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 2. Allow an easy way to get tags from directory names (if the user has them)
7 3. Allow an easy way to remove excess tags, (date-based search)
9 4. Make emacs fast for big search results (see "lazy searching" below)
11 5. Fix Xapian defect #250 so tagging is fast.
13 Emacs interface (notmuch.el)
14 ----------------------------
15 Make the keybindings help ('?') display the summary of each command's
16 documentation, not the function name.
18 Add a global keybinding table for notmuch, and then view-specific
19 tables that add to it.
21 Add a command to archive all threads in a search view.
23 Lazy searching: call "notmuch search" with --first and --max to fill
24 just a screenful of results, and then fill in more as ther user pages
27 Add a '|' binding from the search view.
29 Add a binding to run a search from notmuch-show-mode.
31 When a thread has been entirely read, start out by closing all
32 messages except those that matched the search terms.
34 Add support for choosing from one of the user's configured email
35 addresses for the From line.
37 Make 'notmuch-show-pipe-message have a private history.
39 Add support for a delete keybinding that adds a "deleted" tag to the
40 current message/thread and make searches not return deleted messages
41 by default, (unless the user asks explicitly for deleted messags in
44 Add support to "mute" a thread (add a "muted" tag and then don't
45 display threads in searches by default where any message of the thread
50 Fix configure script to test each compiler warning we want to use.
52 Implement strndup locally (or call talloc_strndup instead).
54 Implement getline locally, (look at gnulib).
58 Fix bash completion to complete multiple search options (both --first
59 and *then* --max-threads), and also complete value for --sort=
60 (oldest-first or newest-first).
62 notmuch command-line tool
63 -------------------------
64 Teach "notmuch search" to return many different kinds of results. Some
67 notmuch search --for threads # Default if no --for is given
68 notmuch search --for messages
69 notmuch search --for tags
70 notmuch search --for addresses
71 notmuch search --for terms
73 Add a "--format" option to "notmuch search", (something printf-like
74 for selecting what gets printed).
76 Add a "--count-only" (or so?) option to "notmuch search" for returning
77 the count of search results.
79 Give "notmuch restore" some progress indicator. Until we get the
80 Xapian bugs fixed that are making this operation slow, we really need
81 to let the user know that things are still moving.
83 Add a "-f <filename>" option to select an alternate configuration
86 Fix notmuch.c to call add_timestamp/get_timestamp with path names
87 relative to the database path. (Otherwise, moving the database to a
88 new directory will result in notmuch creating new timestamp documents
89 and leaving stale ones behind.)
91 Ensure that "notmuch new" is sane if its first, giant indexing session
92 gets interrupted, (that is, ensure that any results indexed so far are
95 Fix notmuch.c to use a DIR prefix for directory timestamps, (the idea
96 being that it can then add other non-directory timestamps such as for
97 noting how far back in the past mail has been indexed, and whether it
98 needs to re-tag messages based on a theoretical "auto-tags"
101 Make "notmuch new" notice when a mail directory has gone more than a
102 month without receiving new mail and use that to trigger the printing
103 of the note that the user might want to mark the directory read-only.
105 Also make "notmuch new" optionally able to just mark those month-old
106 directories read-only on its own. (Could conflict with low-volume
107 lists such as announce lists if they are setup to deliver to their own
110 Allow configuration for filename patterns that should be ignored when
115 Add support for files that are moved or deleted (which obviously need
116 to be handled differently).
118 Actually compile and install a libnotmuch shared library.
120 Fix to use the *last* Message-ID header if multiple such headers are
121 encountered, (I noticed this is one thing that kept me from seeing the
122 same message-ID values as sup).
124 Add support for the user to specify custom headers to be indexed.
126 Add support for configuring "virtual tags" which are a tuple of
127 (tag-name, search-specification). The database is responsible for
128 ensuring that the virtual tag is always consistent.
132 Audit everything for dealing with out-of-memory (and drop xutil.c).
136 Achieve 100% test coverage with the test suite.
138 Investigate why the notmuch database is slightly larger than the sup
139 database for the same corpus of email.
145 replace_document should make minimal changes to database file
146 http://trac.xapian.org/ticket/250
148 It looks like it's going to be easy to fix. Here's the file to
151 xapian-core/backends/flint/flint_database.cc
155 // FIXME - in the case where there is overlap between the new
156 // termlist and the old termlist, it would be better to compare the
157 // two lists, and make the minimum set of modifications required.
158 // This would lead to smaller changesets for replication, and
159 // probably be faster overall
161 So I think this might be as easy as just walking over two
162 sorted lists looking for differences.
164 Note that this is in the currently default "flint" backend,
165 but the Xapian folks are probably more interested in fixing
166 the in-development "chert" backend. So the patch to get
167 upstreamed there will probably also fix:
169 xapian-core/backends/chert/chert_database.cc
171 (I'm hoping the fix will be the same---an identical comment
174 Also, if you want to experiment with the chert backend,
175 compile current Xapian source and run notmuch with
176 XAPIAN_PREFER_CHERT=1. I haven't tried that yet, but there are
177 claims that a chert database can be 40% smaller than an
178 equivalent flint database.
182 "tag:foo and tag:bar and -tag:deleted" goes insane
184 This seems to be triggered by a Boolean operator next to a
185 token starting with a non-word character---suddenly all the
186 Boolean operators get treated as literal tokens)