Your scripts

Why don't we have a scripting language showcase thread? What did you write today, lainons? (And why?)

https://github.com/fukamachi/shelly Because I don't hate myself.

For a few seconds I thought you meant that this thread was for showcasing scripting languages lainons wrote. Considering that there was a compiler thread, it might be nice to have an interpreter thread sometime. Anyway, today I wrote a small chat in Javascript (backend with iojs/nodejs). I did it to remember a bit how websockets work since I wish to make a HTML5 multiplayer game sometime this summer.

>>6182 keep us updated on that game! also, right some things you learn along the way and about your process for the lainzine! also this might interest you http://gameprogrammingpatterns.com/ to read the book ^ wrote http://gameprogrammingpatterns.com/contents.html and >>6015

>>6193 God damn it, I accidentally reposted this and now it's in another thread too.

>>6176 I wrote some crap for reading binary as ASCII. (defparameter code-1 #b0100100100100000011011000110111101110110) (defparameter code-2 #b0110010100100000011110010110111101110101) (defun to-nums (code n) (let (buffer) (dotimes (i n) (push (ldb (byte 8 (* i 8)) code) buffer)) (nreverse buffer))) (format t "~a" (nreverse (append (mapcar #'code-char (to-nums code-2 5)) (mapcar #'code-char (to-nums code-1 5))))) (defparameter code-3 #b010010010010000001101100011011110111011001100101001000000111100101101111011101010010000001110100011011110110111100101110) (format t "~a" (nreverse (mapcar #'code-char (to-nums code-3 15))))

I got sick of using Lua/Scheme for scripting a while ago (Lua is ugly, Guile doesn't work on win32 so I lost portability) so I made a thing myself. Not sure if relevant, but if lainons are interested I'll gladly give advice. https://github.com/chameco/Solid

>>6223 c = 0; while c < 10 do print(c); c = c + 1; end; >implicit variable declarations But why.

Just a simple bash script which I put under startup programs. It checks if it is after 17:00 and if it is then it turns on my keyboard backlight. #!/bin/bash time=$(date +%k%M) if [[ "$time" -ge 1700 && "$time" -le 0759 ]];then echo "Backlight on" xset led 3 else echo "Backlight off" xset -led 3 fi

>>6194 um. are you sure you linked to the right comment? I posted that.

>>6226 You could use redshift hooks instead.

>>6234 nah…I have googled right now to see what redshift means…I always just keep my monitors at full brightness.I like it crisp.

>>6225 Because hash tables, anon. Because hash tables.

>>6241 You better have Lua's 'local' keyword or something like it.

>>6223 >>6225 Why don't you use 'for' instead? You don't need 'while' here since there's no 'guard' variable and you exactly know how many times you want to execute the command (10)

I have just copy-pasted together a script for retrieving Youtube urls from 4chan/8chan threads for sharing music. I can then open the playlist with a program like vlc. import sys import re import urllib.request from collections import OrderedDict if len(sys.argv)<3: print("usage: python3 "+sys.argv[0]+" url playlist.m3u") exit() #I need to fake my user-agent, otherwise I get a 403 req = urllib.request.Request( sys.argv[1], data=None, headers={ 'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36' } ) response = urllib.request.urlopen(req) html = response.read().decode('utf-8') matches = re.findall('https?\\:\\/\\/www\\.youtube\\.com\\/watch\\?v\\=[A-Za-z0-9._%+-]+', html) urls = list(OrderedDict.fromkeys(matches)) f = open(sys.argv[2], 'w') f.write("\n".join(urls)) f.close() Yes, I know, using regex for parsing html is horrible. Please tell me what I should do instead. Other feedback welcome

>>6726 Ooops, formatting error: the print("usage"....) and the exit() lines should be obviously indented. I thought lainchan kept the tabs...

