nitter/src/formatters.nim

138 lines
4.5 KiB
Nim
Raw Normal View History

2019-09-19 03:22:45 +00:00
import strutils, strformat, sequtils, htmlgen, xmltree, times, uri
2019-06-20 14:16:20 +00:00
import regex
2019-08-19 18:53:47 +00:00
import types, utils
2019-06-20 14:16:20 +00:00
2019-06-25 00:38:18 +00:00
from unicode import Rune, `$`
2019-06-20 14:16:20 +00:00
const
2019-06-27 20:30:00 +00:00
urlRegex = re"((https?|ftp)://(-\.)?([^\s/?\.#]+\.?)+([/\?][^\s\)]*)?)"
2019-06-20 14:16:20 +00:00
emailRegex = re"([a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+)"
2019-07-03 10:20:23 +00:00
usernameRegex = re"(^|[^A-z0-9_?])@([A-z0-9_]+)"
2019-06-20 14:16:20 +00:00
picRegex = re"pic.twitter.com/[^ ]+"
ellipsisRegex = re" ?…"
2019-09-19 03:22:45 +00:00
hashtagRegex = re"([^\S])?([#$][A-z0-9]+)"
2019-09-14 06:56:46 +00:00
ytRegex = re"(www.|m.)?youtu(be.com|.be)"
twRegex = re"(www.|mobile.)?twitter.com"
2019-06-25 00:38:18 +00:00
nbsp = $Rune(0x000A0)
2019-09-15 07:57:45 +00:00
const hostname {.strdefine.} = "nitter.net"
2019-06-25 00:38:18 +00:00
proc stripText*(text: string): string =
text.replace(nbsp, " ").strip()
2019-06-20 14:16:20 +00:00
proc shortLink*(text: string; length=28): string =
result = text.replace(re"https?://(www.)?", "")
if result.len > length:
result = result[0 ..< length] & ""
2019-09-15 07:57:45 +00:00
proc toLink*(url, text: string): string =
a(text, href=url)
2019-06-20 14:16:20 +00:00
2019-09-15 07:57:45 +00:00
proc reUrlToShortLink*(m: RegexMatch; s: string): string =
2019-06-20 14:16:20 +00:00
let url = s[m.group(0)[0]]
2019-06-24 06:07:36 +00:00
toLink(url, shortLink(url))
2019-06-20 14:16:20 +00:00
2019-09-15 07:57:45 +00:00
proc reUrlToLink*(m: RegexMatch; s: string): string =
let url = s[m.group(0)[0]]
toLink(url, url.replace(re"https?://(www.)?", ""))
2019-06-20 14:16:20 +00:00
proc reEmailToLink*(m: RegexMatch; s: string): string =
let url = s[m.group(0)[0]]
toLink("mailto://" & url, url)
2019-09-19 03:22:45 +00:00
proc reHashtagToLink*(m: RegexMatch; s: string): string =
result = if m.group(0).len > 0: s[m.group(0)[0]] else: ""
let hash = s[m.group(1)[0]]
let link = toLink("/search?text=" & encodeUrl(hash), hash)
if hash.any(isAlphaAscii):
result &= link
else:
result &= hash
2019-06-20 14:16:20 +00:00
proc reUsernameToLink*(m: RegexMatch; s: string): string =
2019-06-25 01:48:57 +00:00
var username = ""
var pretext = ""
2019-06-20 14:16:20 +00:00
2019-06-25 01:48:57 +00:00
let pre = m.group(0)
let match = m.group(1)
2019-06-20 14:16:20 +00:00
username = s[match[0]]
if pre.len > 0:
pretext = s[pre[0]]
pretext & toLink("/" & username, "@" & username)
2019-09-15 07:57:45 +00:00
proc reUsernameToFullLink*(m: RegexMatch; s: string): string =
result = reUsernameToLink(m, s)
result = result.replace("href=\"/", &"href=\"https://{hostname}/")
proc replaceUrl*(url: string; prefs: Prefs): string =
2019-08-15 16:45:56 +00:00
