Add final state from 2018-10-07
[enwiki-links-graph.git] / convert-to-plain.sh
1 #!/bin/bash
2
3 # Compile flex program to regex-convert the XML to plain text
4 #https://unix.stackexchange.com/a/413684
5 flex -o links-xml-to-plain.c links-xml-to-plain.l
6 gcc -O3  -o links-xml-to-plain links-xml-to-plain.c -lfl
7
8 # Convert to plain text
9 lzop -dc enwiki-links.xml.lzo | pv | tail -c +45 | ./links-xml-to-plain > enwiki-links-plain.txt
10
11 # Extract titles
12 grep ^~~~~ enwiki-links-plain.txt | sed "s/^~~~~//g;te;d;:e" | sort > titles.txt
13
14
15
16 gcc -O3 -g -o links-plain-to-binary links-plain-to-binary.c
17 time ./links-plain-to-binary
18
19 gcc -O3 -g -o links-outgoing-to-incoming links-outgoing-to-incoming.c
20 time ./links-outgoing-to-incoming