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
20 Add a global keybinding table for notmuch, and then view-specific
21 tables that add to it.
23 Add a '|' binding from the search view.
25 Add support for choosing from one of the user's configured email
26 addresses for the From line.
28 Make 'notmuch-show-pipe-message have a private history.
30 Add support for a delete keybinding that adds a "deleted" tag to the
31 current message/thread and make searches not return deleted messages
32 by default, (unless the user asks explicitly for deleted messags in
35 Add keybindings for next/previous thread.
37 Add support to "mute" a thread (add a "muted" tag and then don't
38 display threads in searches by default where any message of the thread
41 Make '=' count from the end rather than from the beginning if more
42 than half-way through the buffer.
44 Fix to automatically wrap long headers (for RFC compliance) before
45 sending. This should probably just be fixed in message-mode itself,
46 (but perhaps we can have a notmuch-message-mode that layers this on
49 Stop hiding the headers so much in the thread-view mode.
51 Allow opening a message in thread-view mode by clicking on either
54 Automatically open a message when navigating to it with N or P.
56 Change 'a' command in thread-view mode to only archive open messages.
58 Add a binding to open all closed messages.
60 Change the 'a'rchive command in the thread view to only archive open
65 Fix bash completion to complete multiple search options (both --first
66 and *then* --max-threads), and also complete value for --sort=
67 (oldest-first or newest-first).
69 notmuch command-line tool
70 -------------------------
71 Add support to "notmuch search" and "notmuch show" to allow for
72 listing of duplicate messages, (distinct filenames with the same
73 Message-ID). I'm not sure what the option should be named. Perhaps
76 Add a -0 option to "notmuch search" so that one can safely deal with
79 notmuch search --output=files -0 <terms> | xargs -0 <command>
81 "notmuch setup" should use realpath() before replacing the
82 configuration file. The ensures that the final target file of any
83 intermediate symbolic links is what is actually replaced, (rather than
86 Replace "notmuch reply" with "notmuch compose --reply <search-terms>".
87 This would enable a plain "notmuch compose" to be used to construct an
88 initial message, (which would then have the properly configured name
89 and email address in the From: line. We could also then easily support
90 "notmuch compose --from <something>" to support getting at alternate
93 Fix the --format=json option to not imply --entire-thread.
95 Implement "notmuch search --exclude-threads=<search-terms>" to allow
96 for excluding muted threads, (and any other negative, thread-based
97 filtering that the user wants to do).
99 Fix "notmuch show" so that the UI doesn't fail to show a thread that
100 is visible in a search buffer, but happens to no longer match the
101 current search. (Perhaps add a --matching=<secondary-search-terms>
102 option (or similar) to "notmuch show".) For now, this is being worked
103 around in the emacs interface by noticing that "notmuch show" returns
104 nothing and re-rerunning the command without the extra arguments.
106 Add a "--format" option to "notmuch search", (something printf-like
107 for selecting what gets printed).
109 Give "notmuch restore" some progress indicator.
111 Fix "notmuch restore" to operate in a single pass much like "notmuch
112 dump" does, rather than doing N searches into the database, each
113 matching 1/N messages.
115 Add a "-f <filename>" option to select an alternate configuration
118 Allow configuration for filename patterns that should be ignored when
121 Replace the "notmuch part --part=id" command with "notmuch show
122 --part=id", (David Edmonson wants to rewrite some of "notmuch show" to
123 provide more MIME-structure information in its output first).
125 Replace the "notmuch search-tags" command with "notmuch search
128 Fix to avoid this ugly message:
130 (process:17197): gmime-CRITICAL **: g_mime_message_get_mime_part: assertion `GMIME_IS_MESSAGE (message)' failed
131 Warning: Not indexing empty mime part.
133 This probably means adding a test case to generate that message,
134 filing an upstream bug against GMime, and then silencing the
135 notmuch-generated portion of the warning (so that once GMime is
136 fixed, this is all silent).
138 Simplify notmuch-reply to simply print the headers (we have the
139 original values) rather than calling GMime (which encodes) and adding
140 the confusing gmime-filter-headers.c code (which decodes).
