This file is indexed.

/usr/share/Yap/rdf_triple.pl is in yap 6.2.2-6+b2.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
/*  Part of SWI-Prolog

    Author:        Jan Wielemaker
    E-mail:        J.Wielemaker@uva.nl
    WWW:           http://www.swi-prolog.org
    Copyright (C): 2002-2010, University of Amsterdam

    This program is free software; you can redistribute it and/or
    modify it under the terms of the GNU General Public License
    as published by the Free Software Foundation; either version 2
    of the License, or (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public
    License along with this library; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

    As a special exception, if you link this library with other files,
    compiled with a Free Software compiler, to produce an executable, this
    library does not by itself cause the resulting executable to be covered
    by the GNU General Public License. This exception does not however
    invalidate any other reasons why the executable file might be covered by
    the GNU General Public License.
*/

:- module(rdf_triple,
	  [ rdf_triples/2,		% +Parsed, -Tripples
	    rdf_triples/3,		% +Parsed, -Tripples, +Tail
	    rdf_reset_ids/0,		% Reset gensym id's
	    rdf_start_file/2,		% +Options, -Cleanup
	    rdf_end_file/1,		% +Cleanup
	    anon_prefix/1		% Prefix for anonynmous resources
	  ]).
:- use_module(library(gensym)).
:- use_module(rdf_parser).

/** <module> Create triples from intermediate representation

Convert the output of xml_to_rdf/3  from   library(rdf)  into  a list of
triples of the format described   below. The intermediate representation
should be regarded a proprietary representation.

	rdf(Subject, Predicate, Object).

Where `Subject' is

	* Atom
	The subject is a resource

	* each(URI)
	URI is the URI of an RDF Bag

	* prefix(Pattern)
	Pattern is the prefix of a fully qualified Subject URI

And `Predicate' is

	* Atom
	The predicate is always a resource

And `Object' is

	* Atom
	URI of Object resource

	* literal(Value)
	Literal value (Either a single atom or parsed XML data)
*/

%%	rdf_triples(+Term, -Triples) is det.
%%	rdf_triples(+Term, -Tridpples, +Tail) is det.
%
%	Convert an object as parsed by rdf.pl into a list of rdf/3
%	triples.  The identifier of the main object created is returned
%	by rdf_triples/3.
%
%	Input is the `content' of the RDF element in the format as
%	generated by load_structure(File, Term, [dialect(xmlns)]).
%	rdf_triples/3 can process both individual descriptions as
%	well as the entire content-list of an RDF element.  The first
%	mode is suitable when using library(sgml) in `call-back' mode.

rdf_triples(RDF, Tripples) :-
	rdf_triples(RDF, Tripples, []).

rdf_triples([]) --> !,
	[].
rdf_triples([H|T]) --> !,
	rdf_triples(H),
	rdf_triples(T).
rdf_triples(Term) -->
	triples(Term, _).

%%	triples(-Triples, -Id, +In, -Tail)
%
%	DGC set processing the output of  xml_to_rdf/3. Id is unified to
%	the identifier of the main description.

triples(description(Type, About, Props), Subject) -->
	{ var(About),
	  share_blank_nodes(true)
	}, !,
	(   { shared_description(description(Type, Props), Subject)
	    }
	->  []
	;   { make_id('__Description', Id)
	    },
	    triples(description(Type, about(Id), Props), Subject),
	    { assert_shared_description(description(Type, Props), Subject)
	    }
	).
triples(description(description, IdAbout, Props), Subject) --> !,
	{ description_id(IdAbout, Subject)
	},
	properties(Props, Subject).
triples(description(TypeURI, IdAbout, Props), Subject) -->
	{ description_id(IdAbout, Subject)
	},
	properties([ rdf:type = TypeURI
		   | Props
		   ], Subject).
triples(unparsed(Data), Id) -->
	{ make_id('__Error', Id),
	  print_message(error, rdf(unparsed(Data)))
	},
	[].


