PDF titel not displayed in middle column

groetker

I experience a somewhat strange behaviour since some days (or weeks?): Whenever I attach a PDF as a "snapshot", the title of the PDF is not shown in the middle column. Instead, the space is blank. On the right hand side, I can see the title.

When I choose "rename PDFs", I do see the new name in the middle column. (I cannot do this though with entries containing more than 1 PDF.)

Also, I am not able any more to open PDFs stored in Zotero (4.0.8) direcly in my browser (firefox 20.0.1). Instead, they are opened by Acrobat. I cannot change the application for PDFs from "Acrobate" to "Acrobate in Firefox".

Any idea?

Thanks
Ralf

aurimas

Whenever I attach a PDF as a "snapshot"

How exactly are you performing this? Also a URL would be helpful even if this problem is not specific to any particular URL.

I cannot change the application for PDFs from "Acrobate" to "Acrobate in Firefox".

Do you mean that there is no "Use Adobe Acrobat (in Firefox)" option? Make sure you have Adobe Acrobat Plugin installed in Firefox (Add-ons -> Plugins)

groetker

1. This one, e.g.:
http://www.bioethics.ualberta.ca/en/~/media/dossetor/Publications/ChoicePrenatalScreening.pdf
It's not URL specific.
I just attach the pdf as snapshot to an existing entry.

2. Yes: in Firefox. The plugin is installed. It works for "PFF Forms" but not for normal PDF (it doesn't give me the option to choose "Adobe in Firefox). I tried re-installing. Also updating my plugins.

aurimas

1. As I was saying before, please describe exactly what steps you are taking to perform this action (I can't replicate it, but there are potentially several ways of doing this). You can also provide a Debug ID for this action.

2. Try re-installing (or just updating) Adobe Reader. If the option is not there in the Firefox Preferences, you will have to ask for assistance on the Firefox website: https://support.mozilla.org/en-US/products/firefox

groetker

The strange thing is: I can view pdfs that are part of an url in my browser. This one, e.g.:
http://static.mozilla.com/moco/en-US/pdf/mozilla_privacypolicy.pdf
Which means: everything is okay with firefox/acrobate.

It's just when I click on PDFs in Zotero that they open in an external application. (Yes: I have activated the option: open pdf within browser within my Zotero preferences.)

The action I perform is just the standard one: I click the icon "Make a new entry out of this website". Then, I choose from the drop down menue "add snapshot".

The problem has never occured in the five years I am using Zotero.

I just replicated the error. Here is the debug id:

1908784281

aurimas

The action I perform is just the standard one: I click the icon "Make a new entry out of this website". Then, I choose from the drop down menue "add snapshot".

OK, I was able to reproduce this. One important detail is that you have to have Adobe Acrobat (or perhaps some other plugin) selected for viewing PDFs in Firefox. In that case, the page that is displaying the PDF does not have a title. So when you attach the PDF as snapshot, Zotero cannot figure out the title and so it leaves it blank. I don't think the correct behavior would be to set the title of the item as the title of the attachment.

If you let Firefox's built-in PDF viewer open PDFs in Firefox, then you will get a page title and when you attach those PDFs as Snapshots they will have a title filled in.

What we could possibly do is set the filename of the PDF to the title.

Having said all of that, this way of collecting items and attachments is generally not recommended, because it results in poorer metadata. You should be using the URL bar icon to import metadata and PDF attachments. Of course not all pages will display the icon in the URL bar or be able to import the PDF in which case this is an OK solution.

I'll have to get back to you (or perhaps Dan or Simon will) regarding PDFs opening in your external viewer.

dstillman

For the file-opening issue: http://www.zotero.org/support/kb/file_handling_issues

groetker

Thanks - deleting mimetypes.rdf as described in the link that Dan provided did work: PDFs openend in Firefox again.

At least it did once... Now, Zotero crashes altogether when I click to open the PDF!

Here's the report related to this: 1642706979

Thanks
Ralf

groetker

... okay: I re-bootet my computer. Now it works again (hopefully it will continue to work.)

I can now describe the bug more exactly.

Take this url:

http://www.who.int/foodsafety/publications/fs_management/No_05_Bisphenol_A_Nov09_en.pdf

When I saved that page a few months ago, Zotero produced an entry with the name "No_05_Bisphenol_A_Nov09_en.pdf (application/pdf-Objekt)". The pdf that I created as a snapshot got the name: "No_05_Bisphenol_A_Nov09_en.pdf".

Today, I save the same page ("make new entry from web-page"). Zotero gives me an entry with no titel! Also, the pdf has no title.

What can I do to have my Zotero working like in the old days?

Best
Ralf

dstillman

Today, I save the same page ("make new entry from web-page"). Zotero gives me an entry with no titel! Also, the pdf has no title.

Does the attachment have a PDF icon? What's the filename in the right-hand pane? If you right-click on the attachment and choose Show File, in the folder that pops up, do you see a single ~181KB PDF file, or something else?

It saves fine for me with the Firefox PDF viewer (pdf.js), which is what you appear to be using. (It's not actually displaying the PDF properly for me when I open it through Zotero, but that's a separate issue. The file and title are fine.)

groetker

... yes: the icon is there.

In the right hand pane the name is: "no_05.Bisphenol_A_Nov09_en.pdf"
Just a usual pdf-title.

If I go for "show file", I find just the usual PDF. Normal size.

