Generalize script for extracting sentences

Can extract text using detex now (as an option).
This commit is contained in:
Filip Gralinski 2022-06-06 19:54:30 +02:00
parent d0acd0c8d4
commit bc7b4fdf89

View File

@ -1,3 +1,15 @@
#!/bin/bash
bash helpers/pdf-to-plain-text.sh "$1" | perl helpers/strip-references.pl | perl -pne 'chomp $_; $_.=" "' | python3 -m syntok.segmenter
input_file="$1"
method="$2"
extract_text() {
if [[ "$method" == "from-tex" ]]
then
detex "$input_file" | egrep '\S' | grep -v 'unsrt' | perl -pne 's/^\s+| +$//g'
else
bash helpers/pdf-to-plain-text.sh "$input_file" | perl helpers/strip-references.pl | perl -pne 'chomp $_; $_.=" "'
fi
}
extract_text | python3 -m syntok.segmenter | egrep '\S'