Remove unnecessary type cast

[line-o]

This is #66 rebased on current master

[duncdrum]

Failing test

[line-o]

Yes, but hard to tell what is failing in latest. It passes fine in exist 6.

[duncdrum]

1) should find article with extended markdown contents and code highlighting

Not sure what is hard to tell. It passes with the typecast

[line-o]

I assume it is not related to the change in this PR but rather a change in latest exist in the last 3 month

[duncdrum]

@line-o Ci runs from today are all green on main and :latest I still think the typecast is functionally necessary. It's possible that this is an instance of a bug in the typing system, or a poor function descriptor somewhere. But it looks very much related

[line-o]

I am not sure what test runs you are referring to @duncdrum
Could you link them?

I have to add that I built function-documentation from master and installed it manually on an instance built on latest develop
#72 is the result of that test

[line-o]

Ah you mean https://github.com/eXist-db/function-documentation/actions/runs/14375034667/job/40305340805
That is indeed odd

[line-o]

even the released version 1.2.1 is broken on exist 7
I am wondering what is tested in CI and why this PR is the one where it starts failing. My luck, I guess.

[duncdrum]

@line-o

function-documentation/.github/workflows/ci.yml

Line 37 in 29ccdd5

wget http://exist-db.org/exist/apps/public-repo/public/templating-1.1.0.xar -O 001.xar

the test is still running with 1.1

[line-o]

Yes, and it should pass.
See, the change is about removing a cast to boolean that is completely unnecessary.
The type is already enforced by the function signature in line 101. $details MUST be an xs:boolean

app:print-module($module, $funcsInModule, boolean($details))

What is happening here is somewhat beyond me.
And on top of all that when building exist from the develop branch you will get templating-1.2.0.xar in autodeplyoment.

[duncdrum]

And on top of all that when building exist from the develop branch you will get templating-1.2.0.xar in autodeplyoment.

On CI this is likely because there hasn't been a push event in the core repo since 1.2.1 was released that would have triggered a new container deploy.

[line-o]

Let me reiterate the failing tests are almost certainly unrelated and need to be fixed with #72
This PR can be merged before that.

[duncdrum]

@line-o i m not following. Without this change test on master are passing on ci, with it they fail.

Can you reproduce the test failure locally? What happens locally if you reverse the commit?

[line-o]

As I wrote in #72 when testing on my local machine if found the released version 1.2.1, master and this PR to be broken on exist 7 but all three run on exist 6

[duncdrum]

@line-o i agree that the typecast should not be necessary in theory, but practice shows it to be necessary. see above. This matches my local testing from when i did the housekeeping PR. Without the cast the app was broken, with it it worked.

I'll try to have a look at what happens on exist 7 by default on the weekend.

[line-o]

🤯 That's a worrying result
@duncdrum thanks for testing

[duncdrum]

yup the flight of the boolean bumblebee