extraction chains

[MrTpat]

So this introduces a concept of "extraction chains" for our job parsing. Certain issues: (#114) are happening because there are parsing changes necessary when LinkedIn changes something on their end. Instead of changing our parsing code each time, we can add a new parser here when we encounter some new linkedin HTML format.

Shoutout to @pafthinder for his initial correct implementation which I've included here. This fixes THAT issue and fixes issues I had been having myself.

[MrTpat]

@feder-cr opened a new PR for V4 as you asked.

[feder-cr]

@MrTpat please explain better the PR

[MrTpat]

@feder-cr was the description of the original PR not sufficient?

[feder-cr]

@MrTpat copy and paste here

[MrTpat]

@feder-cr updated.

[feder-cr]

@MrTpat please on V4

[MrTpat]

@feder-cr it is on v4, but the tests are failing. i don't quite have the time to fix the tests, if someone else wants to pick this work up please go ahead!

[MrTpat]

@feder-cr i fixed the tests. can you merge?

[RushiChaganti]

Hey @MrTpat, I’ve gone over the files you modified, but I believe this method will also fail if LinkedIn updates their class names. So, I think neither your approach nor mine is ideal for the program. Let's have a conversation to figure out a better solution. What do you think? No offense intended.

[MrTpat]

@RushiChaganti the idea is that once linkedin updates their class names, someone can add an extraction chain implementation that adheres to that class change.

[feder-cr]

@MrTpat very good idea, but please on V4

[MrTpat]

@feder-cr im confused what you mean by "please on V4". are you asking me to rebase?

[feder-cr]

i need this on https://github.com/feder-cr/Auto_Jobs_Applier_AIHawk/tree/v4

[MrTpat]

@feder-cr done.

[moudimash99]

@MrTpat I'm interested did this get published in the end?

[surapuramakhil]

@moudimash99 there are not in release yet. PR to track & participate in reviews providing your feedback.

#813