Retrieve Searcher and Search from LRU Cache

We use the new cache template to implement two kind of cache.
1: The Searcher cache is more general in terms of its usage. A Searcher
   can be used for multiple searches without much change to itself. We
   try to retrieve the searcher and perform searches using it whenever
   possible, and if not we put a searcher into the cache. User can
   specify a custom cache length by manipulating the environment
   variable SEARCHER_CACHE_SIZE. It's default value is 10% of all the
   books available.
2: The search cache is much more restricted in terms of usage. It's main
   purpose is to avoid re-searching on the searcher during page changes
   to generate SearchResultSet of various ranges. User can specify a
   custom cache length using the environment variable SEARCH_CACHE_SIZE
   with a default value of 2;

src/server/internalServer.cpp

@@ -58,8 +58,6 @@ extern "C" {
 
 #include <zim/uuid.h>
 #include <zim/error.h>
-#include <zim/search.h>
-#include <zim/suggestion.h>
 #include <zim/entry.h>
 #include <zim/item.h>
 
@@ -80,6 +78,7 @@ extern "C" {
 
 #define MAX_SEARCH_LEN 140
 #define KIWIX_MIN_CONTENT_SIZE_TO_DEFLATE 100
+#define DEFAULT_CACHE_SIZE 2
 
 namespace kiwix {
 
@@ -96,6 +95,18 @@ inline std::string normalizeRootUrl(std::string rootUrl)
   return rootUrl.empty() ? rootUrl : "/" + rootUrl;
 }
 
+// Returns the value of env var `name` if found, otherwise returns defaultVal
+unsigned int getCacheLength(const char* name, unsigned int defaultVal) {
+  try {
+    const char* envString = std::getenv(name);
+    if (envString == nullptr) {
+      throw std::runtime_error("Environment variable not set");
+    }
+    return extractFromString<unsigned int>(envString);
+  } catch (...) {}
+
+  return defaultVal;
+}
 } // unnamed namespace
 
 static IdNameMapper defaultNameMapper;
@@ -134,7 +145,9 @@ InternalServer::InternalServer(Library* library,
   m_ipConnectionLimit(ipConnectionLimit),
   mp_daemon(nullptr),
   mp_library(library),
-  mp_nameMapper(nameMapper ? nameMapper : &defaultNameMapper)
+  mp_nameMapper(nameMapper ? nameMapper : &defaultNameMapper),
+  searcherCache(getCacheLength("SEARCHER_CACHE_SIZE", std::max((unsigned int) (mp_library->getBookCount(true, true)*0.1), 1U))),
+  searchCache(getCacheLength("SEARCH_CACHE_SIZE", DEFAULT_CACHE_SIZE))
 {}
 
 bool InternalServer::start() {
@@ -488,11 +501,11 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
   } catch(const std::out_of_range&) {}
     catch(const std::invalid_argument&) {}
 
-  std::string bookName;
+  std::string bookName, bookId;
   std::shared_ptr<zim::Archive> archive;
   try {
     bookName = request.get_argument("content");
-    const std::string bookId = mp_nameMapper->getIdForName(bookName);
+    bookId = mp_nameMapper->getIdForName(bookName);
     archive = mp_library->getArchiveById(bookId);
   } catch (const std::out_of_range&) {}
 
@@ -509,7 +522,7 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
 
   std::shared_ptr<zim::Searcher> searcher;
   if (archive) {
-    searcher = std::make_shared<zim::Searcher>(*archive);
+    searcher = searcherCache.getOrPut(bookId, [=](){ return std::make_shared<zim::Searcher>(*archive);});
   } else {
     for (auto& bookId: mp_library->filter(kiwix::Filter().local(true).valid(true))) {
       auto currentArchive = mp_library->getArchiveById(bookId);
@@ -540,6 +553,7 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
   }
 
   /* Get the results */
+  std::string queryString;
   try {
     zim::Query query;
     if (patternString.empty()) {
@@ -549,6 +563,7 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
       }
 
       query.setQuery("");
+      queryString = "GEO:" + to_string(latitude) + to_string(longitude) + to_string(distance);
       query.setGeorange(latitude, longitude, distance);
     } else {
       // Execute Ft search
@@ -556,13 +571,16 @@ std::unique_ptr<Response> InternalServer::handle_search(const RequestContext& re
           cout << "Performing query `" << patternString << "'" << endl;
       }
 
-      std::string queryString = removeAccents(patternString);
+      queryString = "FT:" + removeAccents(patternString);
       query.setQuery(queryString);
     }
+    queryString = bookId + queryString;
 
-    zim::Search search = searcher->search(query);
+    std::shared_ptr<zim::Search> search;
-    SearchRenderer renderer(search.getResults(start, pageLength), mp_nameMapper, mp_library, start,
+    search = searchCache.getOrPut(queryString, [=](){ return make_shared<zim::Search>(searcher->search(query));});
-                            search.getEstimatedMatches());
+
+    SearchRenderer renderer(search->getResults(start, pageLength), mp_nameMapper, mp_library, start,
+                            search->getEstimatedMatches());
     renderer.setSearchPattern(patternString);
     renderer.setSearchContent(bookName);
     renderer.setProtocolPrefix(m_root + "/");

src/server/internalServer.h

@@ -28,6 +28,9 @@ extern "C" {
 #include "library.h"
 #include "name_mapper.h"
 
+#include <zim/search.h>
+#include <zim/suggestion.h>
+
 #include <mustache.hpp>
 
 #include <atomic>
@@ -36,9 +39,13 @@ extern "C" {
 #include "server/request_context.h"
 #include "server/response.h"
 
+#include "tools/concurrent_cache.h"
+
 namespace kiwix {
 
 typedef kainjow::mustache::data MustacheData;
+typedef ConcurrentCache<string, std::shared_ptr<zim::Searcher>> SearcherCache;
+typedef ConcurrentCache<string, std::shared_ptr<zim::Search>> SearchCache;
 
 class Entry;
 class OPDSDumper;
@@ -115,6 +122,9 @@ class InternalServer {
     Library* mp_library;
     NameMapper* mp_nameMapper;
 
+    SearcherCache searcherCache;
+    SearchCache searchCache;
+
     std::string m_server_id;
     std::string m_library_id;