1 | C0XF2N ; GPL - Fileman Triples entry point routine ;10/13/11 17:05
|
---|
2 | ;;0.1;C0X;nopatch;noreleasedate;Build 7
|
---|
3 | ;Copyright 2011 George Lilly. Licensed under the terms of the GNU
|
---|
4 | ;General Public License See attached copy of the License.
|
---|
5 | ;
|
---|
6 | ;This program is free software; you can redistribute it and/or modify
|
---|
7 | ;it under the terms of the GNU General Public License as published by
|
---|
8 | ;the Free Software Foundation; either version 2 of the License, or
|
---|
9 | ;(at your option) any later version.
|
---|
10 | ;
|
---|
11 | ;This program is distributed in the hope that it will be useful,
|
---|
12 | ;but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
13 | ;MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
14 | ;GNU General Public License for more details.
|
---|
15 | ;
|
---|
16 | ;You should have received a copy of the GNU General Public License along
|
---|
17 | ;with this program; if not, write to the Free Software Foundation, Inc.,
|
---|
18 | ;51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
|
---|
19 | ;
|
---|
20 | Q
|
---|
21 | ;
|
---|
22 | ; This is based on C0XMAIN but experiments with a fast load for triples
|
---|
23 | ; that will write directly to the fileman global
|
---|
24 | ; The file 172.101 is a F2N design style for triples, which means
|
---|
25 | ; that it is a Flat file with no subfiles, all fields at the root
|
---|
26 | ; ... it is a "2" file solution which means all strings are stored in
|
---|
27 | ; ... strings file and pointed to by the triples file
|
---|
28 | ; ... it is an N file because it has generated Node IDs instead of
|
---|
29 | ; ... DINUM which would use the IEN for the Node ID.
|
---|
30 | ; gpl 11/04/2011
|
---|
31 | ;
|
---|
32 | INITFARY(ZFARY) ; INITIALIZE FILE NUMBERS AND OTHER USEFUL THINGS
|
---|
33 | ; FOR THE DEFAULT TRIPLE STORE. USE OTHER VALUES FOR SUPPORTING ADDITIONAL
|
---|
34 | ; TRIPLE STORES
|
---|
35 | I $D(@ZFARY) Q ; ALREADY INITIALIZED
|
---|
36 | S @ZFARY@("C0XTFN")=172.101 ; TRIPLES FILE NUMBER
|
---|
37 | S @ZFARY@("C0XSFN")=172.201 ; TRIPLES STRINGS FILE NUMBER
|
---|
38 | S @ZFARY@("C0XTN")=$NA(^C0X(101)) ; TRIPLES GLOBAL NAME
|
---|
39 | S @ZFARY@("C0XSN")=$NA(^C0X(201)) ; STRING FILE GLOBAL NAME
|
---|
40 | S @ZFARY@("C0XDIR")="/home/glilly/fmts/trunk/samples/smart-rdf-in/"
|
---|
41 | S @ZFARY@("BLKLOAD")=1 ; this file supports block load
|
---|
42 | S @ZFARY@("FMTSSTYLE")="F2N" ; fileman style
|
---|
43 | S @ZFARY@("REPLYFMT")="JSON"
|
---|
44 | D USEFARY(ZFARY)
|
---|
45 | Q
|
---|
46 | ;
|
---|
47 | USEFARY(ZFARY) ; INITIALIZES VARIABLES SAVED IN ARRAY ZFARY
|
---|
48 | N ZI S ZI=""
|
---|
49 | F S ZI=$O(@ZFARY@(ZI)) Q:ZI="" D
|
---|
50 | . ;N ZX
|
---|
51 | . S ZX="S "_ZI_"="""_@ZFARY@(ZI)_""""
|
---|
52 | . ;W !,ZX
|
---|
53 | . X ZX
|
---|
54 | Q
|
---|
55 | ;
|
---|
56 | FILEIN ; INTERACTIVE ENTRY POINT FOR OPTION TO READ IN A FILE
|
---|
57 | I '$D(C0XFARY) D INITFARY("C0XFARY")
|
---|
58 | D USEFARY("C0XFARY")
|
---|
59 | S DIR(0)="F^3:240"
|
---|
60 | S DIR("A")="File Directory"
|
---|
61 | S DIR("B")=C0XDIR
|
---|
62 | D ^DIR
|
---|
63 | I Y="^" Q ;
|
---|
64 | S C0XDIR=Y
|
---|
65 | S C0XFARY("C0XDIR")=Y
|
---|
66 | S DIR(0)="F^3:240"
|
---|
67 | S DIR("A")="File Name"
|
---|
68 | I '$D(C0XFN) S DIR("B")="qds.rdf"
|
---|
69 | E S DIR("B")=C0XFN
|
---|
70 | D ^DIR
|
---|
71 | I Y="" Q ;
|
---|
72 | I Y="^" Q ;
|
---|
73 | S C0XFN=Y
|
---|
74 | D IMPORT(C0XFN,C0XDIR,,"C0XFARY")
|
---|
75 | K C0XFDA
|
---|
76 | Q
|
---|
77 | ;
|
---|