>>6726 def extractlinks(html): soup = BeautifulSoup(html) anchors = soup.findAll('a') links = [] for a in anchors: links.append(a['href']) return links then filter all the urls with www.youtube.com in them

>>6749 [code]def extractlinks(html): return [a['href'] for a in BeautifulSoup(html).findAll('a')]

A soykafty bash script that's run as a cron job so I can have a batoto rss feed on my browser startpage. #!/bin/bash # Crappy xml parser than only really works for RSS because it breaks on attributes parse_line(){ str="${1/<\/*>/}"; str="${str/<*>/}" echo "$str" } # Fetches feed rss=$(curl $(cat ignore/url)) # Removes pointless CDATA tags rss=${rss//<![CDATA[/} rss=${rss//]]>/} # Get ready to loop title="" istitle=0 link="" finished=0 counter=0 # Clear file first tho echo "" > ignore/batoto.html # More complicated than it needs to be while read -r line; do # Separates into < + tag # Ex: <title> -> <title # Probably could be done a different cuter way parse=$(echo "$line" | cut -d ">" -f 1) # if it's a title make $title equal it or if it's a link make $link equal it -- nothing complicated if [[ $parse == "<title" ]] && [[ "$line" == *"English"* ]]; then title=$(parse_line "$line") title=${title/- English -/-} istitle=1 counter=$[counter+1] elif [[ $parse == "<link" ]] && [[ $istitle -eq 1 ]]; then link=$(parse_line "$line") istitle=0 finished=1 fi # If there's a title and a link, clear both and format it to html then spit it out if [[ $finished -eq 1 ]]; then title1=$(echo "$title" | cut -d "-" -f 1) title1=${title1%?}: title2=$(echo "$title" | cut -d "-" -f 2) echo "<li><a target=\"_blank\" href=\"$link\"><span class=\"bold\">$title1</span>$title2</a></li>" >> ignore/batoto.html finished=0 # If there's 8 entries now, quit out the program if [[ $counter = 8 ]]; then exit; fi fi # pipes content of $rss to the loop done <<< "$rss"

I write and maintain automation scripts for a security software company (won't say who but they are fortune 500) wish I could share them but I can't :3 this is more of a college job to me than anything though, I don't want to do scripts for the rest of my life

require "mechanize" require "nokogiri" unless ARGV.length == 1 puts "Usage: scrapper url" exit end agent = Mechanize.new { |agent| agent.user_agent_alias = "Mac Safari" } url = ARGV[0].chomp.lstrip html_doc = Nokogiri::HTML(agent.get(url).body) files=[] if /https:\/\/boards\.4chan\.org/.match(url) do (html_doc.css('div.file')).each{ |ito| files.push "https:#{ito.css("a")[0]['href']}"} end elsif /https:\/\/7chan\.org/.match(url) do (html_doc.css('p.file_size')).each{ |ito| files.push "#{ito.css("a")[0]['href']}"} end else (html_doc.css('p.fileinfo')).each{ |ito| files.push "#{ito.css("a")[0]['href']}"} end if /\Ahttp/.match(files[0]) do files.each{ |x| `wget #{x}`} end else prefix = (url.split '/')[0..-4].join '/' files.each{ |x| `wget #{prefix}#{x}`} end how in the fuarrrk do i add syntax highlighting for this? Made it a bit ago. Added support for 7chan and 4chan today. Tested on 8chan, 4chan, lainchan and 7chan.

>>8634 #!/Users/normie/.rvm/rubies/ruby-2.2.1/bin/ruby require "mechanize" require "nokogiri" unless ARGV.length == 1 puts "Usage: scrapper url" exit end agent = Mechanize.new { |agent| agent.user_agent_alias = "Mac Safari" } url = ARGV[0].chomp.lstrip html_doc = Nokogiri::HTML(agent.get(url).body) files=[] if /:\/\/boards\.4chan\.org/.match(url) do (html_doc.css('div.file')).each{ |ito| files.push "https:#{ito.css("a")[0]['href']}"} end elsif /:\/\/7chan\.org/.match(url) do (html_doc.css('p.file_size')).each{ |ito| files.push "#{ito.css("a")[0]['href']}"} end elsif /:\/\/boards\.420chan\.org/.match(url) do (html_doc.css('span.filesize')).each{ |ito| files.push "#{ito.css("a")[0]['href']}"} end else (html_doc.css('p.fileinfo')).each{ |ito| files.push "#{ito.css("a")[0]['href']}"} end if /\Ahttp/.match(files[0]) do files.each{ |x| `wget #{x}`} end else prefix = (url.split '/')[0..-4].join '/' files.each{ |x| `wget #{prefix}#{x}`} end made it a tad better, added support for 420chan and https or http links now works. basically just use it as ./scriptname https://desiredurl

