static int
extract_name_and_info( newstr *outtag, newstr *intag )
{
int code = NAME;
newstr_newstrcpy( outtag, intag );
if ( newstr_findreplace( outtag, ":ASIS", "" ) ) code = NAME_ASIS;
if ( newstr_findreplace( outtag, ":CORP", "" ) ) code = NAME_CORP;
return code;
}
/* process_string()
*
* Handle lines like:
*
* '@STRING{TL = {Tetrahedron Lett.}}'
*
* p should point to just after '@STRING'
*
* In BibTeX, if a string is defined several times, the last one is kept.
*
*/
static int
process_string( char *p )
{
int n, status = BIBL_OK;
newstr s1, s2, *t;
newstrs_init( &s1, &s2, NULL );
while ( *p && *p!='{' && *p!='(' ) p++;
if ( *p=='{' || *p=='(' ) p++;
p = process_bibtexline( skip_ws( p ), &s1, &s2, 0, NULL );
if ( p==NULL ) { status = BIBL_ERR_MEMERR; goto out; }
if ( s2.data ) {
newstr_findreplace( &s2, "\\ ", " " );
}
if ( s1.data ) {
n = list_find( &find, s1.data );
if ( n==-1 ) {
t = list_add( &find, &s1 );
if ( t==NULL ) { status = BIBL_ERR_MEMERR; goto out; }
if ( s2.data ) t = list_add( &replace, &s2 );
else t = list_addc( &replace, "" );
if ( t==NULL ) { status = BIBL_ERR_MEMERR; goto out; }
} else {
if ( s2.data ) t = list_set( &replace, n, &s2 );
else t = list_setc( &replace, n, "" );
if ( t==NULL ) { status = BIBL_ERR_MEMERR; goto out; }
}
}
out:
newstrs_free( &s1, &s2, NULL );
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;
}
/*
* process_names( info, newtag, field, level);
*
* split names in author list separated by and's (use '|' character)
* and add names
*/
static void
process_names( fields *info, char *tag, newstr *data, int level, list *asis,
list *corps )
{
newstr_findreplace( data, " and ", "|" );
name_add( info, tag, data->data, level, asis, corps );
}
static void
process_pages( fields *info, newstr *s, int level )
{
char *p, *q;
newstr sp, ep;
newstr_init( &sp );
newstr_init( &ep );
newstr_findreplace( s, " ", "" );
p = q = s->data;
while ( isdigit( *q ) ) q++;
newstr_segcpy( &sp, p, q );
if ( sp.len>0 )
fields_add( info, "PAGESTART", sp.data, level );
p = q;
while ( *p && !isdigit(*p) ) p++;
q = p;
while ( isdigit( *q ) ) q++;
newstr_segcpy( &ep, p, q );
if ( ep.len>0 )
fields_add( info, "PAGEEND", ep.data, level );
newstr_free(&sp);
newstr_free(&ep);
}
int
test_findreplace( newstr *s )
{
int failed = 0;
int numstrings = 1000, i;
char segment[]="0123456789";
for ( i=0; i<numstrings; ++i ) {
newstr_strcpy( s, segment );
newstr_findreplace( s, "234", "" );
}
failed += test_consistency( s, 7, __FUNCTION__ );
for ( i=0; i<numstrings; ++i ) {
newstr_strcpy( s, segment );
newstr_findreplace( s, "234", "223344" );
}
failed += test_consistency( s, 13, __FUNCTION__ );
return failed;
}
static void
bibtexin_cleanref( fields *bibin, param *p )
{
newstr *t, *d;
int i;
for ( i=0; i<bibin->nfields; ++i ) {
t = &( bibin->tag[i] );
d = &( bibin->data[i] );
bibtex_cleandata( d, bibin, p );
if ( !strsearch( t->data, "AUTHORS" ) ) {
newstr_findreplace( d, "\n", " " );
newstr_findreplace( d, "\r", " " );
}
else if ( !strsearch( t->data, "ABSTRACT" ) ||
!strsearch( t->data, "SUMMARY" ) ||
!strsearch( t->data, "NOTE" ) ) {
newstr_findreplace( d, "\n", "" );
newstr_findreplace( d, "\r", "" );
