forked from OhadRau/revery-text-wrap
-
Notifications
You must be signed in to change notification settings - Fork 0
/
tokenize.ml
54 lines (51 loc) · 2.08 KB
/
tokenize.ml
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
let slice ~first ~last string =
let length = last - first in
String.sub string first length
(* Split a string into an AppendList of strings, chopping at every newline and
removing the newline characters. *)
let split_on_newline string =
let lines = Queue.create () in
let first = ref 0 in
for i = 0 to String.length string - 1 do
if string.[i] = '\n' then begin
Queue.add (slice ~first:!first ~last:i string) lines;
first := i + 1
end
done;
(* Append the remaining characters after the last split *)
if !first <= String.length string - 1 then begin
Queue.add (slice ~first:!first ~last:(String.length string) string) lines
end;
lines
(* Tokenize a string into an AppendList of strings, chopping on every instance of whitespace
and every hyphen. Isolate whitespace into its own token and preserve it in the final list. *)
let split_tokens string =
let tokens = Queue.create () in
let first = ref 0 in
for i = 0 to String.length string - 1 do
match string.[i] with
| '\n' | ' ' | '\t' ->
(* Push everything before this character (if there is anything) *)
if !first <> i then begin
Queue.add (slice ~first:!first ~last:i string) tokens
end;
(* Then push this character *)
Queue.add (slice ~first:i ~last:(i+1) string) tokens;
(* And skip it *)
first := i + 1
| '-' ->
(* When a word is already hyphenated, split it into tokens so we don't needlessly
rehyphenate it and end up with a double hyphen *)
Queue.add (slice ~first:!first ~last:(i+1) string) tokens;
first := i + 1
| _ -> ()
done;
(* Append the remaining characters after the last split *)
if !first <= String.length string - 1 then begin
Queue.add (slice ~first:!first ~last:(String.length string) string) tokens
end;
tokens
(* Check if a token is blank/whitespace *)
let is_whitespace str =
(* We only have to check the start b/c the tokenizer won't return anything after the whitespace *)
String.length str = 0 || str.[0] == ' ' || str.[0] == '\t'