X-Git-Url: http://www.club.cc.cmu.edu/~cmccabe/cgi-bin/gitweb.cgi?p=cmccabe-bin;a=blobdiff_plain;f=pdfgrep.sh;fp=pdfgrep.sh;h=330eb925e9b2e94d3556ed68bf17cf9b56b0c499;hp=0000000000000000000000000000000000000000;hb=cfa6933a81a06e229611c39e7bce88e73319db7e;hpb=e3495d89b7c14f67b2bf7b98538ec65b76938355

diff --git a/pdfgrep.sh b/pdfgrep.sh
new file mode 100755
index 0000000..330eb92
--- /dev/null
+++ b/pdfgrep.sh
@@ -0,0 +1,48 @@
+#!/bin/bash
+
+#
+# pdfgrep
+#
+# A script for searching through PDFs to find some words.
+#
+# I guess that everyone knows that PDFs aren't greppable. However, if you use
+# this script, you can grep through PDFs for various key words, just as if
+# they were text.
+#
+# usage: pdfgrep [pattern] [PDFs to search ...]
+#
+# Colin McCabe
+#
+
+die() {
+    echo $1
+    exit 1
+}
+
+ispdf() {
+    echo ${1} | grep -q -i -E '.pdf$'
+}
+
+which pdftotext &> /dev/null || die "you must have pdftotext installed"
+which fold &> /dev/null || die "you must have fold installed"
+
+PATTERN=$1
+shift
+#echo "PATTERN=${PATTERN}"
+
+TMPDIR=`mktemp -d -t pdfgrep.XXXXXXXXXX` || exit 1
+trap "rm -rf ${TMPDIR}; exit" INT TERM EXIT
+
+for PDF in "$@"; do
+    if ispdf "${PDF}"; then
+        #echo "PDF = $PDF"
+        pdftotext "${PDF}" "${TMPDIR}/${PDF}.pre.txt"
+        fold -s -w 120 "${TMPDIR}/${PDF}.pre.txt" > "${TMPDIR}/${PDF}.txt"
+        grep --with-filename -i --color=always \
+            ${PATTERN} "${TMPDIR}/${PDF}.txt" | \
+            sed "s ^${TMPDIR}/  "
+        rm -f "${TMPDIR}/${PDF}.txt"
+    fi
+done
+
+exit 0