}
}
}
static int
test_findreplace( newstr *s )
{
char segment[]="0123456789";
int numstrings = 1000, i;
int failed = 0;
for ( i=0; i<numstrings; ++i ) {
newstr_strcpy( s, segment );
newstr_findreplace( s, "234", "" );
}
if ( string_mismatch( s, 7, "0156789" ) ) failed++;
for ( i=0; i<numstrings; ++i ) {
newstr_strcpy( s, segment );
newstr_findreplace( s, "234", "223344" );
}
if ( string_mismatch( s, 13, "0122334456789" ) ) failed++;
return failed;
}
static int
bibtex_usestrings( newstr *s )
{
int i;
for ( i=0; i<find.n; ++i ) {
if ( !strcasecmp( s->data, (find.str[i]).data ) ) {
newstr_findreplace( s, (find.str[i]).data,
(replace.str[i]).data );
return 1;
}
}
return 0;
}
/* 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 void
copacin_addname( fields *info, char *tag, newstr *name, int level, list *asis,
list *corps )
{
char *usetag = tag, editor[]="EDITOR", *p;
int comma = 0;
if ( strstr( name->data,"[Editor]" ) ) {
newstr_findreplace( name, "[Editor]", "" );
usetag = editor;
}
p = skip_ws( name->data );
while ( *p && !is_ws( *p ) ) {
if ( *p==',' ) comma++;
p++;
}
if ( !comma && is_ws( *p ) ) *p = ',';
name_add( info, usetag, name->data, level, asis, corps );
}
static void
bibtex_addstring( char *p )
{
newstr s1, s2;
newstr_init( &s1 );
newstr_init( &s2 );
p = skip_ws( p );
if ( *p=='(' || *p=='{' ) p++;
p = process_bibtexline( p, &s1, &s2 );
newstr_findreplace( &s2, "\\ ", " " );
bibtex_cleantoken( &s2, NULL );
if ( s1.data ) {
list_add( &find, s1.data );
if ( s2.data ) list_add( &replace, s2.data );
else list_add( &replace, "" );
}
newstr_free( &s1 );
newstr_free( &s2 );
}
static void
name_process( fields *info, char *tag, int level, newstr *inname, list *asis,
list *corps )
{
newstr newtag, outname;
newstr_init( &newtag );
newstr_init( &outname );
if ( name_nomangle( tag, inname->data, &newtag, asis, corps ) ) {
fields_add( info, newtag.data, inname->data, level );
} else {
newstr_findreplace( inname, ".", ". " );
if ( strchr( inname->data, ',' ) )
name_comma( inname->data, &outname );
else
name_nocomma( inname->data, &outname );
if ( outname.len!=0 ) {
fields_add( info, tag, outname.data, level );
}
}
newstr_free( &newtag );
newstr_free( &outname );
}
/*
* return 1 on a nomangle with a newtag value
* return 0 on a name to mangle
*/
static int
name_nomangle( char *tag, char *data, newstr *newtag, list *asis, list *corps )
{
int corp_tag_flag, corp_list_flag;
int asis_tag_flag, asis_list_flag;
name_determine_flags( &corp_tag_flag, &corp_list_flag,
&asis_tag_flag, &asis_list_flag, tag, data, asis, corps );
if ( corp_tag_flag || corp_list_flag || asis_tag_flag || asis_list_flag ) {
newstr_strcpy( newtag, tag );
if ( corp_tag_flag ) { /* do nothing else */
} else if ( corp_list_flag && !asis_tag_flag ) {
newstr_strcat( newtag, ":CORP" );
} else if ( corp_list_flag && asis_tag_flag ) {
newstr_findreplace( newtag, ":ASIS", ":CORP" );
} else if ( asis_tag_flag ) { /* do nothing else */
} else if ( asis_list_flag ) {
newstr_strcat( newtag, ":ASIS" );
}
return 1;
}
else return 0;
}
static int
process_pages( fields *info, newstr *s, int level )
{
int fstatus, status = BIBL_OK;
newstr page;
