PDFJS highlight is not correct.

[unicorn82]

Attach (recommended) or Link to PDF file here: 10pages-noLogo.pdf Configuration:

• Web browser and its version: IE 11, Chrome.
• Operating system and its version: Windows 10 Enterprise
• PDF.js version: https://mozilla.github.io/pdf.js/web/viewer.html
• Is a browser extension: NO

Steps to reproduce the problem:

1. Open PDFJS demo, https://mozilla.github.io/pdf.js/web/viewer.html
2. Click "Open File" button in header and open PDF attached from local
3. Search "engine" and highlight all occurrences.
4. Some keyword cannot be highlighted correctly, for example in page 6. What is the expected behavior? (add screenshot) Keyword need be highlighted correctly.

What went wrong? (add screenshot)

Link to a viewer (if hosted on a site other than mozilla.github.io/pdf.js or as Firefox/Chrome extension):

[unicorn82]

Any comments or suggestion for this issue?

[timvandermeij]

This has to do with the positioning of the text layer. I tihnk there are some similar issues regarding this behavior, but someone needs to check if they share the same cause.

[Ognyshare]

I modified in text_layer_builder.js in the convertMatches functions two while loops.

from:

iIndex += textContentItemsStr[i].length;

to:

iIndex += this.findController.normalize(textContentItemsStr[i]).length;

Since this change I could resolve a simmilar Problem, but it is not perfect because in some cases the match is still shifted one position to the right.

Here is the modified function which I use now in text_layer_build.js : modified_convertMatches.txt

[timvandermeij]

I think this is related to #9448 then.