Grabed documents, but got scrambled mess

NenB

I just spent around 2 hours on online library (wiley), searching for research papers. Added quite a few in zotero library.

But now, when I try to open any of them, I get only scrambled mess like this:

<��d�j���9��!����W��?��x�Y+ endstream endobj 110 0 obj <>stream H��V;o�0��Wx�8����2{K�E��@o��#�D�"e_��)�>~����|��caX~�^��N�UPH�w�f.�Μ,dE����!$q+߱u?MZAk3�4��X*��¸�dѤ��iw��NYQ�єeS�o+��c�0���R�x^��.r�#y�#Q�Ri6

:(

aurimas

Could you link to one or two of the articles that are failing to download properly for you?

NenB

It will be of no use as I access it through proxy access from my University to Wiley library, and obviously I can't share login data.

But this is how typically article for grabbing looks like.

http://shrani.si/f/20/A5/2SauhKLw/journal.jpg

And this is result in zotero library

http://shrani.si/f/3y/Ii/2HF9H2I5/journal2.jpg

Perhaps something have to do with proxy?

I get this popup in ff:

Zotero automatically associated this site with a previously defined proxy. Future requests to URL will be redirected to URL.

NenB

Update: I tried to grab document from another library (Oxford Journals), and got same mess result, so it has most likely have something to do with remote access. Don't know what to do and really try to avoid download every single paper manually.

Access to papers goes like this: I login to my University library, then remote access to another library and login there too. Only when logged to both places, I can access research papers freely. Also, url from remote library changes, with inserting name of my university like this:

http://www.oxfordjournals.org.MYUNIVERSITYLIBRARY/subject/economics/

adamsmith

please provide a URL for a paper as you see it in the URL. We have institutional access to most journals.

To clarify - is this where the PDF should be or the snapshot?

NenB

Here you go, 2 links.

Please edit URL accordingly to your institutional access.

http://onlinelibrary.wiley.com.nukweb.nuk.uni-lj.si/doi/10.1002/tie.20421/pdf

http://oep.oxfordjournals.org.nukweb.nuk.uni-lj.si/content/51/2/284.full.pdf+html?sid=cc297294-90cd-4a98-a301-ac2827fc3a03

aurimas

Are you using Firefox or Standalone (and which browser?) version of Zotero?

I assume these files open in the browser for you, not in your PDF reader, am I right?

If you open the file in the PDF reader, does it work? You can find the file by right clicking on it in Zotero and selecting Show file.

Try fetching this article without going through your university's library/proxy (make sure the url is http://onlinelibrary.wiley.com/doi/10.1002/emmm.201200246/abstract and not replaced by a university proxy). See if it downloads the PDF properly.

If no, then please produce a Report ID for adding the open source article to your library.

NenB

I get the same mess.

Report ID 955923227

aurimas

@Dan, is there anything suspicious in that report?

dstillman

Not really, but this appears to be on a network drive, for what that's worth:

Security Error: Content at zotero://attachment/151/ may not load or link to file:///L:/Path/Firefox/Data/profile/zotero/storage/9TRE48ZD/woliconsnew.gif

I'm still not totally clear on the issue here (or what "I get the same mess" means), though. This doesn't seem to be just a PDF. From the screenshot, this looks like an HTML page with an embedded PDF rather than just a direct PDF download. Is there a PDF in the directory? Does it open separately?

NenB

I get the same mess means that when I try to create new item from this url http://onlinelibrary.wiley.com/doi/10.1002/emmm.201200246/pdf

I get this http://shrani.si/f/J/MH/2q5DvqBW/mess.jpg when access file through ff zotero.

What I noted though, is that pdf file is saved just fine, but for some reason I can't see it properly in ff. It is like zotero opens it like it would be html, instead like pdf.

aurimas

Hence my quoestion above.

If you open the file in the PDF reader, does it work? You can find the file by right clicking on it in Zotero and selecting Show file.

Are you sure your translators are up to date? (Gears menu -> Preferences... -> General tab -> Update Now)

I think this was a problem a while back, but it has been fixed.

Simon

NenB, are you using the "Create Item from Current Page" button to save a snapshot of the page that shows the PDF? If so, does the PDF save if you use the URL bar icon on the abstract page instead?

NenB

@aurimas, when selecting show file it opens folder and pdf there is ok, I can read it in pdf reader. I use 3.0.7 zotero and 12 ff.

@Simon, if I use URL bar icon, it get metadata ok, but when trying to open it is same scrambled mess.

fbennett

If the file is arriving intact, it looks like a problem in your local settings for opening the file. This comment by Dan may be helpful. To trace things back, in Firefox check Edit -> Preferences -> Applications, scroll down to the PDF Document entry and see what it says. Then in your operating system, check the system default (on Windows or Mac -- not sure where those are) or the file manager default (on Linux that's right-click on any PDF in a directory view, then Properties -> Open With), and see what it says there.

