Re: [Patch v2 13/17] notmuch-restore: add support for input format 'batch-tag'

diff --git a/a3/9602fe804fac1b1e019ee6b7218f5f2116ec99 b/a3/9602fe804fac1b1e019ee6b7218f5f2116ec99
new file mode 100644 (file)
index 0000000..4899f09
--- /dev/null
+++ b/a3/9602fe804fac1b1e019ee6b7218f5f2116ec99
@@ -0,0 +1,405 @@
+Return-Path: <jani@nikula.org>\r
+X-Original-To: notmuch@notmuchmail.org\r
+Delivered-To: notmuch@notmuchmail.org\r
+Received: from localhost (localhost [127.0.0.1])\r
+       by olra.theworths.org (Postfix) with ESMTP id 99958431FAF\r
+       for <notmuch@notmuchmail.org>; Sun,  2 Dec 2012 05:29:53 -0800 (PST)\r
+X-Virus-Scanned: Debian amavisd-new at olra.theworths.org\r
+X-Spam-Flag: NO\r
+X-Spam-Score: -0.7\r
+X-Spam-Level: \r
+X-Spam-Status: No, score=-0.7 tagged_above=-999 required=5\r
+       tests=[RCVD_IN_DNSWL_LOW=-0.7] autolearn=disabled\r
+Received: from olra.theworths.org ([127.0.0.1])\r
+       by localhost (olra.theworths.org [127.0.0.1]) (amavisd-new, port 10024)\r
+       with ESMTP id 1GhUazsQcveL for <notmuch@notmuchmail.org>;\r
+       Sun,  2 Dec 2012 05:29:49 -0800 (PST)\r
+Received: from mail-lb0-f181.google.com (mail-lb0-f181.google.com\r
+       [209.85.217.181]) (using TLSv1 with cipher RC4-SHA (128/128 bits))\r
+       (No client certificate requested)\r
+       by olra.theworths.org (Postfix) with ESMTPS id AF574431FAE\r
+       for <notmuch@notmuchmail.org>; Sun,  2 Dec 2012 05:29:48 -0800 (PST)\r
+Received: by mail-lb0-f181.google.com with SMTP id ge1so1763467lbb.26\r
+       for <notmuch@notmuchmail.org>; Sun, 02 Dec 2012 05:29:47 -0800 (PST)\r
+X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;\r
+       d=google.com; s=20120113;\r
+       h=from:to:cc:subject:in-reply-to:references:user-agent:date\r
+       :message-id:mime-version:content-type:x-gm-message-state;\r
+       bh=4pG3dsHLro2a6t1zGdq7KCFVB5k2FuL7Aixh8+mW97k=;\r
+       b=iOG8OULvDABOB7FJf7Wzj4oBmL8rG323NkNMRj+etCZsvLo9H+4gjNIptLqD25o34k\r
+       nBmw7naYTezYcET5v5HTojICOYoTbE01b7mHW3c2rN/OzcU1DpdwMMQSEdOZD5Gn/VJq\r
+       +16FGNtf3QBlUE44eAsx2m0jTWJRL66vXr9Olk+pn+m95Dug3OnMYScjSsbvJxNimj9x\r
+       G//hiLAcxNZxUNSkg5plYVT7RiamSkJpGSpJWseQfgqGMoOkWU/bAv3dXtMm2914uYuf\r
+       bzlgoyR4xVWoX+GVUKRykl6rYYhUb1HdQQCmAYZhlHLXjHLygYFnUrDm2vNQg+0ACfp5\r
+       WYNg==\r
+Received: by 10.152.114.65 with SMTP id je1mr6695753lab.33.1354454987058;\r
+       Sun, 02 Dec 2012 05:29:47 -0800 (PST)\r
+Received: from localhost (dsl-hkibrasgw4-fe51df00-27.dhcp.inet.fi.\r
+       [80.223.81.27])\r
+       by mx.google.com with ESMTPS id b4sm4150390lbi.0.2012.12.02.05.29.45\r
+       (version=SSLv3 cipher=OTHER); Sun, 02 Dec 2012 05:29:46 -0800 (PST)\r
+From: Jani Nikula <jani@nikula.org>\r
+To: david@tethera.net, notmuch@notmuchmail.org\r
+Subject: Re: [Patch v2 13/17] notmuch-restore: add support for input format\r
+       'batch-tag'\r
+In-Reply-To: <1353792017-31459-14-git-send-email-david@tethera.net>\r
+References: <1353792017-31459-1-git-send-email-david@tethera.net>\r
+       <1353792017-31459-14-git-send-email-david@tethera.net>\r
+User-Agent: Notmuch/0.14+124~g3b17402 (http://notmuchmail.org) Emacs/23.4.1\r
