pdf错误：IE9和IE10上的“预期dict对象”

Question

I am not sure if this is an Adobe Acrobat/Reader or IE issue, but I can't seem to get any information from either Adobe support or its user to user forum, so I thought I would see if any IE users can help.
What is happening is some users are attempting to open PDF files located on a website (intranet) via Internet Explorer 9 and also Internet Explorer 10. When the user clicks on the listed PDF file, an Acrobat (or Reader) error "Expected a dict object" appears and the IE screen remains blank. I was able to recreate this error on my PC only on IE9 and IE10 but it works on IE8 and IE11 and the other browsers however. So I am wondering if it is related to IE. If the users open the pdf within either the Acrobat or Reader application, it opens successfully. We have a client who likes to open his PDF files this way, so I would like to resolve this issue (rather that telling him to open the pdf in Acrobat or Reader). Thank you for your answers.

Answer 1

If you have any control over saving the PDF, save or export the PDF itself with the "Optimize for Fast Web View" option or its equivalent turned off.

For my particular situation, I see this problem with PDFs I export from InDesign (CS 1 or 6), and then open in Internet Explorer (11 currently, but in older versions previously) and Acrobat Reader XI (11.0.10 currently).

I could make the problem go away oftentimes by having users clear IE's cache, and then download the PDF again, or change the PDF's filename and have them download the newly renamed version (and that worked often enough, oddly). But, I don't want to have to make end users do such things.

All the solutions I've found otherwise were all about turning off "Allow fast web view" in Reader's preferences and/or disabling the plugin in Internet Explorer - to force the browser to open the file directly in Reader. But, once again, I don't want to make end users monkey with settings like that.

A reference to latter, end-user-oriented solutions: http://www.techyv.com/questions/adobe-reader-expected-dict-object

Answer 2

If, in your testing, the document consistently fails under IE8 or IE9, but consistently succeeds under IE8 or IE11, then there's no randomness about it, but there may be key differences in which helper application (or version of same) is being used by each browser.

Even if, as your wording suggests, you're running your IE8 through IE11 testing on the same machine throughout, I would suggest doublechecking that each browser is calling the same PDF utility.

If you have, for example, the latest version of Reader and an ancient copy of Acrobat sharing space on the same machine, IE8 and IE11 could be finding Reader (which might deal with the document's internals just fine) where IE9 and IE10 could be tied to an elderly Acrobat that has some problem digesting a piece of newer PDF coding.

Conversely, another possibility is that an older Acro/Reader is ignoring a problem that a later release flags as an error to complain about. As Max Wyss suggested, you should download the problem document to see if it's corrupted. Run it through Adobe Acrobat's Pre-flight PDF Syntax analyzer, to see if it will flag any problems in the document. If so, what you may be seeing are different utilities' reactions to a problem in the source document.

Answer 3

I realize this is an old post but I couldn't modify the Adobe posts to correct anyone so this is my best option.

The answers are right but wrong. It is about a corrupt format, not a corrupt file or have anything to do with version difference most likely.

As a developer the hardest part about getting PDF's (and FDFs) to work correctly is field validation. If you have ever looked at the files that fill out PDF forms they are very simple. Field Name, Value.

Another thing that is simple is how easily you can break them. The format of the output is <</T(txtFieldName)/V(Value)>>

but, similar to SQL Injection if the end users put something into confuse it then it breaks and causes the DICT error. Lets say that the field info is:

<</T(txtFieldName)/V(I work for the Central Intelligence Agency (CIA))>>

but your dumb CIA people forget to open or close the CIA parentheses...

<</T(txtFieldName)/V(I work for the Central Intelligence Agency --> CIA))>>

That causes a corrupt file, thereby causing the DICT error.