X-Git-Url: https://git.notmuchmail.org/git?p=notmuch;a=blobdiff_plain;f=lib%2Fdatabase.cc;h=bf56f5206d0f950290ec31637e6fc68516acfe25;hp=bcea88b029e7a097a7887c7570e5715e7a78fc76;hb=406ec4b15d65f1104c7ff3ee654a5e9cd5b39f29;hpb=e2341cbc09b503f996fd46b68f9d96ae6004025b

diff --git a/lib/database.cc b/lib/database.cc
index bcea88b0..bf56f520 100644
--- a/lib/database.cc
+++ b/lib/database.cc
@@ -37,7 +37,7 @@ typedef struct {
 
 /* Here's the current schema for our database:
  *
- * We currently have two different types of documents: mail and timestamps.
+ * We currently have two different types of documents: mail and directory.
  *
  * Mail document
  * -------------
@@ -75,21 +75,24 @@ typedef struct {
  * user in searching. But the database doesn't really care itself
  * about any of these.
  *
- * Timestamp document
+ * Finally, the data portion of a mail document contains the path name
+ * of the mail message (relative to the database path).
+ *
+ * Directory document
  * ------------------
- * A timestamp document is used by a client of the notmuch library to
+ * A directory document is used by a client of the notmuch library to
  * maintain data necessary to allow for efficient polling of mail
- * directories. The notmuch library does no interpretation of
- * timestamps, but merely allows the user to store and retrieve
- * timestamps as name/value pairs.
+ * directories.
  *
- * The timestamp document is indexed with a single prefixed term:
+ * The directory document contains the following terms:
  *
- *	timestamp:	The user's key value (likely a directory name)
+ *	directory:	The directory path (relative to the database path)
+ *	parent:		The document ID of the parent directory document.
+ *			Top-level directories will have a parent value of 0.
  *
  * and has a single value:
  *
- *	TIMESTAMP:	The time_t value from the user.
+ *	TIMESTAMP:	The mtime of the directory (at last scan)
  */
 
 /* With these prefix values we follow the conventions published here:
@@ -111,7 +114,8 @@ prefix_t BOOLEAN_PREFIX_INTERNAL[] = {
     { "type", "T" },
     { "reference", "XREFERENCE" },
     { "replyto", "XREPLYTO" },
-    { "timestamp", "XTIMESTAMP" },
+    { "directory", "XDIRECTORY" },
+    { "parent", "XPARENT" },
 };
 
 prefix_t BOOLEAN_PREFIX_EXTERNAL[] = {
@@ -147,17 +151,20 @@ _find_prefix (const char *name)
 {
     unsigned int i;
 
-    for (i = 0; i < ARRAY_SIZE (BOOLEAN_PREFIX_INTERNAL); i++)
+    for (i = 0; i < ARRAY_SIZE (BOOLEAN_PREFIX_INTERNAL); i++) {
 	if (strcmp (name, BOOLEAN_PREFIX_INTERNAL[i].name) == 0)
 	    return BOOLEAN_PREFIX_INTERNAL[i].prefix;
+    }
 
-    for (i = 0; i < ARRAY_SIZE (BOOLEAN_PREFIX_EXTERNAL); i++)
+    for (i = 0; i < ARRAY_SIZE (BOOLEAN_PREFIX_EXTERNAL); i++) {
 	if (strcmp (name, BOOLEAN_PREFIX_EXTERNAL[i].name) == 0)
 	    return BOOLEAN_PREFIX_EXTERNAL[i].prefix;
+    }
 
-    for (i = 0; i < ARRAY_SIZE (PROBABILISTIC_PREFIX); i++)
+    for (i = 0; i < ARRAY_SIZE (PROBABILISTIC_PREFIX); i++) {
 	if (strcmp (name, PROBABILISTIC_PREFIX[i].name) == 0)
 	    return PROBABILISTIC_PREFIX[i].prefix;
+    }
 
     INTERNAL_ERROR ("No prefix exists for '%s'\n", name);
 
@@ -295,13 +302,14 @@ skip_space_and_comments (const char **str)
 	    int nesting = 1;
 	    s++;
 	    while (*s && nesting) {
-		if (*s == '(')
+		if (*s == '(') {
 		    nesting++;
-		else if (*s == ')')
+		} else if (*s == ')') {
 		    nesting--;
-		else if (*s == '\\')
+		} else if (*s == '\\') {
 		    if (*(s+1))
 			s++;
+		}
 		s++;
 	    }
 	}
@@ -497,11 +505,13 @@ notmuch_database_open (const char *path,
 	notmuch->query_parser = new Xapian::QueryParser;
 	notmuch->term_gen = new Xapian::TermGenerator;
 	notmuch->term_gen->set_stemmer (Xapian::Stem ("english"));
+	notmuch->value_range_processor = new Xapian::NumberValueRangeProcessor (NOTMUCH_VALUE_TIMESTAMP);
 
