Re: [PATCH 08/14] lib: Simplify upgrade code using a transaction

diff --git a/fc/ace1ee0e3e0dad7b16af2d89225739a877b94c b/fc/ace1ee0e3e0dad7b16af2d89225739a877b94c
new file mode 100644 (file)
index 0000000..fa69e82
--- /dev/null
+++ b/fc/ace1ee0e3e0dad7b16af2d89225739a877b94c
@@ -0,0 +1,233 @@
+Return-Path: <amdragon@mit.edu>\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 14E25431FB6\r
+       for <notmuch@notmuchmail.org>; Sun, 27 Jul 2014 09:42:37 -0700 (PDT)\r
+X-Virus-Scanned: Debian amavisd-new at olra.theworths.org\r
+X-Spam-Flag: NO\r
+X-Spam-Score: -2.3\r
+X-Spam-Level: \r
+X-Spam-Status: No, score=-2.3 tagged_above=-999 required=5\r
+       tests=[RCVD_IN_DNSWL_MED=-2.3] 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 ImwWbGeucHTM for <notmuch@notmuchmail.org>;\r
+       Sun, 27 Jul 2014 09:42:29 -0700 (PDT)\r
+Received: from dmz-mailsec-scanner-7.mit.edu (dmz-mailsec-scanner-7.mit.edu\r
+       [18.7.68.36])\r
+       (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))\r
+       (No client certificate requested)\r
+       by olra.theworths.org (Postfix) with ESMTPS id 2E156431FAE\r
+       for <notmuch@notmuchmail.org>; Sun, 27 Jul 2014 09:42:29 -0700 (PDT)\r
+X-AuditID: 12074424-f79146d00000067c-d8-53d52bf43424\r
+Received: from mailhub-auth-3.mit.edu ( [18.9.21.43])\r
+       (using TLS with cipher AES256-SHA (256/256 bits))\r
+       (Client did not present a certificate)\r
+       by dmz-mailsec-scanner-7.mit.edu (Symantec Messaging Gateway) with SMTP\r
+       id 4F.A3.01660.4FB25D35; Sun, 27 Jul 2014 12:42:28 -0400 (EDT)\r
+Received: from outgoing.mit.edu (outgoing-auth-1.mit.edu [18.9.28.11])\r
+       by mailhub-auth-3.mit.edu (8.13.8/8.9.2) with ESMTP id s6RGgRmg012752; \r
+       Sun, 27 Jul 2014 12:42:28 -0400\r
+Received: from awakening.csail.mit.edu (awakening.csail.mit.edu [18.26.4.91])\r
+       (authenticated bits=0)\r
+       (User authenticated as amdragon@ATHENA.MIT.EDU)\r
+       by outgoing.mit.edu (8.13.8/8.12.4) with ESMTP id s6RGgPqo019190\r
+       (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT);\r
+       Sun, 27 Jul 2014 12:42:26 -0400\r
+Received: from amthrax by awakening.csail.mit.edu with local (Exim 4.80)\r
+       (envelope-from <amdragon@mit.edu>)\r
+       id 1XBRWi-00079H-Ri; Sun, 27 Jul 2014 12:42:24 -0400\r
+Date: Sun, 27 Jul 2014 12:42:24 -0400\r
+From: Austin Clements <amdragon@MIT.EDU>\r
+To: Mark Walters <markwalters1009@gmail.com>\r
+Subject: Re: [PATCH 08/14] lib: Simplify upgrade code using a transaction\r
+Message-ID: <20140727164224.GG13893@mit.edu>\r
+References: <1406433173-19169-1-git-send-email-amdragon@mit.edu>\r
+       <1406433173-19169-9-git-send-email-amdragon@mit.edu>\r
+       <87vbqjywhy.fsf@qmul.ac.uk>\r
+MIME-Version: 1.0\r
+Content-Type: text/plain; charset=us-ascii\r
+Content-Disposition: inline\r
+In-Reply-To: <87vbqjywhy.fsf@qmul.ac.uk>\r
+User-Agent: Mutt/1.5.21 (2010-09-15)\r
+X-Brightmail-Tracker:\r
+ H4sIAAAAAAAAA+NgFmpileLIzCtJLcpLzFFi42IR4hTV1v2ifTXYoG++ssXquTwW12/OZHZg\r
+       8tg56y67x7NVt5gDmKK4bFJSczLLUov07RK4Mp69yS44olsxee9nlgbG50pdjJwcEgImElfu\r
+       bmCHsMUkLtxbz9bFyMUhJDCbSeLjwS9QzkZGifkt+5kgnNNMEpN61rJAOEsYJZZeXckC0s8i\r
+       oCpxd+1kNhCbTUBDYtv+5YwgtoiAjsTtQwvAdjALSEt8+90MNImDQ1jAU+LLoUKQMC9QSc+J\r
+       71AzpzJKvH7TxwyREJQ4OfMJC0SvlsSNfy/BekHmLP/HARLmBFp1fO8vsFWiAioSU05uY5vA\r
+       KDQLSfcsJN2zELoXMDKvYpRNya3SzU3MzClOTdYtTk7My0st0jXXy80s0UtNKd3ECApqdheV\r