char *p;
newstr_findreplace( s, " ", "" );
if ( s->len==0 ) return 1;
newstr_init( &page );
p = skip_ws( s->data );
while ( *p && !is_ws(*p) && *p!='-' && *p!='\r' && *p!='\n' && *p!=-30 )
newstr_addchar( &page, *p++ );
if ( page.len>0 ) {
fstatus = fields_add( info, "PAGESTART", page.data, level );
if ( fstatus!=FIELDS_OK ) {
status = BIBL_ERR_MEMERR;
goto out;
}
}
while ( *p && (is_ws(*p) || *p=='-' ) ) p++;
if ( *p && is_utf8_emdash( p ) ) p+=3;
if ( *p && is_utf8_endash( p ) ) p+=3;
newstr_empty( &page );
while ( *p && !is_ws(*p) && *p!='-' && *p!='\r' && *p!='\n' )
newstr_addchar( &page, *p++ );
if ( page.len>0 ) {
fstatus = fields_add( info, "PAGEEND", page.data, level );
if ( fstatus!=FIELDS_OK ) status = BIBL_ERR_MEMERR;
}
out:
newstr_free( &page );
return status;
}
static void
output_names( fields *f, FILE *outptr, int level )
{
convert names[] = {
{ "author", "AUTHOR", 0, MARC_AUTHORITY },
{ "editor", "EDITOR", 0, MARC_AUTHORITY },
{ "annotator", "ANNOTATOR", 0, MARC_AUTHORITY },
{ "artist", "ARTIST", 0, MARC_AUTHORITY },
{ "author", "2ND_AUTHOR", 0, MARC_AUTHORITY },
{ "author", "3RD_AUTHOR", 0, MARC_AUTHORITY },
{ "author", "SUB_AUTHOR", 0, MARC_AUTHORITY },
{ "author", "COMMITTEE", 0, MARC_AUTHORITY },
{ "author", "COURT", 0, MARC_AUTHORITY },
{ "author", "LEGISLATIVEBODY", 0, MARC_AUTHORITY },
{ "author of afterword, colophon, etc.", "AFTERAUTHOR", 0, MARC_AUTHORITY },
{ "author of introduction, etc.", "INTROAUTHOR", 0, MARC_AUTHORITY },
{ "cartographer", "CARTOGRAPHER", 0, MARC_AUTHORITY },
{ "collaborator", "COLLABORATOR", 0, MARC_AUTHORITY },
{ "commentator", "COMMENTATOR", 0, MARC_AUTHORITY },
{ "compiler", "COMPILER", 0, MARC_AUTHORITY },
{ "degree grantor", "DEGREEGRANTOR", 0, MARC_AUTHORITY },
{ "director", "DIRECTOR", 0, MARC_AUTHORITY },
{ "event", "EVENT", 0, NO_AUTHORITY },
{ "inventor", "INVENTOR", 0, MARC_AUTHORITY },
{ "organizer of meeting", "ORGANIZER", 0, MARC_AUTHORITY },
{ "patent holder", "ASSIGNEE", 0, MARC_AUTHORITY },
{ "performer", "PERFORMER", 0, MARC_AUTHORITY },
{ "producer", "PRODUCER", 0, MARC_AUTHORITY },
{ "recipient", "RECIPIENT", 0, MARC_AUTHORITY },
{ "redactor", "REDACTOR", 0, MARC_AUTHORITY },
{ "reporter", "REPORTER", 0, MARC_AUTHORITY },
{ "sponsor", "SPONSOR", 0, MARC_AUTHORITY },
{ "translator", "TRANSLATOR", 0, MARC_AUTHORITY },
{ "writer", "WRITER", 0, MARC_AUTHORITY },
};
int i, n, nfields, ntypes = sizeof( names ) / sizeof( convert );
int f_asis, f_corp, f_conf;
newstr role;
newstr_init( &role );
nfields = fields_num( f );
for ( n=0; n<ntypes; ++n ) {
for ( i=0; i<nfields; ++i ) {
if ( fields_level( f, i )!=level ) continue;
if ( f->data[i].len==0 ) continue;
f_asis = f_corp = f_conf = 0;
newstr_strcpy( &role, f->tag[i].data );
if ( newstr_findreplace( &role, ":ASIS", "" )) f_asis=1;
if ( newstr_findreplace( &role, ":CORP", "" )) f_corp=1;
if ( newstr_findreplace( &role, ":CONF", "" )) f_conf=1;
if ( strcasecmp( role.data, names[n].internal ) )
continue;
if ( f_asis ) {
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, NULL );
output_fil( outptr, lvl2indent(incr_level(level,1)), "namePart", f, i, TAG_OPENCLOSE, TAG_NEWLINE, NULL );
