pogmommy/TwitToMast
Archived
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

attempt at overhauling xpath selectors

Browse source 
attempted to integrate more-static attributes
will likely break next week
This commit is contained in:
Penelope Gomez / Pogmommy 2023-03-02 02:29:47 -07:00
parent a13d99f1c6
commit 7161156a2e
1 changed files with 22 additions and 21 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

43
ref/classes/xpaths.js
View file

@ -1,36 +1,37 @@
class TweetPaths { class TweetPaths {
constructor(orig,i) { constructor(orig,i) {
if (orig == 'home') { if (orig == 'home') {
this.timeLine = "//*[@id='react-root']/div/div/div[2]/main/div/div/div/div/div/div[3]/div/div/section/div/div"; //the immediate parent div of all tweets this.timeLine = "/html//section/div/div"; //FIXED //the immediate parent div of all tweets
} else if (orig == 'thread') { } else if (orig == 'thread') {
this.timeLine = "/html/body/div[1]/div/div/div[2]/main/div/div/div/div[1]/div/section/div/div" //thread tweet xpath this.timeLine = "/html//section/div/div" //FIXED //thread tweet xpath
} }
this.tweet = (`${this.timeLine}/div`); //the div containing individual tweet content: (tweetXpath + '[1]') this.tweet = (`${this.timeLine}/div`); //the div containing individual tweet content: (tweetXpath + '[1]')
this.containsDivs = (`${this.timeLine}[count(div) > 1]`) //timeline conntaining divs this.containsDivs = (`${this.timeLine}[count(div[@data-testid='cellInnerDiv']) > 1]`) //timeline conntaining divs
this.path = `${this.tweet}[${orig == 'home' ? 1 : i}]`; this.path = `${this.tweet}[${orig == 'home' ? 1 : i}][@data-testid='cellInnerDiv']`; //FIXED
//the following xpaths follow an individual tweet xpath: (tweetXpath + '[1]' + variableXPath) //the following xpaths follow an individual tweet xpath: (tweetXpath + '[1]' + variableXPath)
this.urlCard = `${this.path}/div/div/div/article/div/div/div/div[*]/div[*]/div[*]/div[*]/div/div[2]/a` this.tweeterHandle = `${this.path}//article//div[@data-testid='User-Names']//a[not(contains(@href,'status'))]/div/span[starts-with(text(),'@')]` //FIXED /*text label containing tweeter's handle*/ //FIXED
this.tweeterHandle = `${this.path}/div/div/div/article/div/div/div/div[2]/div[2]/div[1]/div/div/div[1]/div/div/div[2]/div/div[1]/a/div/span[contains(text(),'@')]` /*text label containing tweeter's handle*/ this.tweeterName = `${this.path}//article//div[@data-testid='User-Names']/div[1]//a[not(contains(@href,'status'))]//div[1]/span` //FIXED /*text label containing tweeter's name*/ //FIXED
this.tweeterName = `${this.path}/div/div/div/article/div/div/div/div[2]/div[2]/div[1]/div/div/div[1]/div/div/div[1]/div/a/div/div[1]/span` /*text label containing tweeter's name*/ //this.quoteTweetHandle = `${this.path}//article/div/div/div/div[2]/div[2]/div[2]/div[2]/div[*]/div[2]/div/div[1]/div/div/div/div/div/div[2]/div[1]/div/div/div/span`; //xpath to text label that reveals if a tweet is a quote tweet (leads to the quote tweeted user's handle)
this.quoteTweetHandle = `${this.path}/div/div/div/article/div/div/div/div[2]/div[2]/div[2]/div[2]/div[*]/div[2]/div/div[1]/div/div/div/div/div/div[2]/div[1]/div/div/div/span`; //xpath to text label that reveals if a tweet is a quote tweet (leads to the quote tweeted user's handle) this.quoteTweetContent = `${this.path}//article//div[div/span[.="Quote Tweet"]]/div[2]` //FIXED /*xpath to locate entirety of Quote Tweeted Content*/
this.quoteTweetContent = `${this.path}/div/div/div/article/div/div/div/div[2]/div[2]/div[2]/div[2]/div[*]/div[2][div/div[1]/div/div/div/div/div/div[2]/div[1]/div/div/div/span]` /*xpath to locate entirety of Quote Tweeted Content*/
this.ageRestricted = `${this.path}/div/div/div/article//span/span[1]/span[contains(text(),'Age-restricted')]`; //xpath that reveals if tweet is age-restricted (& therefore not visible) this.ageRestricted = `${this.path}//article//span[span[span[contains(text(),'Age-restricted')]]]/a[contains(@href,'notices')]`; //FIXED //xpath that reveals if tweet is age-restricted (& therefore not visible)
this.pinnedTweet = `${this.path}/div/div/div/article/div/div/div/div[1]/div/div/div/div/div[2]/div/div/div/span[contains(text(),'Pinned')]` /*//xpath that reveals if tweet is pinned*/ this.pinnedTweet = `${this.path}//article//div[@data-testid='socialContext']/span[contains(text(),'Pinned')]`; //FIXED /*//xpath that reveals if tweet is pinned*/
