Re: [PATCH v2 3/7] fix thread breakage via ghost-on-removal

diff --git a/df/3ef580593652b2e9b121f880fc9f360cd84a12 b/df/3ef580593652b2e9b121f880fc9f360cd84a12
new file mode 100644 (file)
index 0000000..6826382
--- /dev/null
+++ b/df/3ef580593652b2e9b121f880fc9f360cd84a12
@@ -0,0 +1,171 @@
+Return-Path: <tomi.ollila@iki.fi>\r
+X-Original-To: notmuch@notmuchmail.org\r
+Delivered-To: notmuch@notmuchmail.org\r
+Received: from localhost (localhost [127.0.0.1])\r
+ by arlo.cworth.org (Postfix) with ESMTP id 90E146DE034D\r
+ for <notmuch@notmuchmail.org>; Mon,  4 Apr 2016 23:53:31 -0700 (PDT)\r
+X-Virus-Scanned: Debian amavisd-new at cworth.org\r
+X-Spam-Flag: NO\r
+X-Spam-Score: 0.625\r
+X-Spam-Level: \r
+X-Spam-Status: No, score=0.625 tagged_above=-999 required=5 tests=[AWL=-0.027,\r
+  SPF_NEUTRAL=0.652] autolearn=disabled\r
+Received: from arlo.cworth.org ([127.0.0.1])\r
+ by localhost (arlo.cworth.org [127.0.0.1]) (amavisd-new, port 10024)\r
+ with ESMTP id Kw2KJrRbRnui for <notmuch@notmuchmail.org>;\r
+ Mon,  4 Apr 2016 23:53:21 -0700 (PDT)\r
+Received: from guru.guru-group.fi (guru.guru-group.fi [46.183.73.34])\r
+ by arlo.cworth.org (Postfix) with ESMTP id 9E2F56DE02C2\r
+ for <notmuch@notmuchmail.org>; Mon,  4 Apr 2016 23:53:20 -0700 (PDT)\r
+Received: from guru.guru-group.fi (localhost [IPv6:::1])\r
+ by guru.guru-group.fi (Postfix) with ESMTP id ED20F1000E0;\r
+ Tue,  5 Apr 2016 09:53:34 +0300 (EEST)\r
+From: Tomi Ollila <tomi.ollila@iki.fi>\r
+To: Daniel Kahn Gillmor <dkg@fifthhorseman.net>,\r
+ Notmuch Mail <notmuch@notmuchmail.org>\r
+Subject: Re: [PATCH v2 3/7] fix thread breakage via ghost-on-removal\r
+In-Reply-To: <1459606541-23889-3-git-send-email-dkg@fifthhorseman.net>\r
+References: <1459445693-3900-1-git-send-email-dkg@fifthhorseman.net>\r
+ <1459606541-23889-1-git-send-email-dkg@fifthhorseman.net>\r
+ <1459606541-23889-3-git-send-email-dkg@fifthhorseman.net>\r
+User-Agent: Notmuch/0.21+99~g7cbc880 (http://notmuchmail.org) Emacs/24.3.1\r
+ (x86_64-unknown-linux-gnu)\r
+X-Face: HhBM'cA~<r"^Xv\KRN0P{vn'Y"Kd;zg_y3S[4)KSN~s?O\"QPoL\r
+ $[Xv_BD:i/F$WiEWax}R(MPS`^UaptOGD`*/=@\1lKoVa9tnrg0TW?"r7aRtgk[F\r
+ !)g;OY^,BjTbr)Np:%c_o'jj,Z\r
+Date: Tue, 05 Apr 2016 09:53:34 +0300\r
+Message-ID: <m2lh4sczrl.fsf@guru.guru-group.fi>\r
+MIME-Version: 1.0\r
+Content-Type: text/plain\r
+X-BeenThere: notmuch@notmuchmail.org\r
+X-Mailman-Version: 2.1.20\r
+Precedence: list\r
+List-Id: "Use and development of the notmuch mail system."\r
+ <notmuch.notmuchmail.org>\r
+List-Unsubscribe: <https://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: <https://notmuchmail.org/mailman/listinfo/notmuch>,\r
+ <mailto:notmuch-request@notmuchmail.org?subject=subscribe>\r
+X-List-Received-Date: Tue, 05 Apr 2016 06:53:31 -0000\r
+\r
+On Sat, Apr 02 2016, Daniel Kahn Gillmor <dkg@fifthhorseman.net> wrote:\r
+\r
+> ghost-on-removal the solution to T590-thread-breakage.sh that just\r
+> adds a ghost message after removing each message.\r
+>\r
+> It leaks information about whether we've ever seen a given message id,\r
+> but it's a fairly simple implementation.\r
+>\r
+> Note that _resolve_message_id_to_thread_id also introduces new\r
+> message_ids to the database, so i think just searching for a given\r
+> message ID may introduce the same metadata leakage.\r
+>\r
+> This differs from v1 of this changeset in that we implement the change\r
+> in _notmuch_message_delete, a more "internal" function.\r
+\r
+I fetched your changes from lair.fifthhorseman.net yesterday and diffed\r
+against master; looks pretty good, some quick comments (this email has\r
+most relevant changes related to those changes):\r
+\r
+> ---\r
+>  lib/database.cc |  2 +-\r
+>  lib/message.cc  | 29 ++++++++++++++++++++++++++---\r
+>  2 files changed, 27 insertions(+), 4 deletions(-)\r
+>\r
+> diff --git a/lib/database.cc b/lib/database.cc\r
+> index 3b342f1..d5733c9 100644\r
+> --- a/lib/database.cc\r
+> +++ b/lib/database.cc\r
