Support markdown styling in task descriptions. #121338
Conversation
| @@ -235,7 +235,7 @@ ExtensionsRegistry.registerExtensionPoint({ | |||
| }, | |||
| description: { | |||
| type: 'string', | |||
| description: localize('walkthroughs.tasks.description', "Description of task. Use markdown-style links fo commands or external links: [Title](command:myext.command), [Title](command:toSide:myext.command), or [Title](https://aka.ms)") | |||
| description: localize('walkthroughs.tasks.description', "Description of task. Supports ``preformatted``, __italic__, and **bold** text. Use markdown-style links for commands for external links: [Title](command:myext.command), [Title](command:toSide:myext.command), or [Title](https://aka.ms). Links on their own line will be rendered as buttons.") | |||
There was a problem hiding this comment.
Would pre-formatted text use a single or double backtick? To match markdown, I think it should probably be a single one
There was a problem hiding this comment.
We don't match markdown already, and this is more consistant with the existing... also it seems like markdown is happy with two backtick, even though one also works:
Markdown: one asterisk, two asterisk, one underscore, two underscore, one backtick, two backtick
Us: *one asterisk*, two asterisk, _one underscore_, two underscore, `one backtick`, two backtick
There was a problem hiding this comment.
Ok that's probably ok then and reduces the chance of accidentally creating a code block. I was confused because markdown does support ** and __ (but, as you noted, with different meanings than our formatter)
There was a problem hiding this comment.
Yeah, I prefer our approach TBH, I always end up making weird italic markdown sections when trying to type out mathematic expressions in md-enabled inputs, for instance.
| @@ -214,6 +219,8 @@ function formatTagType(char: string): FormatType { | |||
| return FormatType.Action; | |||
| case ']': | |||
| return FormatType.ActionClose; | |||
| case '`': | |||
There was a problem hiding this comment.
When a backtick happens in a string that does not set supportPreformattedSegments, what happens? How does the parser handler FormatType.Invalid?
There was a problem hiding this comment.
Parser treats invalid tokens as normal chars and appends them to the text content of the node, they'd be invalid tokens before this change too.
Closes #120050