result = url
if prefs.replaceYouTube.len > 0:
result = result.replace(ytRegex, prefs.replaceYouTube)
if prefs.replaceTwitter.len > 0:
result = result.replace(twRegex, prefs.replaceTwitter)
2019-06-20 14:16:20 +00:00
2019-09-15 07:57:45 +00:00
proc linkifyText*(text: string; prefs: Prefs; rss=false): string =
result = xmltree.escape(stripText(text))
result = result.replace(ellipsisRegex, "")
result = result.replace(emailRegex, reEmailToLink)
2019-09-19 03:22:45 +00:00
result = result.replace(hashtagRegex, reHashtagToLink)
2019-09-15 07:57:45 +00:00
if rss:
result = result.replace(urlRegex, reUrlToLink)
result = result.replace(usernameRegex, reUsernameToFullLink)
else:
result = result.replace(urlRegex, reUrlToShortLink)
result = result.replace(usernameRegex, reUsernameToLink)
result = result.replace(re"([^\s\(\n%])<a", "$1 <a")
result = result.replace(re"</a>\s+([;.,!\)'%]|&apos;)", "</a>$1")
result = result.replace(re"^\. <a", ".<a")
result = result.replaceUrl(prefs)
2019-06-20 14:16:20 +00:00
proc stripTwitterUrls*(text: string): string =
result = text
result = result.replace(picRegex, "")
result = result.replace(ellipsisRegex, "")
2019-08-19 18:53:47 +00:00
proc proxifyVideo*(manifest: string; proxy: bool): string =
proc cb(m: RegexMatch; s: string): string =
result = "https://video.twimg.com" & s[m.group(0)[0]]
if proxy: result = getVidUrl(result)
2019-08-19 18:53:47 +00:00
result = manifest.replace(re"(.+(.ts|.m3u8|.vmap))", cb)
2019-06-20 14:16:20 +00:00
proc getUserpic*(userpic: string; style=""): string =
let pic = userpic.replace(re"_(normal|bigger|mini|200x200|400x400)(\.[A-z]+)$", "$2")
2019-06-20 14:16:20 +00:00
pic.replace(re"(\.[A-z]+)$", style & "$1")
proc getUserpic*(profile: Profile; style=""): string =
getUserPic(profile.userpic, style)
2019-08-07 20:27:24 +00:00
proc getVideoEmbed*(id: string): string =
&"https://twitter.com/i/videos/{id}?embed_source=facebook"
2019-08-07 20:02:19 +00:00
2019-06-24 20:40:48 +00:00
proc pageTitle*(profile: Profile): string =
2019-07-31 00:15:43 +00:00
&"{profile.fullname} (@{profile.username})"
2019-06-25 02:52:38 +00:00
2019-08-07 20:02:19 +00:00
proc pageDesc*(profile: Profile): string =
"The latest tweets from " & profile.fullname
proc getJoinDate*(profile: Profile): string =
profile.joinDate.format("'Joined' MMMM YYYY")
proc getJoinDateFull*(profile: Profile): string =
profile.joinDate.format("h:mm tt - d MMM YYYY")
2019-06-25 02:52:38 +00:00
proc getTime*(tweet: Tweet): string =
2019-09-15 09:14:03 +00:00
tweet.time.format("d/M/yyyy', 'HH:mm:ss")
proc getRfc822Time*(tweet: Tweet): string =
tweet.time.format("ddd', 'd MMM yyyy HH:mm:ss 'GMT'")
2019-07-01 21:14:36 +00:00
proc getLink*(tweet: Tweet | Quote): string =
2019-07-01 21:55:19 +00:00
&"/{tweet.profile.username}/status/{tweet.id}"
proc getTombstone*(text: string): string =
text.replace(re"\n* *Learn more", "").stripText().strip(chars={' ', '\n'})