The Firefox PDF-viewer I just installed today or yesterday, to check if that could solve my problem. ...I am just-re-checking. If I am using the pdf.js (and deactivate Adobe) and take a snapshot, then I get as a title something like this: "Microsoft Word - EN_ Note_5_2009_BPA.doc - No_05_Bisphenol_A_Nov09_en.pdf".

If I switch back to the Adobe viewer, I don't get a Titel for the pdf.

In neither case I get a title for for main entry to which the pdf is attached.

As I said: it used to save fine all these years. Until some weeks ago.

Best
Ralf




If I deactivate it, everything stays the same.

aurimas

Take this url:
http://www.who.int/foodsafety/publications/fs_management/No_05_Bisphenol_A_Nov09_en.pdf
[...] Today, I save the same page ("make new entry from web-page"). Zotero gives me an entry with no titel! Also, the pdf has no title.

If you use the "make new entry from web-page" (I assume you're translating, because in english it is actually called Create Web page Item from Current Page) function on that page with either Firefox's pdf.js plugin or Adobe's plugin, you do _not_ get a Zotero (web page) entry. Zotero should just be saving the PDF file without attaching it to a Zotero item. At least that is the behavior I am seeing (I hope Dan can confirm this). We never asked you, but what OS are you using? I am testing this on Windows 7.

The Firefox PDF-viewer I just installed today or yesterday, to check if that could solve my problem. ..

I am not sure why you had to install the Firefox PDF viewer (I assume from here: https://addons.mozilla.org/En-us/firefox/addon/pdfjs/) because as of Firefox 19 it comes bundled with Firefox. Either way, the result of you switching to PDF.js is consistent with the observations I posted above (along with the explanation for why this is happening and what we may be able to do about it)

As I said: it used to save fine all these years. Until some weeks ago.

I don't think anything has changed recently in this area as far as Zotero goes. It seems to me that this change is coming from either Firefox or Adobe Reader

dstillman

If you use the "make new entry from web-page" (I assume you're translating, because in english it is actually called Create Web page Item from Current Page) function on that page with either Firefox's pdf.js plugin or Adobe's plugin, you do _not_ get a Zotero (web page) entry. Zotero should just be saving the PDF file without attaching it to a Zotero item. At least that is the behavior I am seeing (I hope Dan can confirm this).

Yes, with the default settings. With the "Automatically take snapshots when creating items from web pages" preference off, I just get a top-level web page item with no PDF.

Ralf, what is that pref (and "Automatically attach associated PDFs and other files when saving items") set to in the General pane of the Zotero preferences?

aurimas

Yes, with the default settings. With the "Automatically take snapshots when creating items from web pages" preference off, I just get a top-level web page item with no PDF.

I see. You're right.

On a side note, does that behavior make sense? With the pref checked, the button essentially does not do what it says. On the other hand, I don't think it makes much sense to create a web page item in Zotero (and attach the PDF) if you are looking at a PDF. I'm not sure if this needs fixing or what the correct behavior should be if it does, but I thought I'd bring it up.

dstillman

Given that relevant code here contains the lines:

// TODO: this, needless to say, is a temporary hack
if (itemType == 'temporaryPDFHack') {
it's fair to say that this is a somewhat unresolved issue. If I recall, the eventual goal was to improve PDF metadata retrieval (and decrease the dependence on Google Scholar) to the point where we could run it automatically on new PDFs and create parent and child items, but until then (or the ability to run metadata retrieval on a child item and update the parent from the result) the top-level PDF behavior is what we want, so that metadata retrieval can be run on the file.

The behavior with the pref off does not really make sense and probably shouldn't happen. I think it should just create the top-level PDF regardless.

aurimas

The behavior with the pref off does not really make sense and probably shouldn't happen. I think it should just create the top-level PDF regardless.

In that case, may I suggest that the button should read something like "Capture current web page"?

dstillman

I'm not too concerned, particularly given that even if we change the behavior with the pref off this is still theoretically temporary. I'm much more concerned with people confusing that button with the address bar icon, and "Capture current web page" would blur that line even more.

groetker

Thanks ..

I am using Windows XP. And, yes, I am translating because I am using the German edition of Zotero.

I am not concernced with the "pref" issue. (I do have "take snapshots automatically" turned off: but never got confused by the resulting behaviour.)

Rather, my problem is still that titels are missing: the main entry created when I "make new entry from web-page" from an URL like "http://www.who.int/foodsafety/publications/fs_management/No_05_Bisphenol_A_Nov09_en.pdf" has no title. Neither has the pdf that is created when I add a snapshot to the main entry.

Especially when I attach more than 1 PDF to a top level entry and I have to add all titles manually, I'd very much like to have my old functionality back...

groetker

PS: "If you let Firefox's built-in PDF viewer open PDFs in Firefox, then you will get a page title and when you attach those PDFs as Snapshots they will have a title filled in."

It's THIS what isn't working. The PDFs don't have that title. I am not using PDF.js

aurimas

either you are misreading what we're saying or you are not saying what you mean.

Adode plugin = no page title = no title for pdf

Firefox built-in PDF viewer (pdf.js) = page title = title for attached PDF

to use Firefox's built in pdf viewer you have to select "preview in Firefox" for PDF files in the preferences.

groetker

It's THIS what I did not get:

Firefox built-in PDF viewer = pdf.js

(I understood you referred to Adobe as "build in".)

Thanks a lot !
Ralf

wheatman

I had the same issue, and after following this thread, I reset Firefox, reinstalled Zotero and the problem was solved.

http://support.mozilla.org/en-US/kb/reset-firefox-easily-fix-most-problems