test
merging
+72
-56
+14
-10
facet.js
+14
-10
facet.js
···
1
-
function merge_features_to_facets(features) {
1
+
export function merge_features_to_facets(features) {
2
2
// at begin, push its type to stack
3
3
// at end, pop
4
4
// make list of begins and ends
···
6
6
// iterate
7
7
let edges = []
8
8
for (let x of features) {
9
-
edges.push([x.start,'start',x])
10
-
edges.push([x.end,'end',x])
9
+
edges.push([x.start,x])
10
+
edges.push([x.end,x])
11
11
}
12
-
edges.sort((a,b)=>{a[0]-b[0]})
12
+
edges.sort(([a],[b])=>a-b)
13
13
let open = new Set()
14
14
let facets = []
15
15
let last = 0
16
-
for (let [index,type,feature] of edges) {
17
-
facets.push({
18
-
range: [last, index],
19
-
features: [...open],
20
-
})
16
+
for (let [index,feature] of edges) {
17
+
console.log(index)
18
+
if (index > last && open.size) {
19
+
facets.push({
20
+
range: [last, index],
21
+
features: [...open],
22
+
})
23
+
}
21
24
open.delete(feature) || open.add(feature)
22
25
last = index
23
26
}
24
27
return facets
25
28
}
26
29
27
-
spans = [
30
+
/*spans = [
28
31
{type:"abc", start: 1, end: 5},
29
32
{type:"def", start: 3, end: 5},
30
33
{type:"zzz", start: 6, end: 10},
···
38
41
{start: 3, end: 5, types: ["abc","def"]},
39
42
{start: 6, end: 10, types: ["zzz"]},
40
43
]
44
+
*/
+37
-46
parse.js
+37
-46
parse.js
···
1
-
class MarkupLanguage {
2
-
big_regex = /(?:)/g
3
-
constructor() {}
4
-
process_match(match) {
5
-
return [{text:match[0], features:[]}]
1
+
export class MarkupParser {
2
+
constructor(big_regex, process_match) {
3
+
this.process_match = process_match
4
+
this.big_regex = big_regex
6
5
}
7
6
text_to_segments(text) {
8
7
let segments = []
···
21
20
}
22
21
}
23
22
24
-
class Markup1 extends MarkupLanguage {
25
-
big_regex = ()=>{
26
-
let r = String.raw
27
-
let whitespace = r`\s\u00AD\u2060\u200A\u200B\u200C\u200D\u20e2`
28
-
let url = r`[-\w/%&=#+~@$*'!?,.;:]*`
29
-
let url_final = r`[-\w/%&=#+~@$*']`
30
-
return new RegExp([
31
-
r`\b(?<link>https?://${url}${url_final}([(]${url}[)](${url}${url_final})?)?)(?:\[(?<link_text>.*?)\])?`,
32
-
r`(?<=^|\s)[#＃](?<hashtag>(?!\uFE0F)[^${whitespace}]*[^\p{P}${whitespace}])`, // note: must filter out #123
33
-
r`(?<=^|\s|[(])@(?<mention>[-a-zA-Z0-9]+([.][-a-zA-Z0-9]+)+)\b`,
34
-
].join("|"), 'gu')
35
-
}()
36
-
process_match(match) {
37
-
let g = match.groups
38
-
if (g.mention!=null) {
39
-
return [{text:match[0], features:[{
40
-
$type: 'app.bsky.richtext.facet#mention',
41
-
did: g.mention, // hack, DID must be resolved afterwards
23
+
let r = String.raw
24
+
let whitespace = r`\s\u00AD\u2060\u200A\u200B\u200C\u200D\u20e2`
25
+
let url = r`[-\w/%&=#+~@$*'!?,.;:]*`
26
+
let url_final = r`[-\w/%&=#+~@$*']`
27
+
let big_regex = RegExp([
28
+
r`\b(?<link>https?://${url}${url_final}([(]${url}[)](${url}${url_final})?)?)(?:\[(?<link_text>.*?)\])?`,
29
+
r`(?<=^|\s)[#＃](?<hashtag>(?!\uFE0F)[^${whitespace}]*[^\p{P}${whitespace}])`, // note: must filter out #123
30
+
r`(?<=^|\s|[(])@(?<mention>[-a-zA-Z0-9]+([.][-a-zA-Z0-9]+)+)\b`,
31
+
].join("|"), 'gu')
32
+
function process_match(match) {
33
+
let g = match.groups
34
+
if (g.mention!=null) {
35
+
return [{text:match[0], features:[{
36
+
$type: 'app.bsky.richtext.facet#mention',
37
+
did: g.mention, // hack, DID must be resolved afterwards
38
+
}]}]
39
+
}
40
+
if (g.link!=null) {
41
+
if (g.link_text) {
42
+
return [{text:g.link_text, features:[{
43
+
$type: 'app.bsky.richtext.facet#link',
44
+
uri: g.link,
42
45
}]}]
43
-
}
44
-
if (g.link!=null) {
45
-
if (g.link_text) {
46
-
return [{text:g.link_text, features:[{
47
-
$type: 'app.bsky.richtext.facet#link',
48
-
uri: g.link,
49
-
}]}]
50
-
} else {
51
-
return [{text:match[0], features:[{
52
-
$type: 'app.bsky.richtext.facet#link',
53
-
uri: g.link,
54
-
}]}]
55
-
}
56
-
}
57
-
if (g.hashtag!=null && !/^#\d+$/.test(g.hashtag)) {
46
+
} else {
58
47
return [{text:match[0], features:[{
59
-
$type: 'app.bsky.richtext.facet#tag',
60
-
tag: g.link,
48
+
$type: 'app.bsky.richtext.facet#link',
49
+
uri: g.link,
61
50
}]}]
62
51
}
63
-
return super.process_match(match)
64
52
}
53
+
if (g.hashtag!=null && !/^#\d+$/.test(g.hashtag)) {
54
+
return [{text:match[0], features:[{
55
+
$type: 'app.bsky.richtext.facet#tag',
56
+
tag: g.link,
57
+
}]}]
58
+
}
59
+
return [{text:match[0], features:[]}]
65
60
}
66
-
67
-
let text = "abc http://example.com[test] #hello @test.com.xyz"
68
-
69
-
console.log(segments)
70
-
</script>
61
+
export let markup1 = new MarkupParser(big_regex, process_match)
+8
test2.html
+8
test2.html
+13
test3.html
+13
test3.html
···
1
+
<!doctype html>
2
+
3
+
<script type=module>
4
+
import {merge_features_to_facets} from './facet.js'
5
+
6
+
let spans = [
7
+
{type:"abc", start: 1, end: 5},
8
+
{type:"def", start: 3, end: 5},
9
+
{type:"zzz", start: 6, end: 10},
10
+
]
11
+
12
+
console.log(merge_features_to_facets(spans))
13
+
</script>