gigaclone/mkurllist

#!/bin/bash

dir="$PWD"
tmp="$dir/tmp"
rm -r $tmp
mkdir -p $tmp
cd $tmp
{
  extractlog() {
    tar -xzf logs.tgz
    cd log
    for f in ./*;do {
      echo "Working... for $f"
      sed -i '/200\|206\|302\|301/!d;s/\sHTTP.*//g;s/.*\s/http:\/\/profile\.gigaset\.net/g;/http:\/\/profile\.gigaset\.net\"\"/d;s/?.*//g;s/\.net\/device/\.net\/chagall/g;s/^\.$//g' $f
      echo "Remove duplicates for $f"
      awk '!seen[$0]++' $f>>$tmp/tmp
    };done

    md5sum $tmp/logs.tgz >$dir/.md5sum
    rm $tmp/logs.tgz
    rdupes $tmp/tmp
  }
  
  rdupes() {
    [ "$1" != "" ] && {
      echo "Remove all duplicates for $*"
      sed -i 's/\/\//\//g;s/\/$//g' $dir/urllist
      awk '!seen[$0]++' $* $dir/urllist >$dir/final

      mv $dir/final $dir/urllist
      rm $*
    }
    for dl in $(cat $dir/urllist); do {
     a=$(curl -w "%{url_effective}\n" -L -f -s -I -S "$dl" -o /dev/null)
     [ $? = "0" ] && echo "$a" >> $dir/final
    };done
    mv $dir/final $dir/urllist
    awk '!seen[$0]++' $dir/urllist >$dir/final
    mv $dir/final $dir/urllist
  }
  
  extractbin() {
    echo "Extracting urls..."
    cd $tmp
    for f in $(grep "\.bin" $dir/urllist);do {
      [ $(wget -S --spider "$f" 2>&1 | sed '/Length/!d;s/Length\: //g;s/\s.*//g;s/\s//g;s/[^0-9]*//g' | tr -d "\n") -lt 52428800 ] && {
       file=$(wget -qO- "$f" | strings)
       echo "$file" | sed '/http:\/\//!d;/profile.gigaset.net\|update.gigaset.net/!d;s/.*http:\/\//http:\/\//g;s/update\.gigaset/profile\.gigaset/g'>>$tmp/tmp
       baseurl="$(dirname "$f")"/
       url=$(echo "$file" | sed '/http/d;/\.bin/!d')
       for e in $url;do echo "$baseurl$e">>$tmp/tmp;done
      }
    }; done
    rdupes $tmp/tmp
  }
  
  
  treegen() {
    echo "Creating tree..."
    cd $dir
    rm -r tree
    mkdir tree
    cd tree
    for f in $(sed 's/http:\/\/profile.gigaset.net\///g;s/^\/*//g' $dir/urllist);do {
      mkdir -p $(dirname "$f")&>/dev/null
      touch "$f"
    };done
    for f in $(find . -type d);do a=$(echo "$f" | sed 's/\.\///g'); cd $f; index="$(tree -a -T "profile.gigaset.net" -C -H "http://profile.gigaset.net/"$a -L 1 | sed 's/net\/\.\//net\//g;s/class=\"DIR\" href=\"http\:\/\/profile\.gigaset\.net/class=\"DIR\" href=\"http\:\/\/daniil\.it\/gigaclone\/tree/g' | sed 's/<\/a> /<\/a><br>\
/g')"; echo "$index">index.html; cd $OLDPWD; done
    find . -not -iname "index.html" -type f -a -not -name "." -exec rm {} \;
  }
  
  wget http://profile.gigaset.net/logs.tgz -qO logs.tgz
  [ "$(md5sum $tmp/logs.tgz)" != "$(cat $dir/.md5sum)" ] && {
    extractlog
    extractbin
  }
  [ "$1" = 'check' ] && { extractbin; }
  cd $dir
  sort urllist > final
  mv final urllist
  treegen
  
  cd $dir
  rm -rf $tmp/
  mkdir $tmp
  
  git clone https://$GH_TOKEN@github.com/danog/gigaclone.git $tmp/git 
  cd $tmp/git
  rm -r tree 
  cp -a $dir/* .
  rm -rf tmp gigaclone.log
  git add -A
  git commit -m "Updated url list"
  git push origin master &>/dev/null
  
  git checkout gh-pages
  rm -r tree
  cp -a $dir/tree .
  git add -A
  git commit -m "Updated dir tree"
  git push origin gh-pages &>/dev/null
  
  cd $dir
  echo "Clean up."
  rm -rf $tmp
} &> $dir/gigaclone.log