Plots2: Can't publish a newResearch Note

[edieblu]

Steps to reproduce:

Navigate to https://publiclab.org/post and after filling in the required fields, hit publish.

Expected behavior:

The research note is published.

Actual behavior:

Nothing happens. Attaching the console errors here.

[grvsachdeva]

Hi @edieblu, thanks for reporting the issue with such descriptive steps.

@jywarren I think, PublicLab.editor integration is not properly working as the editor instance isn't created resulting in "cannot read property titleModule of undefined".

Thanks!

[jywarren]

Uh oh, maybe we can ask @rexagod or others who've been working on the
editor recently if there have been any changes that could've caused this?

On Sun, Jan 27, 2019, 5:04 PM Gaurav Sachdeva <[email protected]
wrote:

Hi @edieblu https://github.com/edieblu, thanks for reporting the issue
with such descriptive steps.

@jywarren https://github.com/jywarren I think, PublicLab.editor
integration is not properly working as the editor instance isn't created
resulting in "cannot read property titleModule of undefined".

Thanks!

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/plots2/issues/4714#issuecomment-457949450,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AABfJ-G2KtnicWGXOj_shVgoTg4Sqsjkks5vHgZEgaJpZM4aU48i
.

[jywarren]

Can we get a line number for the editor error, I wonder if it's due to the changes to the "steps left" code?

[jywarren]

Hmmm maybe not as that was only 24 h ago...

Fix Publish button disabled state
Closed by jywarren about 24 hours ago

https://github.com/publiclab/PublicLab.Editor/pull/252

[jywarren]

And perhaps we can try to replicate it on stable.publiclab.org...

[jywarren]

Hi @edieblu, do you think you could share the post text in a fist at https://gist.github.com and paste the link here? Sorry for all the trouble, we're just trying to track this down. Thanks!

[jywarren]

@sagarpreet-chadha is there a possibility that this could be related to needing additional dependencies for the new map module? See above errors... I'm trying to reproduce.

[jywarren]

Aha! I believe the error is related to not loading the Google geocoding library. Not sure why it's not loading but it is a dependency of the new Editor map module.

I tracked it to here, but I think that's just the symptom of the true cause: https://github.com/publiclab/plots2/blob/87db5642c4f434e9eaaa3478f79d2584916c6d09/app/views/editor/questionRich.html.erb#L172

[jywarren]

Aha! I believe the error is related to not loading the Google geocoding library. Not sure why it's not loading but it is a dependency of the new Editor map module.

I tracked it to here, but I think that's just the symptom of the true cause: https://github.com/publiclab/plots2/blob/87db5642c4f434e9eaaa3478f79d2584916c6d09/app/views/editor/questionRich.html.erb#L172

OK, it seems to relate to how we don't include the Google API here:

<!-- required for MapModule -->
  | <!-- <script src="https://maps.googleapis.com/maps/api/js?libraries=places&language=en&key=AIzaSyDWgc7p4WWFsO3y0MTe50vF4l4NUPcPuwE"></script> -->

OK, i think it's in /app/views/editor/rich.html.erb but not active. It's in the footer but not being activated early enough as @sagarpreet-chadha has noted. Trying to move it to the top.

[rexagod]

That would explain why none of the footer buttons seem to work.

[jywarren]

This was saved in a db-backed feature -- we have features in both header and footer so i just moved it. Will do same on stable and check there too.

[rexagod]

I did have a hunch about the mapModule causing this since that was the last major addition to the repo and the logs too indicated the same. "google" is not defined was a straight indication to the scripts not being loaded. :thinking:

[jywarren]

OK it's live on production and @edieblue you should be able to post now although i haven't confirmed this myself.

[rexagod]

@jywarren It's working now. :+1:

[jywarren]

Great. Closing this! Also noting that:

1. we should try to do our testing on https://stable.publiclab.org when possible
2. this was a tough one to catch because we are trying to keep the Google Maps API key configurable and not git-tracked, so it didn't show up in development mode.
3. i was thinking of how we could test for this bug in the future but as it's due to a dependency i think i'm less worried about this now.

Finally, if anyone has thoughts on running tests of some kind i haven't thought of that /could/ guard against this in the future, i'm all ears!!!

[edieblu]

Works great @jywarren, thanks for the fix! 👍

[sagarpreet-chadha]

Apologies for late reply 😄 !

I had faced the same error in console when i was working on mapmodule integration , referencing that comment here for future tracking : https://github.com/publiclab/plots2/pull/4608#discussion_r248002791

Instead of adding the if condition in _footer.html.erb , can i add that if condition here ?
When used in footer , the Google API gave - google not found error and when adding an explicit script tag to the page - it gave - 'google api used two times , may give unexpected error' .

So i finally removed the google API script everywhere AND kept it here . What do you think ?

[jywarren]

Thanks and yes I remember you'd warned a bit of this issue so apologies for
not testing it out more thoroughly!

On Tue, Jan 29, 2019, 8:12 AM Sagarpreet Chadha <[email protected]
wrote:

Apologies for late reply 😄 !

I had faced the same error in console when i was working on mapmodule
integration , referencing that comment here for future tracking : #4608
(comment)
https://github.com/publiclab/plots2/pull/4608#discussion_r248002791

Instead of adding the if condition in _footer.html.erb , can i add that if
condition here ?
When used in footer , the Google API gave - google not found error and
when adding an explicit script tag to the page - it gave - 'google api used
two times , may give unexpected error' .

So i finally removed the google API script everywhere AND kept it here .
What do you think ?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/publiclab/plots2/issues/4714#issuecomment-458501605,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AABfJzgdXsPOf2oFqPKghYyd7LjHUfnuks5vICx-gaJpZM4aU48i
.