+       (i686-pc-linux-gnu)\r
+Date: Sun, 02 Dec 2012 15:29:43 +0200\r
+Message-ID: <87zk1wd1ko.fsf@nikula.org>\r
+MIME-Version: 1.0\r
+Content-Type: text/plain; charset=us-ascii\r
+X-Gm-Message-State:\r
+ ALoCoQnkuhDlQ10YCjDeEjMG8q7DRMmvK6smTHhMB09E9uuzdYSSICajMJXHt3UXjYtQLy0EcQ+N\r
+Cc: David Bremner <bremner@debian.org>\r
+X-BeenThere: notmuch@notmuchmail.org\r
+X-Mailman-Version: 2.1.13\r
+Precedence: list\r
+List-Id: "Use and development of the notmuch mail system."\r
+       <notmuch.notmuchmail.org>\r
+List-Unsubscribe: <http://notmuchmail.org/mailman/options/notmuch>,\r
+       <mailto:notmuch-request@notmuchmail.org?subject=unsubscribe>\r
+List-Archive: <http://notmuchmail.org/pipermail/notmuch>\r
+List-Post: <mailto:notmuch@notmuchmail.org>\r
+List-Help: <mailto:notmuch-request@notmuchmail.org?subject=help>\r
+List-Subscribe: <http://notmuchmail.org/mailman/listinfo/notmuch>,\r
+       <mailto:notmuch-request@notmuchmail.org?subject=subscribe>\r
+X-List-Received-Date: Sun, 02 Dec 2012 13:29:53 -0000\r
+\r
+On Sat, 24 Nov 2012, david@tethera.net wrote:\r
+> From: David Bremner <bremner@debian.org>\r
+>\r
+> This is the same as the batch input for notmuch tag, except by default\r
+> it removes all tags before modifying a given message id and only "id:"\r
+> is supported.\r
+> ---\r
+>  notmuch-restore.c |  199 +++++++++++++++++++++++++++++++++--------------------\r
+>  1 file changed, 125 insertions(+), 74 deletions(-)\r
+>\r
+> diff --git a/notmuch-restore.c b/notmuch-restore.c\r
+> index f03dcac..22fcd2d 100644\r
+> --- a/notmuch-restore.c\r
+> +++ b/notmuch-restore.c\r
+> @@ -19,18 +19,22 @@\r
+>   */\r
+>  \r
+>  #include "notmuch-client.h"\r
+> +#include "dump-restore-private.h"\r
+> +#include "tag-util.h"\r
+> +#include "string-util.h"\r
+> +\r
+> +static volatile sig_atomic_t interrupted;\r
+> +static regex_t regex;\r
+>  \r
+>  static int\r
+> -tag_message (notmuch_database_t *notmuch, const char *message_id,\r
+> -         char *file_tags, notmuch_bool_t remove_all,\r
+> -         notmuch_bool_t synchronize_flags)\r
+> +tag_message (unused (void *ctx),\r
+> +         notmuch_database_t *notmuch,\r
+> +         const char *message_id,\r
+> +         tag_op_list_t *tag_ops,\r
+> +         tag_op_flag_t flags)\r
+>  {\r
+>      notmuch_status_t status;\r
+> -    notmuch_tags_t *db_tags;\r
+> -    char *db_tags_str;\r
+>      notmuch_message_t *message = NULL;\r
+> -    const char *tag;\r
+> -    char *next;\r
+>      int ret = 0;\r
+>  \r
+>      status = notmuch_database_find_message (notmuch, message_id, &message);\r
+> @@ -44,55 +48,63 @@ tag_message (notmuch_database_t *notmuch, const char *message_id,\r
+>  \r
+>      /* In order to detect missing messages, this check/optimization is\r
+>       * intentionally done *after* first finding the message. */\r
+> -    if (! remove_all && (file_tags == NULL || *file_tags == '\0'))\r
+> -    goto DONE;\r
+> -\r
+> -    db_tags_str = NULL;\r
+> -    for (db_tags = notmuch_message_get_tags (message);\r
+> -     notmuch_tags_valid (db_tags);\r
+> -     notmuch_tags_move_to_next (db_tags)) {\r
+> -    tag = notmuch_tags_get (db_tags);\r
+> -\r
+> -    if (db_tags_str)\r