this.tweetText = `${this.path}//div[@data-testid='tweetText']`; //xpath that leads to div containing all tweet text this.tweetText = `${this.path}//article//div[@data-testid='tweetText']`; //xpath that leads to div containing all tweet text //FIXED
this.urlCard = `${this.path}//article//a[div[contains(@data-testid,'card')]]` //FIXED
//this.emoji = this.path + "//img"; //xpath that leads to div containing all tweet text //this.emoji = this.path + "//img"; //xpath that leads to div containing all tweet text
this.tweetURL = `${this.path}//div[3]/a[contains(@href, 'status')]`; //xpath to tweet url this.tweetURL = `${this.path}//article//div[@data-testid='User-Names']//div[3]//a[contains(@href,'status')]`; //xpath to tweet url //FIXED
this.video = `${this.path}/div/div/div/article/div/div/div/div[*]/div[*]/div[*]/div[*]/div[1]/div[1]//video`; //xpath that leads to videos that are not parts of quoted content this.video = `${this.path}/html//section/div/div/div[*][@data-testid='cellInnerDiv']//article//div[2]/div[3]/div[1]/div[1]//video`; //FIXED //xpath that leads to videos that are not parts of quoted content
this.singleImage = `${this.path}//div[1]/div/div/div/div/a/div/div[2]/div/img[@alt='Image']`; //xpath to image that reveals if a tweet has one image this.singleImage = `${this.path}//article//div[2]/div[3]/div[1]/div[1]/div/div[1]//a//div[2]/div[@data-testid='tweetPhoto']/img`; //FIXED //xpath to image that reveals if a tweet has one image
this.multiImage = `${this.path}//div[2]/div[2]/div[2]/div[2]/div/div/div/div/div[2]/div/div[1]/div[1]//a/div/div/img[@alt='Image']`; //xpath to image that reveals if a tweet has more than one image //this.singleImage = `${this.path}//div[1]/div/div/div/div/a/div/div[2]/div/img[@alt='Image']`; //xpath to image that reveals if a tweet has one image
this.multiImage = `${this.path}//article//div[3]/div[1]//div[1]/div[2]/div/div[2]/div[1]//a//div[@data-testid='tweetPhoto']/img`; //FIXED //xpath to image that reveals if a tweet has more than one image
if (orig == 'home') { //home timeline only if (orig == 'home') { //home timeline only
this.detectThread = `${this.path}/div/div/div/article/div/a/div/div[2]/div/span`; //xpath to text label that reveals if a tweet is a part of a thread from home timeline this.detectThread = `${this.path}//article//a[contains(@href,'status')]//span[contains(text(),'Show this thread')]`; //xpath to text label that reveals if a tweet is a part of a thread from home timeline //FIXED
this.detectRT = `${this.path}/div/div/div/article/div/div/div/div[1]/div/div/div/div/div[2]/div/div/div/a/span`; //xpath to text label that reveals if a tweet is a retweet this.detectRT = `${this.path}//article//a[span[@data-testid='socialContext']]/span[contains(., 'Retweeted')]`; //xpath to text label that reveals if a tweet is a retweet //FIXED
} else if (orig == 'thread'){ //thread timeline only } else if (orig == 'thread'){ //thread timeline only
this.entryTweet = `${this.path}/div/div/div/article/div/div/div/div[3]/div[*]/div[@role='group']` /*xpath that reveals if tweet is open in thread //openThreadTweetTSXPath*/ this.notEntryTweet = `${this.path}//article//div[@data-testid='User-Names']//time` //FIXED /*xpath that reveals if tweet is open in thread //openThreadTweetTSXPath*/ //fixed
} }
//the following xpaths follow an individual tweet xpath and are used to find all images in a tweet with multiple images: (tweetXpath + '[1]' + multiImage1XPath + x + multiImage2XPath + y + multiImage3XPath) //the following xpaths follow an individual tweet xpath and are used to find all images in a tweet with multiple images: (tweetXpath + '[1]' + multiImage1XPath + x + multiImage2XPath + y + multiImage3XPath)
@ -39,9 +40,9 @@ class TweetPaths {
// 2,1 = second image // 2,1 = second image
// 2,2 = third image // 2,2 = third image
// 1,2 = fourth image // 1,2 = fourth image
this.multiImage1 = "//div[2]/div[2]/div[2]/div[2]/div/div/div/div/div[2]/div/div["; this.multiImage1 = "//article//div[3]/div[1]//div[1]/div[2]/div/div[";
this.multiImage2 = "]/div["; this.multiImage2 = "]/div[";
this.multiImage3 = "]//a/div/div/img[@alt='Image']"; this.multiImage3 = "]//a//div[@data-testid='tweetPhoto']/img";
} }
tweetElement(i, pathFromTweet) { tweetElement(i, pathFromTweet) {
let xPath = (this.path + pathFromTweet); let xPath = (this.path + pathFromTweet);