Update editors/texstudio to 2.12.8

[fernape]

Summary

This comes from PR 227097

Maintainer sent update to 2.12.6 but while processing this PR, 2.12.8 came out
so I went directly for that one.

Maintainer will be notified in the PR.

Changes:

• Update to 2.12.8
• USE_GITHUB since this release is only in GH and not in SF
• Add LICENSE according to project's site (no license file is provided)
• Use DISTVERSION instead of PORTVERSION
• Flavorize port and make qt5 default (as in the old version)

Test Plan

• portlint -AC OK
• run test in 11.1 amd64 OK
• poudriere build logs for {10.3,10.4,11.1}{i386, amd64}, 12i386 availables at:

https://www.dropbox.com/sh/wcrnvdlqc2x6sop/AACj0aClm5_m_60q7cLAUIExa?dl=0

• test in CURRENT is on its way (it will take a bit longer to finish...)

[rigoletto]

Hi.

I was not aware about the change to Github.

EDIT: btw, I am not the maintainer. ^^

Thanks!

[tcberner]

editors/texstudio/Makefile

20 ↗

(On Diff #41305)

^ I think the first flavor should be the default. So if

FLAVORS= qt4 qt5
FLAVOR?=	${FLAVORS:[1]}

40 ↗

(On Diff #41305)

you should set this as
qt5_LIB_DEPENDS=... and qt4_LIB_DEPENDS=..., see 7.2.1 Falvor Helpers

See also 15.7. Flavors -- so you can move those two up a bit :)

[mat]

editors/texstudio/Makefile

20 ↗

(On Diff #41305)

YES, the default flavor must be the first.

47–51 ↗

(On Diff #41305)

This should happen earlier. See Chapter 15. Order of Variables in Port Makefiles.

[fernape]

Addressing issues:

• Move USE block up (mat)
• Make first FLAVOR the default (mat, tcberner): qt5 as in the current port

[mat]

editors/texstudio/Makefile

34 ↗

(On Diff #41372)

This should be in the FLAVORS block up there. See Chapter 15. Order of Variables in Port Makefiles.

Also, why +=?

41 ↗

(On Diff #41372)

This should be in the FLAVORS block up there. See Chapter 15. Order of Variables in Port Makefiles.

Also, why +=?

[fernape]

editors/texstudio/Makefile

34 ↗

(On Diff #41372)

This is a (bad?) habit that I have. Does it harm? because I see the benefit in case someone does a copy-paste. With '=' we would overwrite the previous value. Anyway this was pointed out to me before by tcberner so I'll change it if it is not considered quality code :)

About the other thing, if I move the .if ${FLAVOR} up just below FLAVORS= then portlint complains about "USE_* seen before USES."

[fernape]

Addressing issues:

• Use FLAVOR helpers for LIB_DEPENDS (tcberner)
• Move .if ${FLAVOR}... block up (mat)

[fernape]

Hi,

Are there any more required changes for this port?

[tcberner]

From my side it looks good -- mostly; you can move the qt4_LIB_DEPENDS and qt5_LIB_DEPENDS outside of the if.

As you're adding flavors, you're also required the OK of a portmgr before you can commit it.

[mat]

editors/texstudio/Makefile

39–40 ↗

(On Diff #41453)

This must happen earlier.

34 ↗

(On Diff #41372)

When you use += it means you are adding to the variable, it means it was already defined. If you use += on an undefined variable, nobody understands why you are doing the +=. Most of the time, it is because the variable was defined with some value before, and you needed to append, but the definition is no longer there, and maybe the appending is no longer needed.

So, only, ever, use += for the variables that are set earlier in the Makefile, or for the variables that are allowed to be set in make.conf like CFLAGS/CXXFLAGS/LDFLAGS. In all other case, never use +=.

[fernape]

Move USES earlier.

Pointed out by mat@

portlint -AC OK

[fernape]

editors/texstudio/Makefile

34 ↗

(On Diff #41372)

Understood. Thanks!

[fernape]

Hi there :)

Does this revision need more changes/testing?

[tcberner]

In D15023#332095, @fernape wrote:

Hi there :)

Does this revision need more changes/testing?

@mat: is the falvorization ok?

[mat]

The <flavor>_LIB_DEPENDS should happen in the flavors section, see Chapter 15. Order of Variables in Port Makefiles.

[fernape]

Move <flavor>_LIB_DEPENDS to FLAVOR block

[fernape]

In D15023#332199, @mat wrote:

The <flavor>_LIB_DEPENDS should happen in the flavors section, see Chapter 15. Order of Variables in Port Makefiles.

Thanks Mat!

[mat]

looks good.

[fernape]

In D15023#332850, @mat wrote:

looks good.

I got blocked by the pre-commit hook:

svn: E165001: Commit blocked by pre-commit hook (exit code 1) with output:
Do not commit a port with FLAVORS without first
getting approval from portmgr.

[mat]

well, yes, because you did not add portmgr to the reviewers, so when I approved it, it did not tell you that portmgr approved it :-)

[fernape]

In D15023#333033, @mat wrote:

well, yes, because you did not add portmgr to the reviewers, so when I approved it, it did not tell you that portmgr approved it :-)

OK, I thought it was automatic being you a member of portmgr :-)

Thanks!