[MacPorts] #61868: libusb @1.0.24: cannot parse USB.h with mainline gcc versions due to variant pragma handling not accepted by gcc

[MacPorts]

#61868: libusb @1.0.24: cannot parse USB.h with mainline gcc versions due to
variant pragma handling not accepted by gcc
-----------------------------------+-----------------------
  Reporter:  dgonyier              |      Owner:  michaelld
      Type:  defect                |     Status:  reopened
  Priority:  Normal                |  Milestone:
 Component:  ports                 |    Version:  2.6.4
Resolution:                        |   Keywords:
      Port:  libusb, libusb-devel  |
-----------------------------------+-----------------------

Comment (by mascguy):

 Replying to [comment:48 michaelld]:
 > Thanks @evanmiller! Very interesting & strange solution! This is
 certainly worth considering for libusb itself. I still like the generic
 solution of fixing up LegacySupport to work around this issue generically
 for impacted OSX. I just have not had enough time to finalize it!

 Agreed, this is a very unique solution. Love it!

 Replying to [comment:49 evanmiller]:
 > All right, opened a draft PR over here just for {{{libusb}}} if anyone's
 interested: https://github.com/macports/macports-ports/pull/12677
 >
 > How generic is your LegacySupport solution? At least here on Tiger, it
 appears that other Framework headers are affected by this kind of pragma
 intermingling (see e.g. [https://github.com/phracker/MacOSX-
 SDKs/blob/master/MacOSX10.4u.sdk/System/Library/Frameworks/Kernel.framework/Versions/A/Headers/IOKit/storage/IOAppleLabelScheme.h#L43-L71
 here])

 If we can find a way to (carefully) apply this via LegacySupport somehow,
 that would be awesome! That said, if folks would like to get this fix into
 `libusb` relatively quickly, we can always reconcile later.

 What do you think Michael?

-- 
Ticket URL: <https://trac.macports.org/ticket/61868#comment:51>
MacPorts <https://www.macports.org/>
Ports system for macOS