 	notmuch->query_parser->set_default_op (Xapian::Query::OP_AND);
 	notmuch->query_parser->set_database (*notmuch->xapian_db);
 	notmuch->query_parser->set_stemmer (Xapian::Stem ("english"));
 	notmuch->query_parser->set_stemming_strategy (Xapian::QueryParser::STEM_SOME);
+	notmuch->query_parser->add_valuerangeprocessor (notmuch->value_range_processor);
 
 	for (i = 0; i < ARRAY_SIZE (BOOLEAN_PREFIX_EXTERNAL); i++) {
 	    prefix_t *prefix = &BOOLEAN_PREFIX_EXTERNAL[i];
@@ -518,7 +528,7 @@ notmuch_database_open (const char *path,
 		 error.get_msg().c_str());
 	notmuch = NULL;
     }
-    
+
   DONE:
     if (notmuch_path)
 	free (notmuch_path);
@@ -544,6 +554,7 @@ notmuch_database_close (notmuch_database_t *notmuch)
     delete notmuch->term_gen;
     delete notmuch->query_parser;
     delete notmuch->xapian_db;
+    delete notmuch->value_range_processor;
     talloc_free (notmuch);
 }
 
@@ -554,60 +565,203 @@ notmuch_database_get_path (notmuch_database_t *notmuch)
 }
 
 static notmuch_private_status_t
-find_timestamp_document (notmuch_database_t *notmuch, const char *db_key,
+find_directory_document (notmuch_database_t *notmuch, const char *db_path,
 			 Xapian::Document *doc, unsigned int *doc_id)
 {
-    return find_unique_document (notmuch, "timestamp", db_key, doc, doc_id);
+    return find_unique_document (notmuch, "directory", db_path, doc, doc_id);
 }
 
-/* We allow the user to use arbitrarily long keys for timestamps,
- * (they're for filesystem paths after all, which have no limit we
- * know about). But we have a term-length limit. So if we exceed that,
- * we'll use the SHA-1 of the user's key as the actual key for
- * constructing a database term.
+/* We allow the user to use arbitrarily long paths for directories. But
+ * we have a term-length limit. So if we exceed that, we'll use the
+ * SHA-1 of the path for the database term.
  *
- * Caution: This function returns a newly allocated string which the
- * caller should free() when finished.
+ * Note: This function may return the original value of 'path'. If it
+ * does not, then the caller is responsible to free() the returned
+ * value.
  */
-static char *
-timestamp_db_key (const char *key)
+static const char *
+directory_db_path (const char *path)
 {
-    int term_len = strlen (_find_prefix ("timestamp")) + strlen (key);
+    int term_len = strlen (_find_prefix ("directory")) + strlen (path);
 