		 /*******************************
		 *	    DESCRIPTIONS	*
		 *******************************/

:- thread_local
	node_id/2,			% nodeID --> ID
	unique_id/1.			% known rdf:ID

rdf_reset_node_ids :-
	retractall(node_id(_,_)),
	retractall(unique_id(_)).

description_id(Id, Id) :-
	var(Id), !,
	make_id('__Description', Id).
description_id(about(Id), Id).
description_id(id(Id), Id) :-
	(   unique_id(Id)
	->  print_message(error, rdf(redefined_id(Id)))
	;   assert(unique_id(Id))
	).
description_id(each(Id), each(Id)).
description_id(prefix(Id), prefix(Id)).
description_id(node(NodeID), Id) :-
	(   node_id(NodeID, Id)
	->  true
	;   make_id('__Node', Id),
	    assert(node_id(NodeID, Id))
	).

properties(PlRDF, Subject) -->
	properties(PlRDF, 1, [], [], Subject).

properties([], _, Bag, Bag, _) -->
	[].
properties([H0|T0], N, Bag0, Bag, Subject) -->
	property(H0, N, NN, Bag0, Bag1, Subject),
	properties(T0, NN, Bag1, Bag, Subject).

%%	property(Property, N, NN, Subject)// is det.
%
%	Generate triples for {Subject,  Pred,   Object}.  Also generates
%	triples for Object if necessary.
%
%	@param Property	One of
%
%		* Pred = Object
%		Used for normal statements
%		* id(Id, Pred = Object)
%		Used for reified statements

property(Pred0 = Object, N, NN, BagH, BagT, Subject) --> % inlined object
	triples(Object, Id), !,
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, Id, _, BagH, BagT).
property(Pred0 = collection(Elems), N, NN, BagH, BagT, Subject) --> !,
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, Object, _Id, BagH, BagT),
	collection(Elems, Object).
property(Pred0 = Object, N, NN, BagH, BagT, Subject) --> !,
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, Object, _Id, BagH, BagT).
property(id(Id, Pred0 = Object), N, NN, BagH, BagT, Subject) -->
	triples(Object, ObjectId), !,
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, ObjectId, Id, BagH, BagT).
property(id(Id, Pred0 = collection(Elems)), N, NN, BagH, BagT, Subject) --> !,
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, Object, Id, BagH, BagT),
	collection(Elems, Object).
property(id(Id, Pred0 = Object), N, NN, BagH, BagT, Subject) -->
	{ li_pred(Pred0, Pred, N, NN)
	},
	statement(Subject, Pred, Object, Id, BagH, BagT).

%%	statement(+Subject, +Pred, +Object, +Id, +BagH, -BagT)
%
%	Add a statement to the model. If nonvar(Id), we reinify the
%	statement using the given Id.

statement(Subject, Pred, Object, Id, BagH, BagT) -->
	rdf(Subject, Pred, Object),
	{   BagH = [Id|BagT]
	->  statement_id(Id)
	;   BagT = BagH
	},
	(   { nonvar(Id)
	    }
	->  rdf(Id, rdf:type, rdf:'Statement'),
	    rdf(Id, rdf:subject, Subject),
	    rdf(Id, rdf:predicate, Pred),
	    rdf(Id, rdf:object, Object)
	;   []
	).


statement_id(Id) :-
	nonvar(Id), !.
statement_id(Id) :-
	make_id('__Statement', Id).

%%	li_pred(+Pred, -Pred, +Nth, -NextNth)
%
%	Transform rdf:li predicates into _1, _2, etc.

li_pred(rdf:li, rdf:Pred, N, NN) :- !,
	NN is N + 1,
	atom_concat('_', N, Pred).
li_pred(Pred, Pred, N, N).

