GnuPG 2.1.3 Fails to Compile OS X

[Ben McGinnes]

On 13/04/2015 5:59 pm, Werner Koch wrote:
> On Sun, 12 Apr 2015 00:29, dominyktiller at gmail.com said:
> 
>> =========================================================
>> t-stringhelp.c:488:3: error: function definition is not allowed here
>>   {
>>   ^
> 
> Oh sorry, I didn't spotted the use of a nested function here.
> Fix pushed and attsched.

I tested this in anticipation of an update to the MacPorts package so
it uses the same configure options as the working port of 2.1.2 and
all the same dependencies with the current repo code (commit
2180845959839705200e3172dbafc94b70b9007f).  The configuration
parameters are:

configure --prefix=/opt/local --without-adns
--with-pinentry-pgm=/Applications/MacPorts/pinentry-mac.app/Contents/MacOS/pinentry-mac

It does seem to mostly compile ... it just hates all the translations:

gcc -DHAVE_CONFIG_H -I. -I..  -I../common
-DLOCALEDIR=\"/opt/local/share/locale\"
-DGNUPG_BINDIR="\"/opt/local/bin\""
-DGNUPG_LIBEXECDIR="\"/opt/local/libexec\""
-DGNUPG_LIBDIR="\"/opt/local/lib/gnupg\""
-DGNUPG_DATADIR="\"/opt/local/share/gnupg\""
-DGNUPG_SYSCONFDIR="\"/opt/local/etc/gnupg\""
-DGNUPG_LOCALSTATEDIR="\"/opt/local/var\""
-DGNUPG_DEFAULT_PINENTRY="\"/Applications/MacPorts/pinentry-mac.app/Contents/MacOS/pinentry-mac\""
    -I/opt/local/include -I/usr/local/include -I/opt/local/include
-I/opt/local/include -I/usr/local/include -I/opt/local/include -g -O2
-Wall -Wno-pointer-sign -Wpointer-arith -MT gpgsplit.o -MD -MP -MF
.deps/gpgsplit.Tpo -c -o gpgsplit.o gpgsplit.c
mv -f .deps/gpgsplit.Tpo .deps/gpgsplit.Po
gcc -I/usr/local/include -I/opt/local/include -I/opt/local/include
-I/usr/local/include -I/opt/local/include -g -O2 -Wall -Wno-pointer-sign
-Wpointer-arith   -o gpgsplit gpgsplit.o ../common/libcommon.a
-L/usr/local/lib -lgcrypt -L/opt/local/lib -lgpg-error -L/opt/local/lib
-lgpg-error -lz -lbz2 -L/opt/local/lib -lintl -Wl,-framework
-Wl,CoreFoundation  -L/opt/local/lib -liconv
Making all in po
test ! -f ./gnupg2.pot || \
	  test -z "en at quot.gmo en at boldquot.gmo ca.gmo cs.gmo da.gmo de.gmo
eo.gmo el.gmo es.gmo et.gmo fi.gmo fr.gmo gl.gmo hu.gmo id.gmo it.gmo
ja.gmo nb.gmo pl.gmo pt.gmo ro.gmo ru.gmo sk.gmo sv.gmo tr.gmo uk.gmo
zh_TW.gmo zh_CN.gmo" ||
/Applications/Xcode.app/Contents/Developer/usr/bin/make en at quot.gmo
en at boldquot.gmo ca.gmo cs.gmo da.gmo de.gmo eo.gmo el.gmo es.gmo et.gmo
fi.gmo fr.gmo gl.gmo hu.gmo id.gmo it.gmo ja.gmo nb.gmo pl.gmo pt.gmo
ro.gmo ru.gmo sk.gmo sv.gmo tr.gmo uk.gmo zh_TW.gmo zh_CN.gmo
rm -f en at quot.gmo && /opt/local/bin/msgfmt -c --statistics -o
en at quot.gmo en at quot.po
en at quot.po:54: 'msgid' and 'msgstr' entries do not both end with '\n'
en at quot.po:58: 'msgid' and 'msgstr' entries do not both end with '\n'

... [SNIP] ...

en at quot.po:336: 'msgid' and 'msgstr[0]' entries do not both end with '\n'
en at quot.po:344: 'msgid' and 'msgstr[0]' entries do not both end with '\n'

... [SNIP] ...

en at quot.po:9796: 'msgid' and 'msgstr' entries do not both end with '\n'
en at quot.po:9800: 'msgid' and 'msgstr' entries do not both end with '\n'
/opt/local/bin/msgfmt: found 792 fatal errors
2062 translated messages.
make[3]: *** [en at quot.gmo] Error 1
make[2]: *** [stamp-po] Error 2
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2

Running make check appears to mostly behave up until that same point.
In spite of the number of errors (it's all 2,065 of the msgid lines,
but no need to paste them all in); obviously the cause is somewhere
in those sed or make rules.  I'm just not certain of the right place
to insert the fix (or whether the correct course is to append the \n
to msgid or remove one from msgstr).


Regards,
Ben

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 630 bytes
Desc: OpenPGP digital signature
URL: </pipermail/attachments/20150419/1e9fbd1a/attachment.sig>