+       HYzNh5QOMQpwMCrx8GawXQkWYk0sK67MPcQoycGkJMqrDYwJIb6k/JTKjMTijPii0pzU4kOM\r
+       EhzMSiK8hW+BynlTEiurUovyYVLSHCxK4rxvra2ChQTSE0tSs1NTC1KLYLIyHBxKErx/tICG\r
+       ChalpqdWpGXmlCCkmTg4QYbzAA2frwFUw1tckJhbnJkOkT/FqCglznsSpFkAJJFRmgfXC0s6\r
+       rxjFgV4R5uUAuZsHmLDgul8BDWYCGszifxlkcEkiQkqqgXGO3e1Aee3sjCVHL8xWjPrUqWrh\r
+       /kHf2fb0U8dTplN2C1sW6rodU1Mt/JbQpcz8/2Pqj/5zEWZZigai2/a3anrO2FJcuSjRhkU+\r
+       PONP5sx7J8MjarbufxXkfHSiarDul5QqV5bSz3d19hbeczzKxraq61rC4q4tUfnZPBe3L2ec\r
+       faz3TdDjeCWW4oxEQy3mouJEAEbJL1YVAwAA\r
+Cc: notmuch@notmuchmail.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, 27 Jul 2014 16:42:37 -0000\r
+\r
+Quoth Mark Walters on Jul 27 at 10:35 am:\r
+> \r
+> Hi\r
+> \r
+> On Sun, 27 Jul 2014, Austin Clements <amdragon@MIT.EDU> wrote:\r
+> > Previously, the upgrade was organized as two passes -- an upgrade\r
+> > pass, and a separate cleanup pass -- so the database was always in a\r
+> > valid state.  This change substantially simplifies this code by\r
+> > performing the upgrade in a transaction and combining both passes in\r
+> > to one.  This 1) eliminates a lot of duplicate code between the\r
+> > passes, 2) speeds up the upgrade process, 3) makes progress reporting\r
+> > more accurate, 4) eliminates the potential for stale data if the\r
+> > upgrade is interrupted during the cleanup pass, and 5) makes it easier\r
+> > to reason about the safety of the upgrade code.\r
+> \r
+> I like this but I wonder if it has a side effect: I think with the\r
+> current code the user can interrupt the upgrade (ctrl-c) and continue\r
+> roughly where it left off. This looks like it means the whole upgrade\r
+> needs to be done in one go. Will this be a problem on large mail stores\r
+> (eg rlb with over 1M messages)?\r
+> \r
+> I am not sure what could be done during the interrupted upgrade before\r
+> so maybe this is not a problem.\r
+\r
+I haven't tested this hypothesis, but I don't think a partially\r
+completed upgrade would actually help upon restarting the upgrade.\r
+Since the old upgrade process couldn't safely remove terms/data until\r
+the end of the upgrade, if it were interrupted, the next upgrade would\r
+start right back at the beginning and do everything over again.\r
+\r
+Also, since the old upgrade code had to update the version number\r
+before removing old terms/data, if it was interrupted during the\r
+cleanup process the database would be left with cruft that would\r
+*never* be removed.\r
+\r
+With features we actually have a better chance of making partially\r
+completed upgrades useful: we could commit after each individual\r
+feature gets upgraded.  Of course, that only helps when upgrade has\r
+multiple new features to upgrade to, so it may or may not be useful in\r
+practice depending on how quickly we add new features.\r
+\r
+> Best wishes\r
+> \r
+> Mark\r
+> \r
+> \r
+> > ---\r
+> >  lib/database.cc | 67 ++++++---------------------------------------------------\r
+> >  1 file changed, 7 insertions(+), 60 deletions(-)\r
+> >\r
+> > diff --git a/lib/database.cc b/lib/database.cc\r
+> > index 03eef3e..0be7180 100644\r
+> > --- a/lib/database.cc\r
+> > +++ b/lib/database.cc\r
+> > @@ -1238,6 +1238,9 @@ notmuch_database_upgrade (notmuch_database_t *notmuch,\r
+> >    timer_is_active = TRUE;\r
+> >      }\r
+> >  \r
+> > +    /* Perform the upgrade in a transaction. */\r
+> > +    db->begin_transaction (true);\r
+> > +\r
+> >      /* Before version 1, each message document had its filename in the\r
