-
Notifications
You must be signed in to change notification settings - Fork 0
/
feotweet.ts
653 lines (517 loc) · 21.1 KB
/
feotweet.ts
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
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
import { Config, loadConfig, UserTimeline} from "./priv/config.ts"
import * as twitter from "./priv/twitter.ts"
import { cliffy, feoblog, io, log, toml } from "./priv/deps.ts"
import { htmlToMarkdown } from "./priv/markdown.ts"
import { AttachmentCollector, Attachments, NoOpAttachmentColletor } from "./priv/attachments.ts";
async function main(options: MainOptions): Promise<void> {
logger.debug(() => `Log level: ${logger.levelName}`)
logger.debug(() => `Loading config`)
const config = await loadConfig(options.config)
await syncHomeTimeline(options, config)
await syncUserTimelines(options, config)
logger.debug("Done.")
}
async function syncHomeTimeline(options: MainOptions, config: Config): Promise<void> {
logger.info("Syncing home timeline")
if (!config.twitter.homeTimeline) {
logger.info("No home timeline configured, nothing to do.")
return
}
// Find the last status saved in FeoBlog.
const fbClient = new feoblog.Client({baseURL: config.feoblog.server})
const userID = feoblog.UserID.fromString(config.twitter.homeTimeline.userID)
const lastTimestamp = await getLatestTimestamp(fbClient, userID)
const skipUsers = new Set(config.twitter.homeTimeline.skipUsers.map(name => name.toLowerCase()))
// Collect tweets we haven't saved yet:
const tClient = new twitter.Client(config.twitter)
const newTweets: Tweet[] = []
for await (const tweetJSON of tClient.homeTimeline()) {
const tweet = new Tweet(tweetJSON)
if (!tweet.isPublic) {
logger.debug(() => `skipping private tweet: ${tweet.url}`)
continue
}
if (skipUsers.has(tweet.user.json.screen_name.toLowerCase())) {
logger.debug(() => `Skipping tweet by this user: ${tweet.url}`)
continue
}
if (lastTimestamp && tweet.timestamp <= lastTimestamp) {
break
}
newTweets.push(tweet)
if (newTweets.length >= options.maxTweets) { break }
}
logger.info(() => `Found ${newTweets.length} new tweets`)
// Insert oldest first, so that we can resume if something goes wrong:
newTweets.sort(Tweet.sortByTimestamp)
// Don't copy attachments from the home feed, this could get massive.
const attachments = new NoOpAttachmentColletor()
const privKey = await feoblog.PrivateKey.fromString(config.twitter.homeTimeline.password)
for (const tweet of newTweets) {
const bytes = (await tweet.toItem({attachments})).serialize()
const sig = privKey.sign(bytes)
await fbClient.putItem(userID, sig, bytes)
}
}
async function getLatestTimestamp(client: feoblog.Client, userID: feoblog.UserID): Promise<number|null> {
for await(const entry of client.getUserItems(userID)) {
if (entry.item_type != feoblog.protobuf.ItemType.POST) {
continue
}
return entry.timestamp_ms_utc
}
return null
}
async function syncUserTimelines(options: MainOptions, config: Config): Promise<void> {
const timelines = config.twitter.userTimelines
if (!timelines) { return } // nothing to do
for (const timeline of timelines) {
await syncUserTimeline(timeline, options, config)
}
}
async function syncUserTimeline(timeline: UserTimeline, _options: MainOptions, config: Config): Promise<void> {
logger.info(() => `Syncing timeline for @${timeline.twitterScreenName}`)
const fbClient = new feoblog.Client({baseURL: config.feoblog.server})
const userID = feoblog.UserID.fromString(timeline.userID)
const lastTimestamp = await getLatestTimestamp(fbClient, userID)
// See: https://developer.twitter.com/en/docs/twitter-api/v1/tweets/timelines/api-reference/get-statuses-user_timeline
// Max number supported by the endpoint. Always get the max, because once they fall outside of that range,
// you can never fetch them again.