NenB

But I can open other pdf files just fine. Also pdf in library is embeded, it is not like pdf file alone would open in new window.

aurimas

So you're saying that if you right-click on the PDF attachment in Zotero (should say something like "Full Text PDF" and have a PDF icon), select Show File and open the PDF in your PDF viewer, it works fine (there is no "Wiley bar" at the top, like in the screenshot below)

@aurimas, when selecting show file it opens folder and pdf there is ok, I can read it in pdf reader. I use 3.0.7 zotero and 12 ff.

But if you double-click it in Zotero, it looks like http://shrani.si/f/3y/Ii/2HF9H2I5/journal2.jpg ??

That just doesn't add up, IMO.


And your translators are up to date?

Are you sure your translators are up to date? (Gears menu -> Preferences... -> General tab -> Update Now)

NenB

Ok, I made video. Hope it explains all:

http://tinypic.com/player.php?v=14b3n5z&s=6

aurimas

What you show in the video is capturing the actual webpage, not the article. Though it still should not scramble the PDF that is embedded in the snapshot. There's something messed up with mime types on your firefox setup. Performing the same steps works on my system.

Dan's post from above:

NenB, are you using the "Create Item from Current Page" button to save a snapshot of the page that shows the PDF? If so, does the PDF save if you use the URL bar icon on the abstract page instead?

your response:

@Simon, if I use URL bar icon, it get metadata ok, but when trying to open it is same scrambled mess.

When you do use the url bar icon, how many attachments are you seeing, and do any of them look like a PDF and/or say PDF?

NenB

Just to be clear, we are talking about getting article from exactly this url
http://onlinelibrary.wiley.com/doi/10.1002/emmm.201200246/pdf

In the url address bar it shows journal icon, that is all.

aurimas

That is correct. It should be showing the journal icon. If you click it, it will import the article into Zotero. Is the PDF still scrambled?

NenB

Yes, it is the same mess. I already posted this:

"@Simon, if I use URL bar icon, it get metadata ok, but when trying to open it is same scrambled mess."

aurimas

And then I asked if you see a pdf attachment under that item, or only the snapshot, like you show in the video.

NenB

Oh, I see. If I use url icon bar, I get parent journal icon and 2 subicons. 1 is Full Text PDF and another is snapshot. Snapshot is ok, but when pressing view pdf I get scrambled mess, "open in external viewer" opens pdf normally though.

aurimas

1) Go to Firefox->Options->Options->Applications. What do you have set for "Adobe Acrobat Document" (or if you use something else, try to find what corresponds to a PDF, it might say 'application/pdf')

2) If you right click the PDF attachment (not the snapshot or parent item), select Show File, what is the file extension for the PDF? if you don't see an extension, right-click and select properties. What does it say under "type of file"?

I'm really confused how your snapshot started working, but PDF is not. Maybe Dan or Simon have some new ideas.

NenB

1) http://shrani.si/f/4/12K/2FOyQVBv/adobe.jpg

2) pdf

aurimas

Last question before I run out of ideas, could you post the first 5 lines of the scrambled mess that you are seeing? Open the pdf in Firefox and when you see the page that is scrambled, right-click somewhere on the page, select View Page Source and copy the first few lines from there.

NenB

First 25 lines:

%PDF-1.3
%äãÏÒ
104 0 obj
<<
/Linearized 1.0
/O 108
/H [ 1804 374 ]
/L 267418
/E 59616
/N 3
/T 265293
>>
endobj
xref
104 64
0000000015 00000 n
0000001613 00000 n
0000001718 00000 n
0000001746 00000 n
0000002178 00000 n
0000002476 00000 n
0000002561 00000 n
0000002647 00000 n
0000002733 00000 n
0000002819 00000 n


Later on get to the scrambled data. Sorry to take such an effort to solve this, it is not convenient for me too, to post this rapidly. I will find my way around this issue, but if it can be solved, it would be great.

dstillman

1) All the items you saved via Create New Item from Current Page are, for our present purposes, irrelevant. Those still shouldn't be scrambled, as aurimas said, and that may be due to the same problem, but that's not the way to save articles from sites with translator support (or even without—you should view the PDF without the HTML frame and save it from there), so let's ignore that for the moment.

2) See the top half of http://www.zotero.org/support/kb/file_handling_issues and delete mimetypes.rdf as instructed there.

3) If you still have the problem after deleting mimetypes.rdf, provide a video of saving via the address bar icon and trying to open those. The "Full Text PDF" items from that page in your account look fine on the server, so this is likely a problem with your Firefox installation, but I want to make sure we completely understand what you're seeing.