Is Markdown Validated Differently in V21?

[Eddie]

Hello,

I have a project which includes some markdown files.
After upgrading from V20.1 to V21, I found that Oxygen V21's markdown editor is flagging errors that it did not flag in V20.1.
To be specific, the DITA preview window flags the error. The HTML preview displays the file without error.

The first error I got was at heading level, e.g, if I have:

Code: Select all

#Heading 1 Text

or

Code: Select all

##Heading 2 Text

V21 ignores the heading and looks for the first heading that has a space between the hash mark and the heading. If this heading is not the expected level, I get:

Header level raised from 0 to 3 without intermediate header level

(The numbers change, depending on the actual levels involved.)

After adding spaces between the hash marks and heading text, I get:

No renderer configured for com.vladsch.flexmark.ext.aside.AsideBlock

I've tried searching for specific info about this error online, but can't find anything.

By trial and error, I isolated the cause to pipes ( | ) being used to denote a line block.

(I still have V20.1 installed, so I can confirm the behaviour.)

Anyway, is markdown validated differently in V21? (I'm assuming "Yes".)
And if so, is there anything we can tell our writers so that they know what to expect?

Cheers,
Eddie.

[Radu]

Hi Eddie,

Oxygen 21 now comes bundled with DITA OT 3.2.1 and the publishing plugin from DITA OT which does Markdown to DITA conversions (used also by Oxygen to show the Markdown Preview and for in editor-conversions) was refactored to use the FlexMark project for parsing the Markdown files: https://github.com/vsch/flexmark-java.
About the header problem, indeed the new Markdown parser seems to require a space after the "#" but this seems correct according to the Markdown specification.
About this unhelpful error:

No renderer configured for com.vladsch.flexmark.ext.aside.AsideBlock

this means that somehow the parser has problems interpreting a certain content, can you paste a small sample with how that content looked like?

Regards,
Radu

[Eddie]

Hello Radu,

Thanks for the reply.

Regarding "No renderer configured for com.vladsch.flexmark.ext.aside.AsideBlock", the file in question has 3 lines that start with a pipe, like so:

Code: Select all

| Line 1

| Line 2

| Line 3

If I remove these lines, the file is displayed in DITA Preview. The behaviour is the same if there is only one line starting with a pipe, and also if the pipe is the only character on the line. From what I can gather the pipe is represented in some markdown specs using an "aside" tag or some such.

Cheers,

Eddie.

[Radu]

Hi Eddie,

Based on an older issue I added on the Markdown convertor project:

https://github.com/jelovirt/org.lwdita/issues/38

the convertor in DITA OT 3.3 converts those pipe lines to something like:

Code: Select all

<p>| Line 1 | Line 2 | Line 3</p>

but at least it no longer shows that error. Oxygen 21.1 will probably come bundled with DITA OT 3.3 or newer.

Regards,
Radu

[Eddie]

Hello Radu,

Thanks for the information.

Indeed, as you said in the issue you posted, a line number for the error (or a more simply written message) would be helpful.

Anyway, it's no big deal so we'll just wait for V21.1.

Cheers,
Eddie..