461 lines
14 KiB
Go
461 lines
14 KiB
Go
package scraper
|
||
|
||
import (
|
||
"fmt"
|
||
"html"
|
||
"time"
|
||
"strings"
|
||
"encoding/json"
|
||
"strconv"
|
||
"sort"
|
||
)
|
||
|
||
|
||
type APIMedia struct {
|
||
ID int64 `json:"id_str,string"`
|
||
MediaURLHttps string `json:"media_url_https"`
|
||
Type string `json:"type"`
|
||
URL string `json:"url"`
|
||
OriginalInfo struct {
|
||
Width int `json:"width"`
|
||
Height int `json:"height"`
|
||
} `json:"original_info"`
|
||
}
|
||
|
||
type SortableVariants []struct {
|
||
Bitrate int `json:"bitrate,omitempty"`
|
||
URL string `json:"url"`
|
||
}
|
||
func (v SortableVariants) Len() int { return len(v) }
|
||
func (v SortableVariants) Swap(i, j int) { v[i], v[j] = v[j], v[i] }
|
||
func (v SortableVariants) Less(i, j int) bool { return v[i].Bitrate > v[j].Bitrate }
|
||
|
||
type APIExtendedMedia struct {
|
||
ID int64 `json:"id_str,string"`
|
||
MediaURLHttps string `json:"media_url_https"`
|
||
Type string `json:"type"`
|
||
VideoInfo struct {
|
||
Variants SortableVariants `json:"variants"`
|
||
Duration int `json:"duration_millis"`
|
||
} `json:"video_info"`
|
||
OriginalInfo struct {
|
||
Width int `json:"width"`
|
||
Height int `json:"height"`
|
||
} `json:"original_info"`
|
||
Ext struct {
|
||
MediaStats struct {
|
||
R interface{} `json:"r"`
|
||
} `json:"mediaStats"`
|
||
} `json:"ext"`
|
||
}
|
||
|
||
type APICard struct {
|
||
Name string `json:"name"`
|
||
ShortenedUrl string `json:"url"`
|
||
BindingValues struct {
|
||
Domain struct {
|
||
Value string `json:"string_value"`
|
||
} `json:"domain"`
|
||
Creator struct {
|
||
UserValue struct {
|
||
Value int64 `json:"id_str,string"`
|
||
} `json:"user_value"`
|
||
} `json:"creator"`
|
||
Site struct {
|
||
UserValue struct {
|
||
Value int64 `json:"id_str,string"`
|
||
} `json:"user_value"`
|
||
} `json:"site"`
|
||
Title struct {
|
||
Value string `json:"string_value"`
|
||
} `json:"title"`
|
||
Description struct {
|
||
Value string `json:"string_value"`
|
||
} `json:"description"`
|
||
Thumbnail struct {
|
||
ImageValue struct {
|
||
Url string `json:"url"`
|
||
Width int `json:"width"`
|
||
Height int `json:"height"`
|
||
} `json:"image_value"`
|
||
} `json:"thumbnail_image_large"`
|
||
PlayerImage struct {
|
||
ImageValue struct {
|
||
Url string `json:"url"`
|
||
} `json:"image_value"`
|
||
} `json:"player_image_large"`
|
||
|
||
// For polls
|
||
Choice1 struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice1_label"`
|
||
Choice2 struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice2_label"`
|
||
Choice3 struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice3_label"`
|
||
Choice4 struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice4_label"`
|
||
|
||
Choice1_Count struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice1_count"`
|
||
Choice2_Count struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice2_count"`
|
||
Choice3_Count struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice3_count"`
|
||
Choice4_Count struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"choice4_count"`
|
||
|
||
EndDatetimeUTC struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"end_datetime_utc"`
|
||
CountsAreFinal struct {
