diff options
| author | Nick Mathewson <nickm@torproject.org> | 2018-11-24 10:30:15 -0500 |
|---|---|---|
| committer | Nick Mathewson <nickm@torproject.org> | 2018-11-24 10:30:15 -0500 |
| commit | 0e9a963b6b87282011fe204e81b5c2530153a935 (patch) | |
| tree | 8fe4464dc2e2c850d579db45d1ebb3b366291cab /src/test/test_microdesc.c | |
| parent | 3741f9e524a2d3bd7239ca865d6169fd1e3dddb5 (diff) | |
| download | tor-0e9a963b6b87282011fe204e81b5c2530153a935.tar.gz tor-0e9a963b6b87282011fe204e81b5c2530153a935.zip | |
Revise nodefamily.c to match proposal 298
Prop298 says that family entries should be formatted with
$hexids in uppercase, nicknames in lower case, $hexid~names
truncated, and everything sorted lexically. These changes implement
that ordering for nodefamily.c.
We don't _strictly speaking_ need to nodefamily.c formatting use
this for prop298 microdesc generation, but it seems silly to have
two separate canonicalization algorithms.
Diffstat (limited to 'src/test/test_microdesc.c')
| -rw-r--r-- | src/test/test_microdesc.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/test/test_microdesc.c b/src/test/test_microdesc.c index 3318408d53..debb11155a 100644 --- a/src/test/test_microdesc.c +++ b/src/test/test_microdesc.c @@ -176,7 +176,7 @@ test_md_cache(void *data) tt_ptr_op(md3->family, OP_NE, NULL); encoded_family = nodefamily_format(md3->family); - tt_str_op(encoded_family, OP_EQ, "nodeX nodeY nodeZ"); + tt_str_op(encoded_family, OP_EQ, "nodex nodey nodez"); /* Now rebuild the cache! */ tt_int_op(microdesc_cache_rebuild(mc, 1), OP_EQ, 0); |