[tracker/volume-mountpoints: 8/9] Fixes GB#624346, libtracker-miner: Proper tracker:mountPoint set for volumes

[Aleksander Morgado <aleksm src gnome org>]

commit 33337fdf28a01b3c30977cad7ac77c891cfc5b06
Author: Aleksander Morgado <aleksander lanedo com>
Date:   Thu Jul 15 17:16:51 2010 +0200

    Fixes GB#624346, libtracker-miner: Proper tracker:mountPoint set for volumes
    
      * Before creating a tracker:Volume, we will first check that the specified
        tracker:mountPoint exists in the store. If it does not exist, we will
        create a dummy nfo:Folder for that mount point. But, in order to get it
        populated with the proper information when crawling it, we need to hack
        the mtime and iri caches, so that if cache is empty after querying the
        store with nfo:belongsToContainer; we do a second query using a FILTER
        with fn:start-with(uri). The dummy nfo:Folder we're inserting won't have
        initially the proper nfo:belongsToContainer property, that's why we need
        this hack.

 src/libtracker-miner/tracker-miner-fs.c |  106 +++++++++++++++++++++++++++---
 1 files changed, 95 insertions(+), 11 deletions(-)
---
diff --git a/src/libtracker-miner/tracker-miner-fs.c b/src/libtracker-miner/tracker-miner-fs.c
index 642fe23..cf5385f 100644
--- a/src/libtracker-miner/tracker-miner-fs.c
+++ b/src/libtracker-miner/tracker-miner-fs.c
@@ -1229,6 +1229,7 @@ ensure_iri_cache (TrackerMinerFS *fs,
 	gchar *query, *uri;
 	CacheQueryData data;
 	GFile *parent;
+	guint cache_size;
 
 	g_hash_table_remove_all (fs->private->iri_cache);
 
@@ -1251,7 +1252,6 @@ ensure_iri_cache (TrackerMinerFS *fs,
 	                         "  ?p nie:url \"%s\" "
 	                         "}",
 	                         uri);
-	g_free (uri);
 
 	data.main_loop = g_main_loop_new (NULL, FALSE);
 	data.values = g_hash_table_ref (fs->private->iri_cache);
@@ -1261,24 +1261,67 @@ ensure_iri_cache (TrackerMinerFS *fs,
 	                              NULL,
 	                              cache_query_cb,
 	                              &data);
+	g_free (query);
 
 	g_main_loop_run (data.main_loop);
 
 	g_main_loop_unref (data.main_loop);
 	g_hash_table_unref (data.values);
 
-	if (g_hash_table_size (data.values) == 0 &&
-	    file_is_crawl_directory (fs, file)) {
-		gchar *query_iri;
+	cache_size = g_hash_table_size (fs->private->iri_cache);
 
-		if (item_query_exists (fs, file, &query_iri, NULL)) {
-			g_hash_table_insert (data.values,
-			                     g_object_ref (file), query_iri);
+	if (cache_size == 0) {
+		if (file_is_crawl_directory (fs, file)) {
+			gchar *query_iri;
+
+			if (item_query_exists (fs, file, &query_iri, NULL)) {
+				g_hash_table_insert (data.values,
+				                     g_object_ref (file), query_iri);
+				cache_size++;
+			}
+		} else {
+			/* Quite ugly hack: If mtime_cache is found EMPTY after the query, still, we
+			 * may have a nfo:Folder where nfo:belogsToContainer was not yet set (when
+			 * generating the dummy nfo:Folder for mount points). In this case, make a
+			 * new query not using nfo:belongsToContainer, and using fn:starts-with
+			 * instead. Any better solution is highly appreciated */
+
+			/* Initialize data contents */
+			data.main_loop = g_main_loop_new (NULL, FALSE);
+			data.values = g_hash_table_ref (fs->private->iri_cache);
+
+			g_debug ("Generating iri cache for URI '%s' (fn:starts-with)", uri);
+
+			query = g_strdup_printf ("SELECT ?url ?u "
+			                         "WHERE { ?u a nfo:Folder ; "
+			                         "           nie:url ?url . "
+			                         "        FILTER (fn:starts-with (?url,\"%s\"))"
+			                         "}",
+			                         uri);
+
+			tracker_miner_execute_sparql (TRACKER_MINER (fs),
+			                              query,
+			                              NULL,
+			                              cache_query_cb,
+			                              &data);
+			g_free (query);
+
+			g_main_loop_run (data.main_loop);
+			g_main_loop_unref (data.main_loop);
+			g_hash_table_unref (data.values);
+
+			/* Note that in this case, the cache may be actually populated with items
+			 * which are not direct children of this parent, but doesn't seem a big
+			 * issue right now. In the best case, the dummy item that we created will
+			 * be there with a proper mtime set. */
+			cache_size = g_hash_table_size (fs->private->iri_cache);
 		}
 	}
 
