cli: support --ignored-address for init command

Allow for skipping unwanted addresses (such as 'noreply') from the
contacts cache.

NEWS.org

@@ -8,6 +8,14 @@
     - ~mu~ commands ~extract~, ~verify~ and ~view~ can now read the message from
       standard input; see their man-pages for details
 
+    - ~mu init~ gained the ~--ignored-address~ option for email-addresses / regexps
+      that should _not_ be included in the contacts-cache (i.e., for ~mu cfind~ and
+      Mu4e address completion). See the ~mu-init~ manpage for details.
+
+      It's not unusual for 'noreply`-type e-mail addresses to be the majority in
+      an e-mail corpus, so it useful to get rid of those, with
+      '=--ignored-address=/.*noreply*/'=
+
     - experimental: if you build ~mu~ with [[https://github.com/CLD2Owners/cld2][CLD2]] support (available in many Linux
       distros), ~mu~ will try to detect the language of the body of e-mail
       messages; you can then search by their ISO-639-1 code, e.g. ~mu find

lib/mu-contacts-cache.cc

@@ -188,8 +188,6 @@ ContactsCache::Private::serialize() const
 	dirty_ = 0;
 }
 
-
-
 ContactsCache::ContactsCache(Config& config_db)
 	: priv_{std::make_unique<Private>(config_db)}
 {}
@@ -381,9 +379,6 @@ ContactsCache::is_ignored(const std::string& addr) const
 	return address_matches(addr, priv_->ignored_plain_, priv_->ignored_rx_);
 }
 
-
-
-
 #ifdef BUILD_TESTS
 /*
  * Tests.

man/mu-init.1.org

@@ -21,24 +21,38 @@ starts searching at =<maildir>=. By default, *mu* uses whatever the *MAILDIR*
 environment variable is set to; if it is not set, it tries =~/Maildir= if it
 already exists.
 
-** --my-address=<my-email-address>
+** --my-address=<email-address-or-regex>
 
-specifies that some e-mail addresses are 'my-address' (the option can be used
+specifies that some e-mail address is 'my-address' (the option can be used
 multiple times). Any message in which at least one of the contact fields
 contains such an address is considered a 'personal' messages; this can then be
 used for filtering in *mu-find(1)*, *mu-cfind(1)* and *mu4e*, e.g. to filter-out
 mailing list messages.
 
-=<my-email-address>= can be either a plain e-mail address (such as
+=<email-address-or-regex>= can be either a plain e-mail address (such as
 *foo@example.com*), or a basic PCRE regular-expression (see *pcre(3)* for details),
 wrapped in */* (such as =/foo-.*@example\\.com/=). Depending on your shell, the
 argument may need to be quoted.
 
+** --ignored-address=<email-address-or-regex>
+
+specifies that some e-mail address is to be ignored from the contacts-cache
+(the option can be used multiple times). Such address then cannot be found with
+*mu-cfind(1)* or in the Mu4e contacts cache.
+
+=<my-email-address>= can be either a plain e-mail address or a regexp, just like
+for the =--my-address= option.
+
 ** --reinit
 
-reinitialize the database from an earlier version; that is, create a new
+reinitialize the database from an earlier version; that is, create a new empty
-empty database witht the existing settings. This cannot be combined
+database with the existing settings. This cannot be combined with the other ~init~
-with the other ~init~ options.
+options.
+
+* EXAMPLE
+#+begin_example
+$ mu init --maildir=~/Maildir --my-address=alice@example.com --my-address=bob@example.com --ignored-address='/.*reply.*/'
+#+end_example
 
 #+include: "exit-code.inc" :minlevel 1
 

mu/mu-cmd-info.cc

@@ -49,7 +49,6 @@ Mu::mu_cmd_info(const Mu::Store& store, const Options& opts)
 			return "never";
 		else
 			return time_to_string("%c", t);
-
 	};
 
 	Table info;
@@ -61,8 +60,11 @@ Mu::mu_cmd_info(const Mu::Store& store, const Options& opts)
 	info.add_row({"max-message-size", format("%zu", conf.get<Config::Id::MaxMessageSize>())});
 	info.add_row({"batch-size", format("%zu", conf.get<Config::Id::BatchSize>())});
 	info.add_row({"created", tstamp(conf.get<Config::Id::Created>())});
+
 	for (auto&& c : conf.get<Config::Id::PersonalAddresses>())
 		info.add_row({"personal-address", c});
+	for (auto&& c : conf.get<Config::Id::IgnoredAddresses>())
+		info.add_row({"ignored-address", c});
 
 	info.add_row({"messages in store", format("%zu", store.size())});
 	info.add_row({"last-change", tstamp(store.statistics().last_change)});

mu/mu-cmd-init.cc

@@ -46,10 +46,15 @@ Mu::mu_cmd_init(const Options& opts)
 
 		MemDb mdb;
 		Config conf{mdb};
+
 		if (opts.init.max_msg_size)
 			conf.set<Config::Id::MaxMessageSize>(*opts.init.max_msg_size);
 		if (opts.init.batch_size)
 			conf.set<Config::Id::MaxMessageSize>(*opts.init.batch_size);
+		if (!opts.init.my_addresses.empty())
+			conf.set<Config::Id::PersonalAddresses>(opts.init.my_addresses);
+		if (!opts.init.ignored_addresses.empty())
+			conf.set<Config::Id::IgnoredAddresses>(opts.init.ignored_addresses);
 
 		return Store::make_new(opts.runtime_path(RuntimePath::XapianDb),
 				       opts.init.maildir, conf);
@@ -67,4 +72,3 @@ Mu::mu_cmd_init(const Options& opts)
 
 	return Ok();
 }
-

mu/mu-options.cc

@@ -383,8 +383,12 @@ sub_init(CLI::App& sub, Options& opts)
 		       "Top of the maildir")
 		->type_name("<maildir>");
 	sub.add_option("--my-address", opts.init.my_addresses,
-		       "Personal e-mail addresses")
+		       "Personal e-mail address or regexp")
 		->type_name("<address>");
+	sub.add_option("--ignored-address", opts.init.ignored_addresses,
+		       "Ignored e-mail address or regexp")
+		->type_name("<address>");
+
 	sub.add_option("--max-message-size", opts.init.max_msg_size,
 		       "Maximum allowed message size in bytes");
 	sub.add_option("--batch-size", opts.init.batch_size,
@@ -393,6 +397,7 @@ sub_init(CLI::App& sub, Options& opts)
 		       "Re-initialize database with current settings")
 		->excludes("--maildir")
 		->excludes("--my-address")
+		->excludes("--ignored-address")
 		->excludes("--max-message-size")
 		->excludes("--batch-size");
 }

mu/mu-options.hh

@@ -184,6 +184,8 @@ struct Options {
 	struct Init {
 		std::string	maildir;		/**<  where the mails are */
 		StringVec	my_addresses;		/**<  personal e-mail addresses */
+		StringVec	ignored_addresses;	/**<  addresses to be ignored for
+							 * the contacts-cache */
 		OptSize         max_msg_size;		/**<  max size for message files */
 		OptSize		batch_size;		/**<  db transaction batch size */
 		bool		reinit;			/**<  re-initialize  */