pdf.sum

harvest是一个下载html网页得机器人

#!/bin/sh
#
#  Pdf.sum - takes a filename as the first argument.
#
#  Usage: Pdf.sum filename
#
#  Using Adobe's Acrobat:
#  Create a postscript format file from the input.
#  Pass that filename to the postscript summarizer.
#
#if [ -z $TMPDIR ]; then
#    TMPDIR=/tmp
#fi
#
#acroread -toPostScript -pairs $1 $TMPDIR/$$.ps
#PostScript.sum $TMPDIR/$$.ps
#rm -f $TMPDIR/$$.ps
#
#  Using Derek B. Noonburg's xpdf:
#
#  pdftotext fails to process some pdf files correctly and
#  creates huge text files.
#
#  Use this if your Pdf files are correctly processed by xpdf.
#
#pdftotext $1 - | wrapit "Partial-Text"
#
#  Use this if you have GNU textutils installed.
#
#pdftotext $1 - | head -c 10485760 | wrapit "Partial-Text"
#
#  This should work for all systems.
#
pdftotext $1 - | tr "\t" " " | tr -s " \n\r\f" | fmt -s -w 160 | head -n 65000 | wrapit "Partial-Text"