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
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
|
import {
AppBskyEmbedRecord,
AppBskyEmbedRecordWithMedia,
moderatePost,
AppBskyActorDefs,
AppBskyFeedPost,
AppBskyRichtextFacet,
AppBskyEmbedImages,
} from '@atproto/api'
type ModeratePost = typeof moderatePost
type Options = Parameters<ModeratePost>[1] & {
hiddenPosts?: string[]
mutedWords?: AppBskyActorDefs.MutedWord[]
}
const REGEX = {
LEADING_TRAILING_PUNCTUATION: /(?:^\p{P}+|\p{P}+$)/gu,
ESCAPE: /[[\]{}()*+?.\\^$|\s]/g,
SEPARATORS: /[\/\-\–\—\(\)\[\]\_]+/g,
WORD_BOUNDARY: /[\s\n\t\r\f\v]+?/g,
}
export function hasMutedWord(
mutedWords: AppBskyActorDefs.MutedWord[],
text: string,
facets?: AppBskyRichtextFacet.Main[],
outlineTags?: string[],
) {
const tags = ([] as string[])
.concat(outlineTags || [])
.concat(
facets
?.filter(facet => {
return facet.features.find(feature =>
AppBskyRichtextFacet.isTag(feature),
)
})
.map(t => t.features[0].tag as string) || [],
)
.map(t => t.toLowerCase())
for (const mute of mutedWords) {
const mutedWord = mute.value.toLowerCase()
const postText = text.toLowerCase()
// `content` applies to tags as well
if (tags.includes(mutedWord)) return true
// rest of the checks are for `content` only
if (!mute.targets.includes('content')) continue
// single character, has to use includes
if (mutedWord.length === 1 && postText.includes(mutedWord)) return true
// too long
if (mutedWord.length > postText.length) continue
// exact match
if (mutedWord === postText) return true
// any muted phrase with space or punctuation
if (/(?:\s|\p{P})+?/u.test(mutedWord) && postText.includes(mutedWord))
return true
// check individual character groups
const words = postText.split(REGEX.WORD_BOUNDARY)
for (const word of words) {
if (word === mutedWord) return true
// compare word without leading/trailing punctuation, but allow internal
// punctuation (such as `s@ssy`)
const wordTrimmedPunctuation = word.replace(
REGEX.LEADING_TRAILING_PUNCTUATION,
'',
)
if (mutedWord === wordTrimmedPunctuation) return true
if (mutedWord.length > wordTrimmedPunctuation.length) continue
// handle hyphenated, slash separated words, etc
if (REGEX.SEPARATORS.test(wordTrimmedPunctuation)) {
// check against full normalized phrase
const wordNormalizedSeparators = wordTrimmedPunctuation.replace(
REGEX.SEPARATORS,
' ',
)
const mutedWordNormalizedSeparators = mutedWord.replace(
REGEX.SEPARATORS,
' ',
)
// hyphenated (or other sep) to spaced words
if (wordNormalizedSeparators === mutedWordNormalizedSeparators)
return true
/* Disabled for now e.g. `super-cool` to `supercool`
const wordNormalizedCompressed = wordNormalizedSeparators.replace(
REGEX.WORD_BOUNDARY,
'',
)
const mutedWordNormalizedCompressed =
mutedWordNormalizedSeparators.replace(/\s+?/g, '')
// hyphenated (or other sep) to non-hyphenated contiguous word
if (mutedWordNormalizedCompressed === wordNormalizedCompressed)
return true
*/
// then individual parts of separated phrases/words
const wordParts = wordTrimmedPunctuation.split(REGEX.SEPARATORS)
for (const wp of wordParts) {
// still retain internal punctuation
if (wp === mutedWord) return true
}
}
}
}
return false
}
export function moderatePost_wrapped(
subject: Parameters<ModeratePost>[0],
opts: Options,
) {
const {hiddenPosts = [], mutedWords = [], ...options} = opts
const moderations = moderatePost(subject, options)
if (hiddenPosts.includes(subject.uri)) {
moderations.content.filter = true
moderations.content.blur = true
if (!moderations.content.cause) {
moderations.content.cause = {
// @ts-ignore Temporary extension to the moderation system -prf
type: 'post-hidden',
source: {type: 'user'},
priority: 1,
}
}
}
if (AppBskyFeedPost.isRecord(subject.record)) {
let muted = hasMutedWord(
mutedWords,
subject.record.text,
subject.record.facets || [],
subject.record.tags || [],
)
if (
subject.record.embed &&
AppBskyEmbedImages.isMain(subject.record.embed)
) {
for (const image of subject.record.embed.images) {
muted = muted || hasMutedWord(mutedWords, image.alt, [], [])
}
}
if (muted) {
moderations.content.filter = true
moderations.content.blur = true
if (!moderations.content.cause) {
moderations.content.cause = {
// @ts-ignore Temporary extension to the moderation system -prf
type: 'muted-word',
source: {type: 'user'},
priority: 1,
}
}
}
}
if (subject.embed) {
let embedHidden = false
if (AppBskyEmbedRecord.isViewRecord(subject.embed.record)) {
embedHidden = hiddenPosts.includes(subject.embed.record.uri)
if (AppBskyFeedPost.isRecord(subject.embed.record.value)) {
embedHidden =
embedHidden ||
hasMutedWord(
mutedWords,
subject.embed.record.value.text,
subject.embed.record.value.facets,
subject.embed.record.value.tags,
)
if (AppBskyEmbedImages.isMain(subject.embed.record.value.embed)) {
for (const image of subject.embed.record.value.embed.images) {
embedHidden =
embedHidden || hasMutedWord(mutedWords, image.alt, [], [])
}
}
}
}
if (
AppBskyEmbedRecordWithMedia.isView(subject.embed) &&
AppBskyEmbedRecord.isViewRecord(subject.embed.record.record)
) {
// TODO what
embedHidden = hiddenPosts.includes(subject.embed.record.record.uri)
}
if (embedHidden) {
moderations.embed.filter = true
moderations.embed.blur = true
if (!moderations.embed.cause) {
moderations.embed.cause = {
// @ts-ignore Temporary extension to the moderation system -prf
type: 'post-hidden',
source: {type: 'user'},
priority: 1,
}
}
}
}
return moderations
}
|
