X-Git-Url: https://git.notmuchmail.org/git?p=notmuch;a=blobdiff_plain;f=notmuch-new.c;h=fb59d90c15f4646a9046cceab0bcecc661608ab3;hp=ed3f944f0c9468cde72d2708bf704f7c1a1e057c;hb=fcd433709eff6b7f0fbbd1e5018ca0e37315ce26;hpb=088801a14ab1ccf4c65e5bf7b50b3a1172c477d7 diff --git a/notmuch-new.c b/notmuch-new.c index ed3f944f..fb59d90c 100644 --- a/notmuch-new.c +++ b/notmuch-new.c @@ -24,10 +24,12 @@ typedef struct _filename_node { char *filename; + time_t mtime; struct _filename_node *next; } _filename_node_t; typedef struct _filename_list { + unsigned count; _filename_node_t *head; _filename_node_t **tail; } _filename_list_t; @@ -45,14 +47,18 @@ typedef struct { _filename_list_t *removed_files; _filename_list_t *removed_directories; + _filename_list_t *directory_mtimes; + + notmuch_bool_t synchronize_flags; + _filename_list_t *message_ids_to_sync; } add_files_state_t; -static volatile sig_atomic_t do_add_files_print_progress = 0; +static volatile sig_atomic_t do_print_progress = 0; static void handle_sigalrm (unused (int signal)) { - do_add_files_print_progress = 1; + do_print_progress = 1; } static volatile sig_atomic_t interrupted; @@ -60,10 +66,9 @@ static volatile sig_atomic_t interrupted; static void handle_sigint (unused (int sig)) { - ssize_t ignored; static char msg[] = "Stopping... \n"; - ignored = write(2, msg, sizeof(msg)-1); + write(2, msg, sizeof(msg)-1); interrupted = 1; } @@ -78,47 +83,56 @@ _filename_list_create (const void *ctx) list->head = NULL; list->tail = &list->head; + list->count = 0; return list; } -static void +static _filename_node_t * _filename_list_add (_filename_list_t *list, const char *filename) { _filename_node_t *node = talloc (list, _filename_node_t); + list->count++; + node->filename = talloc_strdup (list, filename); node->next = NULL; *(list->tail) = node; list->tail = &node->next; + + return node; } static void -add_files_print_progress (add_files_state_t *state) +generic_print_progress (const char *action, const char *object, + struct timeval tv_start, unsigned processed, unsigned total) { struct timeval tv_now; double elapsed_overall, rate_overall; gettimeofday (&tv_now, NULL); - elapsed_overall = notmuch_time_elapsed (state->tv_start, tv_now); - rate_overall = (state->processed_files) / elapsed_overall; - - printf ("Processed %d", state->processed_files); + elapsed_overall = notmuch_time_elapsed (tv_start, tv_now); + rate_overall = processed / elapsed_overall; - if (state->total_files) { - double time_remaining; + printf ("%s %d ", action, processed); - time_remaining = ((state->total_files - state->processed_files) / - rate_overall); - printf (" of %d files (", state->total_files); - notmuch_time_print_formatted_seconds (time_remaining); - printf (" remaining). \r"); + if (total) { + printf ("of %d %s", total, object); + if (processed > 0 && elapsed_overall > 0.5) { + double time_remaining = ((total - processed) / rate_overall); + printf (" ("); + notmuch_time_print_formatted_seconds (time_remaining); + printf (" remaining)"); + } } else { - printf (" files (%d files/sec.) \r", (int) rate_overall); + printf ("%s", object); + if (elapsed_overall > 0.5) + printf (" (%d %s/sec.)", (int) rate_overall, object); } + printf (".\033[K\r"); fflush (stdout); } @@ -171,15 +185,20 @@ _entries_resemble_maildir (struct dirent **entries, int count) * * o Ask the filesystem for files and directories within 'path' * (via scandir and stored in fs_entries) - * o Ask the database for files and directories within 'path' - * (db_files and db_subdirs) * * o Pass 1: For each directory in fs_entries, recursively call into * this same function. * - * o Pass 2: If 'fs_mtime' > 'db_mtime', then walk fs_entries - * simultaneously with db_files and db_subdirs. Look for one of - * three interesting cases: + * o Compare fs_mtime to db_mtime. If they are equivalent, terminate + * the algorithm at this point, (this directory has not been + * updated in the filesystem since the last database scan of PASS + * 2). + * + * o Ask the database for files and directories within 'path' + * (db_files and db_subdirs) + * + * o Pass 2: Walk fs_entries simultaneously with db_files and + * db_subdirs. Look for one of three interesting cases: * * 1. Regular file in fs_entries and not in db_files * This is a new file to add_message into the database. @@ -198,6 +217,7 @@ _entries_resemble_maildir (struct dirent **entries, int count) * information is lost from the database). * * o Tell the database to update its time of 'path' to 'fs_mtime' + * if fs_mtime isn't the current wall-clock time. */ static notmuch_status_t add_files_recursive (notmuch_database_t *notmuch, @@ -215,6 +235,7 @@ add_files_recursive (notmuch_database_t *notmuch, notmuch_directory_t *directory; notmuch_filenames_t *db_files = NULL; notmuch_filenames_t *db_subdirs = NULL; + time_t stat_time; struct stat st; notmuch_bool_t is_maildir, new_directory; const char **tag; @@ -224,6 +245,7 @@ add_files_recursive (notmuch_database_t *notmuch, path, strerror (errno)); return NOTMUCH_STATUS_FILE_ERROR; } + stat_time = time (NULL); /* This is not an error since we may have recursed based on a * symlink to a regular file, not a directory, and we don't know @@ -236,15 +258,26 @@ add_files_recursive (notmuch_database_t *notmuch, directory = notmuch_database_get_directory (notmuch, path); db_mtime = notmuch_directory_get_mtime (directory); - if (db_mtime == 0) { - new_directory = TRUE; - db_files = NULL; - db_subdirs = NULL; - } else { - new_directory = FALSE; - db_files = notmuch_directory_get_child_files (directory); - db_subdirs = notmuch_directory_get_child_directories (directory); - } + new_directory = db_mtime ? FALSE : TRUE; + + /* XXX This is a temporary workaround. If we don't update the + * database mtime until after processing messages in this + * directory, then a 0 mtime is *not* sufficient to indicate that + * this directory has no messages or subdirs in the database (for + * example, if an earlier run skipped the mtime update because + * fs_mtime == stat_time, or was interrupted before updating the + * mtime at the end). To address this, we record a (bogus) + * non-zero value before processing any child messages so that a + * later run won't mistake this for a new directory (and, for + * example, fail to detect removed files and subdirs). + * + * A better solution would be for notmuch_database_get_directory + * to indicate if it really created a new directory or not, either + * by a new out-argument, or by recording this information and + * providing an accessor. + */ + if (new_directory) + notmuch_directory_set_mtime (directory, -1); /* If the database knows about this directory, then we sort based * on strcmp to match the database sorting. Otherwise, we can do @@ -305,11 +338,26 @@ add_files_recursive (notmuch_database_t *notmuch, next = NULL; } - /* If this directory hasn't been modified since the last - * "notmuch new", then we can skip the second pass entirely. */ - if (fs_mtime <= db_mtime) + /* If the directory's modification time in the filesystem is the + * same as what we recorded in the database the last time we + * scanned it, then we can skip the second pass entirely. + * + * We test for strict equality here to avoid a bug that can happen + * if the system clock jumps backward, (preventing new mail from + * being discovered until the clock catches up and the directory + * is modified again). + */ + if (fs_mtime == db_mtime) goto DONE; + /* new_directory means a directory that the database has never + * seen before. In that case, we can simply leave db_files and + * db_subdirs NULL. */ + if (!new_directory) { + db_files = notmuch_directory_get_child_files (directory); + db_subdirs = notmuch_directory_get_child_directories (directory); + } + /* Pass 2: Scan for new files, removed files, and removed directories. */ for (i = 0; i < num_fs_entries; i++) { @@ -351,7 +399,7 @@ add_files_recursive (notmuch_database_t *notmuch, /* If we're looking at a symlink, we only want to add it if it * links to a regular file, (and not to a directory, say). * - * Similarly, if the file is of unknown type (due to filesytem + * Similarly, if the file is of unknown type (due to filesystem * limitations), then we also need to look closer. * * In either case, a stat does the trick. @@ -408,13 +456,20 @@ add_files_recursive (notmuch_database_t *notmuch, /* success */ case NOTMUCH_STATUS_SUCCESS: state->added_messages++; + notmuch_message_freeze (message); for (tag=state->new_tags; *tag != NULL; tag++) notmuch_message_add_tag (message, *tag); - notmuch_message_maildir_to_tags (message, next); + if (state->synchronize_flags == TRUE) + notmuch_message_maildir_flags_to_tags (message); + notmuch_message_thaw (message); break; /* Non-fatal issues (go on to next