1 Fix the things that are causing the most pain to new users
2 ----------------------------------------------------------
3 1. A new import is tagging all messages as "inbox" -- total pain
5 Emacs interface (notmuch.el)
6 ----------------------------
7 Add notmuch-bcc and notmuch-cc for setting default Bcc and Cc values,
8 (should affect the message-setup-hook).
10 Switch the notmuch-search view to use "notmuch search --format=json"
11 to fix large classes of bugs regarding poorly-escaped output and lame
12 regular expressions. (The most recently found, unfixed example is the
13 sender's name containing ';' which causes emacs to drop a search
14 result.) This may require removing the outer array from the current
15 "notmuch search --format=json" results.
17 Fix '*' to work by simply calling '+' or '-' on a region consisting of
18 the entire buffer, (this would avoid one race condition---while still
19 leaving other race conditions---but could also potentially make '*' a
20 very expensive operation).
22 Add a global keybinding table for notmuch, and then view-specific
23 tables that add to it.
25 Add a '|' binding from the search view.
27 Add support for choosing from one of the user's configured email
28 addresses for the From line.
30 Make 'notmuch-show-pipe-message have a private history.
32 Add support for a delete keybinding that adds a "deleted" tag to the
33 current message/thread and make searches not return deleted messages
34 by default, (unless the user asks explicitly for deleted messages in
37 Add keybindings for next/previous thread.
39 Add support to "mute" a thread (add a "muted" tag and then don't
40 display threads in searches by default where any message of the thread
43 Make '=' count from the end rather than from the beginning if more
44 than half-way through the buffer.
46 Fix to automatically wrap long headers (for RFC compliance) before
47 sending. This should probably just be fixed in message-mode itself,
48 (but perhaps we can have a notmuch-message-mode that layers this on
51 Stop hiding the headers so much in the thread-view mode.
53 Allow opening a message in thread-view mode by clicking on either
56 Automatically open a message when navigating to it with N or P.
58 Change 'a' command in thread-view mode to only archive open messages.
62 Fix bash completion to complete multiple search options (both --first
63 and *then* --max-threads), and also complete value for --sort=
64 (oldest-first or newest-first).
66 notmuch command-line tool
67 -------------------------
68 Add support to "notmuch search" and "notmuch show" to allow for
69 listing of duplicate messages, (distinct filenames with the same
70 Message-ID). I'm not sure what the option should be named. Perhaps
73 "notmuch setup" should use realpath() before replacing the
74 configuration file. The ensures that the final target file of any
75 intermediate symbolic links is what is actually replaced, (rather than
78 Replace "notmuch reply" with "notmuch compose --reply <search-terms>".
79 This would enable a plain "notmuch compose" to be used to construct an
80 initial message, (which would then have the properly configured name
81 and email address in the From: line. We could also then easily support
82 "notmuch compose --from <something>" to support getting at alternate
85 Implement "notmuch search --exclude-threads=<search-terms>" to allow
86 for excluding muted threads, (and any other negative, thread-based
87 filtering that the user wants to do).
89 Fix "notmuch show" so that the UI doesn't fail to show a thread that
90 is visible in a search buffer, but happens to no longer match the
91 current search. (Perhaps add a --matching=<secondary-search-terms>
92 option (or similar) to "notmuch show".) For now, this is being worked
93 around in the emacs interface by noticing that "notmuch show" returns
94 nothing and re-rerunning the command without the extra arguments.
96 Add a "--format" option to "notmuch search", (something printf-like
97 for selecting what gets printed).
99 Give "notmuch restore" some progress indicator.
101 Fix "notmuch restore" to operate in a single pass much like "notmuch
102 dump" does, rather than doing N searches into the database, each
103 matching 1/N messages.
105 Add a "-f <filename>" option to select an alternate configuration
108 Allow configuration for filename patterns that should be ignored when
111 Fix to avoid this ugly message:
113 (process:17197): gmime-CRITICAL **: g_mime_message_get_mime_part: assertion `GMIME_IS_MESSAGE (message)' failed
114 Warning: Not indexing empty mime part.
116 This probably means adding a test case to generate that message,
117 filing an upstream bug against GMime, and then silencing the
118 notmuch-generated portion of the warning (so that once GMime is
119 fixed, this is all silent).
121 Simplify notmuch-reply to simply print the headers (we have the
122 original values) rather than calling GMime (which encodes) and adding
123 the confusing gmime-filter-headers.c code (which decodes).
125 Properly handle replying to multiple messages. Currently, the JSON
126 reply format only supports a single message, but the default reply
127 format accepts searches returning multiple messages. The expected
128 behavior of replying to multiple messages is not obvious, and there
129 are multiple ideas that might make sense. Some consensus needs to be
130 reached on this issue, and then both reply formats should be updated
135 Add support for custom flag<->tag mappings. In the notmuch
136 configuration file this could be
139 synchronize_flags = R:replied; D*:deleted; S:~unread;
141 In the library interface this could be implemented with an array of
142 structures to define the mapping (flag character, tag name,
143 inverse-sense bit (~ above), and tag-when-any-file-flagged
144 vs. tag-when-all-files-flagged (* above)).