+> -        db_tags_str = talloc_asprintf_append (db_tags_str, " %s", tag);\r
+> -    else\r
+> -        db_tags_str = talloc_strdup (message, tag);\r
+> -    }\r
+> +    if ( (flags & TAG_FLAG_REMOVE_ALL) || (tag_op_list_size (tag_ops)))\r
+\r
+Extra space between ('s, and no need to wrap tag_op_list_size call in\r
+braces.\r
+\r
+> +    tag_op_list_apply (message, tag_ops, flags);\r
+>  \r
+> -    if (((file_tags == NULL || *file_tags == '\0') &&\r
+> -     (db_tags_str == NULL || *db_tags_str == '\0')) ||\r
+> -    (file_tags && db_tags_str && strcmp (file_tags, db_tags_str) == 0))\r
+\r
+This is a necessary optimization you're throwing away, but I'll get back\r
+to this after checking the last patch in the series.\r
+\r
+> -    goto DONE;\r
+> +    if (message)\r
+> +    notmuch_message_destroy (message);\r
+\r
+message != NULL always, you can remove the if.\r
+\r
+>  \r
+> -    notmuch_message_freeze (message);\r
+> +    return ret;\r
+> +}\r
+>  \r
+> -    if (remove_all)\r
+> -    notmuch_message_remove_all_tags (message);\r
+> +static int\r
+> +parse_sup_line (void *ctx, char *line,\r
+> +            char **query_str, tag_op_list_t *tag_ops)\r
+> +{\r
+>  \r
+> -    next = file_tags;\r
+> -    while (next) {\r
+> -    tag = strsep (&next, " ");\r
+> -    if (*tag == '\0')\r
+> -        continue;\r
+> -    status = notmuch_message_add_tag (message, tag);\r
+> -    if (status) {\r
+> -        fprintf (stderr, "Error applying tag %s to message %s:\n",\r
+> -                 tag, message_id);\r
+> -        fprintf (stderr, "%s\n", notmuch_status_to_string (status));\r
+> -        ret = 1;\r
+> -    }\r
+> +    regmatch_t match[3];\r
+> +    char *file_tags;\r
+> +    int rerr;\r
+> +\r
+> +    tag_op_list_reset (tag_ops);\r
+> +\r
+> +    chomp_newline (line);\r
+> +\r
+> +    /* Silently ignore blank lines */\r
+> +    if (line[0] == '\0') {\r
+> +    return 1;\r
+> +    }\r
+> +\r
+> +    rerr = xregexec (&regex, line, 3, match, 0);\r
+> +    if (rerr == REG_NOMATCH) {\r
+> +    fprintf (stderr, "Warning: Ignoring invalid input line: %s\n",\r
+> +             line);\r
+> +    return 1;\r
+>      }\r
+>  \r
+> -    notmuch_message_thaw (message);\r
+> +    *query_str = talloc_strndup (ctx, line + match[1].rm_so,\r
+> +                             match[1].rm_eo - match[1].rm_so);\r
+> +    file_tags = talloc_strndup (ctx, line + match[2].rm_so,\r
+> +                            match[2].rm_eo - match[2].rm_so);\r
+>  \r
+> -    if (synchronize_flags)\r
+> -    notmuch_message_tags_to_maildir_flags (message);\r
+> +    char *tok = file_tags;\r
+> +    size_t tok_len = 0;\r
+>  \r
+> -  DONE:\r
+> -    if (message)\r
+> -    notmuch_message_destroy (message);\r
+> +    tag_op_list_reset (tag_ops);\r
+> +\r
+> +    while ((tok = strtok_len (tok + tok_len, " ", &tok_len)) != NULL) {\r
+> +\r
+> +    if (*(tok + tok_len) != '\0') {\r
+> +        *(tok + tok_len) = '\0';\r
+> +        tok_len++;\r
+> +    }\r
+> +\r
+> +    if (tag_op_list_append (ctx, tag_ops, tok, FALSE))\r
+> +        return -1;\r
+> +    }\r
+> +\r
+> +    return 0;\r
+>  \r
+> -    return ret;\r
+>  }\r
+>  \r
+>  int\r
