Skip to content

Latest commit

 

History

History
84 lines (53 loc) · 3.14 KB

README.md

File metadata and controls

84 lines (53 loc) · 3.14 KB
Raw

DrunkLineCount

Worst -- and most elegant -- ways to count line number of a file.

greg

The clue of this solution is to wrap lines several times reducing the lenght of lines each time. By adding new \n, we add noise to the file. The goal is to detect when the signal / noise ratio (SNR) varies the most to determine the average lenght of lines in a sample. Extrapoling this information leads to an estimation of the total line number in a file.

This program shows estimations "near" as 30% to 10% on files > 1000 lines.

marco

Parce que greg veut un README.

Read the fucking code !

Pour le one-liner perl, juste exécute le

$|=4096: lire par bloc de 4k

s/\n/fonction()/eg: remplace Tous les \n (/g) par le résultat de la fonction (/e dit que la partie de droite est du code exécutable)

Pour le psql

\lo_import :file -_> Charge le fichier pointé par la variable file (passée par le .sh) dans un LOB

\set myoid :LASTOID -> stocke dans la variable myoid le dernier OID retourné par le serveur (par lo_import)

\timing -> affiche les temps d'exécution

-- Fastoche: compter le nombre d'enregistrements retoruné par regexp_split_to_table, splitté par \n -- celle ci reçoit ses données de convert_from (conversion d'une donnée binaire en chaîne, ici par utf8) -- qui reçoit ses données d'un loread d'1Go d'un coup -- sur un descripteur de LOB ouvert par lo_open, en lecture (x'40000') select count(*) from (select regexp_split_to_table(convert_from(loread(lo_open(:myoid,x'40000'::int),1073741800),'utf8'),E'\n')) as tmp;

\timing -> désactive l'affichage des temps d'exécution \lo_unlink :myoid -> détruit le LOB

sim51

The java solution : the beauty of this code is to have an infinity loop that raise an exception with the number of line of the file

The netcat solution : we start a netcat server that count the number of instruction. When we push it a file, every line is consider as an instruction, so we have the number of line of the file

alex

  1. grep_oc

Counts lines with a simple grep : ./grep_oc.sh

  1. ls-l

Counts lines with a sed replacement and a long ls parse with awk : ./grep_oc.sh

  1. sed_bc.sh

Counts lines with a sed replacement and a bc calculation

  1. brainfuck directory :)
  • drafts : with a lot of drafts to work on the main BF programs : countlineinrevhex.bf
  • clean_bf.sh <file.bf> : Simple script that's printing a BF script without comments and newlines
  • bf.rb : A brainfuck interpreter in ruby (original: http://www.stephensykes.com/bf.html) Usage : ruby bf.rb <file.bf>
  • bf_to_c.sh : Brainfuck to C interpreter Usage : ./bf_to_c.sh <file.bf> <file.c> Produce a C code from the BF one and compile it
  • reverse.bf : Reverse a String
  • countlineinrevhex.bf : Counts line and print it in reversing hexa

Usage with ruby BF interpreter (37min for 35k lines): cat | ruby bf.rb countlineinrevhex.bf | ruby bf.rb reverse.bf | cat <(echo -n "ibase=16; ") - <(echo) | bc

Usage with C compiler (18s for 674k lines): ./bf_to_c.sh countlineinrevhex.bf countlineinrevhex.c ./bf_to_c.sh reverse.bf reverse.c cat | ./countlineinrevhex | ./reverse | cat <(echo -n "ibase=16; ") - <(echo) | bc