78 | IMPORT(FNAME,INDIR,INURL,FARY) ; EXTRINSIC THAT READS A FILE FROM THE STANDARD
|
---|
79 | ; DIRECTORY, LOADS IT INTO THE TRIPLESTORE AS TEXT, AND RETURNS THE
|
---|
80 | ; NODE NAME OF THE TEXT TRIPLE
|
---|
81 | ; INDIR IS THE OPTIONAL DIRECTORY (DEFAUTS TO STANDARD DIR)
|
---|
82 | ; INURL IS THE OPTIONAL URI FOR ACCESSING THE FILE FROM THE TRIPLE STORE
|
---|
83 | ; FARY IS THE OPTIONAL FILE ARRAY OF THE TRIPLE STORE TO USE
|
---|
84 | I '$D(FARY) D ;
|
---|
85 | . D INITFARY("C0XFARY")
|
---|
86 | . S FARY="C0XFARY"
|
---|
87 | D USEFARY(FARY)
|
---|
88 | N ZD,ZTMP
|
---|
89 | I '$D(INDIR) S INDIR=C0XDIR ; DIRECTORY OF THE RDF FILE
|
---|
90 | I $G(INURL)="" D ;
|
---|
91 | . ;N ZN2 S ZN2=$P(FNAME,".",1)_"_"_$P(FNAME,".",2) ; REMOVE THE DOT
|
---|
92 | . ;S INURL=FDIR_ZN2
|
---|
93 | . S INURL=INDIR_FNAME
|
---|
94 | N ZTMP
|
---|
95 | S ZTMP=$NA(^TMP("C0X",$J,"FILEIN",1)) ; WHERE TO PUT THE INCOMING FILE
|
---|
96 | K @ZTMP ; MAKE SURE IT'S CLEAR
|
---|
97 | S C0XSTART=$$NOW^XLFDT
|
---|
98 | W !,"STARTED: ",C0XSTART
|
---|
99 | W !,"READING IN: ",FNAME
|
---|
100 | I '$$FILEREAD(ZTMP,INDIR,FNAME,4) D Q ; QUIT IF NO SUCCESS
|
---|
101 | . W !,"ERROR READING FILE: ",INDIR,FNAME
|
---|
102 | S ZRDF=$NA(^TMP("C0X",$J,"FILEIN")) ; WITHOUT THE SUBSCRIPT
|
---|
103 | W !,$O(@ZRDF@(""),-1)," LINES READ"
|
---|
104 | D INSRDF(ZRDF,INURL,FARY) ; IMPORT AND PROCESS THE RDF
|
---|
105 | K INURL
|
---|
106 | K C0XFDA
|
---|
107 | ;K ^TMP("MXMLDOM",$J)
|
---|
108 | Q
|
---|
109 | ;
|
---|
110 | WGET(ZURL,FARY) ; GET FROM THE INTERNET AN RDF FILE AND INSERT IT
|
---|
111 | ;
|
---|
112 | I '$D(FARY) D ;
|
---|
113 | . D INITFARY("C0XFARY")
|
---|
114 | . S FARY="C0XFARY"
|
---|
115 | D USEFARY(FARY)
|
---|
116 | ;N ZLOC,ZTMP
|
---|
117 | K ZTMP
|
---|
118 | S ZLOC=$NA(^TMP("C0X","WGET",$J))
|
---|
119 | K @ZLOC
|
---|
120 | S C0XSTART=$$NOW^XLFDT
|
---|
121 | W !,"STARTED: ",C0XSTART
|
---|
122 | W !,"DOWNLOADING: ",ZURL
|
---|
123 | S OK=$$httpGET^%zewdGTM(ZURL,.ZTMP)
|
---|
124 | M @ZLOC=ZTMP
|
---|
125 | S C0XLINES=$O(@ZLOC@(""),-1)
|
---|
126 | W !,C0XLINES," LINES READ"
|
---|
127 | S C0XDLC=$$NOW^XLFDT ; DOWNLOAD COMPLETE
|
---|
128 | W !,"DOWNLOAD COMPLETE AT ",C0XDLC
|
---|
129 | S C0XDIFF=$$FMDIFF^XLFDT(C0XDLC,C0XSTART,2)
|
---|
130 | W !," ELAPSED TIME: ",C0XDIFF," SECONDS"
|
---|
131 | I C0XDIFF'=0 W !," APPROXIMATELY ",$P(C0XLINES/C0XDIFF,".")," LINES PER SEC"
|
---|
132 | D INSRDF(ZLOC,ZURL,FARY)
|
---|
133 | Q
|
---|
134 | ;
|
---|
135 | INSRDF(ZRDF,ZNAME,FARY) ; INSERT AN RDF FILE INTO THE STORE AND PROCESS
|
---|
136 | ; ZRDF IS PASSED BY NAME
|
---|
137 | I '$D(FARY) D ;
|
---|
138 | . D INITFARY("C0XFARY")
|
---|
139 | . S FARY="C0XFARY"
|
---|
140 | D USEFARY(FARY)
|
---|
141 | S BATCNT=0 ; BATCH COUNTER
|
---|
142 | S BATMAX=10000 ; TRY BATCHES OF THIS SIZE
|
---|
143 | N ZGRAPH,ZSUBJECT
|
---|
144 | S ZGRAPH="_:G"_$$LKY9 ; RANDOM GRAPH NAME
|
---|
145 | S ZSUBJECT=$$ANONS() ; RANDOM ANOYMOUS SUBJECT
|
---|
146 | D ADD(ZGRAPH,ZSUBJECT,"fmts:url",ZNAME,FARY)
|
---|
147 | N ZTXTNM
|
---|
148 | S ZTXTNM="_TXT_INCOMING_RDF_FILE_"_ZNAME_"_"_$$LKY9 ; NAME FOR TEXT NODE
|
---|
149 | D ADD(ZGRAPH,ZSUBJECT,"fmts:rdfSource",ZTXTNM,FARY)
|
---|
150 | D SWUPDIE(.C0XFDA) ; TRY IT OUT
|
---|
151 | K C0XCNT ;RESET FOR NEXT TIME
|
---|
152 | D STORETXT(ZRDF,ZTXTNM,FARY)
|
---|
153 | W !,"ADDED: ",ZGRAPH," ",ZSUBJECT," fmts:rdfSource ",ZTXTNM
|
---|
154 | D PROCESS(.G,ZRDF,ZNAME,ZGRAPH,FARY) ; PARSE AND INSERT THE RDF
|
---|
155 | Q
|
---|
156 | ;
|
---|
157 | STORETXT(ZTXT,ZNAME,FARY) ; STORE TEXT IN THE TRIPLESTORE AT ZNAME
