Accessible Text and logical reading order
First while I am technical I barely know anything about PDF’s or working with them.
Skipping to the question…
What effects the default logical reading order for the accessible text save as for an element to not be where it is expected?
Background: At work we do proofs for mailers. Somewhere in the document we always expect to see
reference #: <some number>
For the most part we extract these by saving as accessible text and then reading the resultant text file. This works well except when it doesn’t…
There is a case where two instances of reference #’s are on the document on a page. One case we pick up but the second is missing the colon (:). In text file.
My question is then what effects the default logical reading order for the accessible text save as for this element to not be where it is expected? Eyeballing it, I don’t see what is different.
Thanks for any insight. I am hoping to be able to ask for a tweak to the source document as I remember other permutations that not using the colon would cause problems.
Submitted February 12, 2019 at 09:46AM by ContentParodox
via reddit
https://www.reddit.com/r/pdf/comments/apxd62/accessible_text_and_logical_reading_order/?utm_source=ifttt
from Medical Release Form
https://medicalreleaseform.tumblr.com/post/182765219526
via
Medical release form
from Tumblr https://jeemiahmelville.tumblr.com/post/182765907012