// The max is documented as 3200, so we shouldn't actually reach this number:
const maxTweets = 5000
const statusLogger = new ThrottledLogger(logger)
const tClient = new twitter.Client(config.twitter)
const newTweets: Tweet[] = []
const timelineOptions = {
skipReplies: timeline.skipReplies,
skipRetweets: timeline.skipRetweets,
copyAttachments: timeline.copyAttachments,
}
for await (const tweetJSON of tClient.userTimeline(timeline.twitterScreenName, timelineOptions)) {
const tweet = new Tweet(tweetJSON)
if (lastTimestamp && tweet.timestamp <= lastTimestamp) {
break
}
newTweets.push(tweet)
statusLogger.info(() => `Loaded ${newTweets.length} tweets.`)
if (newTweets.length >= maxTweets) { break }
}
newTweets.sort(Tweet.sortByTimestamp)
const privKey = await feoblog.PrivateKey.fromString(timeline.password)
// TODO: There are some opportunities for concurrency here:
// 1. Download multiple attachments at once. (in Item.toHTML)
// 2. Download multiple tweets' attachments at once?
// ... though this one may conflict w/ tweet backreferences that I'd like to implement later.
for (const [index, tweet] of newTweets.entries()) {
const collector = timeline.copyAttachments ? new Attachments() : new NoOpAttachmentColletor()
await collector.collect(async (attachments) => {
statusLogger.info(() => `Copying tweet ${index} of ${newTweets.length}`)
const itemBytes = await errorContext(`While copying tweet: ${tweet.url}`, async () => {
return (await tweet.toItem({attachments})).serialize()
})
const sig = privKey.sign(itemBytes)
logger.debug(() => `Copying tweet: ${tweet.url}`)
await fbClient.putItem(userID, sig, itemBytes)
for (const attachment of attachments.attachments) {
logger.debug(() => `PUT-ting file: ${attachment.name} size: ${attachment.size}`)
await attachment.withReader(async (reader) => {
const stream = io.readableStreamFromReader(reader)
await fbClient.putAttachment(userID, sig, attachment.name, attachment.size, stream)
})
}
})
}
}
/** Add some error context when something fails: */
async function errorContext<T>(message: string, callback: () => Promise<T>): Promise<T> {
try {
return await callback()
} catch (cause) {
throw new Exception(message, cause)
}
}
class Exception extends Error {
constructor(readonly message: string, readonly cause: Error) {
// It looks like console.log doesn't use Error.toString(), instead
// it prints error.message and the stack trace. SO cram context into the message:
if (cause) {
message = `message\n${cause}`
}
super(message)
}
}
class ThrottledLogger {
delayMs = 5000
private lastLogMs = 0;
constructor(private logger: log.Logger) {
this.lastLogMs = this.now
}
private get now() { return new Date().valueOf() }
private get elapsed() { return this.now - this.lastLogMs }
private get shouldLog() { return this.elapsed > this.delayMs }
info(message: unknown) {
if (!this.shouldLog) { return }
this.logger.info(message)
this.lastLogMs = this.now
}
}
const STATUS_URL_PAT = /^https:\/\/twitter.com\/[^/]+\/status\/(\d+)$/i
// https://twitter.com/jwz/status/1413750203056721927
/** Used for testing to see how a single tweet would get rendered into Markdown. */
async function example(options: GlobalOptions, url: string) {
const config = await loadConfig(options.config)
const match = STATUS_URL_PAT.exec(url)
if (!match) {
throw {
error: "URL does not appear to be a valid status URL",
url,
expected: STATUS_URL_PAT,
}
}
const id = match[1]
const tClient = new twitter.Client(config.twitter)
const json = await tClient.getStatus(id)
console.log("This JSON:")
// TOML is a nice way to flatten a very deep JSON hierarchy:
// TODO: Why does it force me to cast through unknown first?