     if (term_len > NOTMUCH_TERM_MAX)
-	return notmuch_sha1_of_string (key);
+	return notmuch_sha1_of_string (path);
     else
-	return strdup (key);
+	return path;
+}
+
+/* Given a 'path' (relative to the database path) return the document
+ * ID of the directory document corresponding to the parent directory
+ * of 'path' in 'parent_id'.
+ *
+ * The original 'path' can represent either a regular file or a
+ * directory, (in either case, the document ID of the parent will be
+ * returned). Trailing slashes on 'path' will be ignored, and any
+ * cases of multiple '/' characters appearing in series will be
+ * treated as a single '/'.
+ *
+ * If no directory document exists in the database for the parent, (or
+ * for any of its parents up to the top-level database path), then
+ * directory documents will be created for these (each with an mtime
+ * of 0).
+ *
+ * Return value:
+ *
+ * NOTMUCH_STATUS_SUCCESS: Valid value available in parent_id.
+ *
+ * NOTMUCH_STATUS_XAPIAN_EXCEPTION: A Xapian exception
+ *	occurred and parent_id will be set to (unsigned) -1.
+ */
+notmuch_status_t
+_notmuch_database_find_parent_id (notmuch_database_t *notmuch,
+				  const char *path,
+				  unsigned int *parent_id)
+{
+    const char *slash, *parent_db_path;
+    char *parent_path;
+    notmuch_private_status_t private_status;
+    notmuch_status_t status = NOTMUCH_STATUS_SUCCESS;
+
+    if (path == NULL || *path == '\0') {
+	*parent_id = 0;
+	return NOTMUCH_STATUS_SUCCESS;
+    }
+
+    /* Find the last slash (not counting a trailing slash), if any. */
+
+    slash = path + strlen (path) - 1;
+
+    /* First, skip trailing slashes. */
+    while (slash != path) {
+	if (*slash != '/')
+	    break;
+
+	--slash;
+    }
+
+    /* Then, find a slash. */
+    while (slash != path) {
+	if (*slash == '/')
+	    break;
+
+	--slash;
+    }
+
+    /* Finally, skip multiple slashes. */
+    while (slash != path) {
+	if (*slash != '/')
+	    break;
+
+	--slash;
+    }
+
+    if (slash == path)
+	parent_path = talloc_strdup (notmuch, "");
+    else
+	parent_path = talloc_strndup (notmuch, path, slash - path + 1);
+
+    parent_db_path = directory_db_path (parent_path);
+
+    private_status = find_unique_doc_id (notmuch, "directory",
+					 parent_db_path, parent_id);
+    if (private_status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND) {
+	status = notmuch_database_set_directory_mtime (notmuch,
+						       parent_path, 0);
+	if (status)
+	    return status;
+	private_status = find_unique_doc_id (notmuch, "directory",
+					     parent_db_path, parent_id);
+	status = COERCE_STATUS (private_status, "_find_parent_id");
+    }
+
+    if (parent_db_path != parent_path)
+	free ((char *) parent_db_path);
+
+    talloc_free (parent_path);
+
+    if (status)
+	*parent_id = -1;
+
+    return status;
+}
+
+/* Given a legal 'path' for the database, return the relative path.
+ *
+ * The return value will be a pointer to the originl path contents,
+ * and will be either the original string (if 'path' was relative) or
+ * a portion of the string (if path was absolute and begins with the
+ * database path).
+ */
+const char *
+_notmuch_database_relative_path (notmuch_database_t *notmuch,
+				 const char *path)
+{
+    const char *db_path, *relative;
+    unsigned int db_path_len;
+
+    db_path = notmuch_database_get_path (notmuch);
+    db_path_len = strlen (db_path);
+
+    relative = path;
+
+    if (*relative == '/') {
+	while (*relative == '/' && *(relative+1) == '/')
+	    relative++;
+
+	if (strncmp (relative, db_path, db_path_len) == 0)
+	{
+	    relative += db_path_len;
+	    while (*relative == '/')
+		relative++;
+	}
+    }
+
+    return relative;
 }
 
 notmuch_status_t
-notmuch_database_set_timestamp (notmuch_database_t *notmuch,
-				const char *key, time_t timestamp)
+notmuch_database_set_directory_mtime (notmuch_database_t *notmuch,
+				      const char *path,
+				      time_t mtime)
 {
     Xapian::Document doc;
     Xapian::WritableDatabase *db;
     unsigned int doc_id;
     notmuch_private_status_t status;
     notmuch_status_t ret = NOTMUCH_STATUS_SUCCESS;
-    char *db_key = NULL;
+    const char *db_path = NULL;
+    unsigned int parent_id;
 
     if (notmuch->mode == NOTMUCH_DATABASE_MODE_READ_ONLY) {
 	fprintf (stderr, "Attempted to update a read-only database.\n");
 	return NOTMUCH_STATUS_READONLY_DATABASE;
     }
 
+    path = _notmuch_database_relative_path (notmuch, path);
+
     db = static_cast <Xapian::WritableDatabase *> (notmuch->xapian_db);
-    db_key = timestamp_db_key (key);
+    db_path = directory_db_path (path);
 
     try {
-	status = find_timestamp_document (notmuch, db_key, &doc, &doc_id);
+	status = find_directory_document (notmuch, db_path, &doc, &doc_id);
 
 	doc.add_value (NOTMUCH_VALUE_TIMESTAMP,
-		       Xapian::sortable_serialise (timestamp));
+		       Xapian::sortable_serialise (mtime));
 
 	if (status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND) {
 	    char *term = talloc_asprintf (NULL, "%s%s",
-					  _find_prefix ("timestamp"), db_key);
+					  _find_prefix ("directory"), db_path);
+	    doc.add_term (term);
+	    talloc_free (term);
+
+	    ret = _notmuch_database_find_parent_id (notmuch, path,
+						    &parent_id);
+	    if (ret)
+		return ret;
+
+	    term = talloc_asprintf (NULL, "%s%u",
+				    _find_prefix ("parent"),
+				    parent_id);
 	    doc.add_term (term);
 	    talloc_free (term);
 
@@ -617,43 +771,45 @@ notmuch_database_set_timestamp (notmuch_database_t *notmuch,
 	}
 
     } catch (const Xapian::Error &error) {
-	fprintf (stderr, "A Xapian exception occurred setting timestamp: %s.\n",
+	fprintf (stderr, "A Xapian exception occurred setting directory mtime: %s.\n",
 		 error.get_msg().c_str());
 	notmuch->exception_reported = TRUE;
 	ret = NOTMUCH_STATUS_XAPIAN_EXCEPTION;
     }
 
