me/offline-twitter
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix link parsing for retweets

Browse Source
This commit is contained in:
Alessio 2022-02-01 17:36:23 -08:00
parent 9eb3e42539
commit 32e1558800
1 changed files with 23 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
scraper/api_types_v2.go
View File

@ -136,11 +136,19 @@ func (api_result APIV2Result) ToTweetTrove() TweetTrove {
ret.MergeWith(quoted_trove)
}
// Handle URL cards
// Handle URL cards.
// This should be done in APIV2Tweet (not APIV2Result), but due to the terrible API response structuring (the Card
// should be nested under the APIV2Tweet, but it isn't), it goes here.
if api_result.Result.Legacy.RetweetedStatusResult == nil {
// We have to filter out retweets. For some reason, retweets have a copy of the card in both the retweeting
// and the retweeted TweetResults; it should only be parsed for the real Tweet, not the Retweet
if api_result.Result.Card.Legacy.Name == "summary_large_image" || api_result.Result.Card.Legacy.Name == "player" {
url := api_result.Result.Card.ParseAsUrl()
main_tweet := ret.Tweets[TweetID(api_result.Result.Legacy.ID)]
main_tweet, ok := ret.Tweets[TweetID(api_result.Result.Legacy.ID)]
if !ok {
panic(fmt.Sprintf("Tweet trove didn't contain its own tweet: %d", api_result.Result.Legacy.ID))
}
found := false
for i := range main_tweet.Urls {
if main_tweet.Urls[i].ShortText != url.ShortText {
@ -151,7 +159,8 @@ func (api_result APIV2Result) ToTweetTrove() TweetTrove {
main_tweet.Urls[i] = url
}
if !found {
panic("Tweet trove doesn't contain its own main tweet")
panic(fmt.Sprintf("Couldn't find the url in tweet ID: %d", api_result.Result.Legacy.ID))
}
}
}