package Buscador; use strict; use vars qw($VERSION); use base 'Maypole::Redirect'; #use base 'Apache::MVC'; $VERSION = "0.2"; #use Maypole::Cache; use Buscador::Config; Buscador->config->{cache_options}{class} = "Cache::FileCache"; use Maypole::Constants; our $home; sub debug() {0} BEGIN { require Email::Store; Email::Store->import(Buscador::Config->dsn, verbose => 1 ); use Module::Pluggable search_path => [ "Email::Store" ], sub_name => 'stores', require => 1; use Module::Pluggable::Ordered search_path => [ "Buscador" ]; $home = Buscador::Config->home; Buscador->config->{cache_options}{class} = "Cache::FileCache"; Buscador->config->{model} = "Maypole::Model::CDBI::Plain"; # this is a bit of an egregious hack # perhaps plugins should specifically state whether they # are capable of being set up or not my @stores = sort grep { !/(DBI|Addressing)$/ } grep { !/SUPER$/ } grep { $_->can("table") } Buscador->stores; # not needed any more #@stores = sort qw/ Email::Store::Mail Email::Store::List Email::Store::Date #Email::Store::Entity Email::Store::Entity::Name Email::Store::Attachment #Email::Store::Entity::Address Email::Store::NamedEntity Email::Store::Vote /; Buscador->setup([ @stores ]); }; Buscador->config->{rows_per_page} = 10; Buscador->config->{template_root} = "$home/templates"; Buscador->config->{uri_base} = Buscador::Config->uri; Buscador->config->{img_base} = Buscador::Config->image_uri; Buscador->config->{uri_base} .= "/" unless Buscador->config->{uri_base} =~ m!/$!; Buscador->config->{img_base} .= "/" unless Buscador->config->{img_base} =~ m!/$!; $Email::Store::Plucene::index_path = "$home/emailstore-index"; $Plucene::QueryParser::DefaultOperator = "AND"; sub parse_path { my $self = shift; Buscador->call_plugins("parse_path", $self); $self->SUPER::parse_path(); } 1; __END__ =head1 NAME Buscador - a dynamic mail archiver with a twist =head1 DESCRIPTION Buscador is web based mail archival and retrieval tool based around the concept of Intertwingle : http://www.mozilla.org/blue-sky/misc/199805/intertwingle.html In essence it provides a variety of different views on the mail using a system of plugins. Plugins provided include ones to show thread views, date views, seperation into mailing lists, extraction of named entities and Atom feeds for recent mails, per thread, per list and per person and for handling mailing. =head1 WARNING! Buscador is distinctly B level software and details of its architecture could change at anytime. Be prepared for DB changes and plugin system gutting and rebuilding by reading the safety card in the pocket in front of you and bracing you hands over your head in the event of catastrophic architecture rethinking. =head1 INSTALL =head2 Install dependencies There's a C available from http://thegestalt.org/simon/perl/Bundle-Buscador-0.1.tar.gz however some people have had problems installing some of these. Namely Apache::Request Class::DBI::AsForm Email::MIME Email::MIME::Attachment::Stripper Mail::ListDetector And, in particular C. C installs a lot of weird things such as C, C and C. A cut down version of C without these dependencies is available from http://thegestalt.org/simon/perl/SQL-Translator-0.05-lite.tar.gz =head2 Create config file Make a directory in your web root, cd into it and do % buscador -init this will copy some templates and some images into the directory and then generate a sample config file. You should edit the config file. You might want to move your chrome directory outside your new buscador directory and alter your config accordingly. That way Maypole (which Buscador is based on) doesn't try and first see if there's a table called 'chrome' before passing through to the actual chrome directory and also won't fill your logs with errors. =head2 Import some mails Run % buscador -setup and then % buscador /path/to/mail/folder =head2 Create Apache config Something like SetHandler perl-script PerlHandler Buscador but changed to whatever directory you wnat to install it under. If you're using the default SQLite db remember to make sure that the web server has enough access to read it and get a lock. =head1 PLUGINS The plugin system is based around C. Each plugin get the chance to influence the path being passed in. The order that they are called in is set by the B method, the lower the return value the higher the priority. For example: package Buscador::Foo; # we're middling important sub parse_path_order { 13 } sub parse_path { my ($self, $buscador) = @_; # buscador is an alias for search $buscador->{path} =~ s!/buscador/!/search/!; } 1; however they don't have to touch the path at all and can simply install methods in other namespaces; package Buscador::Bar; # this is where path parsing methods would go package Email::Store::Mail; use Fortune; sub bar :Exported { my $fortune = Fortune ('fortunefile')->read_header()->get_random_fortune(); $r->{template_args}{fortune} = $fortune; $r->{template} = "fortune"; } 1; And now, if we write a 'fortune' template and go to http://example.com/buscador/mail/bar we'll be presented with a fortune. =head1 BUGS Lots and lots. And then some more =over 4 =item UTF8 doesn't work properly. This needs serious testing. =item The templates are ugly. I have a vague design in mind. We could also do with having themes. =item see TODO file =back =head1 AUTHOR Originally Simon Cozens, Esimon@cpan.orgE Now maintained by Simon Wistow, Esimon@thegestalt.orgE =head1 SUPPORT This module is part of the Siesta project - http://siesta.unixbeard.net There is a mailing list at siesta-dev@siesta.unixbeard.net (subscribe at siesta-dev-sub@siesta.unixbeard.net). SVN/DAV access is available via http://siesta.unixbeard.net/svn/trunk/buscador/ =head1 COPYRIGHT AND LICENSE Copyright 2004, 2005 by Simon Cozens and Simon Wistow This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself. =cut