console.log(toml.stringify(json as unknown as Record<string,unknown>))
console.log()
console.log("With this body:")
console.log(json.full_text)
console.log()
const tweet = new Tweet(json)
// TODO: Option to try downloading attachments
// Or read the config for the given screen name?
const collector = new NoOpAttachmentColletor()
await collector.collect(async (attachments) => {
console.log("Would be converted to this HTML:")
console.log(await tweet.toHTML({attachments}))
console.log("Would produce this Markdown:")
console.log()
console.log(await tweet.toMarkdown({attachments}))
for (const a of attachments.attachments) {
console.log(`${a}`, a.size)
}
})
}
export class Tweet {
user: User
timestamp: number
constructor(public json: twitter.TweetJSON) {
this.user = new User(this.json.user)
this.timestamp = Date.parse(json.created_at).valueOf()
}
static sortByTimestamp(a: Tweet, b: Tweet) {
return a.timestamp - b.timestamp
}
get isPublic(): boolean {
// CAN you even RT/QT non-public tweets?
// Just in case:
const qt = this.quotedTweet
if (qt && !qt.isPublic) return false
const rt = this.retweetedTweet
if (rt && !rt.isPublic) return false
return this.user.isPublic
}
get type(): TweetType {
if (this.json.in_reply_to_status_id_str) { return "reply" }
if (this.json.retweeted_status) { return "retweet" }
if (this.json.quoted_status) { return "quoteTweet" }
return "simple"
}
get retweetedTweet(): Tweet|null {
if (this.json.retweeted_status) {
return new Tweet(this.json.retweeted_status)
}
return null
}
get quotedTweet(): Tweet|null {
if (this.json.quoted_status) {
return new Tweet(this.json.quoted_status)
}
return null
}
async toHTML(options: ConvertOptions): Promise<string> {
const rt = this.retweetedTweet
if (rt) {
return [
`<p>${this.user.toHTML()} <a href="${this.url}">retweeted</a>:`,
`<blockquote>`,
await rt.toHTML({...options, attachments: options.attachments.forRetweet()}),
`</blockquote>`
].join("")
}
const lines = [
`<p>${this.user.toHTML()} <a href="${this.url}">wrote</a>:`,
]
if (this.json.in_reply_to_status_id_str) {
const statusID = this.json.in_reply_to_status_id_str
const replyTo = this.json.in_reply_to_screen_name!
const replyToURL = `https://twitter.com/${replyTo}`
const replyToTweetURL = `${replyToURL}/status/${statusID}`
lines[0] = [
`<p>${this.user.toHTML()} <a href="${this.url}">replied</a>`
+ ` to a <a href="${replyToTweetURL}">tweet</a> by <a href="${replyToURL}">@${replyTo}</a>:`,
].join("")
}
lines.push(`<blockquote>`)
lines.push(`<p>${this.getTextAsHTML()}`)
for (const media of this.json.extended_entities?.media || []) {
// This is always a still image:
const imgSrc = await options.attachments.tryAddURL(media.media_url_https, this.url)
// We might link this to a movie if it exists:
let linkHref = imgSrc
let prefix = ""
if (media.video_info) {
const variants = media.video_info.variants
variants.sort((a, b) => (b.bitrate || 0) - (a.bitrate || 0))
const variant = variants[0]
if (variant) {
linkHref = await options.attachments.tryAddURL(variant.url, this.url)
prefix = "Video: "
}
}
lines.push(`<p>${prefix}<a href="${linkHref}"><img src="${imgSrc}"></a>`)
}
lines.push(`</blockquote>`)
const qt = this.quotedTweet
if (qt) {
// let qHTML = qt.toHTML()
// if (qHTML.startsWith("<p>")) {
// qHTML = qHTML.substr(3)
// }
// lines.push(`<p>In reply to:\n<br>${qHTML}`)
// The above still seems to result in a new paragraph in markdown, what's up with that.