|
---|
158 | ;
|
---|
159 | I '$D(FARY) D ;
|
---|
160 | . D INITFARY("C0XFARY")
|
---|
161 | . S FARY="C0XFARY"
|
---|
162 | D USEFARY(FARY)
|
---|
163 | N ZIEN
|
---|
164 | S ZIEN=$$IENOF(ZNAME,FARY) ; GET THE IEN
|
---|
165 | D CLEAN^DILF
|
---|
166 | K ZERR
|
---|
167 | D WP^DIE(C0XSFN,ZIEN_",",1,,ZTXT,"ZERR")
|
---|
168 | I $D(ZERR) D Q ;
|
---|
169 | . W !,"ERROR CREATING WORD PROCESSING FIELD"
|
---|
170 | . S C0XERR="ERROR CREATING WORD PROCESSING FIELD"
|
---|
171 | . D ^%ZTER ; error trap
|
---|
172 | Q
|
---|
173 | ;
|
---|
174 | GETTXT(ZRTN,ZNAME,FARY) ; RETURNS RDF SOURCE OR OTHER TEXT
|
---|
175 | ; ZRTN IS PASSED BY REFERENCE
|
---|
176 | I '$D(FARY) D ;
|
---|
177 | . D INITFARY("C0XFARY")
|
---|
178 | . S FARY="C0XFARY"
|
---|
179 | D USEFARY(FARY)
|
---|
180 | N ZIEN
|
---|
181 | S ZIEN=$$IENOF(ZNAME)
|
---|
182 | S OK=$$GET1^DIQ(C0XSFN,ZIEN_",",1,,"ZRTN")
|
---|
183 | Q
|
---|
184 | ;
|
---|
185 | WHERETXT(ZNAME,FARY) ; EXTRINSIC WHICH RETURNS THE NAME OF THE GLOBAL
|
---|
186 | ; WHERE THE TEXT IS LOCATED. NAME IS THE NAME OF THE STRING
|
---|
187 | I '$D(FARY) D ;
|
---|
188 | . D INITFARY("C0XFARY")
|
---|
189 | . S FARY="C0XFARY"
|
---|
190 | D USEFARY(FARY)
|
---|
191 | N ZIEN
|
---|
192 | S ZIEN=$$IENOF(ZNAME)
|
---|
193 | Q $NA(@C0XSN@(ZIEN,1))
|
---|
194 | ;
|
---|
195 | FILEREAD(ZINTMP,ZDIR,ZFNAME,ZLVL) ; READS A FILE INTO ZINTMP USING FTG^%ZISH
|
---|
196 | ; ZINTMP IS PASSED BY NAME AND INCLUDES THE NEW SUBSCRIPT
|
---|
197 | ; IE ^TMP("C0X","FILEIN",1)
|
---|
198 | ; ZLVL IN THIS CASE WOULD BE 3 INCREMENTING THE 1
|
---|
199 | ; EXTRINSIC WHICH RETURNS THE RESULT OF FTG^%ZISH
|
---|
200 | S OK=$$FTG^%ZISH(ZDIR,FNAME,ZINTMP,ZLVL)
|
---|
201 | Q OK
|
---|
202 | ;
|
---|
203 | TESTPROC ; TEST PROCESS WITH EXISTING SMALL RDF FILE
|
---|
204 | S ZIN=$NA(^TMP("C0X",12226,"FILEIN"))
|
---|
205 | S ZGRAPH="/test/rdfFile"
|
---|
206 | S ZM="/test/rdfFile/meta"
|
---|
207 | D PROCESS(.G,ZIN,ZGRAPH,ZM)
|
---|
208 | Q
|
---|
209 | ;
|
---|
210 | VISTAOWL ;
|
---|
211 | S ZRDF=$NA(^TMP("C0X",542,"FILEIN"))
|
---|
212 | S ZNAME="/home/glilly/vistaowl/VistAOWL.owl"
|
---|
213 | S ZGRAPH="_:G431590209"
|
---|
214 | S FARY="C0XFARY"
|
---|
215 | D INITFARY(FARY)
|
---|
216 | S C0XDOCID=1
|
---|
217 | S BATCNT=0
|
---|
218 | S BATMAX=10000
|
---|
219 | D PROCESS(.G,ZRDF,ZGRAPH,ZNAME,FARY)
|
---|
220 | Q
|
---|
221 | ;
|
---|
222 | PROCESS(ZRTN,ZRDF,ZGRF,ZMETA,FARY) ; PROCESS AN INCOMING RDF FILE
|
---|
223 | ; ZRTN IS PASS BY REFERENCE AND RETURNS MESSAGES ABOUT THE PROCESSING
|
---|
224 | ; ZRDF IS PASSED BY NAME AND IS THE GLOBAL CONTAINING THE RDF FILE
|
---|
225 | ; ZGRF IS THE NAME OF THE GRAPH TO USE IN THE TRIPLE STORE FOR RESULTS
|
---|
226 | ; ZMETA IS OPTIONAL AND IS THE NAME OF THE GRAPH TO STORE METADATA
|
---|
227 | ;
|
---|
228 | I '$D(FARY) D ;
|
---|
229 | . D INITFARY("C0XFARY")
|
---|
230 | . S FARY="C0XFARY"
|
---|
231 | D USEFARY(FARY)
|
---|
232 | ;N BATCNT
|
---|
233 | ;N BATMAX
|
---|
234 | ; -- first parse the rdf file with the MXML parser
|
---|
235 | ;S C0XDOCID=$$PARSE^C0CNHIN(ZRDF,"C0XARRAY") ; PARSE WITH MXML
|
---|
236 | S C0XDLC2=$$NOW^XLFDT ; START OF PARSE
|
---|
237 | S C0XDOCID=$$EN^MXMLDOM(ZRDF,"W") ;
|
---|
238 | ;B
|
---|
239 | K @ZRDF ; DON'T NEED INPUT BUFFER ANYMORE
|
---|
240 | ; -- assign the MXLM dom global name to ZDOM
|
---|
241 | S ZDOM=$NA(^TMP("MXMLDOM",$J,C0XDOCID))
|
---|
242 | ;S ZDOM=$NA(^TMP("MXMLDOM",16850,C0XDOCID)) ;VISTAOWL DOM
|
---|
243 | S C0XNODE=$O(@ZDOM@(""),-1)
|
---|
244 | W !,C0XNODE," XML NODES PARSED"
|
---|