|
||
BooleanValue bool `json:"boolean_value"`
|
||
} `json:"counts_are_final"`
|
||
DurationMinutes struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"duration_minutes"`
|
||
LastUpdatedAt struct {
|
||
StringValue string `json:"string_value"`
|
||
} `json:"last_updated_datetime_utc"`
|
||
} `json:"binding_values"`
|
||
}
|
||
|
||
type APITweet struct {
|
||
ID int64 `json:"id_str,string"`
|
||
ConversationID int64 `json:"conversation_id_str,string"`
|
||
CreatedAt string `json:"created_at"`
|
||
FavoriteCount int `json:"favorite_count"`
|
||
FullText string `json:"full_text"`
|
||
DisplayTextRange []int `json:"display_text_range"`
|
||
Entities struct {
|
||
Hashtags []struct {
|
||
Text string `json:"text"`
|
||
} `json:"hashtags"`
|
||
Media []APIMedia `json:"media"`
|
||
URLs []struct {
|
||
ExpandedURL string `json:"expanded_url"`
|
||
ShortenedUrl string `json:"url"`
|
||
} `json:"urls"`
|
||
Mentions []struct {
|
||
UserName string `json:"screen_name"`
|
||
UserID int64 `json:"id_str,string"`
|
||
} `json:"user_mentions"`
|
||
ReplyMentions string // The leading part of the text which is cut off by "DisplayTextRange"
|
||
} `json:"entities"`
|
||
ExtendedEntities struct {
|
||
Media []APIExtendedMedia `json:"media"`
|
||
} `json:"extended_entities"`
|
||
InReplyToStatusID int64 `json:"in_reply_to_status_id_str,string"`
|
||
InReplyToUserID int64 `json:"in_reply_to_user_id_str,string"`
|
||
InReplyToScreenName string `json:"in_reply_to_screen_name"`
|
||
ReplyCount int `json:"reply_count"`
|
||
RetweetCount int `json:"retweet_count"`
|
||
QuoteCount int `json:"quote_count"`
|
||
RetweetedStatusIDStr string `json:"retweeted_status_id_str"` // Can be empty string
|
||
RetweetedStatusID int64
|
||
QuotedStatusIDStr string `json:"quoted_status_id_str"` // Can be empty string
|
||
QuotedStatusID int64
|
||
QuotedStatusPermalink struct {
|
||
ShortURL string `json:"url"`
|
||
ExpandedURL string `json:"expanded"`
|
||
} `json:"quoted_status_permalink"`
|
||
Time time.Time `json:"time"`
|
||
UserID int64 `json:"user_id_str,string"`
|
||
UserHandle string
|
||
Card APICard `json:"card"`
|
||
TombstoneText string
|
||
}
|
||
|
||
func (t *APITweet) NormalizeContent() {
|
||
id, err := strconv.Atoi(t.QuotedStatusIDStr)
|
||
if err == nil {
|
||
t.QuotedStatusID = int64(id)
|
||
}
|
||
id, err = strconv.Atoi(t.RetweetedStatusIDStr)
|
||
if err == nil {
|
||
t.RetweetedStatusID = int64(id)
|
||
}
|
||
|
||
if (len(t.DisplayTextRange) == 2) {
|
||
t.Entities.ReplyMentions = strings.TrimSpace(string([]rune(t.FullText)[0:t.DisplayTextRange[0]]))
|
||
t.FullText = string([]rune(t.FullText)[t.DisplayTextRange[0]:t.DisplayTextRange[1]])
|
||
}
|
||
|
||
// Handle short links showing up at ends of tweets
|
||
for _, url := range t.Entities.URLs {
|
||
index := strings.Index(t.FullText, url.ShortenedUrl)
|
||
if index == (len(t.FullText) - len(url.ShortenedUrl)) {
|
||
t.FullText = strings.TrimSpace(t.FullText[0:index])
|
||
}
|
||
}
|
||
|
||
// Handle pasted tweet links that turn into quote tweets but still have a link in them
|
||
// This is a separate case from above because we want it gone even if it's in the middle of the tweet
|
||
if t.QuotedStatusID != 0 {
|
||
