zotero is not indexing

bernhard42

hello everybody,

at home i'm running zotero in the following environment:

firefox 3.5.7
mac os 10.6.2
zotero 1.0.10
pdftotext 3.02
pdfinfo 3.02

at work it's a bit different since there some "restrictions":

firefox portable 3.5.7 (from portableapps.com)
windows xp
zotero 1.0.10
pdftotext 3.02
pdfinfo 3.02

i'm carrying the database around on a thumb drive.

in zotero's search preference pane it tells me how many items are indexed, partly indexed and how many are not indexed. the problem is: no matter how many times i click on "rebuild index" (partly or completely) the numbers stay the same. how can i determine where the reason for this odd behaviour is? is it zotero itself or my files?

with the portable version i get the additional error message "<path to pdfinfo-win32.exe>, wrong parameter" and "<path to pdftotext-win32.exe>, wrong parameter". the <...> was added by me. there also appears a command line window saying "program to big for memory" (translated from german).

thats it. hope you understand me.

dstillman

PDFs without embedded text won't be indexed, and documents longer than the page/character limits in the Search pane of the Zotero preferences will be partly indexed.

bernhard42

thanks for reminding me of _that_ one. is there another way to determine if a pdf-file has embedded text than letting zotero try to index it and fail?
i've got some pdfs in my collection that are searchable by acrobat reader but zotero can't index them.
any ideas on how to add a layer of embedded text to the pdf-files other than ocr? (think in online services or tools which can run from a thumb drive).

maybe we should open another thread since my initial question is already answered...

Balint42

Your problem probably has nothing to do with the PDFs but is a zotero bug. I had the "wrong parameter" error too and it was caused by the "pdfinfo-Win32.exe" and "pdftotext-Win32.exe" versions by zotero.
Since you have MacOS Im not sure how much this applies to you, but still I will post how I solved the problem under win32:

Nothing I tried worked, except for this solution: use the original xpdf files. Thus:

1.) Download the xpdf win precompiled libs here: http://foolabs.com/xpdf/download.html
2.) Unpack them anywhere.
3.) Copy "pdfinfo.exe" and "pdftotext.exe" to the zotero dir, mostly sth like "C:\Documents and Settings\User\Application Data\Mozilla\Firefox\Profiles\xyz.default\zotero"
4.) In the zotero dir, delete the old "pdfinfo-Win32.exe" and "pdftotext-Win32.exe" and rename "pdfinfo.exe" and "pdftotext.exe" to "pdfinfo-Win32.exe" and "pdftotext-Win32.exe".
5.) Restart Firefox.

It now should work. Only thing is that a cmd window will now appear each time you index a PDF but never mind that..

amarsan

Thanks, really, Balint, for this guideline.
That solves a bug I didn't know how to deal with !

dstillman

Balint42 and amarsan: What version of Windows are you using? Zotero's versions of the binaries (which turn off the cmd window) obviously work for very many people.

Balint42

Hello,

I am using a quite fresh install of a fully updated win7 32bit. I dont know why it works for most, but there are quite a few reports for this bug if you google..

regards