Wojciech Jaworski / ENIAM

Commit 94377238f7e85dc691625ff9e2d93f1e780d80b0

Authored by Wojciech Jaworski
1 parent 82a1d016

Testy i poprawki

Inline Side-by-side
Showing 30 changed files with 140 additions and 49 deletions
LCGlexicon/ENIAM_LCGlexicon.ml
  View file @9437723
... ... @@ -165,7 +165,7 @@ let make_rules x_flag filename =
165 165 dict_of_grammar lexicon
166 166  
167 167 let find_rules rules cats =
168   - let lex_rules,rules = try StringMap.find rules cats.pos with Not_found -> failwith ("find_rules: unable to find rules for category " ^ cats.pos) in
  168 + let lex_rules,rules = try StringMap.find rules cats.pos with Not_found -> failwith ("find_rules: unable to find rules for category '" ^ cats.pos ^ "' lemma='" ^ cats.lemma ^ "'") in
169 169 (* Printf.printf "find_rules: %s %s |rules|=%d\n" cats.lemma cats.pos (Xlist.size rules); *)
170 170 let rules = try StringMap.find lex_rules cats.lemma @ rules with Not_found -> rules in
171 171 Xlist.fold rules [] (fun rules (selectors,syntax,semantics) ->
... ...
LCGlexicon/ENIAM_LCGlexiconParser.ml
  View file @9437723
... ... @@ -95,6 +95,7 @@ let match_relation = function
95 95  
96 96 let rec split_mid i0 rev = function
97 97 [i,s] -> List.rev ((i,s) :: rev)
  98 + | (i1,s) :: (i2,"|") :: (i3,"|") :: l -> raise (ParseError("split_mid", "duplicated delimeter found", i2))
98 99 | (i1,s) :: (i2,"|") :: l -> split_mid i2 ((i1,s) :: rev) l
99 100 | [] -> raise (ParseError("split_mid", "empty", i0))
100 101 | (i,s) :: l -> raise (ParseError("split_mid", "delimiter not found: " ^ String.concat " " (s :: Xlist.map l snd), i))
... ...
LCGlexicon/ENIAMcategoriesPL.ml
  View file @9437723
... ... @@ -400,8 +400,8 @@ let clarify_categories proper cat proj = function
400 400 else [{empty_cats with lemma=lemma; pos="qub"; pos2="qub"}]
401 401 | lemma,"comp",[] -> [{empty_cats with lemma=lemma; pos="comp"; pos2="comp"}]
402 402 | lemma,"conj",[] -> [{empty_cats with lemma=lemma; pos="conj"; pos2="conj"}]
403   - | lemma,"interj",[] -> [{empty_cats with lemma=lemma; pos="interj"; pos2="interj"}]
404   - | lemma,"sinterj",[] -> [{empty_cats with lemma=lemma; pos="sinterj"; pos2="sinterj"}]
  403 + | lemma,"interj",[] -> [{empty_cats with lemma=lemma; pos="interj"; pos2="interj"; cat=cat; proj=proj}]
  404 + | lemma,"sinterj",[] -> [{empty_cats with lemma=lemma; pos="sinterj"; pos2="sinterj"; (*cat=cat; proj=proj*)}]
405 405 | lemma,"burk",[] -> [{empty_cats with lemma=lemma; pos="burk"; pos2="burk"}]
406 406 | ",","interp",[] -> [{empty_cats with lemma=","; pos="conj"; pos2="conj"}]
407 407 | lemma,"interp",[] -> [{empty_cats with lemma=lemma; pos="interp"; pos2="interp"}]
... ... @@ -617,7 +617,7 @@ let pos_categories = Xlist.fold [
617 617 "part",[Lemma;];
618 618 "comp",[Lemma;];(* ctype *)
619 619 "conj",[Lemma;];(* ctype *)
620   - "interj",[Lemma;];
  620 + "interj",[Lemma;Cat;Proj;];
621 621 "sinterj",[Lemma;];
622 622 "burk",[Lemma;];
623 623 "interp",[Lemma;];
... ...
LCGlexicon/resources/subst_container.dat
  View file @9437723
... ... @@ -488,3 +488,4 @@ mrowie
488 488 rodzaj
489 489 rozdział
490 490 gmach
  491 +zakres
... ...
LCGparser/ENIAM_LCG_XMLof.ml
  View file @9437723
... ... @@ -40,7 +40,7 @@ let rec linear_term = function
40 40 | App(s,t) -> Xml.Element("App",[],[linear_term s;linear_term t])
41 41 | Dot -> Xml.Element("Dot",[],[])
42 42 | Val s -> Xml.Element("Val",[],[Xml.PCData s])
43   - | SetAttr(e,s,t) -> Xml.Element("SetAttr",["label",v],[linear_term s;linear_term t])
  43 + | SetAttr(e,s,t) -> Xml.Element("SetAttr",["label",e],[linear_term s;linear_term t])
44 44 | Fix(s,t) -> Xml.Element("Fix",[],[linear_term s;linear_term t])
45 45 | Empty t -> Xml.Element("Empty",[],[linear_term t])
46 46 | Apply t -> Xml.Element("Apply",[],[linear_term t])
... ... @@ -59,4 +59,3 @@ let linear_term_array a =
59 59 let l = Int.fold 0 (Array.length a - 1) [] (fun l i ->
60 60 Xml.Element("element",["index",string_of_int i],[linear_term a.(i)]) :: l) in
61 61 Xml.Element("array",[],List.rev l)
62   -
... ...
