ReactiveX

ReactiveX

Reactive Extensions for Async Programming

Member Since 8 years ago

Experience Points
0
follower
Lessons Completed
0
follow
Best Reply Awards
42
repos
Activity
Jan
23
2 hours ago
Activity icon
fork

the-widmore forked ReactiveX/RxJavaFX

⚡ RxJava bindings for JavaFX
the-widmore Apache License 2.0 Updated
fork time in 2 minutes ago
started
started time in 7 minutes ago
started
started time in 1 hour ago
Jan
22
1 day ago
started
started time in 3 hours ago
started
started time in 3 hours ago
Activity icon
issue

coveralls issue comment ReactiveX/RxPY

coveralls
coveralls

partially revert #537

The original fix for #525 breaks sequences where some observables emit item faster than other ones. By completing too soon, the remaining observables cannot catchup later. A very simple case is in #578 where the two observables to zip emit their items sequentially.

We can fix both issues by completing whenever an observable completes and there is no queued item. Otherwise we let the remaining observables a chance to emit new items before completion.

coveralls
coveralls

Coverage Status

Coverage increased (+0.007%) to 93.51% when pulling 0a4d40aace278fca502a7febbbb4ebd3472b57f3 on MainRo:bugfix/zip into a4e84d8a488d6c7c75bdb09f6d6f08edcb2b23b0 on ReactiveX:master.

Activity icon
issue

MainRo issue comment ReactiveX/RxPY

MainRo
MainRo

partially revert #537

The original fix for #525 breaks sequences where some observables emit item faster than other ones. By completing too soon, the remaining observables cannot catchup later. A very simple case is in #578 where the two observables to zip emit their items sequentially.

We can fix both issues by completing whenever an observable completes and there is no queued item. Otherwise we let the remaining observables a chance to emit new items before completion.

pull request

MainRo pull request ReactiveX/RxPY

MainRo
MainRo

partially revert #537

The original fix for #525 breaks sequences where some observables emit item faster than other ones. By completing too soon, the remaining observables cannot catchup later. A very simple case is in #578 where the two observables to zip emit their items sequentially.

We can fix both issues by completing whenever an observable completes and there is no queued item. Otherwise we let the remaining observables a chance to emit new items before completion.

started
started time in 3 hours ago
pull request

benlesh merge to ReactiveX/rxjs

benlesh
benlesh

feat(operators): combineAll operator

Description: An operator to combine values from source observables. The values emitted by early observers are saved until the last observer emits, following which all the previously saved values are emitted.

first$: ---a---b---c--
second$: ---------------d--
result$: [ad, bd, cd]
open pull request

benlesh wants to merge ReactiveX/rxjs

benlesh
benlesh

feat(operators): combineAll operator

Description: An operator to combine values from source observables. The values emitted by early observers are saved until the last observer emits, following which all the previously saved values are emitted.

first$: ---a---b---c--
second$: ---------------d--
result$: [ad, bd, cd]
benlesh
benlesh

I see. So what you're trying to solve here is the condition where the first values of the combined observables all show up synchronously, and you might miss a few values from the "earlier" observables while waiting for the subsequent values to arrive?

I'm not sure if we can land this in RxJS core, as it's so similar to combineLatest, and probably confusing. I've seen requests for things like this before, but usually it's for people wanting this:

a$:   ---1----2-----4-----5-----
b$;   ------------3----------6----
out$: ---A----B---C--D-----E--F--- 

A: [1, null]
B: [2, null]
C: [2, 3]
D: [4, 3]
E: [5, 3]
F: [5, 6]

And I think that operator already exists in @cartant's rxjs-etc library.

Activity icon
issue

RalphSteinhagen issue ReactiveX/RxCpp

RalphSteinhagen
RalphSteinhagen

-Werror for unsafe nullptr - `decltype((*(input_type*)nullptr).out((*(Subscriber*)nullptr)))' evaluation

First, thanks for this very interesting and well-thought-through library. :+1:

The shared.cmake prescribes strict type checking via -Werror for both gcc and clang. This fails for at least gcc 11.2.1, clang 13.0.0, and possibly earlier versions for a decltype(..) evaluation relying on an nullptr cast:

[..]/RxCpp/Rx/v2/examples/cep/main.cpp:34:12:   required from here
[..]/RxCpp/Rx/v2/src/rxcpp/rx-coordination.hpp:62:56: error: ‘this’ pointer is null [-Werror=nonnull]
   62 |         using type=decltype((*(input_type*)nullptr).out((*(Subscriber*)nullptr)));
      |                             ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~

Maybe this pointer/type magic could be replaced by a safer alternative.

Since this library aims at newer compiler/C++ standards, maybe the various typedefs could possibly be replaced by using type=....

started
started time in 9 hours ago
started
started time in 10 hours ago
started
started time in 10 hours ago
started
started time in 10 hours ago
started
started time in 11 hours ago
started
started time in 12 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 13 hours ago
started
started time in 14 hours ago
Activity icon
fork

396973118 forked ReactiveX/RxJava

⚡ RxJava – Reactive Extensions for the JVM – a library for composing asynchronous and event-based programs using observable sequences for the Java VM.
396973118 Apache License 2.0 Updated
fork time in 14 hours ago
started
started time in 14 hours ago
started
started time in 15 hours ago
Previous