851321 - Implement Q like promise.done()

Status: RESOLVED WONTFIX

(Add-on SDK Graveyard :: General, defect, P1)

Description

[Erik Vold [:erikvold] (please needinfo? me)]

Promise.and() would log errors in the console and not take a 2nd argument, unlike Promise.then().

Usage would be like so:

promise.then(a).then(b).and(c);

which would log an error in a, b or c

Comment 2

[Irakli Gozalishvili [:irakli] [:gozala] [@gozala]]

We need to implement .done() method similar to one that Q has. It's documented
in details here:

http://documentup.com/kriskowal/q/#tutorial/the-end

Comment 3

[David Teller [:Yoric] - still alive but not very active]

From the document you linked: « This is a stopgap. We are exploring ways to make unhandled errors visible without any explicit handling. »

Comment 4

[David Teller [:Yoric] - still alive but not very active]

See also bug 766078 for a possible implementation strategy. For debugging mode only, of course.

Comment 5

[Joe Walker [:jwalker]]

Note that setting

    promise._reportErrors = true;

Allows a degree of error reporting by default that seems to be working OK for us right now.

Not saying that .done(); isn't useful, just that we're not shouting for it as much.

Comment 6

[Jordan Santell [:jsantell] [@jsantell]]

The current promises implementation is based off of the extremely large Q, and implementing done would require a lot of rewriting -- currently investigating other Promise implementations to leverage rather than pulling out subsets from Q when we need to add a new feature (rsvp, when), while keeping the current API.

Comment 7

[Joe Walker [:jwalker]]

(In reply to Jordan Santell [:jsantell] [@jsantell] from comment #6)
> The current promises implementation is based off of the extremely large Q,
> and implementing done would require a lot of rewriting -- currently
> investigating other Promise implementations to leverage rather than pulling
> out subsets from Q when we need to add a new feature (rsvp, when), while
> keeping the current API.

It's code looks a lot closer to "when" to me. Irakli?

Comment 8

[Jordan Santell [:jsantell] [@jsantell]]

Discussing promises in: https://etherpad.mozilla.org/resolving-promises

Comment 9

[Adrian Aichner [:anaran]]

Hard to believe this is a [good first bug]

Comment 10

[David Teller [:Yoric] - still alive but not very active]

Anyway, the whole module is deprecated.