xidel VS pup

You could try Xidel[1]. It supports JSON, XML and HTML using XPath/XQuery 3.1

It has some extensions to the standard that are pretty nice (JSONiq, CSS selectors, html “template” matching), but you can limit it to just standard XPath/XQuery if you like.

I recommend getting the nightly v .99 build if you give it a try, the stable .98 version is pretty old and I’ve had no issues with .99

1. https://www.videlibri.de/xidel.html

Windows binary of Xidel (https://github.com/benibela/xidel) a commandline tool to download and extract data from HTML pages

Xidel for extraction and pagination

Download an entire subreddit to JSON Lines with Xidel:

I believe that you can't read from a url with batch alone. you'll need a helper app like xidel https://www.videlibri.de/xidel.html

I have not played with this, but it looks like xidel might allow you to do this in Bash. jc also has a URL string parser that could be used in such a script.

I built Xidel for such authenticated requests:

And lnk=$(curl -s https://www.selfstudys.com$url |grep "PDFFlip" | cut -d '"' -f 6) to lnk=$(curl -s https://www.selfstudys.com$url | pup "div#PDFF attr{source}" ) here pup will print content of source attribute from div tag with id PDFF i dont know that much about html & css so this is what i came up with. but i am sure you can also select class & make list of suburls from them. check out the video from bugswriter on pup or read docs from git hub for more info github link: https://github.com/ericchiang/pup

jq is pretty amazing. If you are comfortable with its jquery-like CSS selector syntax, then I should also mention a couple similar cli utilities that apply it to HTML: htmlp and pup.

Regex is not a great tool for parsing web pages. Open up a browser dev tools window and select a bit of the page. Right click > copy... XPath expression or CSS selector. A proper web scraping tool will accept either of those. No muss, no fuss. You can even use simple command line tools: xpath or pup

This is HTML to JSON, written in Rust, and there's also pup[1] which I found out about just the other day on HN[2] which uses a very similar syntax (CSS selectors) but outputs HTML and is written in Go.

I can see room for both though it would interesting to have a more detailed comparison to go on (e.g. types of HTML, speed etc).

[1] https://github.com/ericchiang/pup

[2] https://news.ycombinator.com/item?id=33805732

It looks like the project became inactive for a bit and there are alternatives such as htmlq, etc. https://github.com/ericchiang/pup/issues/150

Another tool worth mentioning is pup - it can produce JSON output which means you can pipe it to jq