+> @@ -2557,7 +2557,7 @@ notmuch_database_remove_message (notmuch_database_t *notmuch,\r
+>  \r
+>      if (status == NOTMUCH_STATUS_SUCCESS && message) {\r
+>          status = _notmuch_message_remove_filename (message, filename);\r
+> -        if (status == NOTMUCH_STATUS_SUCCESS)\r
+> +        if (status == NOTMUCH_STATUS_SUCCESS) \r
+\r
+It looks to be that this change is insignificant enough so it could be\r
+dropped ;)\r
+\r
+>              _notmuch_message_delete (message);\r
+>          else if (status == NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID)\r
+>              _notmuch_message_sync (message);\r
+> diff --git a/lib/message.cc b/lib/message.cc\r
+> index 8d72ea2..e414e9c 100644\r
+> --- a/lib/message.cc\r
+> +++ b/lib/message.cc\r
+> @@ -1037,20 +1037,43 @@ _notmuch_message_sync (notmuch_message_t *message)\r
+>      message->modified = FALSE;\r
+>  }\r
+>  \r
+> -/* Delete a message document from the database. */\r
+> +/* Delete a message document from the database, leaving a ghost\r
+> + * message in its place */\r
+\r
+This comment could tell the condition when ghost message is left --\r
+versus the case all ghost messages are dropped (thread becomes empty of\r
+mail messages).\r
+\r
+>  notmuch_status_t\r
+>  _notmuch_message_delete (notmuch_message_t *message)\r
+>  {\r
+>      notmuch_status_t status;\r
+>      Xapian::WritableDatabase *db;\r
+> +    const char *mid, *tid;\r
+> +    notmuch_message_t *ghost;\r
+> +    notmuch_private_status_t private_status;\r
+> +    notmuch_database_t *notmuch;\r
+> +        \r
+> +    mid = notmuch_message_get_message_id (message);\r
+> +    tid = notmuch_message_get_thread_id (message);\r
+> +    notmuch = message->notmuch;\r
+>  \r
+>      status = _notmuch_database_ensure_writable (message->notmuch);\r
+>      if (status)\r
+>      return status;\r
+>  \r
+> -    db = static_cast <Xapian::WritableDatabase *> (message->notmuch->xapian_db);\r
+> +    db = static_cast <Xapian::WritableDatabase *> (notmuch->xapian_db);\r
+>      db->delete_document (message->doc_id);\r
+> -    return NOTMUCH_STATUS_SUCCESS;\r
+> +        \r
+> +    /* and reintroduce a ghost in its place */\r
+> +    ghost = _notmuch_message_create_for_message_id (notmuch, mid, &private_status);\r
+\r
+In next lines the expected case NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND\r
+could be first. Although the performance difference is negligible to me\r
+it looks silly do this first check and basically always fail there and\r
+then do 'else if' which is likely to succeeed...\r
+(your latest version in lair does not return in this first case but sets\r
+status to NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID. Perhaps later messages in this\r
+thread does the same but those are somewhat out-of-context for this reply).\r
+\r
+> +    if (private_status == NOTMUCH_PRIVATE_STATUS_SUCCESS) {\r
+> +    /* this is deeply weird, and we should not have gotten into\r
+> +       this state.  is there a better error message to return\r
+> +       here? */\r
+> +    return NOTMUCH_STATUS_DUPLICATE_MESSAGE_ID;\r
+> +    } else if (private_status == NOTMUCH_PRIVATE_STATUS_NO_DOCUMENT_FOUND) {\r
+> +    private_status = _notmuch_message_initialize_ghost (ghost, tid);\r
+> +    if (! private_status)\r
+> +        _notmuch_message_sync (ghost);\r
+> +    }\r
+> +    notmuch_message_destroy (ghost);\r
+> +    return COERCE_STATUS (private_status, "Error converting to ghost message");\r
+>  }\r
+\r
+Outside of this patch, but in some of the next messages, adds functions\r
+_notmuch_message_has_term() and _notmuch_message_has_term_st(). Perhaps\r
+the _notmuch_message_has_term() could be left unimplemented?\r
+\r
+>  \r
+>  /* Transform a blank message into a ghost message.  The caller must\r
+> -- \r
+> 2.8.0.rc3\r
+>\r
+> _______________________________________________\r
+> notmuch mailing list\r
+> notmuch@notmuchmail.org\r
+> https://notmuchmail.org/mailman/listinfo/notmuch\r