Font misalignment in converted pdf

Raja Suba

unread,

May 2, 2016, 1:35:14 AM5/2/16

to pdf2htmlEX, coolw...@gmail.com

FONT ISSUES WITH PDF TO HTML CONVERSION

1.All "ti","fi","tt" characters are missing

SAMPLE SCREENSHOT

2. Font overlapping issue

SAMPLE SCREENSHOT

- _NOTE: I don't get this issue with firefox. Getting the above issues in chrome in safari browser_

I AM USING

- Using the **0.13.6** version of pdf2htmlEX

- Using the following command to convert pdf to html

pdf2htmlEX --split-pages 1 --zoom 3 --fit-width 920 --correct-text-visibility 1 --dest-dir $1 $2 2>&1**

TRIED WITH

Using **--fallback 1** option solves all my above problems. But

1. The fallback option reduces the clear visiblilty. Please refer the below screen shots.

SCREENSHOT WITH FALLBACK ENABLED

SCREENSHOT WITH FALLBACK DISABLED

2.Table in the page disappears rather replaced with empty space.

PLEASE CLARIFY

1. Could you explain a bit more on fallback?

2. I have tried the above one (using fallback). Please suggest me if you prefer a different approach.

Kindly help me in this regard.

Lu Wang

unread,

May 2, 2016, 6:09:52 AM5/2/16

to Raja Suba, pdf2htmlEX

Regarding the 1st issue, try if `--tounicode -1` helps.

regards,

- Lu

Message has been deleted

Raja Suba

unread,

May 2, 2016, 8:43:26 AM5/2/16

to pdf2htmlEX, rajasu...@gmail.com

Hi Lu,

Thanks for posting. `--tounicode 1` doesn't help me :( . Please suggest me an alternative to sort this out.

Regards,

Rajasuba S.

Message has been deleted

Raja Suba

unread,

Jun 5, 2016, 3:52:46 AM6/5/16

to pdf2htmlEX, coolw...@gmail.com

Hi,

The above issue occurs only in - webkit web browsers like chrome and safari - which provides support for ligatures - whereas browser like firefox does not.

A ligature is a combination of two or more letters joined as a single glyph
(For more info : https://en.wikipedia.org/wiki/Typographic_ligature)

Reason for the issue

This issue with missing characters is due to ligature support provided by these modern browsers - let me explain how

1. Our tool while converting - it converts characters to glyphs using poppler for rendering - now these browser when they come across characters like tt tf ti ff fi consider them to be ligature and searches for glyphs corresponding to tt and not t t

2. Since they do not have their corresponding glyphs - they just skip the characters and renders the rest - so, we fount the characters missing

This could be solved by

Disabling/ Turning-off the ligature in these browsers - embedding the css in the generating content

Refer:

#543
http://stackoverflow.com/questions/19591746/prevent-ligatures-in-safari-mavericks-ios7-via-css
https://developer.mozilla.org/en/docs/Web/CSS/font-feature-settings
http://caniuse.com/#feat=font-feature

Please correct me if I am wrong.

Lu Wang

unread,

Jun 20, 2016, 5:31:30 PM6/20/16

to Raja Suba, pdf2htmlEX

Maybe you can try to disable ligature with CSS.

regards,

- Lu

Lu Wang

unread,

Jun 21, 2016, 5:17:58 AM6/21/16

to Raja Suba, pdf2htmlEX

Ah, I just read the the last few paragraphs of your previous email, somehow they were folded.

Thanks for your message.

regards,

- Lu

On Tue, Jun 21, 2016 at 5:32 AM, Raja Suba <rajasu...@gmail.com> wrote:

Hi Lu,

Tried the same (diabling ligature with css) and it works fine for me in all the cases.

Thanks and Regards,
Rajasuba S.

David Hedley

unread,

Oct 11, 2016, 5:10:33 AM10/11/16

to pdf2htmlEX, coolw...@gmail.com

Issue 1 is due to dumb Chrome trying to use ligatures in a font that does not contain them.
You can use the javascript I posted https://github.com/coolwanglu/pdf2htmlEX/issues/675
to fix the issue.

Hard to know what the problem with issue 2 is without seeing the PDF, but I would guess it's a word-spacing issue which Chrome/Webkit also has problems with.

You could try my branch https://github.com/davidhedley/pdf2htmlEX in which I have rewritten how correct-text-visibility works and also has a few fixes in it which may work for you

Reply all

Reply to author

Forward