make_stats.sh

   1 #!/bin/sh
   2 # Auto generates stats article.
   3
   4 echo "making stats"
   5
   6 FILE_NAME="wiki_stats.md"
   7
   8 printf "# LRS Wiki Stats\n\nThis is an autogenerated article holding stats about this wiki.\n\n" > $FILE_NAME
   9
  10 printf -- "- number of articles: " >> $FILE_NAME
  11 ls *.md | wc -l >> $FILE_NAME
  12
  13 printf -- "- number of commits: " >> $FILE_NAME
  14 git rev-list --count --all >> $FILE_NAME
  15
  16 printf -- "- total size of all texts in bytes: " >> $FILE_NAME
  17 cat *.md | wc -c >> $FILE_NAME
  18
  19 printf -- "- total number of lines of article texts: " >> $FILE_NAME
  20 cat *.md | wc -l >> $FILE_NAME
  21
  22 printf -- "- number of script lines: " >> $FILE_NAME
  23 cat *.sh | wc -l >> $FILE_NAME
  24
  25 printf -- "- occurences of the word \"person\": " >> $FILE_NAME
  26 grep -o -i "person[s \n\.,]" *.md | wc -l >> $FILE_NAME
  27
  28 printf -- "- occurences of the word \"nigger\": " >> $FILE_NAME
  29 grep -o -i "[^\[]nigger" *.md | wc -l >> $FILE_NAME
  30
  31 printf "\nlongest articles:\n\n" >> $FILE_NAME
  32 ls -1hSs *.md | head -n 20 | sed "s/ *\([^ ]*\) \+\([^ ]*\)\.md/- [\2](\2.md): \1/g" >> $FILE_NAME
  33 printf "\n" >> $FILE_NAME
  34
  35 printf "top 50 5+ letter words:\n\n" >> $FILE_NAME
  36 cat *.md | sed "s/([^ ]\+\.md/ /g" | tr -cs "[:alpha:]" "\n" | \
  37   sed -r "/^.{,4}$/d" | tr "A-Z" "a-z" | sort | uniq -c | sort -nr | \
  38   head -n 50 | sed "s/ \+\([^ ]*\) \+\([^ ]*\)$/- \2 (\1)/g" >> $FILE_NAME
  39 printf "\n" >> $FILE_NAME
  40
  41 printf "latest changes:\n\n\`\`\`\n" >> $FILE_NAME
  42 git log --name-only | head -n 50 | grep "Date:\|.*\.md" | sed "s/\([^ ]*\.md\)/  \1/g" >> $FILE_NAME
  43 printf "\`\`\`\n" >> $FILE_NAME
  44
  45 printf "\nmost wanted pages:\n\n" >> $FILE_NAME
  46
  47 cat ./*.md | sed -n 's/.*\[.*\](\([^\)\(]*\)).*/\1/p' | sort | uniq -c | sort -nr > tmp.txt
  48
  49 rm tmp2.txt
  50 touch tmp2.txt
  51
  52 while read -r line; do
  53   fname=`echo "$line" | grep -o "[^ ]*\.md" -`
  54
  55   if ! [ -f $fname ]; then
  56     printf -- "- [" >> tmp2.txt
  57     printf $fname | sed "s/^\([^ ]*\)\.md/\1](${fname}) (/g" >> tmp2.txt
  58     echo $line | sed "s/^ *\([0-9]*\).*$/\1)/g" >> tmp2.txt
  59   fi
  60 done < tmp.txt
  61
  62 cat tmp2.txt | head -n 20 >> $FILE_NAME
  63 rm tmp.txt tmp2.txt
  64
  65 printf "\nmost popular and lonely pages:\n\n" >> $FILE_NAME
  66
  67 rm tmp.txt
  68 touch tmp.txt
  69
  70 for fname in *.md; do
  71   count=`grep -F -o "($fname)" *.md | wc -l`
  72   echo "$count $fname" >> tmp.txt
  73 done
  74
  75 rm tmp2.txt
  76
  77 cat tmp.txt | sort -n -r | head -n 30 > tmp2.txt
  78 echo "- ..." >> tmp2.txt
  79 cat tmp.txt | sort -n -r | tail -n 30 >> tmp2.txt
  80 cat tmp2.txt | sed "s/ *\([^ ]*\) \+\([^ ]*\)\.md/- [\2](\2.md) (\1) /g" >> $FILE_NAME
  81
  82 rm tmp2.txt
  83 rm tmp.txt
  84
  85 printf "\n" >> $FILE_NAME