file) */ case NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID: - notmuch_message_maildir_to_tags (message, next); + /* Defer sync of maildir flags until after old filenames + * are removed in the case of a rename. */ + if (state->synchronize_flags == TRUE) + _filename_list_add (state->message_ids_to_sync, + notmuch_message_get_message_id (message)); break; case NOTMUCH_STATUS_FILE_NOT_EMAIL: fprintf (stderr, "Note: Ignoring non-mail file: %s\n", @@ -443,9 +498,10 @@ add_files_recursive (notmuch_database_t *notmuch, message = NULL; } - if (do_add_files_print_progress) { - do_add_files_print_progress = 0; - add_files_print_progress (state); + if (do_print_progress) { + do_print_progress = 0; + generic_print_progress ("Processed", "files", state->tv_start, + state->processed_files, state->total_files); } talloc_free (next); @@ -479,11 +535,13 @@ add_files_recursive (notmuch_database_t *notmuch, notmuch_filenames_move_to_next (db_subdirs); } - if (! interrupted) { - status = notmuch_directory_set_mtime (directory, fs_mtime); - if (status && ret == NOTMUCH_STATUS_SUCCESS) - ret = status; - } + /* If the directory's mtime is the same as the wall-clock time + * when we stat'ed the directory, we skip updating the mtime in + * the database because a message could be delivered later in this + * same second. This may lead to unnecessary re-scans, but it + * avoids overlooking messages. */ + if (fs_mtime != stat_time) + _filename_list_add (state->directory_mtimes, path)->mtime = fs_mtime; DONE: if (next) @@ -504,38 +562,56 @@ add_files_recursive (notmuch_database_t *notmuch, return ret; } +static void +setup_progress_printing_timer (void) +{ + struct sigaction action; + struct itimerval timerval; + + /* Setup our handler for SIGALRM */ + memset (&action, 0, sizeof (struct sigaction)); + action.sa_handler = handle_sigalrm; + sigemptyset (&action.sa_mask); + action.sa_flags = SA_RESTART; + sigaction (SIGALRM, &action, NULL); + + /* Then start a timer to send SIGALRM once per second. */ + timerval.it_interval.tv_sec = 1; + timerval.it_interval.tv_usec = 0; + timerval.it_value.tv_sec = 1; + timerval.it_value.tv_usec = 0; + setitimer (ITIMER_REAL, &timerval, NULL); +} + +static void +stop_progress_printing_timer (void) +{ + struct sigaction action; + struct itimerval timerval; + + /* Now stop the timer. */ + timerval.it_interval.tv_sec = 0; + timerval.it_interval.tv_usec = 0; + timerval.it_value.tv_sec = 0; + timerval.it_value.tv_usec = 0; + setitimer (ITIMER_REAL, &timerval, NULL); + + /* And disable the signal handler. */ + action.sa_handler = SIG_IGN; + sigaction (SIGALRM, &action, NULL); +} + + /* This is the top-level entry point for add_files. It does a couple - * of error checks, sets up the progress-printing timer and then calls - * into the recursive function. */ + * of error checks and then calls into the recursive function. */ static notmuch_status_t add_files (notmuch_database_t *notmuch, const char *path, add_files_state_t *state) { notmuch_status_t status; - struct sigaction action; - struct itimerval timerval; - notmuch_bool_t timer_is_active = FALSE; struct stat st; - if (state->output_is_a_tty && ! debugger_is_active () && ! state->verbose) { - /* Setup our handler for SIGALRM */ - memset (&action, 0, sizeof (struct sigaction)); - action.sa_handler = handle_sigalrm; - sigemptyset (&action.sa_mask); - action.sa_flags = SA_RESTART; - sigaction (SIGALRM, &action, NULL); - - /* Then start a timer to send SIGALRM once per second. */ - timerval.it_interval.tv_sec = 1; - timerval.it_interval.tv_usec = 0; - timerval.it_value.tv_sec = 1; - timerval.it_value.tv_usec = 0; - setitimer (ITIMER_REAL, &timerval, NULL); - - timer_is_active = TRUE; - } - if (stat (path, &st)) { fprintf (stderr, "Error reading directory %s: %s\n", path, strerror (errno)); @@ -549,19 +625,6 @@ add_files (notmuch_database_t *notmuch, status = add_files_recursive (notmuch, path, state); - if (timer_is_active) { - /* Now stop the timer. */ - timerval.it_interval.tv_sec = 0; - timerval.it_interval.tv_usec = 0; - timerval.it_value.tv_sec = 0; - timerval.it_value.tv_usec = 0; - setitimer (ITIMER_REAL, &timerval, NULL); - - /* And disable the signal handler. */ - action.sa_handler = SIG_IGN; - sigaction (SIGALRM, &action, NULL); - } - return status; } @@ -710,7 +773,7 @@ notmuch_new_command (void *ctx, int argc, char *argv[]) notmuch_database_t *notmuch; add_files_state_t add_files_state; double elapsed; - struct timeval tv_now; + struct timeval tv_now, tv_start; int ret = 0; struct stat st; const char *db_path; @@ -720,6 +783,7 @@ notmuch_new_command (void *ctx, int argc, char *argv[]) int renamed_files, removed_files; notmuch_status_t status; int i; + notmuch_bool_t timer_is_active = FALSE; add_files_state.verbose = 0; add_files_state.output_is_a_tty = isatty (fileno (stdout)); @@ -732,12 +796,13 @@ notmuch_new_command (void *ctx, int argc, char *argv[]) return 1; } } - config = notmuch_config_open (ctx, NULL, NULL); if (config == NULL) return 1; add_files_state.new_tags = notmuch_config_get_new_tags (config, &add_files_state.new_tags_length); + add_files_state.synchronize_flags = notmuch_config_get_maildir_synchronize_flags (config); + add_files_state.message_ids_to_sync = _filename_list_create (ctx); db_path = notmuch_config_get_database_path (config); dot_notmuch_path = talloc_asprintf (ctx, "%s/%s", db_path, ".notmuch"); @@ -792,26 +857,86 @@ notmuch_new_command (void *ctx, int argc, char *argv[]) add_files_state.removed_files = _filename_list_create (ctx); add_files_state.removed_directories = _filename_list_create (ctx); + add_files_state.directory_mtimes = _filename_list_create (ctx); + + if (! debugger_is_active () && add_files_state.output_is_a_tty + && ! add_files_state.verbose) { + setup_progress_printing_timer (); + timer_is_active = TRUE; + } ret = add_files (notmuch, db_path, &add_files_state); removed_files = 0; renamed_files = 0; - for (f = add_files_state.removed_files->head; f; f = f->next) { + gettimeofday (&tv_start, NULL); + for (f = add_files_state.removed_files->head; f && !interrupted; f = f->next) { status = notmuch_database_remove_message (notmuch, f->filename); if (status == NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID) renamed_files++; else removed_files++; + if (do_print_progress) { + do_print_progress = 0; + generic_print_progress ("Cleaned up", "messages", + tv_start, removed_files + renamed_files, + add_files_state.removed_files->count); + } } - for (f = add_files_state.removed_directories->head; f; f = f->next) { + gettimeofday (&tv_start, NULL); + for (f = add_files_state.removed_directories->head, i = 0; f && !interrupted; f = f->next, i++) { _remove_directory (ctx, notmuch, f->filename, &renamed_files, &removed_files); + if (do_print_progress) { + do_print_progress = 0; + generic_print_progress ("Cleaned up", "directories", + tv_start, i, + add_files_state.removed_directories->count); + } + } + + for (f = add_files_state.directory_mtimes->head; f && !interrupted; f = f->next) { + notmuch_directory_t *directory; + directory = notmuch_database_get_directory (notmuch, f->filename); + if (directory) { + notmuch_directory_set_mtime (directory, f->mtime); + notmuch_directory_destroy (directory); + } } talloc_free (add_files_state.removed_files); talloc_free (add_files_state.removed_directories); + talloc_free (add_files_state.directory_mtimes); + + /* Now that removals are done (hence the database is aware of all + * renames), we can synchronize maildir_flags to tags for all + * messages that had new filenames appear on this run. */ + gettimeofday (&tv_start, NULL); + if (add_files_state.synchronize_flags) { + _filename_node_t *node; + notmuch_message_t *message; + for (node = add_files_state.message_ids_to_sync->head, i = 0; + node; + node = node->next, i++) + { + message = notmuch_database_find_message (notmuch, node->filename); + notmuch_message_maildir_flags_to_tags (message); + notmuch_message_destroy (message); + if (do_print_progress) { + do_print_progress = 0; + generic_print_progress ( + "Synchronized tags for", "messages", + tv_start, i, add_files_state.message_ids_to_sync->count); + } + } + } + + talloc_free (add_files_state.message_ids_to_sync); + add_files_state.message_ids_to_sync = NULL; + + if (timer_is_active) + stop_progress_printing_timer (); gettimeofday (&tv_now, NULL); elapsed = notmuch_time_elapsed (add_files_state.tv_start, @@ -823,10 +948,10 @@ notmuch_new_command (void *ctx, int argc, char *argv[]) "file" : "total files"); notmuch_time_print_formatted_seconds (elapsed); if (elapsed > 1) { - printf (" (%d files/sec.). \n", + printf (" (%d files/sec.).\033[K\n", (int) (add_files_state.processed_files / elapsed)); } else { - printf (". \n"); + printf (".\033[K\n"); } }