On 06/19/2013 04:01 PM, Corey Richardson wrote:
Please discuss, give me your feature requests, comments, etc.
I am unclear why the XML/JSON is part of the parsing/extraction step, it seems like it's on the same level as the generator step. I.e. after the parser/extractor/filter do their thing, the XML generator will take the internal representation and create an XML file for external tools to do what they want with, just like is done with Markdown for pandoc. I don't see what good the XML is with all the extra indentation that is only removed a step below, but without useful things like the brief descriptions. Speaking of brief descriptions... this might not be a concept that is universal to all generators (e.g. sphinx doesn't seem to use them).
I'm not sure what differentiates the filter step from the extraction step. Is it meant to be done in parallel while the rest is done serially? Overall it's not clear which steps are to be done in parallel (which should be expanded upon, as it seems to be one of the main motivations behind this).
Overall I guess what I'm really not clear about is how this is different than what is done today design-wise (aside from the multiple backends bit). It seems to me to be mostly a refactoring effort.
-SL _______________________________________________ Rust-dev mailing list [email protected] https://mail.mozilla.org/listinfo/rust-dev