for _, url := range t.Entities.URLs {
|
||
if url.ShortenedUrl == t.QuotedStatusPermalink.ShortURL {
|
||
t.FullText = strings.ReplaceAll(t.FullText, url.ShortenedUrl, "")
|
||
}
|
||
}
|
||
}
|
||
t.FullText = html.UnescapeString(t.FullText)
|
||
t.FullText = strings.TrimSpace(t.FullText)
|
||
}
|
||
|
||
func (t APITweet) String() string {
|
||
data, err := json.Marshal(t)
|
||
if err != nil {
|
||
panic(err)
|
||
}
|
||
return string(data)
|
||
}
|
||
|
||
|
||
type APIUser struct {
|
||
CreatedAt string `json:"created_at"`
|
||
Description string `json:"description"`
|
||
Entities struct {
|
||
URL struct {
|
||
Urls []struct {
|
||
ExpandedURL string `json:"expanded_url"`
|
||
} `json:"urls"`
|
||
} `json:"url"`
|
||
} `json:"entities"`
|
||
FavouritesCount int `json:"favourites_count"`
|
||
FollowersCount int `json:"followers_count"`
|
||
FriendsCount int `json:"friends_count"`
|
||
ID int64 `json:"id_str,string"`
|
||
ListedCount int `json:"listed_count"`
|
||
Name string `json:"name"`
|
||
Location string `json:"location"`
|
||
PinnedTweetIdsStr []string `json:"pinned_tweet_ids_str"` // Dunno how to type-convert an array
|
||
ProfileBannerURL string `json:"profile_banner_url"`
|
||
ProfileImageURLHTTPS string `json:"profile_image_url_https"`
|
||
Protected bool `json:"protected"`
|
||
ScreenName string `json:"screen_name"`
|
||
StatusesCount int `json:"statuses_count"`
|
||
Verified bool `json:"verified"`
|
||
IsBanned bool
|
||
DoesntExist bool
|
||
}
|
||
|
||
|
||
type UserResponse struct {
|
||
Data struct {
|
||
User struct {
|
||
ID int64 `json:"rest_id,string"`
|
||
Legacy APIUser `json:"legacy"`
|
||
} `json:"user"`
|
||
} `json:"data"`
|
||
Errors []struct {
|
||
Message string `json:"message"`
|
||
Name string `json:"name"`
|
||
Code int `json:"code"`
|
||
} `json:"errors"`
|
||
}
|
||
func (u UserResponse) ConvertToAPIUser() APIUser {
|
||
ret := u.Data.User.Legacy
|
||
ret.ID = u.Data.User.ID
|
||
|
||
// Banned users
|
||
for _, api_error := range u.Errors {
|
||
if api_error.Message == "Authorization: User has been suspended. (63)" {
|
||
ret.IsBanned = true
|
||
} else if api_error.Name == "NotFoundError" {
|
||
ret.DoesntExist = true
|
||
} else {
|
||
panic(fmt.Sprintf("Unknown api error: %q", api_error.Message))
|
||
}
|
||
}
|
||
|
||
return ret
|
||
}
|
||
|
||
type Entry struct {
|
||
EntryID string `json:"entryId"`
|
||
SortIndex int64 `json:"sortIndex,string"`
|
||
Content struct {
|
||
Item struct {
|
||
Content struct {
|
||
Tombstone struct {
|
||
TombstoneInfo struct {
|
||
RichText struct {
|
||
Text string `json:"text"`
|
||
} `json:"richText"`
|
||
} `json:"tombstoneInfo"`
|
||
} `json:"tombstone"`
|
||
Tweet struct {
|
||
ID int64 `json:"id,string"`
|
||
} `json:"tweet"`
|
||
} `json:"content"`
|
||
} `json:"item"`
|
||
Operation struct {
|
||
Cursor struct {
|
||
Value string `json:"value"`
|
||
} `json:"cursor"`
|
||
} `json:"operation"`
|
||
} `json:"content"`
|
||
}
|
||
func (e Entry) GetTombstoneText() string {
|
||
return e.Content.Item.Content.Tombstone.TombstoneInfo.RichText.Text
|
||
}
|
||
type SortableEntries []Entry
|
||
func (e SortableEntries) Len() int { return len(e) }
|
||
func (e SortableEntries) Swap(i, j int) { e[i], e[j] = e[j], e[i] }
