Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Major Issues Reading Pdfs #5099

Closed
nvaccessAuto opened this issue May 19, 2015 · 12 comments
Closed

Major Issues Reading Pdfs #5099

nvaccessAuto opened this issue May 19, 2015 · 12 comments
Labels
Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. bug

Comments

@nvaccessAuto
Copy link

Reported by tweyen on 2015-05-19 18:41
For a year and a half, we have been creating accessible pdfs for our client, who needs to control and lock the content of their pdfs when posted to the web. We experienced a some glitches prior to 2015.1, however, since the update to 2015.1, we have encountered some major new glitches:

1 reading "link" every couple of words when no links exist
2 reading "link" within alternate text for images
3 reading a header from another section before continuing on in the same area (occurs only in a complete read through)
4 link destination error
5 reading "header X" in the middle of a sentence (occurs only in a complete read through)

Attached are some screenshots numbered to correspond with the issues I listed above.

We are setting up our documents in Adobe InDesign and exporting to accessible pdfs. We have not changed how we are setting up documents nor have we encountered any of these problems prior to 2015.1. It appears there is not much discussion about how NVDA reads pdfs but we'd like to get that going because there are significant related issues and we are not finding any support.

Heather Skillicorn

@nvaccessAuto
Copy link
Author

Attachment NVDA errors.jpg added by tweyen on 2015-05-19 18:50
Description:
I don't know how you will read this when the attachments are limited to such a small size

@nvaccessAuto
Copy link
Author

Attachment 1. extraneous links.jpg added by tweyen on 2015-05-19 18:51
Description:

@nvaccessAuto
Copy link
Author

Attachment 2. link in alt text.jpg added by tweyen on 2015-05-19 18:51
Description:

@nvaccessAuto
Copy link
Author

Attachment 4. incorrect links.jpg added by tweyen on 2015-05-19 18:52
Description:

@nvaccessAuto
Copy link
Author

Comment 1 by nvdakor on 2015-05-19 19:01
Hi Heather,
Without sight, we cannot see the light at the end of the tunnel (smiles)...
Do you happen to have an example PDF document the developers can take a look at? If it contains confidential information, please send them directly as an attachment, along with a reference to this ticket number (5099) to mick@nvaccess.org and Jamie@nvaccess.org. If you are sending the PDF directly to Mick and Jamie, please include the following information:

  • What the PDF is about.
  • How NVDA 2014.4 behaves with this document.
  • Inconsistent behavior when reading the document with NVDA 2015.1.
  • Contact information to contact you and your client(s).
    Thanks.
    Sincerely,
    Joseph Lee (nvdakor)

@nvaccessAuto
Copy link
Author

Comment 2 by tweyen on 2015-05-19 19:13
Thank you Joseph, I will contact Mick and Jamie.

@nvaccessAuto
Copy link
Author

Comment 3 by tweyen on 2015-06-23 14:40
Update:

We never heard from Mick and Jamie. However, for problem 1 (reading "link" every couple of words when no links exist), we think the links got corrupted. We deleted all the links and recreated them and the problem was solved. The other four items listed above continue to persist and we are currently creating a new ticket for another issue regarding check boxes.

We are really disappointed in the lack of response and support regarding the issues we are having with NVDA reading pdfs. It is frustrating because our client flags the pdfs for all kinds of things we have no control over. For example, NVDA can not discern acronyms from capitalized text. If you have a sentence in all caps with the word "IT" included, NVDA reads it as "I" "T" even though it should be "it". Or, "PO BOX" reads as "po box" instead of "P" "O". Documents with dashed lists read each item as "dash dash" even though there is only one dash present. I could go on and on with similar problems. In some cases, we have developed tricks to bypass the flaws and quirks of NVDA, but that shouldn't have to happen. I understand the software having a few issues but we've experienced more than just a few and the list keeps growing, all with no response from NVDA.

@nvaccessAuto
Copy link
Author

Comment 4 by nvdakor on 2015-06-23 16:24
Hi Heather,
I'm sorry to hear about this. Perhaps Jamie and Mick are busy these days as they are working on a number of things (Jamie is working on braille display support, while Mick is working on Windows 10 support).
So if I heard you right, you didn't get a response from both of them even after emailing them the PDF, not jpg file? Can you subscribe to NVDA users list (www.freelists.org/list/nvda) and tell us what's going on? That way, I'll forward your email address to the developers so people can contact you directly. One thing to note though: please don't send images, and when asked to send the PDF and the log file, please send us the actual PDF and steps to reproduce this.
As for check box issue, I'll comment on that ticket.
Again sorry for the inconvenience.
Thanks.
//Joseph

@nvaccessAuto
Copy link
Author

Comment 5 by jteh on 2015-06-23 23:35
If you require a more immediate response, we'd be happy to provide information concerning paid support contracts.

The issues regarding capitalised text are not specific to PDF. In addition, they depend on the synthesiser in use, not on NVDA itself. Also, in the examples you gave, there's no absolute guarantee of correctness. For example, the sentence "I WORK IN THE IT INDUSTRY" is a good example of where "IT" should not be interpreted as "it". A screen reader/synthesiser simply cannot get pronunciation right in all cases given the huge number of possibilities. Most screen reader users know and accept this.

@bhavyashah
Copy link

@jcsteh From #5099 (comment), I infer that the scope of this ticket is not specific and thus remains unclear. Could you and @josephsl please take another look at this ticket so as to ascertain what the reported issue or requested feature exactly is?

@Adriani90
Copy link
Collaborator

@Tweyen is this issue solved? Actually it is really hard to reproduce because we do not have a pdf to test. I never had such issues With NVDA yet. But if you test with NVDA2018.4.1 and newest Adobe reader, can you still reproduce it?

@Adriani90 Adriani90 added the Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. label Mar 21, 2023
@Adriani90
Copy link
Collaborator

I am closing this as abandoned, we didn't get any updates from the initial author fro a very long time. Please comment if you are still facing this issues and we can reopen.

@Adriani90 Adriani90 closed this as not planned Won't fix, can't repro, duplicate, stale Mar 21, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Abandoned requested reports or updates are missing since more than 1 year, author or users are not available. bug
Projects
None yet
Development

No branches or pull requests

4 participants