WeBWorK: group WeBWorK representations into a folder#2898
Conversation
|
I see that all of the representation files for the sample chapter and showcase article are committed to the repo. How would you like that handled with this PR? As a separate commit? I'm not sure what makes it easiest for you to do before/after testing that nothing really changed with those things. |
|
For the three examples that have webwork exercises, I moved their repo representations files into the new folders this PR would create. That's on a separate commit, and I'm unsure if you'd prefer that they be merged. |
ed6e14c to
240c487
Compare
|
At first blush, I see some other changes that need to be made. For example a message I've not checked, but will image files be found correctly? I can't tell if some paths would need to change. |
240c487 to
776b95b
Compare
|
OK, that message is reworded. Yes, files are found correctly, since all of the looking for them passes through a publisher variable that is also adjusted here. I've successfully built HTML and PDF with this change, and with all old files in |
|
OK, this looks good. But it will create a headache for the CLI, methinks, and @oscarlevin.
@oscarlevin is already hopping with other changes! What is the outlook, Oscar - should this be scheduled, or sooner is better? @Alex-Jordan - no other changes of this magnitude contemplated? |
|
I'm pretty sure that the CLI looks for each file name that is needed and
regenerates everything if even one is missing. So either folks would need
to move all the files to the new location, or regenerate everything and
would have two copies of each.
So I don't think there is much I would need to change (other than a line or
two changing where to look for the files), but it would be a hassle for
authors.
Not sure how useful it is to have the added file structure. I don't think
there are too many people viewing the contents of generated assets anyway.
I think of that folder as mostly being organized to help the python tools
keep track of things.
…On Sun, Jun 7, 2026, 12:18 PM Rob Beezer ***@***.***> wrote:
*rbeezer* left a comment (PreTeXtBook/pretext#2898)
<#2898 (comment)>
OK, this looks good. But it will create a headache for the CLI, methinks,
and @oscarlevin <https://github.com/oscarlevin>.
- Assets will need a regeneration. I was trying not to put authors
onto a treadmill with repeated instances of this, as we bundled up some
changes all at once already.
- The CLI looks for the one grand file as a hint for re-generation, I
think. Maybe that logic gets more complicated, looking now for some files
at the top-level.
@oscarlevin <https://github.com/oscarlevin> is already hopping with other
changes! What is the outlook, Oscar - should this be scheduled, or sooner
is better? @Alex-Jordan <https://github.com/Alex-Jordan> - no other
changes of this magnitude contemplated?
—
Reply to this email directly, view it on GitHub
<#2898?email_source=notifications&email_token=ABRUBFFEISVLQ7IES6QTZET46WWYDA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRUGM2TSNZXHAZ2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-4643597783>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABRUBFBZPPFRWVXWO4XHJ3L46WWYDAVCNFSM6AAAAACZ5OV7HGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DMNBTGU4TONZYGM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
In terms of timing, if this gets merged in the next few hours, I should be
able to get it out in the CLI today. Otherwise it will have to sit for a
week.
…On Sun, Jun 7, 2026, 12:59 PM Oscar Levin ***@***.***> wrote:
I'm pretty sure that the CLI looks for each file name that is needed and
regenerates everything if even one is missing. So either folks would need
to move all the files to the new location, or regenerate everything and
would have two copies of each.
So I don't think there is much I would need to change (other than a line
or two changing where to look for the files), but it would be a hassle for
authors.
Not sure how useful it is to have the added file structure. I don't think
there are too many people viewing the contents of generated assets anyway.
I think of that folder as mostly being organized to help the python tools
keep track of things.
On Sun, Jun 7, 2026, 12:18 PM Rob Beezer ***@***.***> wrote:
> *rbeezer* left a comment (PreTeXtBook/pretext#2898)
> <#2898 (comment)>
>
> OK, this looks good. But it will create a headache for the CLI, methinks,
> and @oscarlevin <https://github.com/oscarlevin>.
>
> - Assets will need a regeneration. I was trying not to put authors
> onto a treadmill with repeated instances of this, as we bundled up some
> changes all at once already.
> - The CLI looks for the one grand file as a hint for re-generation, I
> think. Maybe that logic gets more complicated, looking now for some files
> at the top-level.
>
> @oscarlevin <https://github.com/oscarlevin> is already hopping with
> other changes! What is the outlook, Oscar - should this be scheduled, or
> sooner is better? @Alex-Jordan <https://github.com/Alex-Jordan> - no
> other changes of this magnitude contemplated?
>
> —
> Reply to this email directly, view it on GitHub
> <#2898?email_source=notifications&email_token=ABRUBFFEISVLQ7IES6QTZET46WWYDA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRUGM2TSNZXHAZ2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-4643597783>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ABRUBFBZPPFRWVXWO4XHJ3L46WWYDAVCNFSM6AAAAACZ5OV7HGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DMNBTGU4TONZYGM>
> .
> You are receiving this because you were mentioned.Message ID:
> ***@***.***>
>
|
|
Well, I've been away, so missed Oscar's window it seems. And this is really only if @Alex-Jordan wants it (which is fine). I like structure. ;-) But we need to do it soon, before @oscarlevin embarks on any kind of caching to differentiate the all-or-none dichotomy. If it sits for new week, then it needs to sit for two, and so should be marked off as a draft. |
|
I think it would be fine to merge, it just might make the nightlies fail
until I get back from vacation.
…On Sun, Jun 7, 2026, 5:53 PM Rob Beezer ***@***.***> wrote:
*rbeezer* left a comment (PreTeXtBook/pretext#2898)
<#2898 (comment)>
Well, I've been away, so missed Oscar's window it seems.
And this is really only if @Alex-Jordan <https://github.com/Alex-Jordan>
wants it (which is fine). I like structure. ;-)
But we need to do it soon, before @oscarlevin
<https://github.com/oscarlevin> embarks on any kind of caching to
differentiate the all-or-none dichotomy. If it sits for new week, then it
needs to sit for two, and so should be marked off as a draft.
—
Reply to this email directly, view it on GitHub
<#2898?email_source=notifications&email_token=ABRUBFF34RZ6NPNOMOAHBI346X56TA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTINRUGQ2DONJSHAZKM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-4644475282>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABRUBFD5Q7QBLVBRDRJN26T46X56TAVCNFSM6AAAAACZ5OV7HGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DMNBUGQ3TKMRYGI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
776b95b to
4b6149b
Compare
|
OK. Now suppose an individual reps file is not inside
|
Previously, the
generated/webwork/folder had a few subfolders (pg,images) and one giant representations file. Now there are representations files for each exercise, and that folder is full of them. The presence of thepgandimagesfolders is drowned out by so many individual representation files.So this change makes a
representationsfolder and puts all of the representations in there. The folder tree structure is now: