shithub: rc

ref: 854c60f24c7671a31f61cef272363df2ddf551f0
dir: /epub2txt/

View raw version 
#!/bin/rc
# epub2txt - convert epub to text
# usage:  epub2txt file.epub
# depend: epub2html

# set some defaults
rfork e
if(! ~ $#* 1) exit usage
keep=yes
name=`{basename $1 | sed 's/\.[Ee][Pp][Uu][Bb]//'}
if(! test -f $name.html){
    keep=no
    epub2html $* || exit $status
}
ifs='
'

# convert extracted epub to text
> $name.txt
for(file in `{awk -F" '/<a/ { print $2 }' $name.html})
    html2ms < $"file | deroff | fmt >> $name.txt
if(~ $keep no)
    rm -rf $name.html $name^_files