|
Rank: Newbie Groups: Member
Joined: 3/11/2021 Posts: 5
|
There is a notable quality degradation/inconsistent appearance in PDF of hr elements with a dotted border style. HTML sample: Quote:<hr style="border-style: none none dotted; border-color: transparent transparent black; padding-top: 2pt; margin-top: 4pt; margin-bottom: -4pt; vertical-align: auto; background-color: transparent;"> PDF rendered by EO (URL to PNG image)PDF rendered by IE (notably better appearance of the dotted boarder) (URL to PNG image).Here is the complete styles information of the hr element (URL to PNG image)Is there any way to control the hr dotted border appearance through EO .PDF options/settings? Modifying HTML is not a desirable approach for us. Are there any plans to make the improvement of hr dotted border appearance? If so when such changes are expected?
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,200
|
Hi,
What version of EO.Pdf do you use? We tested this on the latest version and it renders much thinner lines than in your image. Our latest version uses Google's Chromium browser engine. So it should generate output similar to what you see with Chrome browser on screen.
Thanks!
|
|
Rank: Newbie Groups: Member
Joined: 3/11/2021 Posts: 5
|
We use EO .PDF .NET 21.0.18
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,200
|
That should be new enough. You can try to load your file in Google Chrome and see if you see any visible difference between the rendering on screen and the result PDF. If you still see significant difference, you can send a test project to us and we will be happy to investigate further. See here for more information: https://www.essentialobjects.com/forum/test_project.aspx
|
|
Rank: Newbie Groups: Member
Joined: 3/11/2021 Posts: 5
|
Thank you for your response. Yes, there is quite visible difference between Chrome and PDF. Please, take a look: This is the link to a ZIP file with the source html and side-by-side comparison of dotted line appearance in browser and PDFsI'd be happy to package and send you a test project, if still needed. Please, let me know.
|
|
Rank: Administration Groups: Administration
Joined: 5/27/2007 Posts: 24,200
|
Hi,
We did see the slight difference. There is no way for us to change that. However we are looking into Chromium's source code to see if we can find out more information about this difference.
Because we rely on Chromium browser engine to render the lines, we basically take whatever Chromium browser engine's output as is. Theoretically we could modify Chromium's source code to change its rendering, however generally this is not a workable solution because the browser engine has over 30 million lines of code, and if we were to modify the engine to each user's liking, not only that it will quickly become unmanageable because different users will want different things, but also it will be impossible for us to maintain the stability of the product due to enormous size and complexity of the project. As a result we generally take a hands off approach to the browser engine itself.
In this particular case, the browser engine renders slightly differently result between a screen device and a "PDF" device. You can verify this difference in Chrome browser by comparing the output directly in the browser window and the output in a print preview window. You will see the output in the print preview window is the same as the output in your PDF file. In fact this is the output we are capturing from the browser engine.
The design goal is the output should be the same across all devices (unless explicitly distinguished through CSS media selector). However in practice there are always some differences. One significant difference is the GPU is almost always involved when rendering on screen, but not involved when rendering on non-screen devices. There are other differences too --- such as DPI difference among different devices. So what we are trying to do is while we can not change the output, we might be able to find some clues about specifically which factors caused the difference. That in turn might give you some pointers on where to go next.
How IE renders the page would be out of this equation entirely ---- We are definitely not in a position to modify Google's Chromium browser engine to make it behaves like IE. So that aspect is irrelevant.
Thanks!
|
|