LCGparser/makefile
  View file @9437723
... ... @@ -6,21 +6,21 @@ OCAMLFLAGS=$(INCLUDES) -g
6 6 OCAMLOPTFLAGS=$(INCLUDES) unix.cmxa xml-light.cmxa str.cmxa nums.cmxa zip.cmxa bz2.cmxa xlib.cmxa eniam-lcg-parser.cmxa
7 7 INSTALLDIR=`ocamlc -where`/eniam
8 8  
9   -SOURCES= ENIAM_LCGtypes.ml ENIAM_LCGstringOf.ml ENIAM_LCGrules.ml ENIAM_LCGrenderer.ml ENIAM_LCGchart.ml ENIAM_LCGlatexOf.ml ENIAM_LCGreductions.ml ENIAM_LCGgraphOf.ml
  9 +SOURCES= ENIAM_LCGtypes.ml ENIAM_LCGstringOf.ml ENIAM_LCGrules.ml ENIAM_LCGrenderer.ml ENIAM_LCGchart.ml ENIAM_LCGlatexOf.ml ENIAM_LCGreductions.ml ENIAM_LCGgraphOf.ml ENIAM_LCG_XMLof.ml
10 10  
11 11 all: eniam-lcg-parser.cma eniam-lcg-parser.cmxa
12 12  
13 13 install: all
14 14 mkdir -p $(INSTALLDIR)
15 15 cp eniam-lcg-parser.cmxa eniam-lcg-parser.a eniam-lcg-parser.cma $(INSTALLDIR)
16   - cp ENIAM_LCGtypes.cmi ENIAM_LCGstringOf.cmi ENIAM_LCGrules.cmi ENIAM_LCGrenderer.cmi ENIAM_LCGchart.cmi ENIAM_LCGlatexOf.cmi ENIAM_LCGreductions.cmi ENIAM_LCGgraphOf.cmi $(INSTALLDIR)
17   - cp ENIAM_LCGtypes.cmx ENIAM_LCGstringOf.cmx ENIAM_LCGrules.cmx ENIAM_LCGrenderer.cmx ENIAM_LCGchart.cmx ENIAM_LCGlatexOf.cmx ENIAM_LCGreductions.cmx ENIAM_LCGgraphOf.cmx $(INSTALLDIR)
  16 + cp ENIAM_LCGtypes.cmi ENIAM_LCGstringOf.cmi ENIAM_LCGrules.cmi ENIAM_LCGrenderer.cmi ENIAM_LCGchart.cmi ENIAM_LCGlatexOf.cmi ENIAM_LCGreductions.cmi ENIAM_LCGgraphOf.cmi ENIAM_LCG_XMLof.cmi $(INSTALLDIR)
  17 + cp ENIAM_LCGtypes.cmx ENIAM_LCGstringOf.cmx ENIAM_LCGrules.cmx ENIAM_LCGrenderer.cmx ENIAM_LCGchart.cmx ENIAM_LCGlatexOf.cmx ENIAM_LCGreductions.cmx ENIAM_LCGgraphOf.cmx ENIAM_LCG_XMLof.cmx $(INSTALLDIR)
18 18  
19 19 install-local: all
20 20 mkdir -p $(INSTALLDIR)
21 21 cp eniam-lcg-parser.cmxa eniam-lcg-parser.a eniam-lcg-parser.cma $(INSTALLDIR)
22   - cp ENIAM_LCGtypes.cmi ENIAM_LCGstringOf.cmi ENIAM_LCGrules.cmi ENIAM_LCGrenderer.cmi ENIAM_LCGchart.cmi ENIAM_LCGlatexOf.cmi ENIAM_LCGreductions.cmi ENIAM_LCGgraphOf.cmi $(INSTALLDIR)
23   - cp ENIAM_LCGtypes.cmx ENIAM_LCGstringOf.cmx ENIAM_LCGrules.cmx ENIAM_LCGrenderer.cmx ENIAM_LCGchart.cmx ENIAM_LCGlatexOf.cmx ENIAM_LCGreductions.cmx ENIAM_LCGgraphOf.cmx $(INSTALLDIR)
  22 + cp ENIAM_LCGtypes.cmi ENIAM_LCGstringOf.cmi ENIAM_LCGrules.cmi ENIAM_LCGrenderer.cmi ENIAM_LCGchart.cmi ENIAM_LCGlatexOf.cmi ENIAM_LCGreductions.cmi ENIAM_LCGgraphOf.cmi ENIAM_LCG_XMLof.cmi $(INSTALLDIR)
  23 + cp ENIAM_LCGtypes.cmx ENIAM_LCGstringOf.cmx ENIAM_LCGrules.cmx ENIAM_LCGrenderer.cmx ENIAM_LCGchart.cmx ENIAM_LCGlatexOf.cmx ENIAM_LCGreductions.cmx ENIAM_LCGgraphOf.cmx ENIAM_LCG_XMLof.cmx $(INSTALLDIR)
24 24  
25 25 eniam-lcg-parser.cma: $(SOURCES)
26 26 ocamlc -linkall -a -o eniam-lcg-parser.cma $(OCAMLFLAGS) $^
... ...
exec/ENIAMexec.ml
  View file @9437723
... ... @@ -33,6 +33,7 @@ let translate_mode = function
33 33 | ENIAMsubsyntaxTypes.Mate -> Mate
34 34 | ENIAMsubsyntaxTypes.Swigra -> Swigra
35 35 | ENIAMsubsyntaxTypes.POLFIE -> POLFIE
  36 + | ENIAMsubsyntaxTypes.Error -> Error
36 37  
37 38 let rec translate_sentence = function
38 39 ENIAMsubsyntaxTypes.RawSentence s -> RawSentence s
... ... @@ -53,6 +54,7 @@ let rec translate_paragraph = function
53 54 sentence=translate_sentence p.ENIAMsubsyntaxTypes.sentence}))
54 55 | ENIAMsubsyntaxTypes.AltParagraph l -> AltParagraph(Xlist.map l (fun (mode,paragraph) ->
55 56 translate_mode mode, translate_paragraph paragraph))
  57 + | ENIAMsubsyntaxTypes.ErrorParagraph s -> ErrorParagraph s
56 58  
57 59 let rec translate_text = function
58 60 ENIAMsubsyntaxTypes.RawText s -> RawText s
... ... @@ -64,7 +66,9 @@ let rec translate_text = function
64 66 let clarify_categories cats token =
65 67 match token.ENIAMtokenizerTypes.token with
66 68 ENIAMtokenizerTypes.Lemma(lemma,pos,interp) ->
67   - List.flatten (Xlist.map interp (fun interp -> List.flatten (Xlist.map cats (fun (cat,proj) -> ENIAMcategoriesPL.clarify_categories false cat proj (lemma,pos,interp)))))
  69 + List.flatten (Xlist.map interp (fun interp -> List.flatten (Xlist.map cats (fun (cat,proj) ->
  70 + (* Printf.printf "lemma=%s pos=%s cat=%s proj=%s\n%!" lemma pos cat (String.concat "," proj); *)
  71 + ENIAMcategoriesPL.clarify_categories false cat proj (lemma,pos,interp)))))
68 72 | ENIAMtokenizerTypes.Proper(lemma,pos,interp,senses2) ->
69 73 List.flatten (Xlist.map interp (fun interp -> List.flatten (Xlist.map cats (fun (cat,proj) -> ENIAMcategoriesPL.clarify_categories true cat proj (lemma,pos,interp)))))
70 74 | ENIAMtokenizerTypes.Interp lemma ->
... ...
exec/ENIAMexecTypes.ml
  View file @9437723
... ... @@ -78,7 +78,7 @@ type semantic_processing_result = {
78 78 }
79 79 *)
80 80 type mode =
81   - Raw | Struct | CONLL | ENIAM | Mate | Swigra | POLFIE
  81 + Raw | Struct | CONLL | ENIAM | Mate | Swigra | POLFIE | Error
82 82  
83 83 type sentence =
84 84 RawSentence of string
... ... @@ -98,6 +98,7 @@ and paragraph =
98 98 RawParagraph of string
99 99 | StructParagraph of paragraph_record list (* zdania *)
100 100 | AltParagraph of (mode * paragraph) list
  101 + | ErrorParagraph of string
101 102  
102 103 type text =
103 104 RawText of string
... ... @@ -267,6 +268,7 @@ let rec map_paragraph mode f = function
267 268 let l = Xlist.rev_map l (fun (mode,paragraph) ->
268 269 mode, map_paragraph mode f paragraph) in
269 270 AltParagraph(List.rev l)
  271 + | ErrorParagraph s -> ErrorParagraph s
270 272  
271 273 let rec map_text mode f = function
272 274 RawText s -> RawText s
... ... @@ -295,6 +297,7 @@ let rec fold_paragraph mode s f = function
295 297 | AltParagraph l ->
296 298 Xlist.fold l s (fun s (mode,paragraph) ->
297 299 fold_paragraph mode s f paragraph)
  300 + | ErrorParagraph _ -> s
298 301  
299 302 let rec fold_text mode s f = function
300 303 RawText _ -> s
... ...
