rename R.repeatN and revert argument order#629
Conversation
There was a problem hiding this comment.
I would like to make each of the extensions its own repository with its own npm package. Ideally these would not be Ramda extensions specifically but small, focused JavaScript packages which work well with Ramda but do not depend on it.
For now, we can avoid the naming collision by treating R.lazylist as a namespace.
There was a problem hiding this comment.
i think that makes sense; they can all live under ramda/ group
There was a problem hiding this comment.
they can all live under ramda/ group
They can indeed. This'll imply that they're compatible with, and philosophically consistent with, Ramda itself, while promoting them to first-class citizens with their own issue trackers, release cycles, and top-level namespaces.
There was a problem hiding this comment.
although i think it makes sense to have a ramda/fantasy repo for the algebraic types, rather than each type in its own
There was a problem hiding this comment.
Wonderful! I'm very pleased we agree on this. :)
There was a problem hiding this comment.
Only trouble I see is that we wanted the ability to make custom builds that included some of these extensions as desired. Not sharing a code-base will make this trickier, no?
We needn't conflate the two. If a user wants to use a custom Ramda build such as --category list along with ramda/lazylist, she is of course free to do so. There's no reason to combine these into a single file. That's best left to a dedicated build tool, as one will no doubt want to bundle one's application files as well.
There was a problem hiding this comment.
I'm just having a hard time giving up on what has been a major goal for me in the modularization all along. I know that Mike was worried about bloat and being able to create smaller packages as needed. I was going in the other direction, thinking about how to include those Ramda-parts you wanted in a single simple Ramda file.
That has also, of course, added complexity to the modularization code, and made it much uglier. But it's still hard to consider dropping.
There was a problem hiding this comment.
One benefit of separating the extensions is that it would make it possible to use ramda/lazylist, for example, without ramda/ramda. This is in line with the goal of producing JavaScript files containing only what's necessary.
There was a problem hiding this comment.
Although that may be useful, since I've always thought of these as optional extensions that would be included in certain builds as part of Ramda, that's not much of a pull for me.
|
thread on a tangent! imagine that. this is good to merge IMO |
|
I'll let Scott do the honours on this one. |
|
/ping @CrossEye |
rename R.repeatN and revert argument order
As discussed in #626