245 | S C0XPRS=$$NOW^XLFDT ; PARSE COMPLETE
|
---|
246 | W !,"PARSE COMPLETE AT ",C0XPRS
|
---|
247 | S C0XDIFF=$$FMDIFF^XLFDT(C0XPRS,C0XDLC2,2)
|
---|
248 | W !," ELAPSED TIME: ",C0XDIFF," SECONDS"
|
---|
249 | I C0XDIFF'=0 D ;
|
---|
250 | . W !," APPROXIMATELY ",$P(C0XNODE/C0XDIFF,".")," NODES PER SECOND"
|
---|
251 | ; -- populate the metagraph to point to the graph with status unfinished
|
---|
252 | S METAS=$$ANONS ; GET AN ANONOMOUS RANDOM SUBJECT
|
---|
253 | I '$D(ZMETA) S ZMETA="_:G"_$$LKY9 ; RANDOM GRAPH NAME FOR METAGRAPH
|
---|
254 | D ADD(ZMETA,METAS,"fmts:about",ZGRF,FARY) ; POINT THE META TO THE GRAPH
|
---|
255 | D ADD(ZMETA,METAS,"fmts:status","unfinished",FARY) ; mark as unfinished
|
---|
256 | W !,"INSERTING GRAPH: ",ZGRF
|
---|
257 | ;S C0XDATE=$$FMDTOUTC^C0CUTIL($$NOW^XLFDT,"DT")
|
---|
258 | S C0XDATE=$$NOW^XLFDT
|
---|
259 | D ADD(ZMETA,METAS,"fmts:dateTime",C0XDATE,FARY)
|
---|
260 | D SWUPDIE(.C0XFDA) ; commit the metagraph changes to the triple store
|
---|
261 | ; --
|
---|
262 | ; -- pull out the vocabularies in the RDF statement. marked with xmlns:
|
---|
263 | ; -- put them in a local variable for quick reference
|
---|
264 | ; -- TODO: create a graph for vocabularies and validate incoming against it
|
---|
265 | ;
|
---|
266 | S C0XVOC=""
|
---|
267 | N ZI,ZJ,ZK S ZI=""
|
---|
268 | F S ZI=$O(@ZDOM@(1,"A",ZI)) Q:ZI="" D ; FOR EACH xmlns
|
---|
269 | . S ZVOC=$P(ZI,"xmlns:",2)
|
---|
270 | . I ZVOC'="" S C0XVOC(ZVOC)=$G(@ZDOM@(1,"A",ZI))
|
---|
271 | W !,"VOCABS:"
|
---|
272 | N ZZ S ZZ=""
|
---|
273 | F S ZZ=$O(C0XVOC(ZZ)) Q:ZZ="" W !,ZZ,":",C0XVOC(ZZ)
|
---|
274 | ;
|
---|
275 | ; -- look for children called rdf:Description. quit if none. not an rdf file
|
---|
276 | ;
|
---|
277 | S C0XTYPE("rdf:Description")=1
|
---|
278 | S C0XTYPE("owl:ObjectProperty")=1
|
---|
279 | S C0XTYPE("owl:Ontology")=1
|
---|
280 | S C0XTYPE("owl:Class")=1
|
---|
281 | S C0XTYPE("rdfs:subClassOf")=1
|
---|
282 | S ZI=$O(@ZDOM@(1,"C",""))
|
---|
283 | I '$G(C0XTYPE(@ZDOM@(1,"C",ZI))) D Q ; not an rdf file
|
---|
284 | . W !,"Error. Not an RDF file. Cannot process."
|
---|
285 | . D SHOW(1)
|
---|
286 | ;
|
---|
287 | ; -- now process the rdf description children
|
---|
288 | ;
|
---|
289 | S ZI=""
|
---|
290 | S (C0XSUB,C0XPRE,C0XOBJ)="" ; INITIALIZE subject, object and predicate
|
---|
291 | F S ZI=$O(@ZDOM@(1,"C",ZI)) Q:ZI="" D ;
|
---|
292 | . ; -- we are skipping any child that is not rdf:Description
|
---|
293 | . ; -- TODO: check to see if this is right in general
|
---|
294 | . ;
|
---|
295 | . IF '$G(C0XTYPE(@ZDOM@(1,"C",ZI))) D Q ;
|
---|
296 | . . W !,"SKIPPING NODE: ",ZI
|
---|
297 | . ; -- now looking for the subject for the triples
|
---|
298 | . S ZX=$G(@ZDOM@(ZI,"A","rdf:about"))
|
---|
299 | . I ZX'="" D ; we have the subject
|
---|
300 | . . ;W " about: ",ZX
|
---|
301 | . . S C0XSUB=ZX
|
---|
302 | . E D ;
|
---|
303 | . . S ZX=$G(@ZDOM@(ZI,"A","rdf:nodeID")) ; node id is another style of subject
|
---|
304 | . . I ZX'="" D ;
|
---|
305 | . . . S C0XSUB=ZX
|
---|
306 | . I C0XSUB="" S C0XSUB=$$ANONS ; DEFAULT TO BLANK SUBJECT
|
---|
307 | . ;
|
---|
308 | . ; -- we now have the subject. the children of this node have the rest
|
---|
309 | . ;
|
---|
310 | . S ZJ="" ; for the children of the rdf:Description nodes
|
---|
311 | . F S ZJ=$O(@ZDOM@(ZI,"C",ZJ)) Q:ZJ="" D ; for each child
|
---|
312 | . . S C0XPRE=@ZDOM@(ZJ) ; the predicate without a prefix
|
---|
313 | . . S ZX=$G(@ZDOM@(ZJ,"A","xmlns")) ; name space
|
---|
314 | . . I ZX'="" S C0XPRE=ZX_C0XPRE ; add the namespace prefix