exec/ENIAMexecXMLof.ml
  View file @9437723
... ... @@ -64,6 +64,7 @@ let rec paragraph m = function
64 64 Xml.Element("StructParagraph",set_mode m,Xlist.map sentences (fun p ->
65 65 Xml.Element("Sentence",["id",p.id;"beg",string_of_int p.beg;"len",string_of_int p.len;"next",string_of_int p.next],[sentence "" p.sentence])))
66 66 | AltParagraph l -> Xml.Element("AltParagraph",set_mode m,Xlist.map l (fun (m,t) -> paragraph (ENIAMvisualization.string_of_mode m) t))
  67 + | ErrorParagraph s -> Xml.Element("ErrorParagraph",set_mode m,[Xml.PCData s])
67 68  
68 69 let rec text m = function
69 70 RawText s -> Xml.Element("RawText",set_mode m,[Xml.PCData s])
... ...
exec/ENIAMselectSent.ml
  View file @9437723
... ... @@ -69,6 +69,7 @@ let rec select_sentence_modes_paragraph = function
69 69 let l = Xlist.rev_map l (fun (mode,paragraph) ->
70 70 mode, select_sentence_modes_paragraph paragraph) in
71 71 AltParagraph(List.rev l)
  72 + | ErrorParagraph s -> ErrorParagraph s
72 73  
73 74 let rec select_sentence_modes_text = function
74 75 RawText s -> RawText s
... ... @@ -148,6 +149,7 @@ let rec select_sentences_paragraph mode = function
148 149 let l = Xlist.rev_map l (fun (mode,paragraph) ->
149 150 mode, select_sentences_paragraph mode paragraph) in
150 151 AltParagraph(List.rev l)
  152 + | ErrorParagraph s -> ErrorParagraph s
151 153  
152 154 let rec select_sentences_text mode = function
153 155 RawText s -> RawText s
... ...
exec/ENIAMvisualization.ml
  View file @9437723
... ... @@ -667,6 +667,7 @@ let string_of_mode = function
667 667 | Mate -> "Mate"
668 668 | Swigra -> "Swigra"
669 669 | POLFIE -> "POLFIE"
  670 + | Error -> "Error"
670 671 (*
671 672 (*let rec string_of_sentence = function
672 673 RawSentence s -> sprintf "RawSentence(%s)" s
... ... @@ -1065,9 +1066,27 @@ let html_of_eniam_sentence path file_prefix img verbosity tokens (result : eniam
1065 1066 sprintf "<BR><IMG SRC=\"%s_11_semantic_graph.png\">\n" file_prefix) ^
1066 1067 ""
1067 1068 | SemNotValidated ->
  1069 + if verbosity < 2 then () else (
  1070 + ENIAM_LCGgraphOf.print_dependency_tree path (file_prefix ^ "_6b_dependency_tree") result.dependency_tree6b;
  1071 + ENIAM_LCGgraphOf.print_dependency_tree path (file_prefix ^ "_9_dependency_tree") result.dependency_tree9;
  1072 + ENIAM_LCGlatexOf.print_dependency_tree path (file_prefix ^ "_6b_dependency_tree") "a3" result.dependency_tree6b;
  1073 + ENIAM_LCGlatexOf.print_dependency_tree path (file_prefix ^ "_7_dependency_tree") "a2" result.dependency_tree7;
  1074 + ENIAM_LCGlatexOf.print_references path (file_prefix ^ "_8_dependency_tree") "a3" result.dependency_tree8;
  1075 + ENIAM_LCGlatexOf.print_dependency_tree path (file_prefix ^ "_9_dependency_tree") "a3" result.dependency_tree9;
  1076 + ENIAMsemLatexOf.print_semantic_graph path (file_prefix ^ "_10_semantic_graph") "a3" result.semantic_graph10;
  1077 + ENIAMsemGraphOf.print_semantic_graph2 path (file_prefix ^ "_11_semantic_graph") "" result.semantic_graph11);
1068 1078 if verbosity = 0 then () else (
1069 1079 ENIAMsemGraphOf.print_semantic_graph2 path (file_prefix ^ "_12_semantic_graph") "" result.semantic_graph12);
1070 1080 sprintf "<font color=\"red\">sem_not_validated</font>: %s paths_size=%d chart_size=%d dependency_tree_size=%d\n" result.msg result.paths_size result.chart_size result.dependency_tree_size ^
  1081 + (if verbosity < 2 then "" else
  1082 + sprintf "<BR><A HREF=\"%s_6b_dependency_tree.pdf\">Dependency Tree References 6b</A>\n" file_prefix ^
  1083 + sprintf "<BR><A HREF=\"%s_7_dependency_tree.pdf\">Dependency Tree References 7</A>\n" file_prefix ^
  1084 + sprintf "<BR><A HREF=\"%s_8_dependency_tree.pdf\">Dependency Tree References 8</A>\n" file_prefix ^
  1085 + sprintf "<BR><A HREF=\"%s_9_dependency_tree.pdf\">Dependency Tree References 9</A>\n" file_prefix ^
  1086 + sprintf "<BR><IMG SRC=\"%s_6b_dependency_tree.png\">\n" file_prefix ^
  1087 + sprintf "<BR><IMG SRC=\"%s_9_dependency_tree.png\">\n" file_prefix ^
  1088 + sprintf "<BR><A HREF=\"%s_10_semantic_graph.pdf\">Semantic Graph References 10</A>\n" file_prefix ^
  1089 + sprintf "<BR><IMG SRC=\"%s_11_semantic_graph.png\">\n" file_prefix) ^
1071 1090 (if verbosity = 0 then "" else
1072 1091 sprintf "<BR><IMG SRC=\"%s_12_semantic_graph.png\">\n" file_prefix) ^
1073 1092 ""
... ... @@ -1174,6 +1193,7 @@ let file_prefix_of_mode = function
1174 1193 | Mate -> "M"
1175 1194 | Swigra -> "S"
1176 1195 | POLFIE -> "P"
  1196 + | Error -> "Er"
1177 1197  
1178 1198 let rec html_of_sentence path file_prefix mode img verbosity tokens = function
1179 1199 RawSentence s -> escape_html s
... ... @@ -1204,6 +1224,7 @@ let rec html_of_paragraph path mode img verbosity tokens = function
1204 1224 String.concat "\n" (Xlist.map l (fun (mode,paragraph) ->
1205 1225 sprintf "<tr><td>%s</td><td>%s</td></tr>" (string_of_mode mode) (html_of_paragraph path mode img verbosity tokens paragraph))) ^
1206 1226 "</table>"
  1227 + | ErrorParagraph s -> sprintf "<font color=\"red\">subsyntax_error</font>: %s\n" (escape_html s)
1207 1228  
1208 1229 let rec html_of_text path mode img verbosity tokens = function
1209 1230 RawText s -> escape_html s
... ... @@ -1236,6 +1257,7 @@ let rec find_prev_next_paragraph rev = function
1236 1257 | StructParagraph sentences ->