-    if (db_key)
-	free (db_key);
+    if (db_path != path)
+	free ((char *) db_path);
 
     return ret;
 }
 
 time_t
-notmuch_database_get_timestamp (notmuch_database_t *notmuch, const char *key)
+notmuch_database_get_directory_mtime (notmuch_database_t *notmuch,
+				      const char *path)
 {
     Xapian::Document doc;
     unsigned int doc_id;
     notmuch_private_status_t status;
-    char *db_key = NULL;
+    const char *db_path = NULL;
     time_t ret = 0;
 
-    db_key = timestamp_db_key (key);
+    db_path = directory_db_path (path);
 
     try {
-	status = find_timestamp_document (notmuch, db_key, &doc, &doc_id);
+	status = find_directory_document (notmuch, db_path, &doc, &doc_id);
 
 	if (status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND)
 	    goto DONE;
 
 	ret =  Xapian::sortable_unserialise (doc.get_value (NOTMUCH_VALUE_TIMESTAMP));
     } catch (Xapian::Error &error) {
+	ret = 0;
 	goto DONE;
     }
 
   DONE:
-    if (db_key)
-	free (db_key);
+    if (db_path != path)
+	free ((char *) db_path);
 
     return ret;
 }
@@ -847,12 +1003,11 @@ _notmuch_database_link_message_to_children (notmuch_database_t *notmuch,
  *
  * We first look at 'message_file' and its link-relevant headers
  * (References and In-Reply-To) for message IDs. We also look in the
- * database for existing message that reference 'message'.p
+ * database for existing message that reference 'message'.
  *
- * The end result is to call _notmuch_message_add_thread_id with one
- * or more thread IDs to which this message belongs, (including
- * generating a new thread ID if necessary if the message doesn't
- * connect to any existing threads).
+ * The end result is to call _notmuch_message_ensure_thread_id which
+ * generates a new thread ID if the message doesn't connect to any
+ * existing threads.
  */
 static notmuch_status_t
 _notmuch_database_link_message (notmuch_database_t *notmuch,
@@ -891,7 +1046,7 @@ notmuch_database_add_message (notmuch_database_t *notmuch,
 
     const char *date, *header;
     const char *from, *to, *subject;
-    char *message_id;
+    char *message_id = NULL;
 
     if (message_ret)
 	*message_ret = NULL;
@@ -936,11 +1091,20 @@ notmuch_database_add_message (notmuch_database_t *notmuch,
 	header = notmuch_message_file_get_header (message_file, "message-id");
 	if (header && *header != '\0') {
 	    message_id = _parse_message_id (message_file, header, NULL);
+
 	    /* So the header value isn't RFC-compliant, but it's
 	     * better than no message-id at all. */
 	    if (message_id == NULL)
 		message_id = talloc_strdup (message_file, header);
-	} else {
+
+	    /* Reject a Message ID that's too long. */
+	    if (message_id && strlen (message_id) + 1 > NOTMUCH_TERM_MAX) {
+		talloc_free (message_id);
+		message_id = NULL;
+	    }
+	}
+
+	if (message_id == NULL ) {
 	    /* No message-id at all, let's generate one by taking a
 	     * hash over the file's contents. */
 	    char *sha1 = notmuch_sha1_of_file (filename);
@@ -1012,3 +1176,46 @@ notmuch_database_add_message (notmuch_database_t *notmuch,
 
     return ret;
 }
+
+notmuch_tags_t *
+_notmuch_convert_tags (void *ctx, Xapian::TermIterator &i,
+		       Xapian::TermIterator &end)
+{
+    const char *prefix = _find_prefix ("tag");
+    notmuch_tags_t *tags;
+    std::string tag;
+
+    /* Currently this iteration is written with the assumption that
+     * "tag" has a single-character prefix. */
+    assert (strlen (prefix) == 1);
+
+    tags = _notmuch_tags_create (ctx);
+    if (unlikely (tags == NULL))
+	return NULL;
+
+    i.skip_to (prefix);
+
+    while (i != end) {
+	tag = *i;
+
+	if (tag.empty () || tag[0] != *prefix)
+	    break;
+
+	_notmuch_tags_add_tag (tags, tag.c_str () + 1);
+
+	i++;
+    }
+
+    _notmuch_tags_prepare_iterator (tags);
+
+    return tags;
+}
+
+notmuch_tags_t *
+notmuch_database_get_all_tags (notmuch_database_t *db)
+{
+    Xapian::TermIterator i, end;
+    i = db->xapian_db->allterms_begin();
+    end = db->xapian_db->allterms_end();
+    return _notmuch_convert_tags(db, i, end);
+}