|
---|
315 | . . I C0XPRE[":" D ; expand using vocabulary
|
---|
316 | . . . N ZB,ZA
|
---|
317 | . . . S ZB=$P(C0XPRE,":",1)
|
---|
318 | . . . S ZA=$P(C0XPRE,":",2)
|
---|
319 | . . . I $G(C0XVOC(ZB))'="" D ;
|
---|
320 | . . . . S C0XPRE=C0XVOC(ZB)_ZA ; expanded
|
---|
321 | . . S ZY=$G(@ZDOM@(ZJ,"A","rdf:resource")) ; potential object
|
---|
322 | . . I ZY'="" D Q ;
|
---|
323 | . . . S C0XOBJ=$$EXT^C0XUTIL(ZY) ; object
|
---|
324 | . . . D ADD(ZGRF,C0XSUB,C0XPRE,C0XOBJ) ; finally. our first real triple
|
---|
325 | . . ; -- this is an else because of the quit above
|
---|
326 | . . S ZX=$G(@ZDOM@(ZJ,"A","rdf:nodeID")) ; fishing for nodeId object
|
---|
327 | . . I ZX'="" D Q ; got one
|
---|
328 | . . . S C0XOBJ=ZX ; we are using the incoming nodeIDs as object/subject
|
---|
329 | . . . ; without change... this could be foolish .. look at it again later
|
---|
330 | . . . D ADD(ZGRF,C0XSUB,C0XPRE,C0XOBJ) ; go for it and add a node
|
---|
331 | . . S C0XOBJ=$G(@ZDOM@(ZJ,"T",1)) ; hopefully an object is here
|
---|
332 | . . I C0XOBJ="" D Q ; not a happy situation
|
---|
333 | . . . W !,"ERROR, NO OBJECT FOUND FOR NODE: ",ZJ
|
---|
334 | . . S C0XOBJ=$$EXT^C0XUTIL(C0XOBJ) ; might be namespaced
|
---|
335 | . . D ADD(ZGRF,C0XSUB,C0XPRE,C0XOBJ) ; go for it and add a node
|
---|
336 | S C0XTRP=$$NOW^XLFDT ; PARSE COMPLETE
|
---|
337 | W !,"TRIPLES COMPLETE AT ",C0XTRP
|
---|
338 | S C0XDIFF=$$FMDIFF^XLFDT(C0XTRP,C0XPRS,2)
|
---|
339 | W !," ELAPSED TIME: ",C0XDIFF," SECONDS"
|
---|
340 | I C0XDIFF'=0 D ;
|
---|
341 | . W !," APPROXIMATELY ",$P(C0XCNT/C0XDIFF,".")," TRIPLES PER SECOND"
|
---|
342 | W !,"INSERTING ",C0XCNT," TRIPLES"
|
---|
343 | I $D(C0XFDA) D ;
|
---|
344 | . I $G(BLKLOAD) D ;
|
---|
345 | . . D BULKLOAD(.C0XFDA)
|
---|
346 | . E D ;
|
---|
347 | . . D UPDIE(.C0XFDA) ; commit the updates to the file
|
---|
348 | ; next, mark the graph as finished
|
---|
349 | S C0XINS=$$NOW^XLFDT ; PARSE COMPLETE
|
---|
350 | W !,"INSERTION COMPLETE AT ",C0XPRS
|
---|
351 | S C0XDIFF=$$FMDIFF^XLFDT(C0XINS,C0XTRP,2)
|
---|
352 | W !," ELAPSED TIME: ",C0XDIFF," SECONDS"
|
---|
353 | I C0XDIFF'=0 W !," APPROXIMATELY ",$P(C0XCNT/C0XDIFF,".")," NODES PER SECOND"
|
---|
354 | S C0XEND=$$NOW^XLFDT
|
---|
355 | W !," ENDED AT: ",C0XEND
|
---|
356 | S C0XDIFF=$$FMDIFF^XLFDT(C0XEND,C0XSTART,2)
|
---|
357 | W !," ELAPSED TIME: ",C0XDIFF," SECONDS"
|
---|
358 | I C0XDIFF'=0 W !," APPROXIMATELY ",$P(C0XCNT/C0XDIFF,".")," TRIPLES PER SECOND"
|
---|
359 | Q
|
---|
360 | ;
|
---|
361 | SHOW(ZN) ;
|
---|
362 | I '$D(C0XJOB) S C0XJOB=$J
|
---|
363 | N ZD
|
---|
364 | S ZD=$NA(^TMP("MXMLDOM",C0XJOB,1,ZN))
|
---|
365 | W ZD,"=",@ZD
|
---|
366 | F S ZD=$Q(@ZD) Q:$QS(ZD,4)'=ZN W !,ZD,"=",@ZD
|
---|
367 | ;ZWR ^TMP("MXMLDOM",C0XJOB,1,ZN,*)
|
---|
368 | Q
|
---|
369 | ;
|
---|
370 | ANONS() ; RETURNS AN ANONOMOUS SUBJECT
|
---|
371 | Q "iDPsDPss"_$$LKY9
|
---|
372 | ;
|
---|
373 | NEWG(NGRAPH,NMETA) ; CREATES A NEW META GRAPH, MARKS IT AS UNFINISHED
|
---|
374 | ; THEN CREATES A NEW GRAPH AND POINTS THE METAGRAPH TO IT
|
---|
375 | ; NGRAPH AND NMETA ARE PASSED BY REFERENCE AND ARE THE RETURN
|
---|
376 | S NGRAPH="G"_$$LKY9
|
---|
377 | S NMETA=NGRAPH_"A"
|
---|
378 | Q
|
---|
379 | ;
|
---|
380 | ADD(ZG,ZS,ZP,ZO,FARY) ; ADD A TRIPLE TO THE TRIPLESTORE. ALL VALUES ARE TEXT
|
---|
381 | ; THE FDA IS SET UP BUT THE FILES ARE NOT UPDATED. CALL UPDIE TO COMPLETE
|
---|
382 | I '$D(FARY) D ;
|
---|
383 | . D INITFARY("C0XFARY")
|
---|
384 | . S FARY="C0XFARY"
|
---|
385 | D USEFARY(FARY)
|
---|
386 | I '$D(C0XCNT) S C0XCNT=0
|
