yyyyy</b:Last> will be ignored.
*
* Returns BIBL_ERR_MEMERR on memory error, BIBL_OK otherwise.
*/
static int
wordin_person_last( xml *node, newstr *name )
{
while ( node && !xml_tagexact( node, "b:Last" ) )
node = node->next;
if ( node && node->value->len ) {
newstr_strcpy( name, node->value->data );
if ( newstr_memerr( name ) ) return BIBL_ERR_MEMERR;
}
return BIBL_OK;
}
/* wordin_person_first()
*
* From an xml list, extract the value of any
* <b:First>xxxx</b:First> and append "|xxxx" to name.
*
* Returns BIBL_ERR_MEMERR on memory error, BIBL_OK otherwise
*/
static int
wordin_person_first( xml *node, newstr *name )
{
for ( ; node; node=node->next ) {
if ( !xml_tagexact( node, "b:First" ) ) continue;
if ( node->value->len ) {
if ( name->len ) newstr_addchar( name, '|' );
newstr_strcat( name, node->value->data );
if ( newstr_memerr( name ) ) return BIBL_ERR_MEMERR;
}
}
return BIBL_OK;
}
static newstr *
list_set_cleanup( list *a, int n )
{
if ( newstr_memerr( &(a->str[n]) ) ) return NULL;
if ( a->sorted ) {
if ( n>0 && list_comp_step( a, n-1, n )>0 )
a->sorted = 0;
}
if ( a->sorted ) {
if ( n<a->n-1 && list_comp_step( a, n, n+1 )>0 )
a->sorted = 0;
}
return &(a->str[n]);
}
/* <Pagination>
* <MedlinePgn>12111-6</MedlinePgn>
* </Pagination>
*/
static int
medin_pagination( xml *node, fields *info )
{
int i, fstatus, status;
newstr sp, ep;
char *p, *pp;
if ( xml_tagexact( node, "MedlinePgn" ) && node->value ) {
newstrs_init( &sp, &ep, NULL );
p = newstr_cpytodelim( &sp, xml_data( node ), "-", 1 );
if ( newstr_memerr( &sp ) ) return BIBL_ERR_MEMERR;
if ( sp.len ) {
fstatus = fields_add( info, "PAGESTART", sp.data, 1 );
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
p = newstr_cpytodelim( &ep, p, "", 0 );
if ( newstr_memerr( &ep ) ) return BIBL_ERR_MEMERR;
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];
pp = sp.data;
} else pp = ep.data;
fstatus = fields_add( info, "PAGEEND", pp, 1 );
if ( fstatus!=FIELDS_OK ) return BIBL_ERR_MEMERR;
}
newstrs_free( &sp, &ep, NULL );
}
if ( node->down ) {
status = medin_pagination( node->down, info );
if ( status!=BIBL_OK ) return status;
}
if ( node->next ) {
status = medin_pagination( node->next, info );
if ( status!=BIBL_OK ) return status;
}
return BIBL_OK;
}
/*
* 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;
}
newstr *
list_addvp( list *a, unsigned char mode, void *vp )
{
newstr *s = NULL;
int status;
status = list_ensure_space( a );
if ( status==LIST_OK ) {
s = &( a->str[a->n] );
if ( mode==LIST_CHR )
newstr_strcpy( s, (const char*) vp );
else if ( mode==LIST_STR )
newstr_newstrcpy( s, (newstr*) vp );
else
return NULL;
if ( newstr_memerr( s ) ) return NULL;
a->n++;
if ( a->sorted && a->n > 1 ) {
if ( list_comp_step( a, a->n-2, a->n-1 ) > 0 )
a->sorted = 0;
}
}
return s;
}
/*
* 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;
}
static char *
bibtex_data( char *p, fields *bibin, list *tokens )
{
unsigned int nbracket = 0, nquotes = 0;
char *startp = p;
newstr tok, *t;
newstr_init( &tok );
while ( p && *p ) {
if ( !nquotes && !nbracket ) {
if ( *p==',' || *p=='=' || *p=='}' || *p==')' )
goto out;
}
if ( *p=='\"' && nbracket==0 && ( p==startp || *(p-1)!='\\' ) ) {
nquotes = !nquotes;
newstr_addchar( &tok, *p );
if ( !nquotes ) {
if ( newstr_memerr( &tok ) ) { p=NULL; goto out; }
t = list_add( tokens, &tok );
if ( !t ) { p=NULL; goto out0; }
newstr_empty( &tok );
}
} else if ( *p=='#' && !nquotes && !nbracket ) {
if ( tok.len ) {
if ( newstr_memerr( &tok ) ) { p=NULL; goto out; }
t = list_add( tokens, &tok );
if ( !t ) { p=NULL; goto out0; }
}
newstr_strcpy( &tok, "#" );
t = list_add( tokens, &tok );
if ( !t ) { p=NULL; goto out0; }
newstr_empty( &tok );
} else if ( *p=='{' && !nquotes && ( p==startp || *(p-1)!='\\' ) ) {
nbracket++;
newstr_addchar( &tok, *p );
} else if ( *p=='}' && !nquotes && ( p==startp || *(p-1)!='\\' ) ) {
nbracket--;
newstr_addchar( &tok, *p );
if ( nbracket==0 ) {
if ( newstr_memerr( &tok ) ) { p=NULL; goto out; }
t = list_add( tokens, &tok );
if ( !t ) { p=NULL; goto out; }
newstr_empty( &tok );
}
} else if ( !is_ws( *p ) || nquotes || nbracket ) {
if ( !is_ws( *p ) ) newstr_addchar( &tok, *p );
else {
if ( tok.len!=0 && *p!='\n' && *p!='\r' )
newstr_addchar( &tok, *p );
else if ( tok.len!=0 && (*p=='\n' || *p=='\r')) {
newstr_addchar( &tok, ' ' );
while ( is_ws( *(p+1) ) ) p++;
}
}
} else if ( is_ws( *p ) ) {
if ( tok.len ) {
if ( newstr_memerr( &tok ) ) { p=NULL; goto out; }
t = list_add( tokens, &tok );
if ( !t ) { p=NULL; goto out; }
newstr_empty( &tok );
}
}
p++;
}
out:
if ( nbracket!=0 ) {
fprintf( stderr, "%s: Mismatch in number of brackets in reference.\n", progname );
}
if ( nquotes!=0 ) {
fprintf( stderr, "%s: Mismatch in number of quotes in reference.\n", progname );
}
if ( tok.len ) {
if ( newstr_memerr( &tok ) ) { p = NULL; goto out; }
t = list_add( tokens, &tok );
if ( !t ) p = NULL;
}
out0:
newstr_free( &tok );
return p;
}
