Oxygen XML Forum

Hi,

I hate complaining this and that like a babusha at a bazaar on a rainy day but still, this time, it's about the webhelp topic page layout when block-level comments enabled at the feedback site configuration.

Background:
1/ The DITA 2.0 bookmap I'm using contains MD (90%) files and DITA 2.0 topic/task/ref/concept files (10%)
2/ I generated IDs for all files within the bookmap hierarchy. Elements including p, li, sl, title, ul, ol, pre, etc.
3/ Topic toc was enabled as it's was specified in the .opt file.
4/ Block-level comments feature was enabled.
5/ All the following issues occur on html pages converted from markdown files.

Here's what happened:

1. When clicking a section title at the topic toc on the right side of the page, the page will not be repositioned to the section title itself, but the 1st or 2nd paragraph of list below the title. For example, on this page, https://docs.dolphindb.cn/help/dita/api ... E%E9%A2%98

If I click the last section title on the topic toc, the page will navigate to the second paragraph of this section. However, it should be here: Note: This occurs to pages converted from MD files only. On pages converted from DITA, everything looks great. For example: https://docs.dolphindb.cn/help/dita/iot ... qn_dwf_myb

2. When clicking a section title on the topic toc, sometimes the page will navigate to the section title after the first click, but for the second click on the same section title in topic toc, the page will navigate to the 1st or the 2nd paragraph of that section. For example, on this page, https://docs.dolphindb.cn/help/dita/plu ... #odbcclose if you click "odbc::close" on the topic toc, you will be taken to its title for the first click, then click it again, you'll find an offset similar to the previous issue.
Note: This occurs to pages converted from MD files only too.

3. When opening a page converted from MD, during the first couple of seconds, you will notice a quick progress (not quick enough to hide from my eyes though) that line spacing are increasing among paragraphs, lists, or sections. This could be reproduced on this page: https://docs.dolphindb.cn/help/dita/plu ... EADME.html . My guess was it takes time for the browser to load the frames of various block-level comments within the page, but it occurs to pages converted from markdown files again!

Hi,

We could not reproduce this issue on our side.

For further investigations, it would be very useful to create a minimal valid sample project (DITA Map + markdown topic) that you can send us for analysis. You can upload the sample files here on the forum or you can send the files to support@oxygenxml.com. Also, please specify the
WebHelp version.

Best regards,
Beniamin Savu
Oxygen WebHelp Team
http://www.oxygenxml.com

Hi,

Please find the sample files from attached.

Hi,

Thank you for providing us the sample. We managed to reproduce the problem on our side.

From our investigations it seems that the corresponding HTML elements for the section titles clicked in the topic toc have the id attribute value generated in Chinese characters. This is happening only for the markdown topics, the Dita topics have the id attribute value generated in alphanumeric characters.

I added an issue on our internal issue tracker to be further analyzed by our development team. We will let you know when the problem is fixed.

Best regards,
Beniamin Savu
Oxygen WebHelp Team
http://www.oxygenxml.com

Thank you.

Traditionally, maybe 8 or 9 years ago, I remembered the ID for any element cannot contain CJK characters, that rule was changed in recent major oxygen editor releases, maybe from oxygen 23 or 24, I can't remember. In current release, 25.1, I noticed that ID written in CJK characters will not trigger errors or warnings when editing or building webhelp. (That's an epic improvement especially for east asian users who migrate markdown topics to dita from time to time).

As you noticed, there are some IDs written with Chinese characters in HTML files, that's because of the way how oxygen editor process the markdown topics in a map/bookmap. The conversion from markdown to html, as the Webhelp transformation goes, is that each markdown section segmented by # headings will be converted to topic (nested topics if the original markdown file contains more than 1 headings) first, and then markdown headings will be used as the topic ID accordingly. After, these temporary nested dita topics files are then converted to html. That way, ID with Chinese characters ware included in the html files.

About the automatic lining space increase that you reproduced, is it because of the fact that, for a given font size for certain element in CSS, let's say 14px for the topic2 heading, Chinese characters with font-size=14px tend to looks taller than latin characters with font-size=14px, even if the same font-family is specified for both languages ?

Now here's a test. The following sentence contains characters of 2 languages in the same line, English and Chinese. If you zoom in the page at 200% to 300%, you'll find out the difference in height of characters of English and Chinese.

这是一个English and 中文混杂的sentence.

Hello,

I just wanted to update this thread and let you know that this problem has been resolved in the latest build of the version 26.1 of the software
You can download it from our website:
http://www.oxygenxml.com/download.html
Let us know if you need further assistance.

Regards,
Costin