D1952: bundlespec: move computing the bundle contentops in parsebundlespec

[lothiraldan (Boris Feld)]

lothiraldan added a comment.


  In https://phab.mercurial-scm.org/D1952#38770, @indygreg wrote:
  
  > I like where this is going.
  >
  > It's worth noting that at some point this will reinvent the //capabilities// mechanism of bundle2. Over the wire protocol today, the client submits its bundle2 capabilities and the server emits parts based on the peer's advertised bundle2 feature support. If you squint, this looks a lot like //content options//. Have you given any consideration to merging the two concepts and having e.g. a bundlespec map to a pre-defined set of bundle2 capabilities?
  
  
  Both concepts seem intimately intricated but I was afraid of the complexity of the series that actually merge them and all the potential backward-incompatible implications. Do you have a specific design in mind we could iterate on?
  
  I will update this series according to your feedbacks.

REPOSITORY
  rHG Mercurial

REVISION DETAIL
  https://phab.mercurial-scm.org/D1952

To: lothiraldan, #hg-reviewers, indygreg
Cc: indygreg, mercurial-devel