146 Add an interface to accept a "key" and a byte stream, rather than a
149 Provide a sane syntax for date ranges. First, we don't want to require
150 both endpoints to be specified. For example it would be nice to be
151 able to say things like "since:2009-01-1" or "until:2009-01-1" and
152 have the other endpoint be implicit. Second we'd like to support
153 relative specifications of time such as "since:'2 months ago'". To do
154 any of this we're probably going to need to break down an write our
155 own parser for the query string rather than using Xapian's QueryParser
158 Make failure to read a file (such as a permissions problem) a warning
159 rather than an error (should be similar to the existing warning for a
162 Fix to use the *last* Message-ID header if multiple such headers are
163 encountered, (I noticed this is one thing that kept me from seeing the
164 same message-ID values as sup).
166 Add support for configuring "virtual tags" which are a tuple of
167 (tag-name, search-specification). The database is responsible for
168 ensuring that the virtual tag is always consistent.
170 Indicate to the user if two files with the same message ID have
171 content that is actually different in some interesting way. Perhaps
172 notmuch initially sees all changes as interesting, and quickly learns
173 from the user which changes are not interesting (such as the very
174 common mailing-list footer).
176 Fix notmuch_query_count_messages to share code with
177 notmuch_query_search_messages rather than duplicating code. (And
178 consider renaming it as well.)
180 Provide a mechanism for doing automatic address completion based on
181 notmuch searches. Here was one proposal made in IRC:
183 <cworth> I guess all it would really have to be would be a way
184 to configure a series of searches to try in turn,
185 (presenting ambiguities at a given single level, and
186 advancing to the next level only if one level
187 returned no matches).
188 <cworth> So then I might have a series that looks like this:
189 <cworth> notmuch search --output=address_from tag:address_book_alias
190 <cworth> notmuch search --output=address_to tag:sent
191 <cworth> notmuch search --output=address_from
192 <cworth> I think I might like that quite a bit.
193 <cworth> And then we have a story for an address book for
196 Provide a ~me Xapian synonym for all of the user's configured email
199 Add symbol hiding so that we don't risk leaking any private symbols
200 into the shared-library interface.
202 Audit all libnotmuch entry points to ensure that all Xapian calls are
203 wrapped in a try/catch block.
205 Fix the "count" functionality to be exact as Olly explained in IRC:
207 ojwb> cworth: if you set the check_at_least parameter to the
208 database size, get_matches_estimated() will be exact
210 Fix the threading of a message that has a References: header but no
211 In-Reply-To: header (see id:"87lixxnxpb.fsf@yoom.home.cworth.org").
215 Implement support for "tag:*" to expand to all tags.
217 Fix "notmuch search to:" to be less confusing. Many users expect this
218 to search for all messages with a To: header, but it instead searches
219 for all messages with the word "to". If we don't provide the first
220 behavior, perhaps we should exit on an error when a configured prefix
221 is provided with no value?
223 Support "*" in all cases and not just as a special case. That is, "* "
224 should also work, as well as "* and tag:inbox".
226 Implement a syntax for requesting set-theoertic operations on results
227 of multiple searches. For example, I would like to do:
229 "tag:inbox" SET-SUBTRACT "tag:muted"
233 "tag:notmuch and <date-range>" SET-INTERSECT
234 "tag:notmuch and not (tag:merged or tag:postponed)"
236 See id:3wdpr282yz2.fsf@testarossa.amd.com for more details on the
237 use cases of the above.
241 Store a reference term for every message-id that appears in
242 References. We just started doing this for newly-added documents, but
243 at the next convenient database-schema upgrade, we should go back and
244 fix old messages to be consistent.
246 Start indexing the List-Id header, (and re-index this header for
247 existing messages at the next database upgrade).
249 Add support for the user to specify custom headers to be indexed (and
250 re-index these for existing messages at the next database upgrade).
252 Save filenames for files detected as "not an email file" in the
253 database. This would allow for two things: 1. Optimizing "notmuch new"
254 to not have to look at these files again (since they are potentially
255 large so the detection could be potentially slow). 2. A "notmuch
256 search" syntax could be added to allow the user to find these files,
257 (and perhaps delete them or move them away as appropriate).
259 Fix filesystem/notmuch-new race condition by not updating database
260 mtime for a directory if it is the same as the current mtime.
264 Achieve 100% test coverage with the test suite.
268 Audit everything for dealing with out-of-memory (and drop xutil.c).
270 Investigate why the notmuch database is slightly larger than the sup
271 database for the same corpus of email.
273 Makefile should print message teaching user about LD_LIBRARY_PATH (or
274 similar) if libdir is not set to a directory examined by ldconfig.