From 36e6d11dceb7af9cccdff3c631745427e31b8835 Mon Sep 17 00:00:00 2001 From: Andrew Tridgell Date: Sat, 6 Mar 2010 10:49:21 +1100 Subject: build: add README to explain gen_ndr, and ensure directory exists --- source4/heimdal_build/wscript_build | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) (limited to 'source4/heimdal_build/wscript_build') diff --git a/source4/heimdal_build/wscript_build b/source4/heimdal_build/wscript_build index faea19195e..a29a38b643 100644 --- a/source4/heimdal_build/wscript_build +++ b/source4/heimdal_build/wscript_build @@ -62,7 +62,14 @@ bld(rule="${PYTHON} ${SRC[0].abspath()} ${SRC[2].abspath()} ${SRC[2].parent.absp name = 'HEIMDAL_MAP_TABLE' ) - +# look at the example below - the output dir is lib/gssapi, the source asn1 file is lib/gssapi/spnego/spnego.asn1 +# well ... hmm, it depens if any heimdal C file uses a #include that depends on the path. We can try a different location +# but I'm guessing it will break the heimdal C files +# too hard to use a different output dir from input? +# the restriction is we can't modify the existing C source of heimdal. +# ok, symlink hacks to solve? it should be similar to the gen_ndr hackery of PIDL +# as that is also a different out dir from input file +# ok, but is it OK for PIDL? pidl is putting output in ../gen_ndr/ndr_$NAME.[ch] etc bld.SAMBA_ASN1('HEIMDAL_SPNEGO_ASN1', '../heimdal/lib/gssapi/spnego/spnego.asn1', options='--sequence=MechTypeList', -- cgit