Image-sequencer: `Module 101|ne:1023 not found.` potentially erroring on | character in step parameters

[jywarren]

In working on this: https://github.com/publiclab/image-sequencer-app/issues/6#issuecomment-485018092

I got this error about not being able to parse parameters for the webgl-distort module:

Module 101|ne:1023 not found.

Full error here:

Module 101|ne:1023 not found.
TypeError: Cannot read property 'description' of undefined 
at insertStep (/srv/node_modules/image-sequencer/src/InsertStep.js:23:50) 
at InsertStep (/srv/node_modules/image-sequencer/src/InsertStep.js:61:3) at AddStep (/srv/node_modules/image-sequencer/src/AddStep.js:3:33) 
at Object.addSteps (/srv/node_modules/image-sequencer/src/ImageSequencer.js:75:29) 
at eachStep (/srv/node_modules/image-sequencer/src/Strings.js:132:17) 
at Array.forEach (<anonymous>) 
at Object.importString (/srv/node_modules/image-sequencer/src/Strings.js:131:21) 
at Object.sequencerInstance.loadImages (/srv/index.js:64:27) 
at /srv/node_modules/image-sequencer/src/ImageSequencer.js:179:23 
at /srv/node_modules/image-sequencer/src/ui/LoadImage.js:67:7

[Divy123]

@jywarren can you please provide some description here.
Thanks!!

[VibhorCodecianGupta]

@Divy123 It's an extension to #1033 , i was working on that one. will update as soon as i find a fix

[jywarren]

Indeed it's related, im actually not sure exactly what's happening, but I
would guess maybe it's splitting on the wrong character in the parameters
for webgl-distort?

On Thu, Apr 25, 2019, 1:08 PM Vibhor Gupta [e-mail] wrote:

It's related to #1033
https://github.com/publiclab/image-sequencer/issues/1033 , working on
it. will update as soon as i find a fix

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/image-sequencer/issues/1040#issuecomment-486757785,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAAF6J35SOXAKTFH3BEPVETPSHQMRANCNFSM4HIISR2Q
.

[tech4GT]

Looking into this now!

[tech4GT]

@VibhorCodecianGupta Anything you would like to add here?

[tech4GT]

@jywarren I am seeing this error, do you think this can be linked?

[tech4GT]

Looks like I found one of the issues @publiclab/is-reviewers
The name of fisheye-gl module is being used as "fisheye gl" which is why adding fisheye gl causes an error!

[harshkhandeparkar]

The website doesn't even work. Try the latest example website locally. It
doesn't work. Adding modules throws an error.

On Wed 8 May, 2019, 10:04 PM Varun Gupta, [e-mail] wrote:

Looks like I found one of the issues @publiclab/is-reviewers
https://github.com/orgs/publiclab/teams/is-reviewers
The name of fisheye-gl module is being used as "fisheye gl" which is why
adding fisheye gl causes an error!

—
You are receiving this because you are on a team that was mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/image-sequencer/issues/1040#issuecomment-490558526,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIJI5H3PWVI2R2N723QMMNDPUL6INANCNFSM4HIISR2Q
.

[tech4GT]

Okay this is not specific to fisheye gl

cc @jywarren

[tech4GT]

Yeah I am trying to isolate the error!

[tech4GT]

@HarshKhandeparkar I think we might have to roll back a few commits here the problem appears to be deep rooted.

[tech4GT]

I'll try to figure out which commit added this code.

[harshkhandeparkar]

I think @teisenhower had tried to rearrange the module names. That is maybe when the error started?

[harshkhandeparkar]

I think @teisenhower had tried to rearrange the module names in alphabetical order. That is maybe when the error started?

_Sent from my OnePlus3 using FastHub_

[harshkhandeparkar]

This only happens in the demo though. CLI works. This is a BIG issue. Also urlHash works, so it is only the selector that is broken.

[tech4GT]

Okay, a lot of code has been pushed in since then I think. I'll open a new issue for this then.

[harshkhandeparkar]

:+1:Ok

[tech4GT]

Harsh do you have any insight on the parsing error here?

[tech4GT]

I am trying to replicate it locally.

[harshkhandeparkar]

We really need to refactor our UI code. It is extremely messy and
unreadable. Is there any SoC proposal which mentions such a thing?

On Wed 8 May, 2019, 10:13 PM Varun Gupta, [e-mail] wrote:

I am trying to replicate it locally.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/image-sequencer/issues/1040#issuecomment-490561693,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIJI5H43IGMEO53B66EC5GTPUL7KDANCNFSM4HIISR2Q
.

[tech4GT]

I'm not really involved with soc so I'm not sure, sorry.

[harshkhandeparkar]

Np

On Wed 8 May, 2019, 10:17 PM Varun Gupta, [e-mail] wrote:

I'm not really involved with soc so I'm not sure, sorry.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/image-sequencer/issues/1040#issuecomment-490563023,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AIJI5HZBEXVJ2QV5LWU4XUTPUL7Y7ANCNFSM4HIISR2Q
.

[jywarren]

Hi folks, this issue is quite long now so there are plenty of questions or topics that may need to be broken out into their own so we can follow up with them. I'll focus on the debugging of the parsing error with | characters.

message port closed

I don't think that's it, but it could be - do you think the pipe character is in some way being read as a Linux pipe? But the error is not showing breaking on the pipe, it's breaking on either a . or a :, it looks like.

Harsh - https://sequencer.publiclab.org/examples/#steps=contrast{} is working fine right now. Please open a new issue if you see an unrelated bug, thanks!

fisheye gl - that sounds bad, and would love a PR for this, and improved UI tests should catch something like this. But in the demo right now, it does seem to work for me: https://sequencer.publiclab.org/examples/#steps=contrast{},fisheye-gl{x:1|y:1}

You can see who touched files in the examples folder here: https://github.com/publiclab/image-sequencer/commits/main/examples

[jywarren]

Noting that we published to gh-pages on April 19th (https://github.com/publiclab/image-sequencer/tree/gh-pages), but the commit by teisenhower you're mentioning was April 7th. So it seems unlikely or these errors would be appearing on the online demo right now.

[teisenhower]

Hey all

Sorry I'm just now seeing this.

Not sure if this will help or not but when I worked on that previous dropdown issue I didn't rearrange things to alphabetical order.

Please let me know if there is anything I can do to help with this.

Thanks!

[jywarren]

Thanks Terrence! Sorry, I think it is unrelated to your code, so no
worries! We're just tracking a tough bug and working through all the
possibilities. Thanks so much!

On Wed, May 8, 2019, 7:06 PM Terrence [e-mail] wrote:

Hey all

Sorry I'm just now seeing this.

Not sure if this will help or not but when I worked on that previous
dropdown issue I didn't rearrange things to alphabetical order.

Please let me know if there is anything I can do to help with this.

Thanks!

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/image-sequencer/issues/1040#issuecomment-490683467,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AAAF6JZMCW6QBM6NNXPAUS3PUNMEPANCNFSM4HIISR2Q
.

[harshkhandeparkar]

Sorry @teisenhower.

[teisenhower]

@jywarren No problem!

@HarshKhandeparkar No worries! Just wanted to make sure I could help if possible. Thanks!

[jywarren]

Hi @harshkhandeparkar, I agree with refactoring UI code, but a) let's address in a separate issue, and b) expanded UI tests would be a great first step before potentially breaking UI functionality, what do you think?

[jywarren]

https://codecov.io/gh/publiclab/image-sequencer/tree/main/examples/lib