diff options
Diffstat (limited to 'convert-to-plain.sh')
-rw-r--r-- | convert-to-plain.sh | 20 |
1 files changed, 20 insertions, 0 deletions
diff --git a/convert-to-plain.sh b/convert-to-plain.sh new file mode 100644 index 0000000..e7bd378 --- /dev/null +++ b/convert-to-plain.sh @@ -0,0 +1,20 @@ +#!/bin/bash + +# Compile flex program to regex-convert the XML to plain text +#https://unix.stackexchange.com/a/413684 +flex -o links-xml-to-plain.c links-xml-to-plain.l +gcc -O3 -o links-xml-to-plain links-xml-to-plain.c -lfl + +# Convert to plain text +lzop -dc enwiki-links.xml.lzo | pv | tail -c +45 | ./links-xml-to-plain > enwiki-links-plain.txt + +# Extract titles +grep ^~~~~ enwiki-links-plain.txt | sed "s/^~~~~//g;te;d;:e" | sort > titles.txt + + + +gcc -O3 -g -o links-plain-to-binary links-plain-to-binary.c +time ./links-plain-to-binary + +gcc -O3 -g -o links-outgoing-to-incoming links-outgoing-to-incoming.c +time ./links-outgoing-to-incoming |