Add cctalk parsers by frank-dspeed · Pull Request #1342 · serialport/node-serialport

frank-dspeed · 2017-09-22T10:31:13Z

this adds the discussed cctalk parser
#1240

frank-dspeed · 2017-09-22T10:35:39Z

This is by the way 100% tested on real hardware in production its based on work:
https://github.com/direktspeed/serialport-parsers-cctalk

Only made compatible because i use some extra dependencies because of the ES7 and will switch to debug-extra https://github.com/direktspeed/debug-extra
that i created to debug streams and that stuff
its like fs extra extending debug with some stream and promise stuff

codecov-io · 2017-09-22T11:35:54Z

Codecov Report

Merging #1342 into master will increase coverage by 0.6%.
The diff coverage is 100%.

@@            Coverage Diff            @@
##           master    #1342     +/-   ##
=========================================
+ Coverage   78.97%   79.58%   +0.6%     
=========================================
  Files          20       20             
  Lines         899      862     -37     
  Branches      164      162      -2     
=========================================
- Hits          710      686     -24     
+ Misses        189      176     -13

Impacted Files	Coverage Δ
lib/parsers/index.js	`100% <ø> (ø)`	⬆️
lib/parsers/cctalk.js	`100% <100%> (ø)`
lib/bindings/auto-detect.js	`69.23% <0%> (-30.77%)`	⬇️
lib/bindings/win32.js
lib/serialport.js	`97.16% <0%> (+0.02%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4850901...7382fe4. Read the comment docs.

reconbot

The lint failure on Travis is weird, I'll look into it. This just needs at least a basic unit test (eg, this binary in gets these data events) so people not familiar with it's operations can ensure they don't break it and it's good to merge =)

reconbot · 2017-09-22T12:51:24Z

+  }
+  _transform(buffer, _, cb) {
+    debug('parser set')(this.buffer, this.cursor);
+    debug('parser set')(this.buffer.toString('hex'), buffer.buffer, this.cursor);


I'd either push this into a conditional debug.enabled or use a custom formatter so toString doesn't get called when not debugging.

reconbot · 2017-09-22T13:42:00Z

You can now rebase and travis will pass linting

…ode-serialport into add_cctalk_parsers

frank-dspeed · 2017-09-24T09:32:22Z

@reconbot can you please contribute to the tests?

we need to test the following Arrays

write [2,0,1,254,217] // its length 5
NOTE: the secund byte 0 indicates that the length is 5+0=5
answer [1,0,2,0,217]

write [2,2,1,254,1,1,217] // Length 7
NOTE: the secund 2 indicates that the length is 5+2= 7
answer [1,0,2,0,217]

frank-dspeed · 2017-09-24T10:04:05Z

@reconbot and please look into the current travis error i don't know any more whats going on as
it complains about buffer set wich is not really the same like buffer = that is now deprecated.

i don't know why they deprecated that because it makes sense to set some arrays at array position.

reconbot · 2017-09-24T19:11:20Z

The emits data events every 5 bytes doesn't look like it does anything relevant as we're not writing cctalk data. I'm also getting this error locally.

frank-dspeed · 2017-09-24T19:12:31Z

@reconbot yes the tests are not done when i remove them and do it localy i get totall mess :)

     .isOpen
        ✓ is a read only property
        ✓ returns false when the port is created
        ✓ returns false when the port is opening
        1) returns false when the port is opening


  151 passing (237ms)
  11 pending
  1 failing

  1) SerialPort property .isOpen returns false when the port is opening:
     TypeError: Cannot read property 'then' of undefined
      at SerialPort.open (lib/serialport.js:232:46)
      at Context.it (test/serialport.js:238:14)

reconbot · 2017-09-24T19:13:30Z

@@ -76,6 +76,7 @@ parser.on('data', console.log);



Add some docs up here ☝️

reconbot · 2017-09-24T19:14:01Z

    "chai-subset": "^1.5.0",
    "conventional-changelog-cli": "^1.3.2",
-    "eslint": "^4.5.0",
+    "eslint": "^4.7.2",


