Author: Alexander Barkov
Email: [EMAIL PROTECTED]
Message:
We won't fix this in 3.1.17 sources.
Quick workaround for this:
Create a shell script:
#!/bin/sh
echo "<body>"
pdftotext $1 -
echo "</body>"
And use it instead of direct execution of pdftotext.
Don't forget to change parser's output format from text/plain
to text/html in Mime indexer,conf command.
Thanks for reporting!
BTW. in 3.2.0 everything works fine.
> In the meantime we found out something. The problem seems to happen if we use MySQL
>with "pdf2text $1 -" (outputting text/plain) but not using Oracle with
>Oracle's InterMedia-Text Parser ctxhx (outputting text/html). With MySQL (3.23.32)
>and "Index Yes" the TXT column is filled after parsing; with "Index
>No" its not. With Oracle (8.1.7.0.0) using ctxhx and "Index No" the
>TXT column is filled properly, using "pdf2text $1 -" its not. Both external
>parsers write to stdout.
Reply: <http://www.mnogosearch.org/board/message.php?id=2804>
___________________________________________
If you want to unsubscribe send "unsubscribe general"
to [EMAIL PROTECTED]
