Open domenic opened 10 years ago
Thanks for the plural/multiple. That was a global search/replace error, when I decided to use the word plural instead of multiple.
Yeah, the latter part of the document is older prose, which needs more of my attention for both consistency (because my ideas evolved) and attention (because it is simply later).
I am not sure I should s/resolve/fulfill/ in the last remaining task. There is a notion of "continue" that captures all kinds of resolution, which I perhaps should clarify instead.
Otherwise, I concur will all of these ideas and have a push coming soon.
{ value: x, done: true }
vs.{ value: y, done: false }
.) The introduction to iterators does not mention return values, but eventually you get around to it in the section on generator functions. Perhaps that is where they belong? But if so the way in which generators hijack the iteration object to express not only sequence values but also return values is worth calling out.yield
does, but not whatreturn
andthrow
do.then
method internally creates a new deferred..." is confusing; it's very unclear what its relation to the immediately-previous code sampel, and itsgetUserFromDatabase
, are.forEach
would return a task. Since streams are unicast, it stands to reason..." would be better as "Since streams are unicast, asynchronousforEach
would return a task. It stands to reason..."for (title of titles)
andfor (quote of quotes)
are missingvar
s.Clock
examples should use named option arguments; I have no idea whatnew Clock(1000, 500)
does (without looking at the accompanying clock implementation).<
s instead of<
s and so on. This is starting to get deeper into the document, where you haven't proofread as closely, I imagine.