Date range no longer updating from URL

[trvrb]

On current master the date slider and the redux date range is no longer properly loading from the URL. Go here to see the behavior:

http://www.nextstrain.org/zika?dmax=2015-06-21&dmin=2015-02-25

Date range should be subsetted, but it is not.

[jameshadfield]

comment out https://github.com/nextstrain/auspice/blob/master/src/actions/loadData.js#L77 and https://github.com/nextstrain/auspice/blob/master/src/reducers/controls.js#L128 (and L129) and the old behavior is restored

[jameshadfield]

@colinmegill I wonder if we should look into async await (or similar) for the JSON requests - it would definitely make race conditions like this easier to handle...

[colinmegill]

Regardless of whether async / await is used, the underlying infrastructure will be this: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise/all which is a safe bet to go ahead and use.

…

On Wed, May 3, 2017 at 11:08 PM james hadfield ***@***.***> wrote: @colinmegill <https://github.com/colinmegill> I wonder if we should look into async await (or similar) - it would definitely make race conditions like easier to handle... — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#320 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABsDGaZMkAi5KhsK16w0j-OEEOtht86tks5r2Wt4gaJpZM4NP2J-> .

[colinmegill]

Here's the polyfill: https://github.com/stefanpenner/es6-promise And the compat table: https://caniuse.com/#search=promise

…

On Thu, May 4, 2017 at 7:48 AM Colin Megill ***@***.***> wrote: Regardless of whether async / await is used, the underlying infrastructure will be this: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise/all which is a safe bet to go ahead and use. On Wed, May 3, 2017 at 11:08 PM james hadfield ***@***.***> wrote: > @colinmegill <https://github.com/colinmegill> I wonder if we should look > into async await (or similar) - it would definitely make race conditions > like easier to handle... > > — > You are receiving this because you were mentioned. > > > Reply to this email directly, view it on GitHub > <#320 (comment)>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/ABsDGaZMkAi5KhsK16w0j-OEEOtht86tks5r2Wt4gaJpZM4NP2J-> > . >

[colinmegill]

Here's an example from the creator of Redux: reduxjs/redux#723

…

On Thu, May 4, 2017 at 7:49 AM Colin Megill ***@***.***> wrote: Here's the polyfill: https://github.com/stefanpenner/es6-promise And the compat table: https://caniuse.com/#search=promise On Thu, May 4, 2017 at 7:48 AM Colin Megill ***@***.***> wrote: > Regardless of whether async / await is used, the underlying > infrastructure will be this: > https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise/all which > is a safe bet to go ahead and use. > > On Wed, May 3, 2017 at 11:08 PM james hadfield ***@***.***> > wrote: > >> @colinmegill <https://github.com/colinmegill> I wonder if we should >> look into async await (or similar) - it would definitely make race >> conditions like easier to handle... >> >> — >> You are receiving this because you were mentioned. >> >> >> Reply to this email directly, view it on GitHub >> <#320 (comment)>, >> or mute the thread >> <https://github.com/notifications/unsubscribe-auth/ABsDGaZMkAi5KhsK16w0j-OEEOtht86tks5r2Wt4gaJpZM4NP2J-> >> . >> >

[trvrb]

Sounds like this is more complicated than I initially imagined.

[jameshadfield]

Thanks @colinmegill - From the docs:

The Promise.all() method returns a single Promise that resolves when all of the promises in the iterable argument have resolved, or rejects with the reason of the first promise that rejects.

But we want to allow some 404s (e.g. zika doesn't have frequencies.json). This kind of functionality isn't even in ES6 AFAIK. Seems like Q.allSettled(promises) from https://github.com/kriskowal/q is what we're after.

Or maybe we can do promise.all on the meta + tree, and allow sequences, frequencies, entropy etc to load similarly to now... i'll try this approach today.