On Sat, 31 Mar 2018, Martin Husemann wrote:
> On Fri, Mar 30, 2018 at 07:33:06PM -0400, Wes Frazier wrote:
> > FYI,
> >
> > I was able to get xfce4-gtk2-engine to build by changed the text
> > encoding of the files in the gtk-3.0 folder to UTF8. Some script it
> > calls for building purposes now seems to expect utf8 encoding.
>
> This is a python2.7 issue, it works fine when defaulting python to 3.6.
> The script is glib-mkenums, it tries to parse special annotations in
> C / C++ comments and fails to read the comments if source encoding is
> not UTF8.
>
> Apparently the unicode library changed slightly and the fallback to some
> replacement/default char is not working in older python.
In each affected file, the author's name appears in at least two places
but only the instances using byte 0xd6 cause a problem. The other
instance(s) are encoded in a two-byte sequence: 0xc3 0x96 and do not
cause a problem.
The sequences display as follows on 'urxvt' (via 'cat'):
0xd6: umlaut-O
0xc3 0x96: tilde-A
But in 'less', the "0xc3 0x96" sequence shows "tilde-A" followed by
a highlighted "<96>" which I presume it cannot interpret. It would
appear to be suppressed entirely by 'urxvt' when displayed with 'cat'.
My interrim solution was to create local patches removing the offending
instance(s) in the various files.
--
|/"\ John D. Baker, KN5UKS NetBSD Darwin/MacOS X
|\ / jdbaker[snail]mylinuxisp[flyspeck]com OpenBSD FreeBSD
| X No HTML/proprietary data in email. BSD just sits there and works!
|/ \ GPGkeyID: D703 4A7E 479F 63F8 D3F4 BD99 9572 8F23 E4AD 1645