Issue 1924: New configuration value for permitting crawler access

diff --git a/doc/htconfig.md b/doc/htconfig.md
index d46abb3a0bd3a64b99f3da732865d15cdff5f497..4764c287c8977c9d2efbbfcbcfb6a2afccf1f92b 100644 (file)
--- a/doc/htconfig.md
+++ b/doc/htconfig.md
@@ -44,6 +44,8 @@ line to your .htconfig.php:
 * ostatus_poll_timeframe - Defines how old an item can be to try to complete the conversation with it.
 * paranoia (Boolean) - Log out users if their IP address changed.
 * permit_crawling (Boolean) - Restricts the search for not logged in users to one search per minute.
+* free_crawls - Number of "free" searches when "permit_crawling" is activated (Default value is 10)
+* crawl_permit_period - Period in seconds between allowed searches when the number of free searches is reached and "permit_crawling" is activated (Default value is 60)
 * png_quality - Default value is 8.
 * proc_windows (Boolean) - Should be enabled if Friendica is running under Windows.
 * proxy_cache_time - Time after which the cache is cleared. Default value is one day.

diff --git a/mod/search.php b/mod/search.php
index c15dfae3fea8d981cf74efe685f9fef59078413f..7c78339c70b1638c271f90dcef86caccfbc0bd3b 100644 (file)
--- a/mod/search.php
+++ b/mod/search.php
@@ -104,20 +104,30 @@ function search_content(&$a) {
        }
 
        if (get_config('system','permit_crawling') AND !local_user()) {
-               // To-Do:
-               // - 10 requests are "free", after the 11th only a call per minute is allowed
+               // Default values:
+               // 10 requests are "free", after the 11th only a call per minute is allowed
+
+               $free_crawls = intval(get_config('system','free_crawls'));
+               if ($free_crawls == 0)
+                       $free_crawls = 10;
+
+               $crawl_permit_period = intval(get_config('system','crawl_permit_period'));
+               if ($crawl_permit_period == 0)
+                       $crawl_permit_period = 10;
 
                $remote = $_SERVER["REMOTE_ADDR"];
                $result = Cache::get("remote_search:".$remote);
                if (!is_null($result)) {
-                       if ($result > (time() - 60)) {
+                       $resultdata = json_decode($result);
+                       if (($resultdata->time > (time() - $crawl_permit_period)) AND ($resultdata->accesses > $free_crawls)) {
                                http_status_exit(429,
                                                array("title" => t("Too Many Requests"),
                                                        "description" => t("Only one search per minute is permitted for not logged in users.")));
                                killme();
                        }
-               }
-               Cache::set("remote_search:".$remote, time(), CACHE_HOUR);
+                       Cache::set("remote_search:".$remote, json_encode(array("time" => time(), "accesses" => $resultdata->accesses + 1)), CACHE_HOUR);
+               } else
+                       Cache::set("remote_search:".$remote, json_encode(array("time" => time(), "accesses" => 1)), CACHE_HOUR);
        }
 
        nav_set_selected('search');