+> @@ -100,16 +112,19 @@ notmuch_restore_command (unused (void *ctx), int argc, char *argv[])\r
+>  {\r
+>      notmuch_config_t *config;\r
+>      notmuch_database_t *notmuch;\r
+> -    notmuch_bool_t synchronize_flags;\r
+>      notmuch_bool_t accumulate = FALSE;\r
+> +    tag_op_flag_t flags = 0;\r
+> +    tag_op_list_t *tag_ops;\r
+> +\r
+>      char *input_file_name = NULL;\r
+>      FILE *input = stdin;\r
+>      char *line = NULL;\r
+>      size_t line_size;\r
+>      ssize_t line_len;\r
+> -    regex_t regex;\r
+> -    int rerr;\r
+> +\r
+> +    int ret = 0;\r
+>      int opt_index;\r
+> +    int input_format = DUMP_FORMAT_AUTO;\r
+>  \r
+>      config = notmuch_config_open (ctx, NULL, NULL);\r
+>      if (config == NULL)\r
+> @@ -119,9 +134,15 @@ notmuch_restore_command (unused (void *ctx), int argc, char *argv[])\r
+>                             NOTMUCH_DATABASE_MODE_READ_WRITE, &notmuch))\r
+>      return 1;\r
+>  \r
+> -    synchronize_flags = notmuch_config_get_maildir_synchronize_flags (config);\r
+> +    if (notmuch_config_get_maildir_synchronize_flags (config))\r
+> +    flags |= TAG_FLAG_MAILDIR_SYNC;\r
+>  \r
+>      notmuch_opt_desc_t options[] = {\r
+> +    { NOTMUCH_OPT_KEYWORD, &input_format, "format", 'f',\r
+> +      (notmuch_keyword_t []){ { "auto", DUMP_FORMAT_AUTO },\r
+> +                              { "batch-tag", DUMP_FORMAT_BATCH_TAG },\r
+> +                              { "sup", DUMP_FORMAT_SUP },\r
+> +                              { 0, 0 } } },\r
+>      { NOTMUCH_OPT_STRING, &input_file_name, "input", 'i', 0 },\r
+>      { NOTMUCH_OPT_BOOLEAN,  &accumulate, "accumulate", 'a', 0 },\r
+>      { 0, 0, 0, 0, 0 }\r
+> @@ -134,6 +155,9 @@ notmuch_restore_command (unused (void *ctx), int argc, char *argv[])\r
+>      return 1;\r
+>      }\r
+>  \r
+> +    if (! accumulate)\r
+> +    flags |= TAG_FLAG_REMOVE_ALL;\r
+> +\r
+>      if (input_file_name) {\r
+>      input = fopen (input_file_name, "r");\r
+>      if (input == NULL) {\r
+> @@ -154,35 +178,61 @@ notmuch_restore_command (unused (void *ctx), int argc, char *argv[])\r
+>       * non-space characters for the message-id, then one or more\r
+>       * spaces, then a list of space-separated tags as a sequence of\r
+>       * characters within literal '(' and ')'. */\r
+> -    if ( xregcomp (&regex,\r
+> -               "^([^ ]+) \\(([^)]*)\\)$",\r
+> -               REG_EXTENDED) )\r
+> -    INTERNAL_ERROR ("compile time constant regex failed.");\r
+> +    char *p;\r
+>  \r
+> -    while ((line_len = getline (&line, &line_size, input)) != -1) {\r
+> -    regmatch_t match[3];\r
+> -    char *message_id, *file_tags;\r
+> +    line_len = getline (&line, &line_size, input);\r
+> +    if (line_len == 0)\r
+> +    return 0;\r
+>  \r
+> -    chomp_newline (line);\r
+> +    for (p = line; *p; p++) {\r
+> +    if (*p == '(')\r
+> +        input_format = DUMP_FORMAT_SUP;\r
+> +    }\r
+>  \r
+> -    rerr = xregexec (&regex, line, 3, match, 0);\r
+> -    if (rerr == REG_NOMATCH) {\r
+> -        fprintf (stderr, "Warning: Ignoring invalid input line: %s\n",\r
+> -                 line);\r
+> -        continue;\r
+> +    if (input_format == DUMP_FORMAT_AUTO)\r
+> +    input_format = DUMP_FORMAT_BATCH_TAG;\r
+> +\r
+> +    if (input_format == DUMP_FORMAT_SUP)\r
+> +    if ( xregcomp (&regex,\r
+> +                   "^([^ ]+) \\(([^)]*)\\)$",\r
+> +                   REG_EXTENDED) )\r
+> +        INTERNAL_ERROR ("compile time constant regex failed.");\r
+> +\r
+> +    tag_ops = tag_op_list_create (ctx);\r
+> +    if (tag_ops == NULL) {\r
+> +    fprintf (stderr, "Out of memory.\n");\r
+> +    return 1;\r
+> +    }\r
+\r
+Tag op list creation could be moved earlier to keep the parsing stuff\r
+closer together.\r
+\r
+> +\r
+> +    do {\r
+> +    char *query_string;\r
+> +\r
+> +    if (input_format == DUMP_FORMAT_SUP) {\r
+> +        ret =  parse_sup_line (ctx, line, &query_string, tag_ops);\r
+> +    } else {\r
+> +        ret =  parse_tag_line (ctx, line, TAG_FLAG_BE_GENEROUS,\r
+> +                               &query_string, tag_ops);\r
+\r
+Extra spaces after each = above.\r
+\r
+> +\r
+> +        if (ret == 0) {\r
+> +            if ( strncmp ("id:", query_string, 3) != 0) {\r
+> +                fprintf (stderr, "Unsupported query: %s\n", query_string);\r
+> +                continue;\r
+> +            }\r
+\r
+There should probably be a comment somewhere here saying that everything\r
+after id: is taken to be the message id, i.e. a query of "id:foo AND\r
+tag:bar" will not do what you want.\r
+\r
+An option is to add a TAG_FLAG to accept only message id as the query in\r
+parse_tag_line. It could trim trailing whitespace (if it doesn't\r
+already) and look for spaces in between individual search terms *before*\r
+hex decoding the query string, and fail if it finds any, and then check\r
+for id: prefix after hex decode.\r
+\r
+> +            /* delete id: from front of string; tag_message expects a\r
+> +             * raw message-id */\r
+> +            query_string = query_string + 3;\r
+> +        }\r
+>      }\r
+>  \r
+> -    message_id = xstrndup (line + match[1].rm_so,\r
+> -                           match[1].rm_eo - match[1].rm_so);\r
+> -    file_tags = xstrndup (line + match[2].rm_so,\r
+> -                          match[2].rm_eo - match[2].rm_so);\r
+> +    if (ret > 0)\r
+> +        continue;\r
+>  \r
+> -    tag_message (notmuch, message_id, file_tags, ! accumulate,\r
+> -                 synchronize_flags);\r
+> +    if (ret < 0 || tag_message (ctx, notmuch, query_string,\r
+> +                                tag_ops, flags))\r
+> +        break;\r
+> +\r
+> +    }  while ((line_len = getline (&line, &line_size, input)) != -1);\r
+>  \r
+> -    free (message_id);\r
+> -    free (file_tags);\r
+> -    }\r
+>  \r
+>      regfree (&regex);\r
+\r
+Only do this for sup format.\r
+\r
+BR,\r
+Jani.\r
+\r
+>  \r
+> @@ -190,8 +240,9 @@ notmuch_restore_command (unused (void *ctx), int argc, char *argv[])\r
+>      free (line);\r
+>  \r
+>      notmuch_database_destroy (notmuch);\r
+> +\r
+>      if (input != stdin)\r
+>      fclose (input);\r
+>  \r
+> -    return 0;\r
+> +    return ret;\r
+>  }\r
+> -- \r
+> 1.7.10.4\r
+>\r
+> _______________________________________________\r
+> notmuch mailing list\r
+> notmuch@notmuchmail.org\r
+> http://notmuchmail.org/mailman/listinfo/notmuch\r