+> >       * data field. Copy that into the new format by calling\r
+> >       * notmuch_message_add_filename.\r
+> > @@ -1265,6 +1268,7 @@ notmuch_database_upgrade (notmuch_database_t *notmuch,\r
+> >        filename = _notmuch_message_talloc_copy_data (message);\r
+> >        if (filename && *filename != '\0') {\r
+> >            _notmuch_message_add_filename (message, filename);\r
+> > +          _notmuch_message_clear_data (message);\r
+> >            _notmuch_message_sync (message);\r
+> >        }\r
+> >        talloc_free (filename);\r
+> > @@ -1312,6 +1316,8 @@ notmuch_database_upgrade (notmuch_database_t *notmuch,\r
+> >                                                   NOTMUCH_FIND_CREATE, &status);\r
+> >            notmuch_directory_set_mtime (directory, mtime);\r
+> >            notmuch_directory_destroy (directory);\r
+> > +\r
+> > +          db->delete_document (*p);\r
+> >        }\r
+> >    }\r
+> >      }\r
+> > @@ -1353,67 +1359,8 @@ notmuch_database_upgrade (notmuch_database_t *notmuch,\r
+> >      notmuch->features |= NOTMUCH_FEATURES_CURRENT;\r
+> >      db->set_metadata ("features", _print_features (local, notmuch->features));\r
+> >      db->set_metadata ("version", STRINGIFY (NOTMUCH_DATABASE_VERSION));\r
+> > -    db->flush ();\r
+> > -\r
+> > -    /* Now that the upgrade is complete we can remove the old data\r
+> > -     * and documents that are no longer needed. */\r
+> > -    if (version < 1) {\r
+> > -  notmuch_query_t *query = notmuch_query_create (notmuch, "");\r
+> > -  notmuch_messages_t *messages;\r
+> > -  notmuch_message_t *message;\r
+> > -  char *filename;\r
+> > -\r
+> > -  for (messages = notmuch_query_search_messages (query);\r
+> > -       notmuch_messages_valid (messages);\r
+> > -       notmuch_messages_move_to_next (messages))\r
+> > -  {\r
+> > -      if (do_progress_notify) {\r
+> > -          progress_notify (closure, (double) count / total);\r
+> > -          do_progress_notify = 0;\r
+> > -      }\r
+> > -\r
+> > -      message = notmuch_messages_get (messages);\r
+> > -\r
+> > -      filename = _notmuch_message_talloc_copy_data (message);\r
+> > -      if (filename && *filename != '\0') {\r
+> > -          _notmuch_message_clear_data (message);\r
+> > -          _notmuch_message_sync (message);\r
+> > -      }\r
+> > -      talloc_free (filename);\r
+> > -\r
+> > -      notmuch_message_destroy (message);\r
+> > -  }\r
+> >  \r
+> > -  notmuch_query_destroy (query);\r
+> > -    }\r
+> > -\r
+> > -    if (version < 1) {\r
+> > -  Xapian::TermIterator t, t_end;\r
+> > -\r
+> > -  t_end = notmuch->xapian_db->allterms_end ("XTIMESTAMP");\r
+> > -\r
+> > -  for (t = notmuch->xapian_db->allterms_begin ("XTIMESTAMP");\r
+> > -       t != t_end;\r
+> > -       t++)\r
+> > -  {\r
+> > -      Xapian::PostingIterator p, p_end;\r
+> > -      std::string term = *t;\r
+> > -\r
+> > -      p_end = notmuch->xapian_db->postlist_end (term);\r
+> > -\r
+> > -      for (p = notmuch->xapian_db->postlist_begin (term);\r
+> > -           p != p_end;\r
+> > -           p++)\r
+> > -      {\r
+> > -          if (do_progress_notify) {\r
+> > -              progress_notify (closure, (double) count / total);\r
+> > -              do_progress_notify = 0;\r
+> > -          }\r
+> > -\r
+> > -          db->delete_document (*p);\r
+> > -      }\r
+> > -  }\r
+> > -    }\r
+> > +    db->commit_transaction ();\r
+> >  \r
+> >      if (timer_is_active) {\r
+> >    /* Now stop the timer. */\r
+> >\r