Script to play mpv, in Firefox + Vimperator, I can just hit a key, and it'll play videos (mainly youtube) on the page in mpv instead of using flash or mozplugger. Hit 'Q' in Vimperator: nnoremap Q y:execute "!urxvtc -e '/home/user/.local/bin/yt'" #!/usr/bin/ruby def print_help puts "Available flag options: \n\ flag | intention | command \n\ ------------------------------------------------------------------ \n\ -nv | 'no video' | /usr/local/bin/mpv --no-video \n\ -c | 'check' | youtube-dl --list-formats --no-playlist \n\ -md | 'music download | youtube-dl -x --audio-format best -f best \n\ -q | 'query' | /usr/local/bin/mpv --ytdl-format= \n\ -d | 'download' | youtube-dl --no-playlist \n\ none | 'default' | /usr/local/bin/mpv --ytdl-format=18" end link = `xclip -out` if ARGV[0] == '-h' print_help elsif link =~ /youtube/ puts "#{link} contains yt" if ARGV[0] == '-nv' puts 'now playing: no video option' exec "/usr/local/bin/mpv --no-video \"#{link}\"" elsif ARGV[0] == '-c' puts 'now checking formats' exec "youtube-dl --list-formats --no-playlist \"#{link}\"" elsif ARGV[0] == '-md' puts 'now downloading (music only)' exec "youtube-dl -x --audio-format best -f best \"#{link}\"" elsif ARGV[0] == '-q' puts 'now playing with chosen quality of ' + ARGV[1] exec '/usr/local/bin/mpv --ytdl-format=' + ARGV[1] + " \"#{link}\"" elsif ARGV[0] == '-d' puts 'now downloading (video)' exec "youtube-dl --no-playlist \"#{link}\"" else puts 'now playing: low quality' exec "/usr/local/bin/mpv --ytdl-format=18 \"#{link}\"" end else puts "#{link} is not a youtube link" print_help end puts link

Selects a random wallpaper through Nitrogen #!/usr/bin/env python3 ## sets a random wallpaper through Nitrogen ## Nitrogen: http://projects.l3ib.org/nitrogen/ # the program needs a file in the home directory called '.rwall.conf', # create the file and add the path of where you keep your wallpaper images # EXAMPLE: path = /home/user/Pictures/Wallpapers/ # and run the program, or make it execute at the startup of your window manager import os, random, io, sys import subprocess def main(): conffile = ".rwall.conf" confpath = os.path.expanduser("~/.rwall.conf") filespath = None # get conf file or create it if os.path.exists(confpath): pass else: # write defaults f = open(confpath, 'w') f.write("path = /path/to/your/wallpapers/\n") f.close() # read conf file and set vars accordingly with open(confpath, 'r') as r_config: config_lines = r_config.readlines() for var in config_lines: var = var.strip().split() filespath = var[2] # get path string if filespath == "/path/to/your/wallpapers/": sys.exit("Did not supply a valid directory in " + confpath) files = os.listdir(filespath) randomfile = random.choice(files) chosenpath = os.path.join(filespath, randomfile) process = subprocess.call(['nitrogen', '--set-scaled', chosenpath]) main()

Today I made it way easier to run my tiny personal wiki software from a local machine instead of a whole webserver setup. I've been working on it on and off for a few years. It's all in one script file (except templates and styles etc.) and written in Python. Pages/articles are written in and rendered from ReStructured Text and saved in flatfiles. It supports simple password authentication, all of the configuration is at the top of the script, and you can spawn an instance at home easily now using thttpd and the example spawn script. Project page and description: https://github.com/pariahsoft/PyWW Main script: https://github.com/pariahsoft/PyWW/blob/master/index.cgi (Yes CGI for ease of personal use.) Demo Site (wiped every 24 hours): http://pariahsoft.com/pyww/ The code is too long to paste here but it's less than 300 lines.

>>8636 > youtube > instead of using flash or mozplugger. what's wrong with html5?

I made a sh script I know it isn't ruby or some other scripting language with better features but here you go. It's for downloading/streaming songs off the nigge.rs website (comedic songs, mostly.) #!/bin/sh #Licensed under the GPLv3. Requires mpv or mplayer. usage() { echo "Usage: nigge.rs (-d) songname" echo "or, for a list of all songs," echo "nigge.rs -q" } if [ $# -eq 0 ]; then usage; exit 1 fi download=false if [ "$1" = -d ]; then download=true shift fi query=false if [ "$1" = -q ]; then query=true download=false stream=false shift fi dl() { wget http://nigge.rs/media/$1/$1.ogg ;} stream() { mpv http://nigge.rs/media/$1/$1.ogg || mplayer http://nigge.rs/id/$1/$1.ogg ;} if $download; then dl $1 & stream $1 elif $stream; then stream $1 else shift fi if $query; then wget http://nigge.rs awk -F '[<>]' '/<a / { gsub(/<b>/, ""); sub(/ .*/, "", $3); print $3 } ' index.html rm -f index.html fi

>>8977 I fell for the mpv meme, and wrote this before HTML5 on YouTube became a thing. Just inertia and ignorance, really.