1237 1258 Xlist.fold sentences rev (fun rev p -> find_prev_next_sentence p.id p.file_prefix rev p.sentence)
1238 1259 | AltParagraph l -> Xlist.fold l rev (fun rev (mode,paragraph) -> find_prev_next_paragraph rev paragraph)
  1260 + | ErrorParagraph s -> rev
1239 1261  
1240 1262 let rec make_prev_next_map map prev = function
1241 1263 [x] -> StringMap.add map x (prev,"")
... ... @@ -1295,6 +1317,7 @@ let rec print_main_result_paragraph cg_bin_path path id tokens prev_next_map = f
1295 1317 | StructParagraph sentences ->
1296 1318 Xlist.iter sentences (fun p -> print_main_result_sentence cg_bin_path path id p.file_prefix tokens p.id prev_next_map p.sentence)
1297 1319 | AltParagraph l -> Xlist.iter l (fun (mode,paragraph) -> print_main_result_paragraph cg_bin_path path id tokens prev_next_map paragraph)
  1320 + | ErrorParagraph s -> print_not_parsed_main_result cg_bin_path path id "Er" s 0 prev_next_map
1298 1321  
1299 1322 let rec print_main_result_text cg_bin_path path id tokens = function
1300 1323 RawText s -> ()
... ... @@ -1354,6 +1377,7 @@ let rec print_main_result_first_page_paragraph cg_bin_path path id tokens prev_n
1354 1377 let p = List.hd sentences in
1355 1378 print_main_result_first_page_sentence cg_bin_path path id p.file_prefix tokens p.id prev_next_map p.sentence
1356 1379 | AltParagraph l -> Xlist.iter l (fun (mode,paragraph) -> print_main_result_first_page_paragraph cg_bin_path path id tokens prev_next_map paragraph)
  1380 + | ErrorParagraph s -> print_not_parsed_main_result cg_bin_path path id "Er" s 0 prev_next_map
1357 1381  
1358 1382 let rec print_main_result_first_page_text cg_bin_path path id tokens = function
1359 1383 RawText s -> ()
... ...
integration/ENIAMpreIntegration.ml
  View file @9437723
... ... @@ -331,6 +331,7 @@ let rec parse_paragraph mode tokens = function
331 331 let l = Xlist.rev_map l (fun (mode,paragraph) ->
332 332 mode, parse_paragraph mode tokens paragraph) in
333 333 AltParagraph(List.rev l)
  334 + | ErrorParagraph s -> ErrorParagraph s
334 335  
335 336 let rec parse_text mode tokens = function
336 337 RawText s -> RawText s
... ...
lexSemantics/ENIAMlexSemantics.ml
  View file @9437723
... ... @@ -124,6 +124,7 @@ let rec split_tokens_into_groups_paragraph a = function
124 124 Xlist.iter sentences (fun p -> split_tokens_into_groups_sentence a p.sentence)
125 125 | AltParagraph l -> Xlist.iter l (fun (mode,paragraph) ->
126 126 split_tokens_into_groups_paragraph a paragraph)
  127 + | ErrorParagraph s -> ()
127 128  
128 129 let rec split_tokens_into_groups_text a = function
129 130 RawText s -> ()
... ... @@ -461,7 +462,7 @@ let rec create_tokens_for_artificial_nodes_rec tokens lex_sems = function
461 462 let id = ExtArray.add tokens empty_token_env in
462 463 let lex_sem = {empty_lex_sem with frames=[{empty_frame with meanings=[t.lemma, [t.lemma,0], unknown_meaning_weight]}]} in
463 464 let id2 = ExtArray.add lex_sems lex_sem in
464   - if id <>id2 then failwith "create_tokens_for_artificial_nodes_rec" else
  465 + if id <> id2 then failwith "create_tokens_for_artificial_nodes_rec: tokens inconsistent with lex_sems" else
465 466 let t = if t.symbol = Dot then
466 467 {t with symbol = match t.pos with
467 468 "<root>" -> Tuple[Val "<root>"]
... ...
lexSemantics/ENIAMvalence.ml
  View file @9437723
... ... @@ -235,6 +235,12 @@ let transform_qub_pos lemma = function
235 235 | QUB as morf -> [morf]
236 236 | pos -> failwith ("transform_qub_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
237 237  
  238 +let transform_interj_phrase lemma = function
  239 + | phrase -> failwith ("transform_interj_phrase: " ^ lemma ^ " " ^ ENIAMwalStringOf.phrase phrase)
  240 +
  241 +let transform_interj_pos lemma = function
  242 + | pos -> failwith ("transform_interj_pos: " ^ lemma ^ " " ^ ENIAMwalStringOf.pos pos)
  243 +
238 244 let transform_siebie_phrase lemma = function
239 245 | phrase -> failwith ("transform_siebie_phrase: " ^ lemma ^ " " ^ ENIAMwalStringOf.phrase phrase)
240 246  
... ... @@ -363,31 +369,31 @@ let transform_preps morf =
363 369 | SimpleLexArg(lex,PREP c) -> if is_compar lex then SimpleLexArg(lex,COMPAR c) else SimpleLexArg(lex,PREP c)
364 370 | PrepNP(psem,prep,c) -> if is_compar prep then ComparP(prep,c) else PrepNP(psem,prep,c)
365 371 | PrepAdjP(prep,c) -> if is_compar prep then ComparP(prep,c) else PrepAdjP(prep,c)
366   - | PrepNCP(prep,case,ctype,comp) as morf -> if is_compar prep then failwith "transform_preps" else morf
  372 + | PrepNCP(prep,case,ctype,comp) as morf -> if is_compar prep then failwith "transform_preps 1" else morf
367 373 | morf -> morf in
368 374 match morf with
369 375 | ComparP(prep,Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> ComparP(prep,Case case))
370   - | ComparP _ -> failwith "transform_preps"
  376 + | ComparP _ -> failwith "transform_preps 2"
371 377 | LexArg(id,lex,COMPAR Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> LexArg(id,lex,COMPAR (Case case)))
372 378 | SimpleLexArg(lex,COMPAR Str) -> Xlist.map ["nom";"gen";"dat";"acc";"inst";"postp"] (fun case -> SimpleLexArg(lex,COMPAR (Case case)))
373 379 | LexArg(id,lex,COMPAR (Case _)) as morf -> [morf]
374 380 | SimpleLexArg(lex,COMPAR (Case _)) as morf -> [morf]
375   - | LexArg(id,lex,COMPAR _) -> failwith "transform_preps"