// May as well just:
lines.push("<p>with quote tweet:")
lines.push(await qt.toHTML({...options, attachments: options.attachments.forQuoteTweet()}))
}
const html = lines.join("")
return html
}
// the twitter full_text can include short URLs like https://t.co/shortCode
// when there is embedded media, or quote tweets. This removes those,
// since we're going to render them anyway.
// This also replaces short URLs with full URLs.
getText(): string {
let text = this.json.full_text
// remove quote-tweet URLs, they're redundant with what we display:
const qt = this.quotedTweet
if (qt) {
const shortURL = this.findQTShortURL()
if (!shortURL) {
logger.warning(() => `No URL for quote tweet: ${this.url} ${qt.url}`)
logger.warning(() => `entities: ${JSON.stringify(this.json.entities, null, 4)}`)
logger.warning(() => `text: ${text}`)
logger.warning(() => "Continuing without deleting the URL.")
} else {
text = text.replaceAll(shortURL, "")
}
}
for (const urlMeta of this.json.entities?.urls || []) {
text = text.replaceAll(urlMeta.url, urlMeta.expanded_url)
}
// Remove media links. We'll display & link them in toHTML().
for (const media of this.json.extended_entities?.media || []) {
text = text.replaceAll(media.url, "")
}
return text
}
// Quoted tweets contain a (shortened) URL of the quoted tweet at the end of their text.
// This lets us find it, so we can remove it.
private findQTShortURL(): string|undefined {
const qt = this.quotedTweet
if (!qt) return undefined;
// You can't just use simple string matches.
// 1. Sometimes the entity URL contains a ?get=parameter
// 2. Sometimes the user has changed their display name, so the URL
// inside the tweet contains an old name which no longer matches
// what's returned by the API.
// Instead, rely on the globally unique status ID to find the URL.
const statusID = this.statusIDFromURL(qt.url)
if (!statusID) {
// This would seem to indicate Twitter has broken their API, so fail hard:
throw new Error(`Tweet ${this.url} has quote tweet URL (${qt.url}) which is not a status URL!?`)
}
const meta = this.json.entities?.urls.find(it => this.statusIDFromURL(it.expanded_url) === statusID)
if (meta) return meta.url
return undefined
}
private statusIDFromURL(url: string): string|undefined {
const match = Tweet.STATUS_PAT.exec(url)
if (!match) return undefined
return match[1]
}
private static STATUS_PAT = /\/status\/(\d+)/i
getTextAsHTML(): string {
let text = this.getText()
// Could break URLs. Hmm. Not sure what we can do here.
// Maybe only quote them at the beginning of words? TODO
// .replaceAll("&", "&")
text = text.replaceAll("<", "<")
// Link URLs:
text = replaceAll(text, URL_PAT, (match) => {
const url = match[0]
return `<a href="${url}">${url}</a>`
})
// Link Mentions
text = replaceAll(text, MENTION_PAT, (match) => {
const userName = match[1]
return `<a href="https://twitter.com/${userName}">@${userName}</a>`
})
// Fix line returns last, so as not to mess with whitespace matching in our regexes:
return text.replaceAll("\n", "\n<br>")
}
get url(): string {
return `https://twitter.com/${this.json.user.screen_name}/status/${this.json.id_str}`
}
async toMarkdown(options: ConvertOptions): Promise<string> {
return htmlToMarkdown(await this.toHTML(options))
}
async toItem(options: ConvertOptions): Promise<feoblog.protobuf.Item> {
const item = new feoblog.protobuf.Item({
timestamp_ms_utc: this.timestamp,
// I didn't see TZ offsets in the twitter JSON data, so UTC for everyone.