144 Add support for custom flag<->tag mappings. In the notmuch
145 configuration file this could be
148 synchronize_flags = R:replied; D*:deleted; S:~unread;
150 In the library interface this could be implemented with an array of
151 structures to define the mapping (flag character, tag name,
152 inverse-sense bit (~ above), and tag-when-any-file-flagged
153 vs. tag-when-all-files-flagged (* above)).
155 Add an interface to accept a "key" and a byte stream, rather than a
158 Provide a sane syntax for date ranges. First, we don't want to require
159 both endpoints to be specified. For example it would be nice to be
160 able to say things like "since:2009-01-1" or "until:2009-01-1" and
161 have the other endpoint be implicit. Second we'd like to support
162 relative specifications of time such as "since:'2 months ago'". To do
163 any of this we're probably going to need to break down an write our
164 own parser for the query string rather than using Xapian's QueryParser
167 Make failure to read a file (such as a permissions problem) a warning
168 rather than an error (should be similar to the existing warning for a
171 Fix to use the *last* Message-ID header if multiple such headers are
172 encountered, (I noticed this is one thing that kept me from seeing the
173 same message-ID values as sup).
175 Add support for configuring "virtual tags" which are a tuple of
176 (tag-name, search-specification). The database is responsible for
177 ensuring that the virtual tag is always consistent.
179 Indicate to the user if two files with the same message ID have
180 content that is actually different in some interesting way. Perhaps
181 notmuch initially sees all changes as interesting, and quickly learns
182 from the user which changes are not interesting (such as the very
183 common mailing-list footer).
185 Fix notmuch_query_count_messages to share code with
186 notmuch_query_search_messages rather than duplicating code. (And
187 consider renaming it as well.)
189 Provide a mechanism for doing automatic address completion based on
190 notmuch searches. Here was one proposal made in IRC:
192 <cworth> I guess all it would really have to be would be a way
193 to configure a series of searches to try in turn,
194 (presenting ambiguities at a given single level, and
195 advancing to the next level only if one level
196 returned no matches).
197 <cworth> So then I might have a series that looks like this:
198 <cworth> notmuch search --output=address_from tag:address_book_alias
199 <cworth> notmuch search --output=address_to tag:sent
200 <cworth> notmuch search --output=address_from
201 <cworth> I think I might like that quite a bit.
202 <cworth> And then we have a story for an address book for
205 Provide a ~me Xapian synonym for all of the user's configured email
208 Add symbol hiding so that we don't risk leaking any private symbols
209 into the shared-library interface.
211 Audit all libnotmuch entry points to ensure that all Xapian calls are
212 wrapped in a try/catch block.
214 Fix the "count" functionality to be exact as Olly explained in IRC:
216 ojwb> cworth: if you set the check_at_least parameter to the
217 database size, get_matches_estimated() will be exact
221 Implement support for "tag:*" to expand to all tags.
223 Fix "notmuch search to:" to be less confusing. Many users expect this
224 to search for all messages with a To: header, but it instead searches
225 for all messages with the word "to". If we don't provide the first
226 behavior, perhaps we should exit on an error when a configured prefix
227 is provided with no value?
229 Support "*" in all cases and not just as a special case. That is, "* "
230 should also work, as well as "* and tag:inbox".
232 Implement a syntax for requesting set-theoertic operations on results
233 of multiple searches. For example, I would like to do:
235 "tag:inbox" SET-SUBTRACT "tag:muted"
239 "tag:notmuch and <date-range>" SET-INTERSECT
240 "tag:notmuch and not (tag:merged or tag:postponed)"
242 See id:3wdpr282yz2.fsf@testarossa.amd.com for more details on the
243 use cases of the above.
247 Store a reference term for every message-id that appears in
248 References. We just started doing this for newly-added documents, but
249 at the next convenient database-schema upgrade, we should go back and
250 fix old messages to be consistent.
252 Start indexing the List-Id header, (and re-index this header for
253 existing messages at the next database upgrade).
255 Add support for the user to specify custom headers to be indexed (and
256 re-index these for existing messages at the next database upgrade).
260 Achieve 100% test coverage with the test suite.
264 Audit everything for dealing with out-of-memory (and drop xutil.c).
266 Investigate why the notmuch database is slightly larger than the sup
267 database for the same corpus of email.
269 Makefile should print message teaching user about LD_LIBRARY_PATH (or
270 similar) if libdir is not set to a directory examined by ldconfig.