-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathtokenisationPath
31 lines (26 loc) · 1.41 KB
/
tokenisationPath
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
The classes and functions used to tokenise a document in OSCAR3. My comments are in the functions under @lh359
ProcessingDocumentFactory.java:
-------------------------------
ProcessingDocument makeTokenisedDocument(ITokeniser tokeniser, Document sourceDoc, boolean tokeniseForNEs, boolean mergeNEs, boolean runGenia)
|
V
ProcessingDocument makeTokenisedDocument(ITokeniser tokeniser, Document sourceDoc, boolean tokeniseForNEs, boolean mergeNEs, boolean runGenia, Document safDoc)
|
-------> XMLStrings.getInstance().getChemicalPlaces(procDoc.doc);
|
-------> tokeniser.tokenise(text, procDoc, offset, safDoc != null ? safDoc.getRootElement() : e, tokeniseForNEs, mergeNEs);
Tokeniser.java:
---------------
public TokenSequence tokenise(String s, ProcessingDocument doc, int offset,
Element elem, boolean tokeniseForNEs, boolean mergeNEs)
|
V
List<Token> splitToken(Token token)
|
V
List<Token> rawSplitToken(Token token)
|
-------------------- --------------------
| |
V V
List<Token> splitAt(Token token, int splitOffset) List<Token> splitAt(Token token, int splitOffset0, int splitOffset1)