This is so weird, we totally have this in master

reconbot · 2017-09-24T19:15:27Z

+  }
+  _transform(buffer, _, cb) {
+    // TODO: Better Faster es7 no supported by node 4
+    const array = Array.prototype.slice.call(buffer, 0); // [...buffer];


Array.from(buffer)

but this seems weird that casting from a buffer to an array to a Uint8Array is the right way forward.

this is because buffers don't support the needed methods they are really static we need a event chain so i choosed to go over to array

reconbot · 2017-09-24T19:21:25Z

These tests don't pass but its a start

'use strict';
/* eslint-disable no-new */

const Buffer = require('safe-buffer').Buffer;
const sinon = require('sinon');
const CCTalkParser = require('../lib/parsers/cctalk');

describe('CCTalkParser', () => {
  it('emits data for a default length message', () => {
    const data = Buffer.from([2, 0, 1, 254, 217]);
    const spy = sinon.spy();
    const parser = new CCTalkParser();
    parser.on('data', spy);
    parser.write(data);
    assert.equal(spy.callCount, 1);
    assert.deepEqual(spy.callArgWith(Buffer.from([1, 0, 2, 0, 217])));
  });

  it('emits data for a 7 byte length message', () => {
    const parser = new CCTalkParser();
    const spy = sinon.spy();
    parser.on('data', spy);
    parser.write([2, 2, 1, 254, 1, 1, 217]);
    assert.equal(spy.callCount, 1);
    assert.deepEqual(spy.getCall(0).args[0], Buffer.from([1, 0, 2, 0, 217]));
  });
});

frank-dspeed · 2017-09-24T19:32:14Z

@reconbot i think you got it wrong because the answer should come from the device not from the parser the parser will simply emit the complet message written to it
so when we don't use your magic mockup framework to emit on that write the correct answer it can't work

frank-dspeed · 2017-09-24T19:33:54Z

[2, 0, 1, 254, 217] writen to the mockup device should let the mock up device write [1, 0, 2, 0, 217]

for the parser its enought if he emits 2 commands from this write [2, 0, 1, 254, 217,2,2,1,254,1,1,217]

first is 5 length secund is 7 length

reconbot · 2017-09-24T19:34:53Z

We're testing the wrong thing then, the parser is a stream, the stream has a behavior, and we need to ensure when we write to it, we get the right behavior. Nothing more.

…

On Sun, Sep 24, 2017, 3:32 PM Frank Lemanschik ***@***.***> wrote: @reconbot <https://github.com/reconbot> i think you got it wrong because the answer should come from the device not from the parser the parser will simply emit the complet message written to it so when we don't use your magic mockup framework to emit on that write the correct answer it can't work — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#1342 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AABlbjoeVSVD_Rj2WWI-T5V3y8D7SxCEks5slq4_gaJpZM4Pghel> .

frank-dspeed · 2017-09-24T19:36:34Z

@reconbot for the parser its enought if he emits 2 commands from this write [2, 0, 1, 254, 217,2,2,1,254,1,1,217]
so we need to check if he emits 2 times first time 2, 0, 1, 254, 217 secund time 2,2,1,254,1,1,217 thats it

reconbot · 2017-09-24T20:46:21Z

Perfect, and sinon is pretty magical and has good docs.

frank-dspeed · 2017-09-24T21:04:24Z

@reconbot but i don't understand it i will write a more good test

var c = 0
parser.write(Buffer.from([2, 0, 1, 254, 217,2,2,1,254,1,1,217])
parser.on('data',(data)=>{
if (data.length === 5 || data.length === 7) {
  c++
if (c === 2) test passed
} else {
console.log('NOT GOOD')
}

})

frank-dspeed

All Done! LGTM

frank-dspeed · 2017-09-30T08:06:22Z

@reconbot anything left undone ? it looks good to me?

frank-dspeed

All Done

reconbot · 2017-10-01T21:03:44Z

Yeah this does look good. =)

Sorry got busy with a big launch at work tomorrow.