})
item.post = new feoblog.protobuf.Post({
body: await this.toMarkdown(options)
})
if (options.attachments.attachments.length > 0) {
item.post.attachments = new feoblog.protobuf.Attachments({
file: options.attachments.attachments.map((a) => {
return new feoblog.protobuf.File({
hash: a.hash,
size: a.size,
name: a.name,
})
}),
})
}
return item
}
}
interface ConvertOptions {
/** Collects attachments to attach to the Post. */
attachments: AttachmentCollector
}
// Replace all non-overlapping regexes with some other value.
function replaceAll(text: string, pat: RegExp, mapper: (match: RegExpMatchArray) => string): string {
const matches = [... text.matchAll(pat)].reverse()
for (const match of matches) {
text = replaceMatch(text, match, mapper(match))
}
return text
}
// Really? There's not a built-in way to do this?
function replaceMatch(original: string, match: RegExpMatchArray, newValue: string) {
if (match.index === undefined) {
// When can this be undefined? Why does TypeScript say it can be?
throw {
error: "No match index",
match
}
}
const start = match.index
const len = match[0].length
return (
original.substr(0, start)
+ newValue
+ original.substr(start + len)
)
}
// Note: can't use \b before @ because @ is not a "word character", so does not make a word boundary.
const MENTION_PAT = /(?<=^[.]?|\s)@([a-z0-9_]{2,15})/gi
const URL_PAT = /\bhttps?:\/\/[^)"\s]+/g
type TweetType = "simple"|"reply"|"retweet"|"quoteTweet"
class User {
constructor(public json: twitter.UserJSON) {}
get isPublic() {
return !this.json.protected
}
toHTML(): string {
const name = this.json.name
const screenName = this.json.screen_name
let html =`<a href="${this.url}">@${screenName}</a>`
if (name && name.toLowerCase() != screenName.toLowerCase()) {
html += ` ("${this.json.name}")`
}
return html
}
get url() {
return `https://twitter.com/${this.json.screen_name}`
}
}
type ValueHandler<In,Out> = (value: In) => Out
// std log's loglevels are non-contiguous. Here we just order them for ourselves:
const logLevels = [
log.LogLevels.DEBUG,
log.LogLevels.INFO,
log.LogLevels.WARNING,
log.LogLevels.ERROR,
log.LogLevels.CRITICAL
]
// a cliffy option handler to increment the log level
function incLogLevel(incAmount: number): ValueHandler<boolean,void> {
const handler = (_value: boolean) => {
const levelNum = logLevels.findIndex(it => it === logger.level)
if (levelNum < 0) {
throw new Error(`Initial logLevel was not found: ${logger.levelName}`)
}
const newLevel = logLevels[levelNum + incAmount]
if (newLevel) {
logger.level = newLevel
}
}
return handler
}
const CLI_OPTIONS = (
new cliffy.Command<void>()
.name("feotweet")
.description("A tool to sync a twitter feed to FeoBlog")
.globalOption<{config: string}>("--config", "Config file to use", {default: "./feotweet.toml"})
.globalOption<void>("-v, --verbose", "Increase log verbosity", {
collect: true,
value: incLogLevel(-1)
})
.globalOption<void>("-q, --quiet", "Decrease log verbosity", {
collect: true,
value: incLogLevel(1)
})
.option<{maxTweets: number}>("--maxTweets", "Max # of tweets to read from Twitter", {default: 100})
.action(main)
)
CLI_OPTIONS.command("example")
.description("Convert one tweet from its URL")
.hidden()
.arguments<[url: string]>("<url:string>")
.action(example)
interface GlobalOptions {
config: string
}
interface MainOptions extends GlobalOptions {
maxTweets: number
}
await log.setup({
handlers: {
// Set the handler to the lowest level, so it'll
// log everything we throw at it. We'll limit our
// logs via our logger.level.
// (Otherwise, log defaults to "INFO")
default: new log.handlers.ConsoleHandler("DEBUG")
},
loggers: {
default: {
level: "INFO",
handlers: ["default"]
}
}
})
const logger = log.getLogger()
if (import.meta.main) {
try {
await CLI_OPTIONS.parse(Deno.args)
} catch (error) {
logger.error(error)
Deno.exit(1)
}
}