%%	collection(+Elems, -Id)
%
%	Handle the elements of a collection and return the identifier
%	for the whole collection in Id.

collection([], Nil) -->
	{ global_ref(rdf:nil, Nil)
	}.
collection([H|T], Id) -->
	triples(H, HId),
	{ make_id('__List', Id)
	},
	rdf(Id, rdf:type, rdf:'List'),
	rdf(Id, rdf:first, HId),
	rdf(Id, rdf:rest, TId),
	collection(T, TId).


rdf(S0, P0, O0) -->
	{ global_ref(S0, S),
	  global_ref(P0, P),
	  global_obj(O0, O)
	},
	[ rdf(S, P, O) ].


global_ref(In, Out) :-
	(   nonvar(In),
	    In = NS:Local
	->  (   NS == rdf,
	        rdf_name_space(RDF)
	    ->	atom_concat(RDF, Local, Out)
	    ;	atom_concat(NS, Local, Out)
	    )
	;   Out = In
	).

global_obj(V, V) :-
	var(V), !.
global_obj(literal(type(Local, X)), literal(type(Global, X))) :- !,
	global_ref(Local, Global).
global_obj(literal(X), literal(X)) :- !.
global_obj(Local, Global) :-
	global_ref(Local, Global).


		 /*******************************
		 *	       SHARING		*
		 *******************************/

:- thread_local
	shared_description/3,		% +Hash, +Term, -Subject
	share_blank_nodes/1,		% Boolean
	shared_nodes/1.			% counter

reset_shared_descriptions :-
	retractall(shared_description(_,_,_)),
	retractall(shared_nodes(_)).

shared_description(Term, Subject) :-
	term_hash(Term, Hash),
	shared_description(Hash, Term, Subject),
	(   retract(shared_nodes(N))
	->  N1 is N + 1
	;   N1 = 1
	),
	assert(shared_nodes(N1)).


assert_shared_description(Term, Subject) :-
	term_hash(Term, Hash),
	assert(shared_description(Hash, Term, Subject)).


		 /*******************************
		 *	      START/END		*
		 *******************************/

%%	rdf_start_file(+Options, -Cleanup) is det.
%
%	Initialise for the translation of a file.

rdf_start_file(Options, Cleanup) :-
	rdf_reset_node_ids,		% play safe
	reset_shared_descriptions,
	set_bnode_sharing(Options, C1),
	set_anon_prefix(Options, C2),
	add_cleanup(C1, C2, Cleanup).

%%	rdf_end_file(:Cleanup) is det.
%
%	Cleanup reaching the end of an RDF file.

rdf_end_file(Cleanup) :-
	rdf_reset_node_ids,
	(   shared_nodes(N)
	->  print_message(informational, rdf(shared_blank_nodes(N)))
	;   true
	),
	reset_shared_descriptions,
	Cleanup.

set_bnode_sharing(Options, erase(Ref)) :-
	option(blank_nodes(Share), Options, noshare),
	(   Share == share
	->  assert(share_blank_nodes(true), Ref), !
	;   Share == noshare
	->  fail			% next clause
	;   throw(error(domain_error(share, Share), _))
	).
set_bnode_sharing(_, true).

set_anon_prefix(Options, erase(Ref)) :-
	option(base_uri(BaseURI), Options),
	nonvar(BaseURI), !,
	atomic_list_concat(['__', BaseURI, '#'], AnonBase),
	asserta(anon_prefix(AnonBase), Ref).
set_anon_prefix(_, true).

add_cleanup(true, X, X) :- !.
add_cleanup(X, true, X) :- !.
add_cleanup(X, Y, (X, Y)).


		 /*******************************
		 *	       UTIL		*
		 *******************************/

%%	anon_prefix(-Prefix) is semidet.
%
%	If defined, it is the prefix used to generate a blank node.

:- thread_local
	anon_prefix/1.

make_id(For, ID) :-
	anon_prefix(Prefix), !,
	atom_concat(Prefix, For, Base),
	gensym(Base, ID).
make_id(For, ID) :-
	gensym(For, ID).

anon_base('__Description').
anon_base('__Statement').
anon_base('__List').
anon_base('__Node').

%%	rdf_reset_ids is det.
%
%	Utility predicate to reset the gensym counters for the various
%	generated identifiers.  This simplifies debugging and matching
%	output with the stored desired output (see rdf_test.pl).

rdf_reset_ids :-
	anon_prefix(Prefix), !,
	(   anon_base(Base),
	    atom_concat(Prefix, Base, X),
	    reset_gensym(X),
	    fail
	;   true
	).
rdf_reset_ids :-
	(   anon_base(Base),
	    reset_gensym(Base),
	    fail
	;   true
	).