+ Notmuch requires either [maildir
+ flag](http://cr.yp.to/proto/maildir.html) or a "mh"-style
+ maildirectory to operate on. Basically any setup in which each mail
+ is in a file of its own will work. Here are some generally
+ well-regarded mail retrieval tools:
+
+ * [offlineimap](https://github.com/nicolas33/offlineimap/) -
+ quite useful and widely tested, it also offers a handy hook that
+ will come in useful a bit later in our setup. Also supports
+ "presynchook" and "postsynchook" command that will get run
+ whenever you sync. Point _postsynchook_ to a script that gets run
+ on every sync and that will do the automatic updating and tagging
+ of your notmuch database.
+
+ * [mbsync](http://isync.sourceforge.net/)
+
+ * [getmail](http://pyropus.ca/software/getmail/)
+
+ * [fetchmail](http://fetchmail.berlios.de/)
+
+ See the [[initial_tagging]] page for more info on intial tagging of messages.
+
+* <span id="print_filenames">**Print only filenames of a search (python bindings)**</span>
+
+ Notmuch includes [python bindings](http://pypi.python.org/pypi/notmuch)
+ to the notmuch shared library. Extensive API documentation
+ [is available](http://packages.python.org/notmuch).
+
+ The bindings are very simple to use. As an example, given you have
+ the python bindings installed (or simply set your PYTHONPATH
+ environment variable to point to the .../bindings/python directory),
+ this script will print the filenames of a matching search: