Backport: Several geoip changes/fixes as requested.

svn:r14782

ChangeLog

@@ -13,6 +13,10 @@ Changes in version 0.2.0.27-rc - 2008-05-??
       fails do not leave it unattached and ask the controller to deal.  Fixes
       the second part of bug 681.
 
+  o Minor features:
+    - Allow comments in geoip file.
+    - Make bridge authorities never serve extrainfo docs.
+
   o New files:
     - A new contrib/tor-exit-notice.html file that exit relay operators
       can put on their website to help reduce abuse queries.

src/or/directory.c

@@ -2461,7 +2461,7 @@ directory_handle_command_get(dir_connection_t *conn, const char *headers,
   }
 
   if (!strcmpstart(url,"/tor/server/") ||
-      !strcmpstart(url,"/tor/extra/")) {
+      (!options->BridgeAuthoritativeDir && !strcmpstart(url,"/tor/extra/"))) {
     int res;
     const char *msg;
     const char *request_type = NULL;

src/or/geoip.c

@@ -76,6 +76,10 @@ geoip_parse_entry(const char *line)
     geoip_entries = smartlist_create();
     country_idxplus1_by_lc_code = strmap_new();
   }
+  while (TOR_ISSPACE(*line))
+    ++line;
+  if (*line == '#')
+    return 0;
   if (sscanf(line,"%u,%u,%2s", &low, &high, b) == 3) {
     geoip_add_entry(low, high, b);
     return 0;
@@ -277,12 +281,12 @@ geoip_remove_old_clients(time_t cutoff)
 }
 
 /** Do not mention any country from which fewer than this number of IPs have
- * connected.  This avoids reporting information that could deanonymize
- * users. */
-#define MIN_IPS_TO_NOTE_COUNTRY 8
+ * connected.  This conceivably avoids reporting information that could
+ * deanonymize users, though analysis is lacking. */
+#define MIN_IPS_TO_NOTE_COUNTRY 0
 /** Do not report any geoip data at all if we have fewer than this number of
  * IPs to report about. */
-#define MIN_IPS_TO_NOTE_ANYTHING 16
+#define MIN_IPS_TO_NOTE_ANYTHING 0
 /** When reporting geoip data about countries, round down to the nearest
  * multiple of this value. */
 #define IP_GRANULARITY 8
@@ -344,8 +348,10 @@ geoip_get_client_history(time_t now)
       ++total;
     }
     /* Don't record anything if we haven't seen enough IPs. */
+#if MIN_IPS_TO_NOTE_ANYTHING > 0
     if (total < MIN_IPS_TO_NOTE_ANYTHING)
       goto done;
+#endif
     /* Make a list of c_hist_t */
     entries = smartlist_create();
     for (i = 0; i < n_countries; ++i) {
@@ -353,7 +359,11 @@ geoip_get_client_history(time_t now)
       const char *countrycode;
       c_hist_t *ent;
       /* Only report a country if it has a minimum number of IPs. */
+#if MIN_IPS_TO_NOTE_COUNTRY > 0
       if (c >= MIN_IPS_TO_NOTE_COUNTRY) {
+#else
+      if (1) {
+#endif
         /* Round up to the next multiple of IP_GRANULARITY */
         c += IP_GRANULARITY-1;
         c -= c % IP_GRANULARITY;
@@ -375,7 +385,9 @@ geoip_get_client_history(time_t now)
         smartlist_add(chunks, tor_strdup(buf));
       });
     result = smartlist_join_strings(chunks, ",", 0, NULL);
+#if MIN_IPS_TO_NOTE_ANYTHING > 0
   done:
+#endif
     tor_free(counts);
     if (chunks) {
       SMARTLIST_FOREACH(chunks, char *, c, tor_free(c));

src/or/router.c

@@ -1822,7 +1822,14 @@ extrainfo_dump_to_string(char *s, size_t maxlen, extrainfo_t *extrainfo,
     return -1;
 
   if (options->BridgeRelay && options->BridgeRecordUsageByCountry) {
-    char *geoip_summary = geoip_get_client_history(time(NULL));
+    static time_t last_purged_at = 0;
+    char *geoip_summary;
+    time_t now = time(NULL);
+    if (now > last_purged_at+48*60*60) {
+      geoip_remove_old_clients(now-48*60*60);
+      last_purged_at = now;
+    }
+    geoip_summary = geoip_get_client_history(time(NULL));
     if (geoip_summary) {
       char geoip_start[ISO_TIME_LEN+1];
       format_iso_time(geoip_start, geoip_get_history_start());