376   - | SimpleLexArg(lex,COMPAR _) -> failwith "transform_preps"
  381 + | LexArg(id,lex,COMPAR _) -> failwith "transform_preps 3"
  382 + | SimpleLexArg(lex,COMPAR _) -> failwith "transform_preps 4"
377 383 | PrepNP(sem,"per",Str) -> [PrepNP(sem,"per",Case "nom");PrepNP(sem,"per",Case "voc")] (* FIXME: voc do poprawienie w leksykonie *)
378 384 | PrepNP(_,_,Case _) as morf -> [morf]
379 385 | PrepAdjP(_,Case _) as morf -> [morf]
380 386 | PrepNCP(_,Case _,_,_) as morf -> [morf]
381   - | PrepNP(_,"_",CaseUndef) as morf -> [morf]
382   - | PrepNP _ -> failwith "transform_preps"
383   - | PrepAdjP _ -> failwith "transform_preps"
384   - | PrepNCP _ -> failwith "transform_preps"
  387 + | PrepNP(_,_,CaseUndef) as morf -> [morf]
  388 + | PrepNP _ as morf -> failwith ("transform_preps 5: " ^ ENIAMwalStringOf.phrase morf)
  389 + | PrepAdjP _ -> failwith "transform_preps 6"
  390 + | PrepNCP _ -> failwith "transform_preps 7"
385 391 | LexArg(id,"w",PREP Str) -> [LexArg(id,"w",PREP (Case "acc"));LexArg(id,"w",PREP (Case "loc"));]
386 392 | SimpleLexArg("w",PREP Str) -> [SimpleLexArg("w",PREP (Case "acc"));SimpleLexArg("w",PREP (Case "loc"))]
387 393 | LexArg(id,lex,PREP (Case _)) as morf -> [morf]
388 394 | SimpleLexArg(lex,PREP (Case _)) as morf -> [morf]
389   - | LexArg(id,lex,PREP _) -> failwith "transform_preps"
390   - | SimpleLexArg(lex,PREP _) -> failwith "transform_preps"
  395 + | LexArg(id,lex,PREP _) -> failwith "transform_preps 8"
  396 + | SimpleLexArg(lex,PREP _) -> failwith "transform_preps 9"
391 397 | morf -> [morf]
392 398  
393 399 let transform_pers_schema lemma negation mood schema =
... ... @@ -488,6 +494,7 @@ let transform_schema pos lemma schema =
488 494 | "comp" -> transform_comp_phrase,transform_comp_pos
489 495 | "qub" -> transform_qub_phrase,transform_qub_pos
490 496 | "siebie" -> transform_siebie_phrase,transform_siebie_pos
  497 + | "interj" -> transform_interj_phrase,transform_interj_pos
491 498 | _ -> failwith "transform_schema"
492 499 in
493 500 Xlist.map schema (fun s ->
... ... @@ -539,6 +546,9 @@ let transform_entry pos lemma negation pred aspect schema =
539 546 if negation <> NegationUndef || aspect <> AspectUndef then failwith ("transform_entry 5");
540 547 Xlist.map ["congr";"rec"] (fun acm ->
541 548 [Acm,Eq,[acm]],transform_num_schema acm schema)) else
  549 + if pos = "interj" then (
  550 + if negation <> NegationUndef || pred <> PredFalse || aspect <> AspectUndef then failwith ("transform_entry 6");
  551 + [[],transform_schema "interj" lemma schema]) else
542 552 List.flatten (Xlist.map (expand_negation negation) (fun negation ->
543 553 let sel = [Negation,Eq,[ENIAMwalStringOf.negation negation]] @ aspect_sel aspect in
544 554 if pos = "fin" || pos = "bedzie" then
... ...
lexSemantics/ENIAMwalReduce.ml
  View file @9437723
... ... @@ -90,6 +90,7 @@ let select_comprep_adjuncts lexemes =
90 90 not (StringSet.is_empty (StringSet.intersection reqs lexemes)) then s :: l else l)
91 91 with Not_found -> l)
92 92  
  93 +(* FIXME: trzeba zanalizować interację tej procedury z Pro w schemacie w wersji z walentym i z semantyką dziedzinową *)
93 94 let set_necessary pos schema =
94 95 Xlist.map schema (fun p ->
95 96 let nec =
... ... @@ -101,6 +102,8 @@ let set_necessary pos schema =
101 102 | _ -> b) then Req else
102 103 if p.gf <> SUBJ && p.cr = [] (*&& p.ce = []*) then Opt else
103 104 if p.gf = SUBJ && pos = "impt" then ProNG else
  105 + if p.gf = SUBJ && pos = "pact" then Opt else
  106 + if p.gf = OBJ && pos = "ppas" then Opt else
104 107 if Xlist.fold p.morfs false (fun b -> function
105 108 NP NomAgr -> true
106 109 | NCP(NomAgr,_,_) -> true
... ...
lexSemantics/ENIAMwalRenderer.ml
  View file @9437723
... ... @@ -183,9 +183,11 @@ let render_phrase_cat cat = function
183 183 | NP CaseUndef -> Tensor[Atom "np"; Atom cat; Top; Top; Top; Top]
184 184 | PrepNP(Psem,"",CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "sem"; Top; Top]
185 185 | PrepNP(Psem,"_",CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "sem"; Top; Top]
  186 + | PrepNP(Psem,prep,CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "sem"; Atom prep; Top]
186 187 | PrepNP(Psem,prep,Case case) -> Tensor[Atom "prepnp"; Atom cat; Atom "sem"; Atom prep; Atom case]
187 188 | PrepNP(Pnosem,"",CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "nosem"; Top; Top]
188 189 | PrepNP(Pnosem,"_",CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "nosem"; Top; Top]
  190 + | PrepNP(Pnosem,prep,CaseUndef) -> Tensor[Atom "prepnp"; Atom cat; Atom "nosem"; Atom prep; Top]
189 191 | PrepNP(Pnosem,prep,Case case) -> Tensor[Atom "prepnp"; Atom cat; Atom "nosem"; Atom prep; Atom case]
190 192 | AdjP(Case case) -> Tensor[Atom "adjp"; Atom cat; Top; Atom case; Top]
191 193 (* | AdjP NomAgr -> Tensor[Atom "adjp"; AVar "number"; Atom "nom"; AVar "gender"]*)
... ...
morphology/resources/alt_supplement.tab
  View file @9437723
... ... @@ -4,4 +4,12 @@ sobie siebie siebie:dat.loc
4 4 sobą siebie siebie:inst
5 5 to to pred
6 6 yay yay interj
  7 +świetnie świetnie interj
  8 +doskonale doskonale interj
  9 +idealnie idealnie interj
  10 +zdecydowanie zdecydowanie interj
  11 +ok ok interj
  12 +super super interj
  13 +dobrze dobrze interj
  14 +dzięki dzięki interj
7 15  
... ...
