[MacPorts] #72414: qt6: successful build/install after, "qt6 is known to fail" (was: successful build/install after, "qt6 is known to fail")
MacPorts
noreply at macports.org
Mon Apr 28 01:46:33 UTC 2025
#72414: qt6: successful build/install after, "qt6 is known to fail"
----------------------+---------------------------------
Reporter: tdprime | Owner: MarcusCalhoun-Lopez
Type: defect | Status: assigned
Priority: Low | Milestone:
Component: ports | Version: 2.10.5
Resolution: | Keywords:
Port: qt6 |
----------------------+---------------------------------
Changes (by ryandesign):
* cc: reneeotten (added)
* owner: (none) => MarcusCalhoun-Lopez
* status: new => assigned
Old description:
> This started with,\\
> $ sudo port install qt6\\
> Password:\\
> qt6 is known to fail. Try to install anyway? [y/N]:
>
> Very mysterious. There really should have been some explanation.
>
> I aborted. Checked, "port info qt6". Then macports-users. Although this
> was seen and reported, there was no identified cause.
>
> Very curious, I dove into the Portfile. "known_fail" was there. But, I
> didn't see the messages about SQL. Not quite sure what to make of this.
> Could be a red herring.
>
> After that brief investigation, I decided to let the install run its
> course. Expecting there would be a failure. Surprised when the whole
> thing finished successfully.
>
> So... what next? I see two things that need to be done. Most important is
> better messaging. If a failure is predicted, it should say why. If
> possible, reference a ticket ID containing more information. Second,
> refine the test that triggered the alert. Perhaps the specific conditions
> need to be better understood.
New description:
This started with,
{{{
$ sudo port install qt6
Password:
qt6 is known to fail. Try to install anyway? [y/N]:
}}}
Very mysterious. There really should have been some explanation.
I aborted. Checked, `port info qt6`. Then macports-users. Although this
was seen and reported, there was no identified cause.
Very curious, I dove into the Portfile. "known_fail" was there. But, I
didn't see the messages about SQL. Not quite sure what to make of this.
Could be a red herring.
After that brief investigation, I decided to let the install run its
course. Expecting there would be a failure. Surprised when the whole thing
finished successfully.
So... what next? I see two things that need to be done. Most important is
better messaging. If a failure is predicted, it should say why. If
possible, reference a ticket ID containing more information. Second,
refine the test that triggered the alert. Perhaps the specific conditions
need to be better understood.
--
Comment:
Your request for better messaging should be filed separately as a feature
request against MacPorts base. Ports used to implement their own messaging
by using `return -code error "some message"` but then the `known_fail`
feature was implemented in MacPorts base to streamline the process,
however the ability to specify a custom message to explain the cause of
the expected failure was not added to this feature.
Later, the `platforms` keyword was enhanced so that supported OS version
ranges could be expressed and so that it would automatically set
`known_fail` on unsupported platforms, but I don't know if it causes a
message to appear explaining the supported (or unsupported) OS versions
and it does not support a port specifying a custom message.
I'm not sure what you mean when you speak of SQL here.
--
Ticket URL: <https://trac.macports.org/ticket/72414#comment:1>
MacPorts <https://www.macports.org/>
Ports system for macOS
More information about the macports-tickets
mailing list