HTML ids are coming out the same for mutliple webwork exercises on a page

[Alex Jordan]

If you go to:

https://pretextbook.org/examples/webwork/sample-chapter/html/sample-ww-chapter-4-5.html

then click to activate exercise #4, focus will jump to #1 and #1 will become live. This is because all the divs on this page that wrap around the WW exercises have the same id, "-ww-inner". But according to commit b4c10e5, the ID should be coming out as:

concat(@assembly-id, '-ww-inner')

So the @assembly-id is blank. I know Claude did things with @assembly-id, but I haven't dug deep here to pinpoint the full explanation. Rob, when you can come to this, could you just ask Claude if it made a mistake here? Point out that these html id's should not be coming out all the same? I need to put off other WW things (version 2.21, the LaTeX macros stuff, sidebyside, clearing away b-human-readable) until this is figured out because it's obstructing testing. But I'll stress here that I'm in no rush. Whatever timeline works out.

[Rob Beezer]

See

https://github.com/PreTeXtBook/pretext/pull/3145

On 8/9/26 16:52, Alex Jordan wrote:
> If you go to:
> https://pretextbook.org/examples/webwork/sample-chapter/html/sample-ww-

> chapter-4-5.html <https://pretextbook.org/examples/webwork/sample-chapter/html/
> sample-ww-chapter-4-5.html>

> then click to activate exercise #4, focus will jump to #1 and #1 will become
> live. This is because all the divs on this page that wrap around the WW
> exercises have the same id, "-ww-inner". But according to commit b4c10e5, the ID
> should be coming out as:
> concat(@assembly-id, '-ww-inner')
>
> So the @assembly-id is blank. I know Claude did things with @assembly-id, but I
> haven't dug deep here to pinpoint the full explanation. Rob, when you can come
> to this, could you just ask Claude if it made a mistake here? Point out that
> these html id's should not be coming out all the same? I need to put off other
> WW things (version 2.21, the LaTeX macros stuff, sidebyside, clearing away b-
> human-readable) until this is figured out because it's obstructing testing. But
> I'll stress here that I'm in no rush. Whatever timeline works out.
>

> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/
> CA%2BR-jrdVMQaZdg9S3M%3Dyo-i67NcJQfLwo-RbTRjAV3yhf50dkQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/pretext-dev/CA%2BR-jrdVMQaZdg9S3M%3Dyo-
> i67NcJQfLwo-RbTRjAV3yhf50dkQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.

[Rob Beezer]

On fone, so partially working from memory. Yes, I use Claude a lot now, but in a very guided fashion. So let's not think Claude started this, or that it is all its fault. And for this, I likely would have fixed the problem on my own easily, but testing the bugfix was so much faster with Claude.

The assembly stylesheet adds many extra attributes. To avoid confusion (and abuse!) they were all put into a (private, internal) "pi" namespace. And any attributes just needed for assembly get removed before the tree hits any conversion. Like pi:assembly-id being a good example.

Now WW ships the resulting material from assembly to a WW server, and it comes back in a #webwork-reps. And, here is the key part, that has new attributes, with new values, *with the same names*, though now they should be not in the pi namespace. So poor Claude got a bit confused, and I think I've always had some level of misunderstanding about these myself. My fault for not noticing that Claude had not tested with the WW sample chapter and not understanding the problem sooner.

Thanks for the reports, Alex. Claude went looking for a third instance and claims to have come up empty. So, in a perverse way, I think the namespace addition has done its job, even if the route was not so smooth.

Rob

[Matt Boelkins]

After upgrading PreTeXt on the CLI and rebuilding, Mitch and I are continuing to have this issue with ACM and ACS/APC respectively.  

@oscar, could you please update the CLI at your earliest convenience to incorporate Rob's PR from 8/10?

Thanks,

Matt

[Oscar Levin]

Yes!  The new version (2.49.0) should be available in a few minutes.

[Alex Jordan]

This is fixed now, but the sample chapter and showcase article at pretextbook.org need to be rebuilt and reposted.

To unsubscribe from this group and stop receiving emails from it, send an email to pretext-dev...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/6e68ead3-432b-4b57-9bc4-6b1750c7fe7an%40googlegroups.com.

[Rob Beezer]

On 8/21/26 09:42, Alex Jordan wrote:
> This is fixed now, but the sample chapter and showcase article at

> pretextbook.org <http://pretextbook.org> need to be rebuilt and reposted.

Done, just now.