semantics/ENIAMsemGraph.ml
  View file @9437723
... ... @@ -25,7 +25,7 @@ let empty_concept =
25 25 {c_sense=Dot;c_name=Dot;(* c_variable: string; c_visible_var: bool;*) c_quant=Dot; c_local_quant=true; (*c_modalities: (string * type_term) list;
26 26 c_left_input_pos: int; c_right_input_pos: int;*) c_relations=Dot; c_variable="",""; c_pos=(-1); c_cat=Dot; c_label=""; c_def_label=""}
27 27  
28   -let empty_context = {cx_sense=Dot; cx_contents=Dot; cx_relations=Dot; cx_variable="",""; cx_pos=(-1); cx_cat=Dot}
  28 +let empty_context = {cx_sense=Dot; cx_contents=Dot; cx_relations=Dot; cx_variable="",""; cx_pos=(-1); cx_cat=Dot; cx_label=""; cx_def_label=""}
29 29  
30 30 let rec make_args_list = function
31 31 Tuple l -> List.flatten (Xlist.map l make_args_list)
... ... @@ -154,7 +154,7 @@ let create_normal_concept tokens lex_sems t cat proj =
154 154 | "NEGATION",Val "neg" -> {c with c_quant=Tuple[c.c_quant;Val "nie"]}
155 155 | e,t -> failwith ("create_normal_concept verb: " ^ e)) in
156 156 let id = ExtArray.add tokens ENIAMtokenizerTypes.empty_token_env in
157   - let _ = ExtArray.add lex_sems in
  157 + let _ = ExtArray.add lex_sems ENIAMlexSemanticsTypes.empty_lex_sem in
158 158 let cx = {empty_context with cx_contents=add_proj proj c; cx_variable=string_of_int id,""; cx_pos=c.c_pos; cx_cat=Val "Situation"} in
159 159 make_relation t (Context cx) else
160 160 if t.pos = "adj" || t.pos = "adjc" || t.pos = "adjp" || t.pos = "adja" || t.pos = "pact" || t.pos = "ppas" || t.pos = "apron" || t.pos = "ordnum" || t.pos = "roman-adj" then
... ... @@ -249,12 +249,15 @@ let create_normal_concept tokens lex_sems t cat proj =
249 249 if t.pos = "comp" then
250 250 make_relation t (SetContextName(c.c_sense,RemoveRelation("CORE","",c.c_relations))) else
251 251 if t.pos = "conj" then
252   - let c = {empty_context with cx_sense=t.meaning; cx_contents=t.args; cx_variable=c.c_variable; cx_pos=c.c_pos} in
  252 + let c = {empty_context with cx_sense=t.meaning; cx_contents=t.args; cx_variable=c.c_variable; cx_pos=c.c_pos; cx_cat=c.c_cat; cx_def_label=c.c_def_label; cx_label=c.c_label} in
253 253 let c = Xlist.fold t.attrs c (fun c -> function
254 254 | "NUM",_ -> c
255 255 | "CASE",_ -> c
256 256 | "GEND",_ -> c
257 257 | "PERS",_ -> c
  258 + | "ASPECT",_ -> c
  259 + | "controller",_ -> c
  260 + | "controllee",_ -> c
258 261 | e,t -> failwith ("create_normal_concept conj: " ^ e)) in
259 262 ManageCoordination({t with attrs=[]; args=Dot},Context c) else
260 263 (* if t.pos = "interj" then
... ... @@ -264,7 +267,10 @@ let create_normal_concept tokens lex_sems t cat proj =
264 267 if t.pos = "sinterj" || t.pos = "interj" then
265 268 let c = Xlist.fold t.attrs c (fun c -> function
266 269 | e,t -> failwith ("create_normal_concept sinterj: " ^ e)) in
267   - make_relation t (Concept c) else
  270 + let id = ExtArray.add tokens ENIAMtokenizerTypes.empty_token_env in
  271 + let _ = ExtArray.add lex_sems ENIAMlexSemanticsTypes.empty_lex_sem in
  272 + let cx = {empty_context with cx_contents=add_proj proj c; cx_variable=string_of_int id,""; cx_pos=c.c_pos; cx_cat=Val "Situation"} in
  273 + make_relation t (Context cx) else
268 274 if t.lemma = "<root>" then t.args else
269 275 if t.lemma = "<merge>" then RemoveRelation("null","",t.args) else
270 276 if t.pos = "interp" && t.lemma = "?" && t.args = Dot then SingleRelation(Val "int") else
... ... @@ -511,7 +517,7 @@ let rec reduce_tree = function
511 517 (match reduce_tree t with
512 518 Relation(r,a,t) ->
513 519 if (r = r0 && a = a0) || r0 = "" then t else
514   - Context{empty_context with cx_contents=
  520 + Context{empty_context with cx_cat=Val "Situation"; cx_contents=
515 521 Concept{empty_concept with c_relations=Relation(r,a,t)}; (*cx_variable=string_of_int id,""; cx_pos=c.c_pos*)}
516 522 (* | TripleRelation(r,a,s,t) ->
517 523 Context{empty_context with cx_contents=
... ... @@ -625,6 +631,7 @@ let rec simplify_tree = function
625 631 let l = Xlist.fold l [] (fun l t ->
626 632 match simplify_tree t with
627 633 Dot -> l
  634 + | Tuple l2 -> l2 @ l
628 635 | t -> t :: l) in
629 636 (match l with
630 637 [] -> Dot
... ... @@ -655,7 +662,8 @@ let rec simplify_tree = function
655 662 c_cat = simplify_tree (Variant(e,lt3))}
656 663 | Context c ->
657 664 let lt1,lt2,lt3 = Xlist.fold l ([],[],[]) (fun (lt1,lt2,lt3) -> function
658   - i,Context c2 -> if c.cx_sense = c2.cx_sense then (i,c2.cx_contents) :: lt1, (i,c2.cx_relations) :: lt2, (i,c2.cx_cat) :: lt3 else raise Not_found
  665 + i,Context c2 -> if c.cx_sense = c2.cx_sense && c.cx_label = c2.cx_label &&
  666 + c.cx_def_label = c2.cx_def_label then (i,c2.cx_contents) :: lt1, (i,c2.cx_relations) :: lt2, (i,c2.cx_cat) :: lt3 else raise Not_found
659 667 | _ -> raise Not_found) in
660 668 let e = if e = "" then ENIAM_LCGreductions.get_variant_label () else e in
661 669 Context{c with
... ...
semantics/ENIAMsemGraphOf.ml
  View file @9437723
... ... @@ -225,7 +225,9 @@ let rec print_graph2_rec file edge_rev edge_label edge_style edge_head upper = f
225 225 | Context t ->
226 226 let id = !id_counter in
227 227 incr id_counter;
228   - fprintf file " subgraph cluster%d {\nlabel=\"%s%s\"\n" id
  228 + fprintf file " subgraph cluster%d {\nlabel=\"%s%s%s%s\"\n" id
  229 + (if t.cx_label="" then "" else "?" ^ t.cx_label ^ " ")
  230 + (if t.cx_def_label="" then "" else "*" ^ t.cx_def_label ^ " ")
229 231 (if t.cx_cat=Dot then "" else escape_string (ENIAMsemStringOf.linear_term 0 t.cx_cat ^ " "))
230 232 (if t.cx_sense = Dot then "" else escape_string (ENIAMsemStringOf.linear_term 0 t.cx_sense));
231 233 let iid = print_graph2_rec file false "" "" "" 0 t.cx_contents in
... ...