---|
387 | N ZNODE
|
---|
388 | S ZNODE="N"_$$LKY17
|
---|
389 | N ZNARY ; GET READY TO CALL IENOFA
|
---|
390 | S ZNARY("ZG",ZG)=""
|
---|
391 | S ZNARY("ZS",ZS)=""
|
---|
392 | S ZNARY("ZP",ZP)=""
|
---|
393 | S ZNARY("ZO",ZO)=""
|
---|
394 | D IENOFA(.ZIENS,.ZNARY,FARY) ; RESOLVE/ADD STRINGS
|
---|
395 | ;S ZGIEN=$$IENOF(ZG) ; LAYGO TO GET IEN
|
---|
396 | ;S ZSIEN=$$IENOF(ZS)
|
---|
397 | ;S ZPIEN=$$IENOF(ZP)
|
---|
398 | ;S ZOIEN=$$IENOF(ZO)
|
---|
399 | ;I $D(C0XFDA) D UPDIE ; ADD THE STRINGS IF NEEDED
|
---|
400 | I '$D(BATCNT) S BATCNT=0
|
---|
401 | S BATCNT=BATCNT+1
|
---|
402 | S C0XCNT=C0XCNT+1
|
---|
403 | I $G(BLKLOAD)=1 D ; we are using bulk load
|
---|
404 | . S C0XFDA(C0XTFN,BATCNT,.01)=ZNODE
|
---|
405 | . S C0XFDA(C0XTFN,BATCNT,.02)=$O(ZIENS("IEN","ZG",""))
|
---|
406 | . S C0XFDA(C0XTFN,BATCNT,.03)=$O(ZIENS("IEN","ZS",""))
|
---|
407 | . S C0XFDA(C0XTFN,BATCNT,.04)=$O(ZIENS("IEN","ZP",""))
|
---|
408 | . S C0XFDA(C0XTFN,BATCNT,.05)=$O(ZIENS("IEN","ZO",""))
|
---|
409 | E D ;
|
---|
410 | . S C0XFDA(C0XTFN,"?+"_BATCNT_",",.01)=ZNODE
|
---|
411 | . S C0XFDA(C0XTFN,"?+"_BATCNT_",",.02)=$O(ZIENS("IEN","ZG",""))
|
---|
412 | . S C0XFDA(C0XTFN,"?+"_BATCNT_",",.03)=$O(ZIENS("IEN","ZS",""))
|
---|
413 | . S C0XFDA(C0XTFN,"?+"_ATCNT_",",.04)=$O(ZIENS("IEN","ZP",""))
|
---|
414 | . S C0XFDA(C0XTFN,"?+"_BATCNT_",",.05)=$O(ZIENS("IEN","ZO",""))
|
---|
415 | I '$D(BATMAX) S BATMAX=10000
|
---|
416 | I BATCNT=BATMAX D ; BATCH IS DONE
|
---|
417 | . I $G(BLKLOAD) D ; bulk load
|
---|
418 | . . D BULKLOAD(.C0XFDA) ; bulk load the batch
|
---|
419 | . E D ; no bulk load
|
---|
420 | . . D UPDIE(.C0XFDA)
|
---|
421 | . K C0XFDA
|
---|
422 | . S BATCNT=0 ; RESET COUNTER
|
---|
423 | ; REMEMBER TO CALL UPDIE WHEN YOU'RE DONE
|
---|
424 | Q
|
---|
425 | ;
|
---|
426 | LKY9() ;EXTRINIC THAT RETURNS A RANDOM 9 DIGIT NUMBER. USED FOR GENERATING
|
---|
427 | ; UNIQUE NODE AND GRAPH NAMES
|
---|
428 | N ZN,ZI
|
---|
429 | S ZN=""
|
---|
430 | F ZI=1:1:9 D ;
|
---|
431 | . S ZN=ZN_$R(10)
|
---|
432 | Q ZN
|
---|
433 | ;
|
---|
434 | LKY17() ;EXTRINIC THAT RETURNS A RANDOM 9 DIGIT NUMBER. USED FOR GENERATING
|
---|
435 | ; UNIQUE NODE AND GRAPH NAMES
|
---|
436 | N ZN,ZI
|
---|
437 | S ZN=""
|
---|
438 | F ZI=1:1:17 D ;
|
---|
439 | . S ZN=ZN_$R(10)
|
---|
440 | Q ZN
|
---|
441 | ;
|
---|
442 | IENOF(ZSTRING,FARY) ; EXTRINSIC WHICH RETURNS THE IEN OF ZS IN THE STRINGS FILE
|
---|
443 | I '$D(FARY) D ;
|
---|
444 | . D INITFARY("C0XFARY")
|
---|
445 | . S FARY="C0XFARY"
|
---|
446 | N ZIEN
|
---|
447 | I $G(ZSTRING)="" Q "" ; NO STRING
|
---|
448 | S ZIEN=$O(@C0XSN@("B",ZSTRING,""))
|
---|
449 | I ZIEN="" D ;
|
---|
450 | . S C0XFDA2(C0XSFN,"+1,",.01)=ZSTRING
|
---|
451 | . D UPDIE(.C0XFDA2)
|
---|
452 | . S ZIEN=$O(@C0XSN@("B",ZSTRING,""))
|
---|
453 | . K C0XFDA2
|
---|
454 | Q ZIEN
|
---|
455 | ;
|
---|
456 | IENOFA(ZOUTARY,INARY,FARY) ; RESOLVE STRINGS TO IEN IN STRINGS FILE
|
---|
457 | ; OR ADD THEM IF
|
---|
458 | ; MISSING. ZINARY AND ZOUTARY ARE PASSED BY REFERENCE
|
---|
459 | ; ZINARY LOOKS LIKE ZINARY("VAR","VAL")=""
|
---|
460 | ; RETURNS IN ZOUTARY OF THE FORM ZOUTARY("IEN","VAR",IEN)=""
|
---|
461 | I '$D(FARY) D ;
|
---|
462 | . D INITFARY("C0XFARY")
|
---|
463 | . S FARY="C0XFARY"
|
---|
464 | K ZOUTARY ; START WITH CLEAN RESULTS
|
---|
465 | K C0XFDA2 ; USE A SEPARATE FDA FOR THIS
|
---|
466 | I '$D(C0XVOC) D VOCINIT^C0XUTIL
|
---|
467 | N ZINARY
|
---|
468 | N ZI S ZI=""
|
---|
469 | F S ZI=$O(INARY(ZI)) Q:ZI="" D ;
|
---|
470 | . N ZK
|
---|
471 | . S ZK=$O(INARY(ZI,""))
|
---|
