ENIAMtokenizerTypes.ml
5.12 KB
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
(*
* ENIAMtokenizer, a tokenizer for Polish
* Copyright (C) 2016 Wojciech Jaworski <wjaworski atSPAMfree mimuw dot edu dot pl>
* Copyright (C) 2016 Institute of Computer Science Polish Academy of Sciences
*
* This library is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This library 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 Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*)
open Xstd
(* Długość pojedynczego znaku w tekście *)
let factor = 100
type token =
SmallLetter of string * string (* uppercase * lowercase *)
| CapLetter of string * string (* uppercase * lowercase *)
| AllSmall of string * string * string (* lowercase * firstcap * lowercase *)
| AllCap of string * string * string (* uppercase * firstcap * lowercase *)
| FirstCap of string * string * string (* uppercase * firstcap * lowercase *)
| SomeCap of string * string * string (* uppercase * orig * lowercase *)
| RomanDig of string * string (* value * cat *)
| Interp of string (* orth *)
| Symbol of string (* orth *)
| Dig of string * string (* value * cat *)
| Other of string (* orth *)
| Lemma of string * string * string list list list (* lemma * cat * interp *)
(* | Proper of string * string * string list list list * string list (* lemma * cat * interp * senses *) *)
(* | Sense of string * string * string list list list * (string * string * string list) list (* lemma * cat * interp * senses *) *)
| Compound of string * token list (* sense * components *)
| Tokens of string * int list (*cat * token id list *)
type letter_size = SL | CL | AS | FC | AC | SC
type attr =
FC | CS | MaybeCS | HasAglSuffix | MWE | LemmNotVal | TokNotFound | NotValProper | LemmLowercase | Roman | Capitalics
| SentBeg | SentEnd | SentBegEnd
| BrevLemma of string
| Disamb of string * string * string list list
(* Tekst reprezentuję jako zbiór obiektów typu token_record zawierających
informacje o poszczególnych tokenach *)
type token_env = {
orth: string; (* sekwencja znaków pierwotnego tekstu składająca się na token *)
corr_orth: string; (* sekwencja znaków pierwotnego tekstu składająca się na token z poprawionymi błędami *)
beg: int; (* pozycja początkowa tokenu względem początku akapitu *)
len: int; (* długość tokenu *)
next: int; (* pozycja początkowa następnego tokenu względem początku akapitu *)
token: token; (* treść tokenu *)
attrs: attr list; (* dodatkowe atrybuty *)
weight: float;
lemma_frequency: float;
morf_frequency: float;
tagger_output: (string * float * bool * bool) list;
cat: string;
}
(* Tokeny umieszczone są w strukturze danych umożliwiającej efektywne wyszukiwanie ich sekwencji,
struktura danych sama z siebie nie wnosi informacji *)
type tokens =
| Token of token_env
| Variant of tokens list
| Seq of tokens list
type pat = L | CL | SL | (*SL2 |*) D of string | C of string | S of string | RD of string | O of string | T of string | I of string
let empty_token_env = {
orth="";corr_orth="";beg=0;len=0;next=0; token=Symbol ""; attrs=[]; weight=0.; lemma_frequency=0.; morf_frequency=0.;
tagger_output=[];cat="X"}
let internet_mode = ref false
let resource_path =
try Sys.getenv "ENIAM_RESOURCE_PATH"
with Not_found ->
if Sys.file_exists "/usr/share/eniam" then "/usr/share/eniam" else
if Sys.file_exists "/usr/local/share/eniam" then "/usr/local/share/eniam" else
if Sys.file_exists "resources" then "resources" else
failwith "resource directory does not exists"
let data_path =
try Sys.getenv "ENIAM_USER_DATA_PATH"
with Not_found -> "data"
let mte_filename = resource_path ^ "/tokenizer/mte_20151215.tab"
let mte_filename2 = resource_path ^ "/tokenizer/mte.tab"
let known_lemmata_filename = resource_path ^ "/tokenizer/known_lemmata.tab"
let known_orths_filename = resource_path ^ "/tokenizer/known_orths.tab"
let user_known_lemmata_filename = data_path ^ "/known_lemmata.tab"
let user_known_orths_filename = data_path ^ "/known_orths.tab"
let top_level_domains_filename = resource_path ^ "/tokenizer/top-level-domains.tab"
module OrderedTokenEnv = struct
type t = token_env
let compare = compare
end
module TokenEnvSet = Xset.Make(OrderedTokenEnv)
module OrderedAttr = struct
type t = attr
let compare = compare
end
module AttrQMap = Xmap.MakeQ(OrderedAttr)
type ont = {number: string; gender: string; no_sgjp: bool; poss_ndm: bool; ont_cat: string}
module OrderedOnt = struct
type t = ont
let compare = compare
end
module OntSet = Xset.Make(OrderedOnt)
let known_lemmata = ref (StringMap.empty : OntSet.t StringMap.t StringMap.t)
let known_orths = ref StringSet.empty
let theories_paths = ref ([] : string list)