[tor-commits] [tor/master] guards_choose_dirguard(): replace one XXXX with another.

[nickm at torproject.org]

commit 17c3faa2e393c59e9ee4aeca6986b0905d17f3b5
Author: Nick Mathewson <nickm at torproject.org>
Date:   Wed Nov 30 07:51:40 2016 -0500

    guards_choose_dirguard(): replace one XXXX with another.
    
    I had been asking myself, "hey, doesn't the new code need to look at
    this "info" parameter? The old code did!"  But it turns out that the
    old code hasn't, since 05f7336624d6a47b3.
    
    So instead of "support this!" the comment now says "we can remove
    this!"
---
 src/or/entrynodes.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/src/or/entrynodes.c b/src/or/entrynodes.c
index 9630f17..dd3a890 100644
--- a/src/or/entrynodes.c
+++ b/src/or/entrynodes.c
@@ -4779,7 +4779,13 @@ guards_choose_dirguard(dirinfo_type_t info,
   if (get_options()->UseDeprecatedGuardAlgorithm) {
     return choose_random_dirguard(info);
   } else {
-    // XXXX prop271 look at info?
+    /* XXXX prop271 We don't need to look at the dirinfo_type_t here,
+     * apparently. If you look at the old implementation, and you follow info
+     * downwards through choose_random_dirguard(), into
+     * choose_random_entry_impl(), into populate_live_entry_guards()... you
+     * find out that it isn't even used, and hasn't been since 0.2.7.1-alpha,
+     * when we realized that every Tor on the network would support
+     * microdescriptors. -NM */
     const node_t *r = NULL;
     if (entry_guard_pick_for_circuit(get_guard_selection_info(),
                                      &r,