472 | . S ZINARY($$EXT^C0XUTIL(ZI),$$EXT^C0XUTIL(ZK))=""
|
---|
473 | N ZV,ZIEN
|
---|
474 | N ZCNT S ZCNT=0
|
---|
475 | F S ZI=$O(ZINARY(ZI)) Q:ZI="" D ; LOOK FOR MISSING STRINGS
|
---|
476 | . S ZV=$O(ZINARY(ZI,""))
|
---|
477 | . I ZV["^" S ZV=$TR(ZV,"^","|")
|
---|
478 | . I $O(@C0XSN@("B",ZV,""))="" D ;
|
---|
479 | . . S ZCNT=ZCNT+1
|
---|
480 | . . S C0XFDA2(C0XSFN,"+"_ZCNT_",",.01)=ZV
|
---|
481 | I $D(C0XFDA2) D ;
|
---|
482 | . D UPDIE(.C0XFDA2) ; ADD MISSING STRINGS
|
---|
483 | . K C0XFDA2 ; CLEAN UP
|
---|
484 | F S ZI=$O(ZINARY(ZI)) Q:ZI="" D ; NOT GET ALL IENS
|
---|
485 | . S ZV=$O(ZINARY(ZI,""))
|
---|
486 | . I ZV["^" S ZV=$TR(ZV,"^","|")
|
---|
487 | . S ZIEN=$O(@C0XSN@("B",ZV,"")) ; THEY SHOULD BE THERE NOW
|
---|
488 | . I ZIEN="" D ;
|
---|
489 | . . W !,"ERROR ADDING STRING: ",ZV
|
---|
490 | . . B
|
---|
491 | . S ZOUTARY("IEN",ZI,ZIEN)=""
|
---|
492 | Q
|
---|
493 | ;
|
---|
494 | BULKLOAD(ZBFDA) ; BULK LOADER FOR LOADING TRIPLES INTO FILE 172.101
|
---|
495 | ; USING GLOBAL SETS INSTEAD OF UPDATE^DIE
|
---|
496 | ; QUITS IF FILE IS NOT 172.101
|
---|
497 | ; EXPECTS AN FDA WITHOUT STRINGS FOR THE IENS, STARTING AT 1
|
---|
498 | ; QUITS IF FIRST ENTRY IS NOT IENS 1
|
---|
499 | ; ASSUMES THAT THE LAST IENS IS THE COUNT OF ENTRIES
|
---|
500 | ; ZBFDA IS PASSED BY REFERENCE
|
---|
501 | ;
|
---|
502 | ; -- reserves a block of iens from file 172.101 by locking the zero node
|
---|
503 | ; -- ^C0X(101,0) and adding the count of entries to piece 2 and 3
|
---|
504 | ; -- then unlocking to minimize the duration of the lock
|
---|
505 | ;
|
---|
506 | W !,"USING BULKLOAD"
|
---|
507 | I '$D(ZBFDA) Q ; EMPTY FDA
|
---|
508 | I $O(ZBFDA(""))'=172.101 Q ; WRONG FILE
|
---|
509 | N ZCNT,ZP3,ZP4
|
---|
510 | ; -- find the number of nodes to insert
|
---|
511 | S ZCNT=$O(ZBFDA(172.101,""),-1)
|
---|
512 | I ZCNT="" D Q ;
|
---|
513 | . W !,"ERROR IN BULK LOAD - INVALID NODE COUNT"
|
---|
514 | . B
|
---|
515 | ; -- lock the zero node and reserve a block of iens to insert
|
---|
516 | W !,"LOCKING ZERO NODE"
|
---|
517 | LOCK +^C0X(101,0)
|
---|
518 | S ZP3=$P(^C0X(101,0),U,3)
|
---|
519 | S ZP4=$P(^C0X(101,0),U,4)
|
---|
520 | S $P(^C0X(101,0),U,3)=ZP3+ZCNT+1
|
---|
521 | S $P(^C0X(101,0),U,4)=ZP4+ZCNT+1
|
---|
522 | LOCK -^C0X(101,0)
|
---|
523 | N ZI,ZN,ZG,ZS,ZP,ZO,ZIEN,ZBASE
|
---|
524 | S ZBASE=ZP3 ; the last ien in the file
|
---|
525 | W !,"ZERO NODE UNLOCKED, IENS RESERVED=",ZCNT
|
---|
526 | W !,$$NOW^XLFDT
|
---|
527 | S ZI=""
|
---|
528 | F S ZI=$O(ZBFDA(172.101,ZI)) Q:ZI="" D ;
|
---|
529 | . S ZN=$G(ZBFDA(172.101,ZI,.01)) ; node name
|
---|
530 | . I ZN="" D BLKERR Q ;
|
---|
531 | . S ZG=$G(ZBFDA(172.101,ZI,.02)) ; graph pointer
|
---|
532 | . I ZG="" D BLKERR Q ;
|
---|
533 | . S ZS=$G(ZBFDA(172.101,ZI,.03)) ; subject pointer
|
---|
534 | . I ZS="" D BLKERR Q ;
|
---|
535 | . S ZP=$G(ZBFDA(172.101,ZI,.04)) ; predicate pointer
|
---|
536 | . I ZP="" D BLKERR Q ;
|
---|
537 | . S ZO=$G(ZBFDA(172.101,ZI,.05)) ; object pointer
|
---|
538 | . I ZO="" D BLKERR Q ;
|
---|
539 | . S ZIEN=ZI+ZBASE ; the new ien
|
---|
540 | . S ^C0X(101,ZIEN,0)=ZN_U_ZG_U_ZS_U_ZP_U_ZO ; set the zero node
|
---|
541 | . D INDEX(ZIEN,ZN,ZG,ZS,ZP,ZO)
|
---|
542 | Q
|
---|
543 | ;
|
---|
544 | INDEX(ZIEN,ZN,ZG,ZS,ZP,ZO) ; HARD SET THE INDEX FOR ONE ENTRY
|
---|
545 | S ^C0X(101,"B",ZN,ZIEN)="" ; the B index
|
---|
546 | S ^C0X(101,"G",ZG,ZIEN)="" ; the G for Graph index
|
---|
547 | S ^C0X(101,"SPO",ZS,ZP,ZO,ZIEN)=""
|
---|
548 | S ^C0X(101,"SOP",ZS,ZO,ZP,ZIEN)=""
|
---|
549 | S ^C0X(101,"OPS",ZO,ZP,ZS,ZIEN)=""
|
---|
