#headless-chrome #web #web-page #grep #browser #terminal #read

app webgrep

grep the web: a full-browser-spec search-focused ultra-simple way to read the web without having to leave the terminal

14 unstable releases (3 breaking)

0.4.3 Jul 26, 2024
0.4.2 Jul 26, 2024
0.3.4 Jul 26, 2024
0.2.2 Jul 5, 2024
0.1.2 Jul 2, 2024

#1117 in Command line utilities

Download history 154/week @ 2024-07-28 7/week @ 2024-09-15 60/week @ 2024-09-29 1/week @ 2024-10-06

576 downloads per month

MIT license

29KB
398 lines

webgrep: the simplest way to surf the web

webgrep is the fastest way to read the web right from your terminal, with all the features of regular grep. webgrep supports full browser webpage rendering using headless-chrome, as well as more basic http request/html parsing. it can even read pdfs. webgrep can render javascript, search links trees recursively, read webpages with regex, and more. (note i know the order for source and search is reversed from original grep, but its so much better trust me)

how to use:

download it with cargo, then use it like any other unix command: cargo install webgrep webgrep <search (url)> <search or regex expression> recommended: use headless-chrome (see how below), since many sites only work with javascript/other browser-required features. to use headless-chrome with webgrep, either use the -c/--chrome flag, or set the following enviromental variable: (note, using chrome is much slower [~2sec/request vs .25sec/request]) export WEBGREP_ALWAYS_USE_CHROME=1

flags:

  • -r/--recursive OPTIONAL, iterate through/search all links on the base url for a set depth (see below), provide a value for depth (how many branches of the link tree to explore)
  • -p/--pathcontains OPTIONAL, plaintext string that must be in a path for it to be explored (useful to prevent recursive search from going too far astray)
  • -o/--samehost OPTIONAL, only explore webpages on the same host (ie same domain)
  • -c/--chrome OPTIONAL, use chrome instead of just basic http request (needed for sites with js)
  • -i/--insensitive OPTIONAL, search case-insensitivly
  • -b/--break_when_found OPTIONAL, stop recursion when search nets result

Examples:

recursively search through a college course catalog for a keyword (case insensitive):

webgrep https://https://oxy.smartcatalogiq.com/en/2021-2022/catalog/course-descriptions/ "(?i)ethical" --samehost -p course-descriptions

read a react.js webpage using headless chrome:

webgrep https://quinnpatwardhan.com/aboutme frisb -c (or, with env variable set): webgrep https://quinnpatwardhan.com/aboutme frisb

bugs/contributing

please report any bugs you encounter on the 'issues' page. we welcome any contributions, just make a pr!

Dependencies

~34–51MB
~811K SLoC