tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

msword docs: avoid generating an error for files containing only a picture (empty antiword output)

Browse Source
This commit is contained in:
Jean-Francois Dockes 2014-07-06 16:24:11 +02:00
parent d8fe30750a
commit 25271db690
1 changed files with 25 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

41
src/filters/rcldoc
View File

@ -90,23 +90,30 @@ umask 77
checkcmds awk antiword iconv
# output the result
# We need to do some strange stuff to retrieve the status from antiword. Things
# would be simpler if we relied on using bash.
# Explanations:
#http://stackoverflow.com/questions/1221833/bash-pipe-output-and-capture-exit-status
stdintoexitstatus() {
read exitstatus
return $exitstatus
}
# The strange 'BEGIN' setup is to prevent 'file' from thinking this file
# is an awk program
$decoder "$infile" |
(((($decoder "$infile"; echo $? >&3) |
awk 'BEGIN'\
' {
cont = ""
gotdata = 0
}
{
if (!($0 ~ /^[ ]*$/)) {
if (!($0 ~ /^[ ]*$/) && gotdata == 0) {
print "<html><head><title></title>"
print "<meta http-equiv=\"Content-Type\" content=\"text/html;charset=UTF-8\">"
print "</head>\n<body>\n<p>"
gotdata = 1
if (gotdata == 0) {
print "<html><head><title></title>"
print "<meta http-equiv=\"Content-Type\" content=\"text/html;charset=UTF-8\">"
print "</head>\n<body>\n<p>"
}
}
$0 = cont $0
cont = ""
@ -124,17 +131,19 @@ awk 'BEGIN'\
print "</p><hr><p>"
next
}
gsub(/&/, "\\&amp;", $0)
gsub(/</, "\\&lt;", $0)
gsub(/>/, "\\&gt;", $0)
print $0 "<br>"
if (gotdata == 1) {
gsub(/&/, "\\&amp;", $0)
gsub(/</, "\\&lt;", $0)
gsub(/>/, "\\&gt;", $0)
print $0 "<br>"
}
}
END {
if (gotdata == 0)
exit(1)
print "</p></body></html>"
}'
if (gotdata == 1)
print "</p></body></html>"
}' >&4) 3>&1) | stdintoexitstatus) 4>&1
# Antiword rarely fails, we try to catch the most common reasons: