X-Git-Url: https://git.notmuchmail.org/git?p=notmuch;a=blobdiff_plain;f=lib%2Fdatabase.cc;h=5d300732d4a635a369b569ddbe452f39bff73f59;hp=726c5a98a02b1e153358c0daff820dac8cac3cc3;hb=d7e5f5827e21be7dd8993e5a877bdb73cdb64325;hpb=2ce25b93a72b4a8d6daa5321f9ef7df0772a789f diff --git a/lib/database.cc b/lib/database.cc index 726c5a98..5d300732 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 * ------------- @@ -63,6 +63,11 @@ typedef struct { * * tag: Any tags associated with this message by the user. * + * direntry: A colon-separated pair of values (INTEGER:STRING), + * where INTEGER is the document ID of a directory + * document, and STRING is the name of a file within + * that directory for this mail message. + * * A mail document also has two values: * * TIMESTAMP: The time_t value corresponding to the message's @@ -75,21 +80,30 @@ typedef struct { * user in searching. But the database doesn't really care itself * about any of these. * - * Timestamp document + * The data portion of a mail document is empty. + * + * 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 directory document contains the following terms: * - * The timestamp document is indexed with a single prefixed term: + * directory: The directory path (relative to the database path) + * Or the SHA1 sum of the directory path (if the + * path itself is too long to fit in a Xapian + * term). * - * timestamp: The user's key value (likely a directory name) + * 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) + * + * The data portion of a directory document contains the path of the + * directory (relative to the datbase path). */ /* With these prefix values we follow the conventions published here: @@ -111,7 +125,9 @@ prefix_t BOOLEAN_PREFIX_INTERNAL[] = { { "type", "T" }, { "reference", "XREFERENCE" }, { "replyto", "XREPLYTO" }, - { "timestamp", "XTIMESTAMP" }, + { "directory", "XDIRECTORY" }, + { "direntry", "XDIRENTRY" }, + { "parent", "XPARENT" }, }; prefix_t BOOLEAN_PREFIX_EXTERNAL[] = { @@ -147,17 +163,20 @@ _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); @@ -172,6 +191,8 @@ notmuch_status_to_string (notmuch_status_t status) return "No error occurred"; case NOTMUCH_STATUS_OUT_OF_MEMORY: return "Out of memory"; + case NOTMUCH_STATUS_READONLY_DATABASE: + return "The database is read-only"; case NOTMUCH_STATUS_XAPIAN_EXCEPTION: return "A Xapian exception occurred"; case NOTMUCH_STATUS_FILE_ERROR: @@ -225,10 +246,19 @@ find_unique_doc_id (notmuch_database_t *notmuch, if (i == end) { *doc_id = 0; return NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND; - } else { - *doc_id = *i; - return NOTMUCH_PRIVATE_STATUS_SUCCESS; } + + *doc_id = *i; + +#if DEBUG_DATABASE_SANITY + i++; + + if (i != end) + INTERNAL_ERROR ("Term %s:%s is not unique as expected.\n", + prefix_name, value); +#endif + + return NOTMUCH_PRIVATE_STATUS_SUCCESS; } static Xapian::Document @@ -293,13 +323,14 @@ 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++; } } @@ -323,7 +354,7 @@ _parse_message_id (void *ctx, const char *message_id, const char **next) const char *s, *end; char *result; - if (message_id == NULL) + if (message_id == NULL || *message_id == '\0') return NULL; s = message_id; @@ -391,7 +422,7 @@ parse_references (void *ctx, { char *ref; - if (refs == NULL) + if (refs == NULL || *refs == '\0') return; while (*refs) { @@ -438,7 +469,8 @@ notmuch_database_create (const char *path) goto DONE; } - notmuch = notmuch_database_open (path); + notmuch = notmuch_database_open (path, + NOTMUCH_DATABASE_MODE_READ_WRITE); DONE: if (notmuch_path) @@ -448,7 +480,8 @@ notmuch_database_create (const char *path) } notmuch_database_t * -notmuch_database_open (const char *path) +notmuch_database_open (const char *path, + notmuch_database_mode_t mode) { notmuch_database_t *notmuch = NULL; char *notmuch_path = NULL, *xapian_path = NULL; @@ -476,22 +509,30 @@ notmuch_database_open (const char *path) } notmuch = talloc (NULL, notmuch_database_t); + notmuch->exception_reported = FALSE; notmuch->path = talloc_strdup (notmuch, path); if (notmuch->path[strlen (notmuch->path) - 1] == '/') notmuch->path[strlen (notmuch->path) - 1] = '\0'; + notmuch->mode = mode; try { - notmuch->xapian_db = new Xapian::WritableDatabase (xapian_path, - Xapian::DB_CREATE_OR_OPEN); + if (mode == NOTMUCH_DATABASE_MODE_READ_WRITE) { + notmuch->xapian_db = new Xapian::WritableDatabase (xapian_path, + Xapian::DB_CREATE_OR_OPEN); + } else { + notmuch->xapian_db = new Xapian::Database (xapian_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]; @@ -504,11 +545,11 @@ notmuch_database_open (const char *path) notmuch->query_parser->add_prefix (prefix->name, prefix->prefix); } } catch (const Xapian::Error &error) { - fprintf (stderr, "A Xapian exception occurred: %s\n", + fprintf (stderr, "A Xapian exception occurred opening database: %s\n", error.get_msg().c_str()); notmuch = NULL; } - + DONE: if (notmuch_path) free (notmuch_path); @@ -521,11 +562,20 @@ notmuch_database_open (const char *path) void notmuch_database_close (notmuch_database_t *notmuch) { - notmuch->xapian_db->flush (); + try { + if (notmuch->mode == NOTMUCH_DATABASE_MODE_READ_WRITE) + (static_cast (notmuch->xapian_db))->flush (); + } catch (const Xapian::Error &error) { + if (! notmuch->exception_reported) { + fprintf (stderr, "Error: A Xapian exception occurred flushing database: %s\n", + error.get_msg().c_str()); + } + } delete notmuch->term_gen; delete notmuch->query_parser; delete notmuch->xapian_db; + delete notmuch->value_range_processor; talloc_free (notmuch); } @@ -536,98 +586,322 @@ 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, split it into two parts: the directory part is all + * components except for the last, and the basename is that last + * component. Getting the return-value for either part is optional + * (the caller can pass NULL). + * + * The original 'path' can represent either a regular file or a + * directory---the splitting will be carried out in the same way in + * either case. Trailing slashes on 'path' will be ignored, and any + * cases of multiple '/' characters appearing in series will be + * treated as a single '/'. + * + * Allocation (if any) will have 'ctx' as the talloc owner. But + * pointers will be returned within the original path string whenever + * possible. + * + * Note: If 'path' is non-empty and contains no non-trailing slash, + * (that is, consists of a filename with no parent directory), then + * the directory returned will be an empty string. However, if 'path' + * is an empty string, then both directory and basename will be + * returned as NULL. + */ +notmuch_status_t +_notmuch_database_split_path (void *ctx, + const char *path, + const char **directory, + const char **basename) +{ + const char *slash; + + if (path == NULL || *path == '\0') { + if (directory) + *directory = NULL; + if (basename) + *basename = NULL; + 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; + + if (basename) + *basename = slash; + + --slash; + } + + /* Finally, skip multiple slashes. */ + while (slash != path) { + if (*slash != '/') + break; + + --slash; + } + + if (slash == path) { + if (directory) + *directory = talloc_strdup (ctx, ""); + if (basename) + *basename = path; + } else { + if (directory) + *directory = talloc_strndup (ctx, path, slash - path + 1); + } + + return NOTMUCH_STATUS_SUCCESS; +} + +notmuch_status_t +_notmuch_database_find_directory_id (notmuch_database_t *notmuch, + const char *path, + unsigned int *directory_id) +{ + notmuch_private_status_t private_status; + notmuch_status_t status = NOTMUCH_STATUS_SUCCESS; + const char *db_path; + + if (path == NULL) { + *directory_id = 0; + return NOTMUCH_STATUS_SUCCESS; + } + + db_path = directory_db_path (path); + + private_status = find_unique_doc_id (notmuch, "directory", + db_path, directory_id); + if (private_status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND) { + status = notmuch_database_set_directory_mtime (notmuch, + path, 0); + if (status) + goto DONE; + + private_status = find_unique_doc_id (notmuch, "directory", + db_path, directory_id); + status = COERCE_STATUS (private_status, "_find_directory_id"); + } + + DONE: + if (db_path != path) + free ((char *) db_path); + + if (status) + *directory_id = -1; + + return status; } +const char * +_notmuch_database_get_directory_path (void *ctx, + notmuch_database_t *notmuch, + unsigned int doc_id) +{ + Xapian::Document document; + + document = find_document_for_doc_id (notmuch, doc_id); + + return talloc_strdup (ctx, document.get_data ().c_str ()); +} + +/* Given a legal 'filename' for the database, (either relative to + * database path or absolute with initial components identical to + * database path), return a new string (with 'ctx' as the talloc + * owner) suitable for use as a direntry term value. + */ notmuch_status_t -notmuch_database_set_timestamp (notmuch_database_t *notmuch, - const char *key, time_t timestamp) +_notmuch_database_filename_to_direntry (void *ctx, + notmuch_database_t *notmuch, + const char *filename, + char **direntry) +{ + const char *relative, *directory, *basename; + Xapian::docid directory_id; + notmuch_status_t status; + + relative = _notmuch_database_relative_path (notmuch, filename); + + status = _notmuch_database_split_path (ctx, relative, + &directory, &basename); + if (status) + return status; + + status = _notmuch_database_find_directory_id (notmuch, directory, + &directory_id); + if (status) + return status; + + *direntry = talloc_asprintf (ctx, "%u:%s", directory_id, basename); + + return NOTMUCH_STATUS_SUCCESS; +} + +/* 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_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 *parent, *db_path = NULL; + unsigned int parent_id; + void *local = talloc_new (notmuch); + + 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_key = timestamp_db_key (key); + db = static_cast (notmuch->xapian_db); + 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); + char *term = talloc_asprintf (local, "%s%s", + _find_prefix ("directory"), db_path); doc.add_term (term); - talloc_free (term); - notmuch->xapian_db->add_document (doc); + doc.set_data (path); + + _notmuch_database_split_path (local, path, &parent, NULL); + + _notmuch_database_find_directory_id (notmuch, parent, &parent_id); + + term = talloc_asprintf (local, "%s%u", + _find_prefix ("parent"), + parent_id); + doc.add_term (term); + + db->add_document (doc); } else { - notmuch->xapian_db->replace_document (doc_id, doc); + db->replace_document (doc_id, doc); } - } catch (Xapian::Error &error) { - fprintf (stderr, "A Xapian exception occurred: %s.\n", + } catch (const Xapian::Error &error) { + 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); + + talloc_free (local); 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; } @@ -821,12 +1095,11 @@ _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, @@ -865,7 +1138,7 @@ 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; @@ -896,9 +1169,9 @@ notmuch_database_add_message (notmuch_database_t *notmuch, subject = notmuch_message_file_get_header (message_file, "subject"); to = notmuch_message_file_get_header (message_file, "to"); - if (from == NULL && - subject == NULL && - to == NULL) + if ((from == NULL || *from == '\0') && + (subject == NULL || *subject == '\0') && + (to == NULL || *to == '\0')) { ret = NOTMUCH_STATUS_FILE_NOT_EMAIL; goto DONE; @@ -908,13 +1181,22 @@ notmuch_database_add_message (notmuch_database_t *notmuch, * is to find a message ID (or else create one ourselves). */ header = notmuch_message_file_get_header (message_file, "message-id"); - if (header) { + 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); @@ -940,31 +1222,36 @@ notmuch_database_add_message (notmuch_database_t *notmuch, talloc_free (message_id); - if (message == NULL) + if (message == NULL) { + ret = COERCE_STATUS (private_status, + "Unexpected status value from _notmuch_message_create_for_message_id"); goto DONE; + } + + _notmuch_message_add_filename (message, filename); /* Is this a newly created message object? */ if (private_status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND) { - _notmuch_message_set_filename (message, filename); _notmuch_message_add_term (message, "type", "mail"); - } else { - ret = NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID; - goto DONE; - } - ret = _notmuch_database_link_message (notmuch, message, message_file); - if (ret) - goto DONE; + ret = _notmuch_database_link_message (notmuch, message, + message_file); + if (ret) + goto DONE; - date = notmuch_message_file_get_header (message_file, "date"); - _notmuch_message_set_date (message, date); + date = notmuch_message_file_get_header (message_file, "date"); + _notmuch_message_set_date (message, date); - _notmuch_message_index_file (message, filename); + _notmuch_message_index_file (message, filename); + } else { + ret = NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID; + } _notmuch_message_sync (message); } catch (const Xapian::Error &error) { - fprintf (stderr, "A Xapian exception occurred: %s.\n", - error.get_msg().c_str()); + fprintf (stderr, "A Xapian exception occurred adding message: %s.\n", + error.get_description().c_str()); + notmuch->exception_reported = TRUE; ret = NOTMUCH_STATUS_XAPIAN_EXCEPTION; goto DONE; } @@ -982,3 +1269,46 @@ 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); +}