semantics/ENIAMsemLatexOf.ml
  View file @9437723
... ... @@ -66,7 +66,7 @@ let rec linear_term c = function
66 66 | Context c ->
67 67 "{\\left[\\begin{array}{ll}" ^
68 68 (String.concat "\\\\ " (Xlist.map ([
69   - "SENSE",c.cx_sense;"CAT",c.cx_cat;
  69 + "SENSE",c.cx_sense;"CAT",c.cx_cat;"LABEL",Val c.cx_label;"DEF-LABEL",Val c.cx_def_label;
70 70 "VARIABLE",Val (fst c.cx_variable ^ "_" ^ snd c.cx_variable);"POS",Val (string_of_int c.cx_pos);
71 71 "RELATIONS",c.cx_relations;"CONTENTS",c.cx_contents]) (fun (e,t) ->
72 72 "\\text{" ^ (Xlatex.escape_string e) ^ "} & " ^ (linear_term 0 t)))) ^ "\\end{array}\\right]}"
... ...
semantics/ENIAMsemStringOf.ml
  View file @9437723
... ... @@ -43,7 +43,7 @@ let rec linear_term c = function
43 43 | Context c ->
44 44 "[" ^
45 45 (String.concat "; " (Xlist.map ([
46   - "SENSE",c.cx_sense;"CAT",c.cx_cat;
  46 + "SENSE",c.cx_sense;"CAT",c.cx_cat;"LABEL",Val c.cx_label;"DEF-LABEL",Val c.cx_def_label;
47 47 "VARIABLE",Val (fst c.cx_variable ^ "_" ^ snd c.cx_variable);"POS",Val (string_of_int c.cx_pos);
48 48 "RELATIONS",c.cx_relations;"CONTENTS",c.cx_contents]) (fun (e,t) ->
49 49 e ^ ": " ^ (linear_term 0 t)))) ^ "]"
... ...
semantics/ENIAMsemTypes.ml
  View file @9437723
... ... @@ -57,7 +57,7 @@ and concept =
57 57 c_pos: int; c_cat: linear_term}
58 58  
59 59 and context =
60   - {cx_sense: linear_term; cx_contents: linear_term;
  60 + {cx_sense: linear_term; cx_contents: linear_term; cx_label: string; cx_def_label: string;
61 61 cx_relations: linear_term; cx_variable: (string * string); cx_pos: int; cx_cat: linear_term}
62 62  
63 63 and linear_term =
... ...
semantics/ENIAMsemValence.ml
  View file @9437723
... ... @@ -220,6 +220,19 @@ let string_of_arg arg =
220 220 let string_of_position p =
221 221 (string_of_argdir p.dir) ^ String.concat "+" (StringSet.to_list p.morfs)
222 222  
  223 +let manage_arg p t =
  224 + let t = SetAttr("gf",Val (ENIAMwalStringOf.gf p.gf),t) in
  225 + let t =
  226 + if p.gf = ENIAMwalTypes.SUBJ || p.gf = ENIAMwalTypes.OBJ || p.gf = ENIAMwalTypes.ARG then
  227 + SetAttr("role",p.role,SetAttr("role-attr",p.role_attr,SetAttr("selprefs",p.selprefs,t)))
  228 + else if p.gf = ENIAMwalTypes.CORE then SetAttr("selprefs",p.selprefs,t)
  229 + else if p.gf = ENIAMwalTypes.ADJUNCT || p.gf = ENIAMwalTypes.NOSEM || p.gf = ENIAMwalTypes.CORE then t
  230 + else failwith "manage_arg: ni 2" in
  231 + let t = Xlist.fold p.cr t (fun t cr -> SetAttr("controller",Val cr,t)) in
  232 + let t = Xlist.fold p.ce t (fun t ce -> SetAttr("controllee",Val ce,t)) in
  233 + let t = if p.gf = ENIAMwalTypes.NOSEM then Dot else t in
  234 + t
  235 +
223 236 let rec match_arg_positions lemma arg rev = function
224 237 p :: positions ->
225 238 (* Printf.printf "match_arg_positions 1: arg=%s rev=[%s] positions=%s :: [%s]\n%!" (string_of_arg arg) (String.concat "; " (Xlist.map rev string_of_position)) (string_of_position p) (String.concat "; " (Xlist.map positions string_of_position)); *)
... ... @@ -228,16 +241,13 @@ let rec match_arg_positions lemma arg rev = function
228 241 (match l with
229 242 [] -> (*print_endline "match_arg_positions: not matched";*) match_arg_positions lemma arg (p :: rev) positions
230 243 | [t] ->
231   - let t = SetAttr("gf",Val (ENIAMwalStringOf.gf p.gf),t) in
232   - let t =
233   - if p.gf = ENIAMwalTypes.SUBJ || p.gf = ENIAMwalTypes.OBJ || p.gf = ENIAMwalTypes.ARG then
234   - SetAttr("role",p.role,SetAttr("role-attr",p.role_attr,SetAttr("selprefs",p.selprefs,t)))
235   - else if p.gf = ENIAMwalTypes.CORE then SetAttr("selprefs",p.selprefs,t)
236   - else if p.gf = ENIAMwalTypes.ADJUNCT || p.gf = ENIAMwalTypes.NOSEM || p.gf = ENIAMwalTypes.CORE then t
237   - else failwith "match_arg_positions: ni 2" in
238   - let t = Xlist.fold p.cr t (fun t cr -> SetAttr("controller",Val cr,t)) in
239   - let t = Xlist.fold p.ce t (fun t ce -> SetAttr("controllee",Val ce,t)) in
240   - let t = if p.gf = ENIAMwalTypes.NOSEM then Dot else t in
  244 + let t = manage_arg p t in
  245 + if p.is_multi then (t, rev @ (p :: positions)) :: (match_arg_positions lemma arg (p :: rev) positions)
  246 + else (t, rev @ positions) :: (match_arg_positions lemma arg (p :: rev) positions)
  247 + | [t1;t2] -> (* FIXME: przydałoby się to uogólnić na listę dowolnej długości *)
  248 + let t1 = manage_arg p t1 in
  249 + let t2 = manage_arg p t2 in
  250 + let t = Variant("",["1",t1;"2",t2]) in
241 251 if p.is_multi then (t, rev @ (p :: positions)) :: (match_arg_positions lemma arg (p :: rev) positions)
242 252 else (t, rev @ positions) :: (match_arg_positions lemma arg (p :: rev) positions)
243 253 | _ -> failwith ("match_arg_positions: lemma=" ^ lemma ^ " arg=" ^ string_of_arg arg ^ " position=" ^ string_of_position p))
... ...