reconbot · 2017-10-07T14:09:38Z

released with serialport@6.0.0-beta3 🎉

frank-dspeed added 2 commits September 22, 2017 11:28

Added Working CCTalk Parser

de63a9b

Added Working CCTalk Parser

7e6f6e8

frank-dspeed mentioned this pull request Sep 22, 2017

Data Event: Does not Receive Message in One Data Block => Message is Mixed Up With Sent Message #837

Closed

Apply ESLint Rule SET

50c7ec9

Improved Const Position and Name for Reading

e498b84

reconbot requested changes Sep 22, 2017

View reviewed changes

chore(packages): upgrade eslint and sinon (serialport#1343)

c590021

frank-dspeed added 9 commits September 23, 2017 13:33

Remove not needed debug

401bd30

add basic test

b41b161

Added Working CCTalk Parser

21b5ba3

Added Working CCTalk Parser

d99dcdd

Apply ESLint Rule SET

0659493

Improved Const Position and Name for Reading

00e25bb

Remove not needed debug

604198d

add basic test

897d530

Merge branch 'add_cctalk_parsers' of https://github.com/direktspeed/n…

6ce2e77

…ode-serialport into add_cctalk_parsers

Deprecating buffer.set Hope God will allow that

d421232

frank-dspeed added 3 commits September 24, 2017 13:59

New Version array based

2b00ea9

New Version array based faster

fcdd6d6

Disable tests

cf972f4

reconbot reviewed Sep 24, 2017

View reviewed changes

Fix Flatten Array

b0f9311

frank-dspeed added 7 commits September 25, 2017 12:35

Fixed: parser and Tests also updated to use array

9fa9817

Fixed: parser and Tests also updated to use array

cd5d899

Fixed: parser and Tests also updated to use array

e43809c

Fixed: parser and Tests also updated to use array

72f078a

Fixed: parser and Tests also updated to use array

a48513c

Fixed: node < 6

926f893

Finish: written basic documentation

d0fb367

frank-dspeed commented Sep 25, 2017

View reviewed changes

frank-dspeed added 2 commits September 25, 2017 22:31

Fix: written basic documentation

d5bfcfa

compiled docs for cctalk parser

7382fe4

frank-dspeed commented Oct 1, 2017

View reviewed changes

reconbot approved these changes Oct 1, 2017

View reviewed changes

reconbot merged commit bcb492f into serialport:master Oct 1, 2017

lock Bot locked and limited conversation to collaborators Apr 5, 2018

Uh oh!

Uh oh!

Conversation

frank-dspeed commented Sep 22, 2017

Uh oh!

frank-dspeed commented Sep 22, 2017

Uh oh!

codecov-io commented Sep 22, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

reconbot left a comment

Choose a reason for hiding this comment

Uh oh!

reconbot Sep 22, 2017

Choose a reason for hiding this comment

Uh oh!

reconbot commented Sep 22, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

reconbot commented Sep 24, 2017

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

reconbot Sep 24, 2017

Choose a reason for hiding this comment

Uh oh!

reconbot Sep 24, 2017

Choose a reason for hiding this comment

Uh oh!

reconbot Sep 24, 2017

Choose a reason for hiding this comment

Uh oh!

frank-dspeed Sep 25, 2017

Choose a reason for hiding this comment

Uh oh!

reconbot commented Sep 24, 2017

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

frank-dspeed commented Sep 24, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

reconbot commented Sep 24, 2017 via email

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

reconbot commented Sep 24, 2017

Uh oh!

frank-dspeed commented Sep 24, 2017

Uh oh!

frank-dspeed left a comment

Choose a reason for hiding this comment

Uh oh!

frank-dspeed commented Sep 30, 2017

Uh oh!

frank-dspeed left a comment

Choose a reason for hiding this comment

Uh oh!

reconbot commented Oct 1, 2017

Uh oh!

reconbot commented Oct 7, 2017

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

3 participants

codecov-io commented Sep 22, 2017 •

edited

Loading

reconbot commented Sep 22, 2017 •

edited

Loading

frank-dspeed commented Sep 24, 2017 •

edited

Loading