Pdf metadata retrieval not working (done all the regular tests)

apant

So I have gone through the past posts and checked the following -

- Firefox 3.5.5 privacy pane accepts cookies.

- pdftotext version 3.02 is installed

- pdfinfo version 3.02 is intalled

- resolver set to: http://worldcatlibraries.org/registry/gateway

- tried out the sample pdf that works and posted here - http://forums.zotero.org/discussion/9694/retrieve-metadata-for-pdfs-not-working/

Suggestions?

apant

Also using Mac OS X - with Zotero 2.0b7.4. thanks.

dstillman

Provide a Debug ID for the retrieval attempt.

apant

Dan, thanks for the quick response.

The Debug ID is D1895324320.

dstillman

How about a Report ID?

apant

here you go -

Report ID - 549110552

thanks.

dstillman

Are you getting an error, or is nothing happening?

Also, just to be sure, you sent in that Report ID after triggering the error and before restarting Firefox, right?

There are no Zotero errors in there, but there are some Firefox errors. Those may or may not have anything to do with the problem, but you should be able to tell by restarting Firefox, seeing what's in Report Errors (or making sure it's greyed out), reproducing the error, and checking Report Errors again.

dstillman

Pointing another thread here.

GorWi

Hello Dan,

here is the report ID:

94610619. "Report error" was already active BEFORE I tried to retrieve the metadata, so apparently something slready does not work during startup. Still, this ID is AFTER metadata retrieval. The firefox error console showed the same error before and after; I guess that this means that the report ID is valid for both cases, yes ?

Cheers,
E

dstillman

It just means that error isn't related to this issue.

dstillman

For some reason the indexing cache file isn't being created for you. Are you able to run pdftotext manually from the command line with the same arguments as in the debug output?

GorWi

Hello Dan,

first I'd like to say thank you !

Now, I tried to run the pdftotext command, copied with all options from the debug report, run in the profile directory - runs without an error. Opening the text file showed me that the conversion was successful.
One thing I noticed - I am running 64bit linux, and the pdftotext installed by zotero is called "pdfinfo-Linux-x86_64", so apparently the one included in Ubuntu is run. Just in case I also ran the one installed by zotero with the same options, also no errors and correct conversion.

Cheers,
E

GorWi

Oh, I should add that still the metadata retrieval does not work - I tried once more. Problem sadly still not solved, only conversion by hand works.

Cheers,
E

dstillman

so apparently the one included in Ubuntu is run

No, that's just the debug output. The one installed by Zotero is run.

I'm guessing you're getting an SELinux, AppArmor, or other security-related error trying to run an executable via Firefox. You'll need to check various system logs (or disable those things) to figure out what's going on. (This doesn't particularly help apant on OS X, but it's almost certainly your issue.)

GorWi

Ok, will do that. As soon as I know if what exactly solves the issue, I'll post it here for anyone else with the same problem. Thank you again for your time and advice !

Cheers,
E

GorWi

Disabling Apparmor solves the problem - great ! Thank you!

E