550 | S ^C0X(101,"OSP",ZO,ZS,ZP,ZIEN)=""
|
---|
551 | S ^C0X(101,"PSO",ZP,ZS,ZO,ZIEN)=""
|
---|
552 | S ^C0X(101,"POS",ZP,ZO,ZS,ZIEN)=""
|
---|
553 | S ^C0X(101,"GOPS",ZG,ZO,ZP,ZS,ZIEN)=""
|
---|
554 | S ^C0X(101,"GOSP",ZG,ZO,ZS,ZP,ZIEN)=""
|
---|
555 | S ^C0X(101,"GPSO",ZG,ZP,ZS,ZO,ZIEN)=""
|
---|
556 | S ^C0X(101,"GPOS",ZG,ZP,ZO,ZS,ZIEN)=""
|
---|
557 | S ^C0X(101,"GSPO",ZG,ZS,ZP,ZO,ZIEN)=""
|
---|
558 | S ^C0X(101,"GSOP",ZG,ZS,ZO,ZP,ZIEN)=""
|
---|
559 | Q
|
---|
560 | ;
|
---|
561 | REINDEX ; REINDEX THE ^C0X(101, TRIPLE STORE
|
---|
562 | K ^C0X(101,"B")
|
---|
563 | K ^C0X(101,"G")
|
---|
564 | K ^C0X(101,"SPO")
|
---|
565 | K ^C0X(101,"SOP")
|
---|
566 | K ^C0X(101,"OPS")
|
---|
567 | K ^C0X(101,"OSP")
|
---|
568 | K ^C0X(101,"PSO")
|
---|
569 | K ^C0X(101,"POS")
|
---|
570 | K ^C0X(101,"GOPS")
|
---|
571 | K ^C0X(101,"GOSP")
|
---|
572 | K ^C0X(101,"GPSO")
|
---|
573 | K ^C0X(101,"GPOS")
|
---|
574 | K ^C0X(101,"GSPO")
|
---|
575 | K ^C0X(101,"GSOP")
|
---|
576 | N ZIEN,ZZ
|
---|
577 | S ZIEN=0
|
---|
578 | F S ZIEN=$O(^C0X(101,ZIEN)) Q:+ZIEN=0 D ; FOR EACH NODE
|
---|
579 | . S ZZ=$G(^C0X(101,ZIEN,0))
|
---|
580 | . I ZZ="" D Q ;
|
---|
581 | . . W !,"ERROR REINDEXING NODE ",ZI
|
---|
582 | . S ZN=$P(ZZ,"^",1)
|
---|
583 | . S ZG=$P(ZZ,"^",2)
|
---|
584 | . S ZS=$P(ZZ,"^",3)
|
---|
585 | . S ZP=$P(ZZ,"^",4)
|
---|
586 | . S ZO=$P(ZZ,"^",5)
|
---|
587 | . D INDEX(ZIEN,ZN,ZG,ZS,ZP,ZO)
|
---|
588 | Q
|
---|
589 | ;
|
---|
590 | BLKERR ;
|
---|
591 | W !,"ERROR IN BULK LOAD"
|
---|
592 | S C0XERR="ERROR IN BULK LOAD"
|
---|
593 | S C0XLOC=ZBFDA(ZI)
|
---|
594 | D ^%ZTER ; report the error
|
---|
595 | B
|
---|
596 | Q
|
---|
597 | ;
|
---|
598 | DELGRAPH(ZGRF,FARY) ; delete a graph from the triplestore
|
---|
599 | ; (doesn't delete strings)
|
---|
600 | ;
|
---|
601 | I '$D(FARY) D ;
|
---|
602 | . D INITFARY("C0XFARY")
|
---|
603 | . S FARY="C0XFARY"
|
---|
604 | D USEFARY(FARY)
|
---|
605 | N ZGRAPH
|
---|
606 | D TING(.ZGRAPH,ZGRF,FARY)
|
---|
607 | I '$D(ZGRAPH) D Q ;
|
---|
608 | . W !,"NO TRIPLES IN GRAPH"
|
---|
609 | K C0XFDA
|
---|
610 | N ZI S ZI=""
|
---|
611 | F S ZI=$O(ZGRAPH(ZI)) Q:ZI="" D ;
|
---|
612 | . S C0XFDA(C0XTFN,ZI_",",.01)="@"
|
---|
613 | D UPDIE(.C0XFDA)
|
---|
614 | Q
|
---|
615 | ;
|
---|
616 | TING(ZRTN,ZGRF,FARY) ; return the iens for graph ZGRF
|
---|
617 | ; ZRTN is passed by reference
|
---|
618 | I '$D(FARY) D ;
|
---|
619 | . D INITFARY("C0XFARY")
|
---|
620 | . S FARY="C0XFARY"
|
---|
621 | D USEFARY(FARY)
|
---|
622 | K ZRTN
|
---|
623 | N ZI,ZG S ZI=""
|
---|
624 | S ZG=$$IENOF(ZGRF)
|
---|
625 | I ZG="" D Q ;
|
---|
626 | . W !,"ERROR GRAPH NOT FOUND"
|
---|
627 | I '$D(@C0XTN@("G",ZG)) Q ;
|
---|
628 | F S ZI=$O(@C0XTN@("G",ZG,ZI)) Q:ZI="" D ;
|
---|
629 | . S ZRTN(ZI)=""
|
---|
630 | Q
|
---|
631 | ;
|
---|
632 | SWUPDIE(ZFDA) ; SWITCH BETWEEN UPDIE AND BULKLOAD
|
---|
633 | . I $G(BLKLOAD) D ; bulk load
|
---|
634 | . . D BULKLOAD(.ZFDA) ; bulk load the batch
|
---|
635 | . E D ; no bulk load
|
---|
636 | . . D UPDIE(.ZFDA)
|
---|
637 | . K ZFDA
|
---|
638 | Q
|
---|
639 | ;
|
---|
640 | UPDIE(ZFDA) ; INTERNAL ROUTINE TO CALL UPDATE^DIE AND CHECK FOR ERRORS
|
---|
641 | ; ZFDA IS PASSED BY REFERENCE
|
---|
642 | ;ZWR ZFDA
|
---|
643 | ;B
|
---|
644 | K ZERR
|
---|
645 | D CLEAN^DILF
|
---|
646 | D UPDATE^DIE("","ZFDA","","ZERR")
|
---|
647 | I $D(ZERR) S ZZERR=ZZERR ; ZZERR DOESN'T EXIST,
|
---|
648 | ; INVOKE THE ERROR TRAP IF TASKED
|
---|
649 | ;. W "ERROR",!
|
---|
650 | ;. ZWR ZERR
|
---|
651 | ;. B
|
---|
652 | K ZFDA
|
---|
653 | Q
|
---|
654 | ;
|
---|