Formatter

[rakiru]

Convert a message such as "{BOLD}foo{BOLD_END}" into the correct formatting for the given protocol.

This should probably be a class instantiated as proto_instance.formatter (Mumble could use a generic HTML one then, for instance, or a subclass of one, and protocols with no formatting could use one that does nothing (or markdown-style)). Actual format of messages still to be discussed.

Should missing formatting support (such as a protocol not support italics) do nothing, or attempt to format it another way? Should a formatter that can support certain formatting, but not natively (such as Mumble/HTML not having colour reverse support) do nothing, or attempt the formatting its own way?

Want to back this issue? Post a bounty on it! We accept bounties via Bountysource.

[rakiru]

Also, all formatters should subclass a generic formatter.

This generic formatter could actual parse/tokenise the input message and call other methods to handle it (either on a per-token basis or as a list of tokens), and by default simply return an empty string (or markdown) for formatting charaters/strings.

It could also contain generic formatting functions, such as limiting a string length and appending "...", etc. On this note, see #11 (some of that stuff could be done with the python standard library).

[gdude2002]

Well, if a protocol has no formatter, supply a generic one that removes formatting tokens.

Are you good at tokenizing and such? 'cause I'm really not. xD

[rakiru]

How hard can it be? :P

[gdude2002]

It's something I was never able to wrap my head around :P

[gdude2002]

It just occurred to me that formatters should go both ways. That is to say..

• Tokens to formatting
• Formatting to tokens

We should definitely plan for this as some protocols are going to have odd formatting.

[gdude2002]

Raising this to blocker status, we really need this one.