On Sun, Apr 16, 2023, 01:37 Csepp <raingloom@riseup.net> wrote:

>
> Théo Maxime Tyburn <theo.tyburn@gmail.com> writes:
>
> > Csepp <raingloom@riseup.net> writes:
> >
> >> Théo Maxime Tyburn <theo.tyburn@gmail.com> writes:
> >>
> >>> ```
> >>> checking for OpenEXR >= 1.0.6... no
> >>> [...]
> >>> Delegate library configuration:
> >>> OpenEXR           --with-openexr=yes          no
> >>> ```
> >>
> >> If we are going by semver, then 1.0.6 is incompatible with 3.x.  Maybe
> >> it expects an older version?
> >
> > Hum. But it seems to look for a version that is greater or equal
> > 1.0.6. Shouldn't 3.x be matched ? Maybe I misunderstand semver.
> > You mean 3.x could have incompatibilities so that is what is is not
> > matched ?
>
> A major version bump is supposed to mean a backwards incompatible
> change, so depending on how the greater-than operator works, 1.x might
> not be "smaller" than 3.x.  It's not a total ordering, I think it's a...
> semilattice?  Or something along those lines.
>

Ah I see. That would make sense.

Though semver.org says:

```
Precedence is determined by the first difference when comparing each of
these identifiers from left to right as follows: Major, minor, and patch
versions are always compared numerically.

Example: 1.0.0 < 2.0.0 < 2.1.0 < 2.1.1.
```

But maybe the imagemagick config also specifies a maximum version and
doesn't say it out loud in the printed output. I'll have a look

>