+	g_debug ("Populated IRI cache with '%u' items", cache_size);
+
 	g_object_unref (parent);
-	g_free (query);
+	g_free (uri);
 }
 
 static const gchar *
@@ -2385,6 +2428,7 @@ ensure_mtime_cache (TrackerMinerFS *fs,
 	gchar *query, *uri;
 	CacheQueryData data;
 	GFile *parent;
+	guint cache_size;
 
 	if (G_UNLIKELY (!fs->private->mtime_cache)) {
 		fs->private->mtime_cache = g_hash_table_new_full (g_file_hash,
@@ -2452,14 +2496,55 @@ ensure_mtime_cache (TrackerMinerFS *fs,
 		                              cache_query_cb,
 		                              &data);
 		g_free (query);
-
-
 		g_main_loop_run (data.main_loop);
 	}
 
 	g_main_loop_unref (data.main_loop);
 	g_hash_table_unref (data.values);
 
+	cache_size = g_hash_table_size (fs->private->mtime_cache);
+
+	/* Quite ugly hack: If mtime_cache is found EMPTY after the query, still, we
+	 * may have a nfo:Folder where nfo:belogsToContainer was not yet set (when
+	 * generating the dummy nfo:Folder for mount points). In this case, make a
+	 * new query not using nfo:belongsToContainer, and using fn:starts-with
+	 * instead. Any better solution is highly appreciated */
+	if (parent && cache_size == 0) {
+		/* Initialize data contents */
+		data.main_loop = g_main_loop_new (NULL, FALSE);
+		data.values = g_hash_table_ref (fs->private->mtime_cache);
+		uri = g_file_get_uri (parent);
+
+		g_debug ("Generating mtime cache for URI '%s' (fn:starts-with)", uri);
+
+		query = g_strdup_printf ("SELECT ?url ?last "
+		                         "WHERE { ?u a nfo:Folder ; "
+		                         "           nie:url ?url ; "
+		                         "           nfo:fileLastModified ?last . "
+		                         "        FILTER (fn:starts-with (?url,\"%s\"))"
+		                         "}",
+		                         uri);
+		g_free (uri);
+
+		tracker_miner_execute_sparql (TRACKER_MINER (fs),
+		                              query,
+		                              NULL,
+		                              cache_query_cb,
+		                              &data);
+		g_free (query);
+		g_main_loop_run (data.main_loop);
+		g_main_loop_unref (data.main_loop);
+		g_hash_table_unref (data.values);
+
+		/* Note that in this case, the cache may be actually populated with items
+		 * which are not direct children of this parent, but doesn't seem a big
+		 * issue right now. In the best case, the dummy item that we created will
+		 * be there with a proper mtime set. */
+		cache_size = g_hash_table_size (fs->private->mtime_cache);
+	}
+
+	g_debug ("Populated mtime cache with '%u' items", cache_size);
+
 	/* Iterate repopulated HT and add all to the check_removed HT */
 	g_hash_table_foreach (fs->private->mtime_cache,
 	                      add_to_check_removed_cb,
@@ -2545,7 +2630,6 @@ should_check_file (TrackerMinerFS *fs,
 	} else {
 		g_signal_emit (fs, signals[CHECK_FILE], 0, file, &should_check);
 	}
-
 	return should_check;
 }