profile
viewpoint

ea167/android-layout-wrap-content-updater 26

Update dynamically View sizes when their layout_width or layout_height="wrap_content"

ea167/code-fest 1

Hour of Code / Hausner Code Fest: Intro to AI & DeepLearning

ea167/hour-of-code-rpi-node-sphero 1

Hour of Code Project for 6th-8th grade Students, involving Raspberry Pi, Node.JS, and Sphero

ea167/aws-es-kibana 0

AWS ElasticSearch Kibana Proxy

ea167/ea167.github.io 0

Fast.ai Blog Template (test)

ea167/flow 0

Adds static typing to JavaScript to improve developer productivity and code quality.

ea167/jupyker 0

Docker image for AI: Jupyter Notebook, Keras, Tensorflow, Scikit-learn, Python, and more, on Ubuntu

ea167/jupyker-cpu 0

Same as jupyker repo, but for CPU without gpu support

ea167/jupyker-nlp 0

Jupiter notebook + Tensorflow + Keras + NLP: NLTK, Spacy

starteddturing/node-gstreamer-superficial

started time in 2 hours

issue openedversatica/mediasoup-client

Mediasoup-client lib cannot be used as is without transpilation

Hello,

Among thousands of dependencies we use, Mediasoup-client is the only one that is not supported by recent browsers without transpilation. The lib seems to use JS that is too advanced even for recent browsers.

Setting, in Babel 7, target browsers as

	"safari": 	"13",
	"edge": 	"81",
	"chrome": 	"81",
	"firefox": 	"78"

WebPack still fails with several errors like

ERROR in ../audio-bo-client/~/mediasoup-client/lib/handlers/Chrome70.js
Module parse failed: /Users/evenium/DEV/GIT/audio-bo-client/node_modules/mediasoup-client/lib/handlers/Chrome70.js Unexpected token (107:12)
You may need an appropriate loader to handle this file type.
|             rtcpMuxPolicy: 'require',
|             sdpSemantics: 'unified-plan',
|             ...additionalSettings
|         }, proprietaryConstraints);
|         // Handle RTCPeerConnection connection status.

Specifically, it seems that it uses the "object rest spread operator" that is not yet supported widely (at least from Babel's point of view!).

For the server-side, it makes sense to use all the advanced JS features, but for the client-side, would it be possible to use Typescript transpilation settings that are more conservative?

Thank you for your awesome work!

created time in 15 days

startedcoturn/coturn

started time in 24 days

startedarut/nginx-rtmp-module

started time in a month

startedPAIR-code/lit

started time in a month

startedlibuv/libuv

started time in a month

startedaws/aws-graviton-getting-started

started time in a month

push eventea167/ea167.github.io

Eric Amram

commit sha 52b8cf4c4b3dbf0b645be6ea9ad77eaac3d0c0c2

Update Gemfile.lock Security

view details

push time in a month

startedversatica/mediasoup-sdp-bridge

started time in a month

startedQData/TextAttack

started time in a month

startedfluent-ffmpeg/node-fluent-ffmpeg

started time in 2 months

startedvideo-dev/hls.js

started time in 2 months

startedlawl/NoiseTorch

started time in 2 months

issue commentflowtype/flow-bin

$FlowFixMe not taken into account anymore in some situations with 0.129.0

Thanks for the link @jbrown215.

I feel deeply sad that such a great project gets ruined by a few poor decisions on crucial matters. Good luck for the future.

ea167

comment created time in 2 months

issue commentflowtype/flow-bin

$FlowFixMe not taken into account anymore in some situations with 0.129.0

Do you realize how disastrous and inconvenient this is? How much garbage it adds to the code files??

You should check the blog post Flow: $FlowFixMe command should be $FlowFixYourself to understand how wrong you are with your $FlowFixMe approach.

Unfortunately, this adds up to the too many reasons to migrate to Typescript. As for me, I'll revert to 0.126.0 and won't update anymore.

ea167

comment created time in 2 months

issue commentflowtype/flow-bin

$FlowFixMe not taken into account anymore in some situations with 0.129.0

Ok, I've identified one issue. Actually it happens when there is already another comment before, like:

// My comment before       // $FlowFixMe
delete this.myattrib;

(Was working with version 0.126 and before)

Please note that it would be much more convenient to have a marker on the same line like

delete this.myattrib;    // $FlowIgnore
ea167

comment created time in 2 months

issue commentflowtype/flow-bin

$FlowFixMe not taken into account anymore in some situations with 0.129.0

Hello @dsainati1,

My project is huge, I'll try to extract a case. Note that I upgraded from 0.126.0

If I revert to 0.126.0, all the reported errors disappear, so there is definitely something in the following versions.

ea167

comment created time in 2 months

issue openedflowtype/flow-bin

$FlowFixMe not taken into account anymore in some situations with 0.129.0

Version 0.129.0 reports errors despite $FlowFixMe

For instance:

// $FlowFixMe
delete this._isPlayBack;

reports:

Cannot delete `this._isPlayBack` because  undefined [1] is incompatible with  boolean [2]. Flow(incompatible-type)
myfile.js[864, 3]: [1] undefined
myfile.js[144, 29]: [2] boolean

I have dozens of errors like that on code that was clean since I upgraded to 0.129.0. Am I missing something?

created time in 2 months

push eventEvRestricted/node-webrtc

ea167

commit sha d065b397232cb2fe932df1ba8836bc2e3b821151

bf

view details

push time in 2 months

push eventEvRestricted/node-webrtc

ea167

commit sha 08749291f8fcda01c373552114b09d0f6540d51d

bf

view details

push time in 2 months

push eventEvRestricted/node-webrtc

ea167

commit sha 9f2b4a2137ad9a48817c094429ee7b92b22e48c0

cmt

view details

push time in 2 months

push eventEvRestricted/node-webrtc

Eric Amram

commit sha dfea9cac73e35e998dffda2831bb85c681b576a2

"name": "wrtc-ev-h264" + Restore the 0.4.6 version number

view details

push time in 2 months

push eventEvRestricted/node-webrtc

Eric Amram

commit sha b0830a61b8a462a62d5bbdaee46ef747f271242b

disable download, force build from source

view details

push time in 2 months

push eventEvRestricted/node-webrtc

Eric Amram

commit sha 4cd17acee8464081ad2d7c2c59d7191ca6e1ff50

Enable H264 codec in libwebrtc Added proprietary_codecs=true ffmpeg_branding="Chrome" rtc_use_h264=true

view details

push time in 2 months

startedxiph/rnnoise

started time in 2 months

issue commentansible/ansible

Docs not clear: git clone to existing non-empty

The current behaviour breaks Ansible rules: Ansible is supposed to be idempotent, and this behaviour is not.

As @TheLastProject said, there should be an update parameter that defaults to yes.

I was also surprised by the need of including the repo name in dest, which is not the default way of how the command-line git works. Overall it makes the use of the git module error-prone and counter-intuitive.

ansibot

comment created time in 2 months

startedtyiannak/pyAudioAnalysis

started time in 3 months

startedFFmpeg/FFmpeg

started time in 3 months

startedZenika/alpine-chrome

started time in 3 months

startedpuppeteer/puppeteer

started time in 3 months

startedmicrosoft/computervision-recipes

started time in 3 months

startedcastorini/honkling

started time in 3 months

more