} else if ( f_corp ) {
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, "type", "corporate", NULL );
output_fil( outptr, lvl2indent(incr_level(level,1)), "namePart", f, i, TAG_OPENCLOSE, TAG_NEWLINE, NULL );
} else if ( f_conf ) {
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_OPEN, TAG_NEWLINE, "type", "conference", NULL );
output_fil( outptr, lvl2indent(incr_level(level,1)), "namePart", f, i, TAG_OPENCLOSE, TAG_NEWLINE, NULL );
} else {
output_name(outptr, f->data[i].data, level);
}
output_tag( outptr, lvl2indent(incr_level(level,1)), "role", NULL, TAG_OPEN, TAG_NEWLINE, NULL );
if ( names[n].code & MARC_AUTHORITY )
output_tag( outptr, lvl2indent(incr_level(level,2)), "roleTerm", names[n].mods, TAG_OPENCLOSE, TAG_NEWLINE, "authority", "marcrelator", "type", "text", NULL );
else
output_tag( outptr, lvl2indent(incr_level(level,2)), "roleTerm", names[n].mods, TAG_OPENCLOSE, TAG_NEWLINE, "type", "text", NULL );
output_tag( outptr, lvl2indent(incr_level(level,1)), "role", NULL, TAG_CLOSE, TAG_NEWLINE, NULL );
output_tag( outptr, lvl2indent(level), "name", NULL, TAG_CLOSE, TAG_NEWLINE, NULL );
fields_setused( f, i );
}
}
newstr_free( &role );
}
static void
bibtex_cleantoken( newstr *s, param *p )
{
if ( p && p->latexin==0 ) return;
/* 'textcomp' annotations */
newstr_findreplace( s, "\\textit", "" );
newstr_findreplace( s, "\\textbf", "" );
newstr_findreplace( s, "\\textsl", "" );
newstr_findreplace( s, "\\textsc", "" );
newstr_findreplace( s, "\\textsf", "" );
newstr_findreplace( s, "\\texttt", "" );
newstr_findreplace( s, "\\textsubscript", "" );
newstr_findreplace( s, "\\textsuperscript", "" );
newstr_findreplace( s, "\\emph", "" );
newstr_findreplace( s, "\\url", "" );
/* Other text annotations */
newstr_findreplace( s, "\\it ", "" );
newstr_findreplace( s, "\\em ", "" );
newstr_findreplace( s, "\\%", "%" );
newstr_findreplace( s, "\\$", "$" );
newstr_findreplace( s, "{", "" );
newstr_findreplace( s, "}", "" );
while ( newstr_findreplace( s, " ", " " ) ) {}
/* 'textcomp' annotations that we don't want to substitute on output*/
newstr_findreplace( s, "\\textdollar", "$" );
newstr_findreplace( s, "\\textunderscore", "_" );
}
static void
bibtex_cleantoken( newstr *s )
{
/* 'textcomp' annotations */
newstr_findreplace( s, "\\textit", "" );
newstr_findreplace( s, "\\textbf", "" );
newstr_findreplace( s, "\\textsl", "" );
newstr_findreplace( s, "\\textsc", "" );
newstr_findreplace( s, "\\textsf", "" );
newstr_findreplace( s, "\\texttt", "" );
newstr_findreplace( s, "\\textsubscript", "" );
newstr_findreplace( s, "\\textsuperscript", "" );
newstr_findreplace( s, "\\emph", "" );
newstr_findreplace( s, "\\url", "" );
newstr_findreplace( s, "\\mbox", "" );
/* Other text annotations */
newstr_findreplace( s, "\\it ", "" );
newstr_findreplace( s, "\\em ", "" );
newstr_findreplace( s, "\\%", "%" );
newstr_findreplace( s, "\\$", "$" );
while ( newstr_findreplace( s, " ", " " ) ) {}
/* 'textcomp' annotations that we don't want to substitute on output*/
newstr_findreplace( s, "\\textdollar", "$" );
newstr_findreplace( s, "\\textunderscore", "_" );
bibtex_process_bracket( s );
bibtex_process_tilde( s );
}