Unable to disambiguate emphasis delimiter

[pcj]

AST nodes do not provide start or end indices into the source and Emphasis.java and StrongEmphasis.java do not provide delimiter infomation. Therefore, for the input:

Hello *Italic* **Bold** _Emph_ __Strong__ ~~Strike~~!

It does not currently seem possible to parse this, build an AST, and emit the same markdown back out, as the metadata about the delimiter character is lost.

[robinst]

It would be cool to have a CommonMark renderer (cmark has one too)! There's probably more issues related to that, so I've created #12 to track these.

Would you be interested in contributing to this? This issue for example should be pretty simple to implement. Basically put the delimiter char on the node types (see EmphasisDelimiterProcessor).

(Excuse my not being responsive by the way, I'm currently on holidays.)

[pcj]

I thought that opener.getLiteral().charAt(0) would be sufficient in EmphasisDelimiterProcessor. Sometimes the opener literal is null however. Surprised me.

[robinst]

Yeah, that's a bit surprising. The reason is that the opener and closer nodes are truncated before calling the processor in InlineParserImpl. Maybe it would make sense to do this after calling the processor.

In this case though, you can call getDelimiterChar() in EmphasisDelimiterProcessor to get the character.

[robinst]

Fixed with #36.

[robinst]

Released in 0.5.0.