semantics/ENIAMsemXMLof.ml
  View file @9437723
... ... @@ -49,8 +49,8 @@ let rec linear_term = function
49 49 Xml.Element("relations",[],[linear_term c.c_relations]);
50 50 Xml.Element("cat",[],[linear_term c.c_cat])])
51 51 | Context c ->
52   - Xml.Element("Context",
53   - ["variable",fst c.cx_variable ^ "_" ^ snd c.cx_variable;"pos",string_of_int c.cx_pos],
  52 + Xml.Element("Context",["label",c.cx_label;"def_label",c.cx_def_label;
  53 + "variable",fst c.cx_variable ^ "_" ^ snd c.cx_variable;"pos",string_of_int c.cx_pos],
54 54 [Xml.Element("sense",[],[linear_term c.cx_sense]);
55 55 Xml.Element("contents",[],[linear_term c.cx_contents]);
56 56 Xml.Element("relations",[],[linear_term c.cx_relations]);
... ...
subsyntax/ENIAMsubsyntax.ml
  View file @9437723
... ... @@ -357,11 +357,14 @@ let parse_text_tokens tokens query =
357 357 let paragraphs = List.rev (Xlist.fold paragraphs [] (fun l -> function "" -> l | s -> s :: l)) in
358 358 let n = if Xlist.size paragraphs = 1 then 0 else 1 in
359 359 let paragraphs,_ = Xlist.fold paragraphs ([],n) (fun (paragraphs,n) paragraph ->
  360 + try
360 361 let paths = parse paragraph in
361 362 (* print_endline "parse_text 1"; *)
362 363 let pid = if n = 0 then "" else string_of_int n ^ "_" in
363 364 let sentences = ENIAMsentences.split_into_sentences pid paragraph tokens paths in
364   - (AltParagraph[Raw,RawParagraph paragraph; Struct,StructParagraph sentences]) :: paragraphs, n+1) in
  365 + (AltParagraph[Raw,RawParagraph paragraph; Struct,StructParagraph sentences]) :: paragraphs, n+1
  366 + with e ->
  367 + (AltParagraph[Raw,RawParagraph paragraph; Error,ErrorParagraph (Printexc.to_string e)]) :: paragraphs, n+1) in
365 368 AltText[Raw,RawText query; Struct,StructText(List.rev paragraphs)], tokens
366 369  
367 370 let parse_text query =
... ...
subsyntax/ENIAMsubsyntaxHTMLof.ml
  View file @9437723
... ... @@ -104,6 +104,7 @@ let rec html_of_paragraph tokens = function
104 104 String.concat "\n" (Xlist.map l (fun (mode,paragraph) ->
105 105 sprintf "<tr><td>%s</td><td>%s</td></tr>" (ENIAMsubsyntaxStringOf.mode mode) (html_of_paragraph tokens paragraph))) ^
106 106 "</table>"
  107 + | ErrorParagraph s -> (*print_endline "ErrorParagraph";*) s
107 108  
108 109 let rec html_of_text tokens = function
109 110 RawText s -> s
... ...
subsyntax/ENIAMsubsyntaxStringOf.ml
  View file @9437723
... ... @@ -28,6 +28,7 @@ let mode = function
28 28 | Mate -> "Mate"
29 29 | Swigra -> "Swigra"
30 30 | POLFIE -> "POLFIE"
  31 + | Error -> "Error"
31 32  
32 33 let token_extarray t =
33 34 String.concat "\n" (List.rev (Int.fold 0 (ExtArray.size t - 1) [] (fun l id ->
... ... @@ -73,6 +74,7 @@ let rec paragraph spaces t = function
73 74 | AltParagraph l ->
74 75 String.concat "\n" (Xlist.map l (fun (m,p) ->
75 76 sprintf "%sAltParagraph mode=%s %s" spaces (mode m) (paragraph "" t p)))
  77 + | ErrorParagraph s -> spaces ^ "ErrorParagraph: " ^ s
76 78  
77 79 let rec text spaces t = function
78 80 RawText s -> spaces ^ "RawText: " ^ s
... ...
subsyntax/ENIAMsubsyntaxTypes.ml
  View file @9437723
... ... @@ -20,7 +20,7 @@
20 20 open ENIAMtokenizerTypes
21 21  
22 22 type mode =
23   - Raw | Struct | CONLL | ENIAM | Mate | Swigra | POLFIE
  23 + Raw | Struct | CONLL | ENIAM | Mate | Swigra | POLFIE | Error
24 24  
25 25 type sentence =
26 26 RawSentence of string
... ... @@ -38,6 +38,7 @@ and paragraph =
38 38 RawParagraph of string
39 39 | StructParagraph of sentence_env list (* zdania *)
40 40 | AltParagraph of (mode * paragraph) list
  41 + | ErrorParagraph of string
41 42  
42 43 type text =
43 44 RawText of string
... ... @@ -81,6 +82,7 @@ let int_of_mode = function
81 82 | Mate -> 4
82 83 | Swigra -> 5
83 84 | POLFIE -> 6
  85 + | Error -> 7
84 86  
85 87 let compare_mode x y =
86 88 compare (int_of_mode x) (int_of_mode y)
... ... @@ -109,6 +111,7 @@ let rec map_paragraph mode f = function
109 111 let l = Xlist.rev_map l (fun (mode,paragraph) ->
110 112 mode, map_paragraph mode f paragraph) in
111 113 AltParagraph(List.rev l)
  114 + | ErrorParagraph s -> ErrorParagraph s
112 115  
113 116 let rec map_text mode f = function
114 117 RawText s -> RawText s
... ... @@ -137,6 +140,7 @@ let rec fold_paragraph mode s f = function
137 140 | AltParagraph l ->
138 141 Xlist.fold l s (fun s (mode,paragraph) ->
139 142 fold_paragraph mode s f paragraph)
  143 + | ErrorParagraph _ -> s
140 144  
141 145 let rec fold_text mode s f = function
142 146 RawText _ -> s
... ...
subsyntax/ENIAMsubsyntaxXMLof.ml
  View file @9437723
... ... @@ -57,6 +57,7 @@ let rec paragraph m = function
57 57 Xml.Element("StructParagraph",set_mode m,Xlist.map sentences (fun p ->
58 58 Xml.Element("Sentence",["id",p.id;"beg",string_of_int p.beg;"len",string_of_int p.len;"next",string_of_int p.next],[sentence "" p.sentence])))
59 59 | AltParagraph l -> Xml.Element("AltParagraph",set_mode m,Xlist.map l (fun (m,t) -> paragraph (ENIAMsubsyntaxStringOf.mode m) t))
  60 + | ErrorParagraph s -> Xml.Element("ErrorParagraph",set_mode m,[Xml.PCData s])
60 61  
61 62 let rec text m = function
62 63 RawText s -> Xml.Element("RawText",set_mode m,[Xml.PCData s])
... ...
subsyntax/resources/brev.tab
  View file @9437723
... ... @@ -708,7 +708,7 @@ wz. wzorowy subst:_:_:m3
708 708 wz . wzorowy adj:_:_:_:pos
709 709 wzgl . względnie adv:pos
710 710 x raz subst:_:_:m3
711   -X październik subst:_:_:m3
  711 +#X październik subst:_:_:m3
712 712 x . książę subst:_:_:m1
713 713 x . ksiądz subst:_:_:m1
714 714 z . zeszyt subst:_:_:m3
... ...