Warning: Odd "Save to DEVONthink" behavior in Tige

Scenario: PDF files on the USEPA website opened in Safari using Acrobat 7 Reader plug-in (rather than Safari’s lame reader, LOL), followed by executing the “Save to DEVONthink” print command. Repeat with several PDF files.

Result: Several PDFs successfully load into DT, words are counted/recorded properly. At some point a PDF is loaded into DT and the only words counted are in the title. However, the PDF has text it isn’t just an image. For some reason the program fails to read the words. I’ve no idea if this is something happening in Acrobat Reader plug-in or in DT but it has happened to me twice after successfully doing several PDFs prior to the failure.

Workaround: Save the PDF to disk from Acrobat Reader then drag and drop the PDF onto DT. This always works and the words are counted.

Bill, have you or Christian or anyone else seen this or have any idea why it might be happening?



My 20" iMac with 2 GB RAM is supposed to ship by Tuesday, and I’m keeping the TiBook on 10.3.9 for ‘old’ OS support for a while – so I’m not running under Tiger yet.

I use PDF Browser Plugin instead of Acrobat’s plugin.

Some of EPA’s PDF files have pretty well done bookmarking, and once in a while they contain working hyperlinks. To retain such attributes of PDFs, I save them to disk rather than ‘print’ them into DEVONthink. I’ve set up Folder Actions folders (using scripts supplied on the DEVONthink download disk image) to give me the option of importing as plain text or Index import to DEVONthink. So saving the PDF file to the desired Folder Actions folder automatically imports it into DEVONthink.

All this to say that I haven’t hit your problem, and hope I don’t experience it when I start running under Tiger. :slight_smile:

… wondering how this issue has played out with folks’ experience… I continue to have the same problem: importing pdf’s (whether by “saving” to a folder with Action Import; or printing to said folder; or using “Save as Devon.script”) often (always??) results in DT entries that show text, and count the words, but don’t “know” the words, meaning I can’t find specific words using Search functions, either within DT or within Preview. Now, as it happens, I haven’t used Acrobat in quite awhile. If it were true that the problem could be resolved by changing from Preview to Acrobat as my default viewer, that would be a start. I used to used the PDF Browser Plugin under Panther, but Tiger has built-in functionality for handling PDFs and I haven’t fiddled with adding in a separate plugin on top of that – though I would if it would resolve this “obscure-text-upon-download” problem.

ChemBob or dclayton:

Please send me the URLs of several of the PDF files with which you are having problems.

I’m up and running in Tiger now, so ready to test this.