|
||
func (e SortableEntries) Less(i, j int) bool { return e[i].SortIndex > e[j].SortIndex }
|
||
|
||
type TweetResponse struct {
|
||
GlobalObjects struct {
|
||
Tweets map[string]APITweet `json:"tweets"`
|
||
Users map[string]APIUser `json:"users"`
|
||
} `json:"globalObjects"`
|
||
Timeline struct {
|
||
Instructions []struct {
|
||
AddEntries struct {
|
||
Entries SortableEntries `json:"entries"`
|
||
} `json:"addEntries"`
|
||
ReplaceEntry struct {
|
||
Entry Entry
|
||
} `json:"replaceEntry"`
|
||
} `json:"instructions"`
|
||
} `json:"timeline"`
|
||
}
|
||
|
||
var tombstone_types = map[string]string{
|
||
"This Tweet was deleted by the Tweet author. Learn more": "deleted",
|
||
"This Tweet is from a suspended account. Learn more": "suspended",
|
||
"You’re unable to view this Tweet because this account owner limits who can view their Tweets. Learn more": "hidden",
|
||
"This Tweet is unavailable. Learn more": "unavailable",
|
||
"This Tweet violated the Twitter Rules. Learn more": "violated",
|
||
"This Tweet is from an account that no longer exists. Learn more": "no longer exists",
|
||
"Age-restricted adult content. This content might not be appropriate for people under 18 years old. To view this media, " +
|
||
"you’ll need to log in to Twitter. Learn more": "age-restricted",
|
||
}
|
||
/**
|
||
* Insert tweets into GlobalObjects for each tombstone. Returns a list of users that need to
|
||
* be fetched for tombstones.
|
||
*/
|
||
func (t *TweetResponse) HandleTombstones() []UserHandle {
|
||
ret := []UserHandle{}
|
||
|
||
// Handle tombstones in quote-tweets
|
||
for _, api_tweet := range t.GlobalObjects.Tweets {
|
||
// Ignore if tweet doesn't have a quoted tweet
|
||
if api_tweet.QuotedStatusIDStr == "" {
|
||
continue
|
||
}
|
||
// Ignore if quoted tweet is in the Global Objects (i.e., not a tombstone)
|
||
if _, ok := t.GlobalObjects.Tweets[api_tweet.QuotedStatusIDStr]; ok {
|
||
continue
|
||
}
|
||
|
||
user_handle, err := ParseHandleFromTweetUrl(api_tweet.QuotedStatusPermalink.ExpandedURL)
|
||
if err != nil {
|
||
panic(err)
|
||
}
|
||
|
||
var tombstoned_tweet APITweet
|
||
tombstoned_tweet.ID = int64(int_or_panic(api_tweet.QuotedStatusIDStr))
|
||
tombstoned_tweet.UserHandle = string(user_handle)
|
||
tombstoned_tweet.TombstoneText = "unavailable"
|
||
|
||
ret = append(ret, user_handle)
|
||
fmt.Printf("Adding quoted tombstoned tweet: TweetID %d, handle %q\n", tombstoned_tweet.ID, tombstoned_tweet.UserHandle)
|
||
|
||
t.GlobalObjects.Tweets[api_tweet.QuotedStatusIDStr] = tombstoned_tweet
|
||
}
|
||
|
||
// Handle tombstones in the conversation flow
|
||
entries := t.Timeline.Instructions[0].AddEntries.Entries
|
||
sort.Sort(entries)
|
||
for i, entry := range entries {
|
||
if entry.GetTombstoneText() != "" {
|
||
// Try to reconstruct the tombstone tweet
|
||
var tombstoned_tweet APITweet
|
||
tombstoned_tweet.ID = int64(i) // Set a default to prevent clobbering other tombstones
|
||
if i + 1 < len(entries) && entries[i+1].Content.Item.Content.Tweet.ID != 0 {
|
||
next_tweet_id := entries[i+1].Content.Item.Content.Tweet.ID
|
||
api_tweet, ok := t.GlobalObjects.Tweets[fmt.Sprint(next_tweet_id)]
|
||
if !ok {
|
||
panic("Weird situation!")
