请教一个小问题：如何把html网页文件输出成txt文本格式？
如何把html网页文件输出成txt文本格式？要求段落格式不变。望能人解答指点！ #! /bin/bash# #############################################################################
NAME_="html2txt"
PURPOSE_="convert html file to ascii text; write the converted file to disk"
SYNOPSIS_="$NAME_ [-vhmlr] <file> [file...]"
REQUIRES_="standard GNU commands, lynx"
VERSION_="1.1"
DATE_="2004-04-18; last update: 2005-03-03"
AUTHOR_="Dawid Michalczyk <[email protected]>"
URL_="www.comp.eonworks.com"
CATEGORY_="text"
PLATFORM_="Linux"
SHELL_="bash"
DISTRIBUTE_="yes"
# #############################################################################
# This program is distributed under the terms of the GNU General Public License
usage () {
echo >&2 "$NAME_ $VERSION_ - $PURPOSE_
Usage: $SYNOPSIS_
Requires: $REQUIRES_
Options:
-r, remove input file after conversion
-v, verbose
-h, usage and options (help)
-m, manual
-l, see this script"
exit 1
}
manual () { echo >&2 "
NAME
$NAME_ $VERSION_ - $PURPOSE_
SYNOPSIS
$SYNOPSIS_
DESCRIPTION
$NAME_ converts ascii files with html content to plain text. It replaces the
previous suffix, if any, with a \"txt\" suffix. It skips the following files:
- binary files
- directories
- files that already have the same name as <input_file>.txt
Option -r, removes the input file after conversion.
EXAMPLES
Use find with xargs to run the script recursively on multiple files. For
example, to convert all html files to text recursively:
find . -name \"*.html\" | xargs $NAME_
"; exit 1; }
# arg check
[ $# -eq 0 ] && { echo >&2 missing argument, type $NAME_ -h for help; exit 1; }
# var initializing
rm_html=
verbose=
# option and argument handling
while getopts vhmlr options; do
case $options in
r) rm_html=on ;;
v) verbose=on ;;
h) usage ;;
m) manual | more; exit 1 ;;
l) more $0 ;;
\?) echo invalid or missing argument, type $NAME_ -h for help; exit 1 ;;
esac
done
shift $(( $OPTIND - 1 ))
# check if required command is in $PATH variable
which lynx &> /dev/null
[[ $? != 0 ]] && { echo >&2 the required \"lynx\" command is not in your PATH variable; exit 1; }
# main
for a in "$@"; do
file $a | grep -q text # testing if input file is an ascii file
[ $? == 0 ] && is_text=0 || is_text=1
if [ -f $a ] && [[ ${a#*.} != txt ]] && (( $is_text == 0 )); then
if [ -f ${a%.*}.txt ]; then
echo ${NAME_}: skipping: ${a%.*}.txt file already exist
continue
else
[[ $verbose ]] && echo "${NAME_}: converting: $a -> ${a%.*}.txt"
lynx -dump $a > ${a%.*}.txt
[[ $? == 0 ]] && stat=0 || stat=1
[[ $stat == 0 ]] && [[ $verbose ]] && [[ $rm_html ]] && echo ${NAME_}: removing: $a
[[ $stat == 0 ]] && [[ $rm_html ]] && rm -f -- $a
fi
else
[[ $is_text == 1 ]] && echo ${NAME_}: skipping: $a not an ascii file
fi
done
这shell“长” 多谢版主，我down下去好好研究。。。 curl?
页:
[1]