>>8997 >>8977 >what's wrong with html5 >[nothing] html5 is not very secure and makes you very trackable: 1. Browsers are leaky buckets security wise due to their big attack surface. Some media decoder is always bugged and allows people with 0days to crawl out of the sandboxes (e.g. recent pdf issue with adobe plugin on windows, of course plugins on windows are especially bad in that respect) 2. the javascript API of browsers (which you need for youtube etc. to work) knows all kinds of soykaf about your computer, which you might not necessarily want. these are two things that are not great about html5, just for completion's sake

How would you lainons approach downloaded content from a series of URLs that were all indexed on a single URL. So the structure would go something like this: Index URL w/ links -- Link1 -Embedded Content we want -- Link 2 -More juicy content .... -- Link n - n content Yeah, we want to download that specific bit of content without having to follow/download every damn HTML page. Ideas?

>>9064 With python script using httplib2 and BeatifulSoup.

W H E R E  I S  P A G E  W I T H  M A R K U P H E R E I S P A G E W I T H M A R K U P

>>9064 curling the page into some perl regexes

Bash function to show a notification window at or after a specified time period. at-msg() { read -p "Examples: ‘+10 mins’, ‘19:30’."$'\n'"When? > " when read -p "Text message: > " msg date --date="$when" +"%H:%M" >/dev/null || return 4 at "`date --date="$when" +"%H:%M"`" <<<"DISPLAY=$DISPLAY Xdialog --msgbox \"\n $msg \n\" 200x100" } How to use: $ at-msg Examples: ‘+10 mins’, ‘19:30’. When? > +20 min Text message: > Kettle! warning: commands will be executed using /bin/sh job 2017 at Sat Sep 12 22:38:00 2015 Requires atd running and Xdialog installed.

Bash function to copy the current playlist from mpd to a specified folder (music player or an SD card) # Copies current MPD playlist to a specified folder. # $1 — where to copy copy-playlist() { err() { echo "$1" >&2; echo $2; } # $1 — message; $2 — return code local dest="$1" cur_pl='current.m3u' pl_dir pl library_path got_a_sane_reply pl_dir=`sed -nr 's/^\s*playlist_directory\s+"(.+)"\s*$/\1/p' ~/.mpd/mpd.conf` library_path=`sed -nr 's/^\s*music_directory\s+"(.+)"\s*$/\1/p' ~/.mpd/mpd.conf` eval [ -d "$pl_dir" ] \ || return `err 'Playlist directory is indeterminable.' 3` eval [ -d "$library_path" ] \ || return `err 'Path to music library is indeterminable.' 4` pl="$pl_dir/$cur_pl" mpc save "$cur_pl" # MPD_HOST must be set in the environment [ -f "$pl" ] || return `err 'Playlist wasn’t saved' 5` [ -w "$dest" ] && { while read filepath; do filepath="${filepath/$library_path/}" cp -v "$library_path/$filepath" "$dest" || { unset got_a_sane_reply until [ -v got_a_sane_reply ]; do read -n1 -p 'Error on copying. Continue? [Y/n] > '; echo case "$REPLY" in y|Y|'') got_a_sane_reply=t;; # And skip further messages n|N) break 2;; *) echo 'Y or N.';; esac done } done < "$pl" : }|| return `err "Destination dir ‘$dest’ is not writable." 6` }

>>9361 > stripping tabs from <code> This engine never fails to amaze me with each step I get to know it.

>>9362 I'm pretty sure tabs are just a different width here. This is why you should always use spaces. fuarrrk tabs.

>>9398 fuck spaces

while true; do espeak "rofl"; done

Here's something I never bother saving I use for when I add new books or whatnot to my organized book directory:(map 'string (lambda (c) (case c (#\ #\-) (t (char-downcase c)))) "")Example:CL-USER> (map 'string (lambda (c) (case c (#\ #\-) (t (char-downcase c)))) "The GNU C Reference Manual") "the-gnu-c-reference-manual" CL-USER>