|
||
}
|
||
tombstoned_tweet.ID = api_tweet.InReplyToStatusID
|
||
tombstoned_tweet.UserID = api_tweet.InReplyToUserID
|
||
ret = append(ret, UserHandle(api_tweet.InReplyToScreenName))
|
||
}
|
||
if i - 1 >= 0 && entries[i-1].Content.Item.Content.Tweet.ID != 0 {
|
||
prev_tweet_id := entries[i-1].Content.Item.Content.Tweet.ID
|
||
_, ok := t.GlobalObjects.Tweets[fmt.Sprint(prev_tweet_id)]
|
||
if !ok {
|
||
panic("Weird situation 2!")
|
||
}
|
||
tombstoned_tweet.InReplyToStatusID = prev_tweet_id
|
||
}
|
||
|
||
short_text, ok := tombstone_types[entry.GetTombstoneText()]
|
||
if !ok {
|
||
panic(fmt.Sprintf("Unknown tombstone text: %s", entry.GetTombstoneText()))
|
||
}
|
||
tombstoned_tweet.TombstoneText = short_text
|
||
|
||
// Add the tombstoned tweet to GlobalObjects
|
||
t.GlobalObjects.Tweets[fmt.Sprint(tombstoned_tweet.ID)] = tombstoned_tweet
|
||
}
|
||
}
|
||
|
||
return ret
|
||
}
|
||
|
||
func (t *TweetResponse) GetCursor() string {
|
||
entries := t.Timeline.Instructions[0].AddEntries.Entries
|
||
if len(entries) > 0 {
|
||
last_entry := entries[len(entries) - 1]
|
||
if strings.Contains(last_entry.EntryID, "cursor") {
|
||
return last_entry.Content.Operation.Cursor.Value
|
||
}
|
||
}
|
||
|
||
// Next, try the other format ("replaceEntry")
|
||
instructions := t.Timeline.Instructions
|
||
last_replace_entry := instructions[len(instructions) - 1].ReplaceEntry.Entry
|
||
if strings.Contains(last_replace_entry.EntryID, "cursor") {
|
||
return last_replace_entry.Content.Operation.Cursor.Value
|
||
}
|
||
return ""
|
||
}
|
||
|
||
/**
|
||
* Test for one case of end-of-feed. Cursor increments on each request for some reason, but
|
||
* there's no new content. This seems to happen when there's a pinned tweet.
|
||
*
|
||
* In this case, we look for an "entries" object that has only cursors in it, and no tweets.
|
||
*/
|
||
func (t *TweetResponse) IsEndOfFeed() bool {
|
||
entries := t.Timeline.Instructions[0].AddEntries.Entries
|
||
if len(entries) > 2 {
|
||
return false
|
||
}
|
||
for _, e := range entries {
|
||
if !strings.Contains(e.EntryID, "cursor") {
|
||
return false
|
||
}
|
||
}
|
||
return true
|
||
}
|
||
|
||
|
||
func idstr_to_int(idstr string) int64 {
|
||
id, err := strconv.Atoi(idstr)
|
||
if err != nil {
|
||
panic(err)
|
||
}
|
||
return int64(id)
|
||
}
|