DBDefs.pm.sample

#!/usr/bin/env perl

use strict;
use warnings;

package DBDefs;
use parent 'DBDefs::Default';
use MusicBrainz::Server::Replication qw( :replication_type );
use MusicBrainz::Server::DatabaseConnectionFactory;

# Commented-out lines should generally have sane defaults; anything that's uncommented
# probably needs personal attention.

################################################################################
# Directories
################################################################################

# The server root, i.e. the parent directory of admin, bin, lib, root, etc.
# By default, this uses the path of lib/DBDefs/Default.pm, minus '/lib/DBDefs/Default.pm'
# sub MB_SERVER_ROOT    { '/home/httpd/musicbrainz/musicbrainz-server' }
# Where static files are located
# sub STATIC_FILES_DIR { my $self= shift; $self->MB_SERVER_ROOT . '/root/static' }

################################################################################
# The Database
################################################################################

# Configuring databases here is required; there are no defaults.
MusicBrainz::Server::DatabaseConnectionFactory->register_databases(
    # How to connect when we need read-write access to the database
    READWRITE => {
        database    => 'musicbrainz_db',
        username    => 'musicbrainz',
        password        => 'musicbrainz',
#       host            => '',
#       port            => '',
    },
    # How to connect to a test database
    TEST => {
        database    => 'musicbrainz_test',
        username    => 'musicbrainz',
        password        => 'musicbrainz',
#       host            => '',
#       port            => '',
    },
    # How to connect to a Selenium test database. This database is created
    # (and dropped) automatically by t/selenium.mjs, and uses the TEST
    # database above as a template.
    SELENIUM => {
        database    => 'musicbrainz_selenium',
        schema      => 'musicbrainz',
        username    => 'musicbrainz',
#       host        => '',
#       port        => '',
    },
    # How to connect for read-only access.  See "REPLICATION_TYPE" (below)
    READONLY => {
        database    => 'musicbrainz_db',
        username    => 'musicbrainz',
        password        => 'musicbrainz',
#       host            => '',
#       port            => '',
    },
    # How to connect for administrative access
    SYSTEM    => {
        database    => 'template1',
        username    => 'postgres',
#       password        => '',
#       host            => '',
#       port            => '',
    },
    # How to connect when running maintenance scripts located under admin/.
    # This defaults to READWRITE if left undefined, but should be configured if
    # READWRITE points to a connection pooler that doesn't support session-based features.
#   MAINTENANCE    => {
#       database    => 'musicbrainz_db',
#       username    => 'musicbrainz',
#       password        => 'musicbrainz',
#       host            => '',
#       port            => '5432',
#   },
    # How to connect when running development scripts that query the production database.
#   PROD_STANDBY    => {
#       database    => 'musicbrainz_db',
#       username    => 'musicbrainz_ro',
#       password        => '',
#       host            => '',
#       port            => '5432',
#   },
);

# The schema sequence number.  Must match the value in
# replication_control.current_schema_sequence.
# This is required, there is no default in order to prevent it changing without
# manual intervention.
sub DB_SCHEMA_SEQUENCE { 27 }

# What type of server is this?
# * RT_MASTER - This is a master replication server.  Changes are allowed, and
#               they result in replication packets being produced.
# * RT_MIRROR - This is a mirror replication server.  After loading a snapshot
#               produced by a master, the only changes allowed are those made
#               by applying the next replication packet in turn.  If the mirror
#               server is not going to be used for development work, change
#               DB_STAGING_SERVER to 0.
#
#               A READONLY database connection must be configured if you
#               choose RT_MIRROR, as well as the usual READWRITE.
# * RT_STANDALONE - This server neither generates nor uses replication
#               packets.  Changes to the database are allowed.
# sub REPLICATION_TYPE { RT_STANDALONE }

# If you plan to use the RT_MIRROR setting (replicated data from MusicBrainz' Live Data Feed)
# you must sign in at https://metabrainz.org and generate an access token to access
# the replication packets. Enter the access token below:
# NOTE: DO NOT EXPOSE THIS ACCESS TOKEN PUBLICLY!
#
# sub REPLICATION_ACCESS_TOKEN { '' }

################################################################################
# GPG Signature
################################################################################

# Location of the public key file
# sub GPG_PUB_KEY { '' }

# Define how validation deals with the missing signature file:
#   FAIL    - validation fails if signature file is missing
#   PASS    - validation passes if signature file is missing
# sub GPG_MISSING_SIGNATURE_MODE { 'PASS' }

# Key identifiers (compatible with --recipient in GPG) for
# signatures and encryption of data dumps and packets. Should
# only be required on the master server.
# sub GPG_SIGN_KEY { '' }
# sub GPG_ENCRYPT_KEY { '' }

################################################################################
# HTTP Server Names
################################################################################

# The host names used by the server.
# To use a port number other than 80, add it like so: "myhost:8000"
# Additionally you should set the environment variable
# MUSICBRAINZ_USE_PROXY=1 when using a reverse proxy to make the server
# aware of it when generating things like the canonical url in catalyst.
sub WEB_SERVER                { 'www.musicbrainz.example.com' }
# Relevant only if SSL redirects are enabled
# sub WEB_SERVER_SSL            { 'localhost' }
# sub SEARCH_SERVER             { 'search.musicbrainz.org' }
# sub SEARCH_ENGINE             { 'SOLR' }
# Used, for example, to have emails sent from the beta server list the
# main server
# sub WEB_SERVER_USED_IN_EMAIL  { my $self = shift; $self->WEB_SERVER }

# Used for automatic beta redirection. Enabled if BETA_REDIRECT_HOSTNAME
# is truthy.
# sub IS_BETA                   { 0 }
# sub BETA_REDIRECT_HOSTNAME    { '' }

################################################################################
# Mail Settings
################################################################################

# sub SMTP_SERVER { 'localhost' }

# This value should be set to some secret value for your server.  Any old
# string of stuff should do; something suitably long and random, like for
# passwords.  However you MUST change it from the default
# value (the empty string).  This is so an attacker can't just look in CVS and
# see the default secret value, and then use it to attack your server.

# sub SMTP_SECRET_CHECKSUM { '' }
# sub EMAIL_VERIFICATION_TIMEOUT { 604800 } # one week

################################################################################
# Server Settings
################################################################################

# Set this to 0 if this is the master MusicBrainz server or a mirror server.
# Keeping this defined enables the banner that is shown across the top of each
# page, as well as some testing features that are only enabled when not on
# the live server.
# sub DB_STAGING_SERVER { 1 }

# This description is shown in the banner when DB_STAGING_SERVER is enabled.
# If left empty the default value will be shown.
# sub DB_STAGING_SERVER_DESCRIPTION { '' }

# Only change this if running a non-sanitized database on a staging server,
# e.g. http://beta.musicbrainz.org.
# * It shows a banner informing that 'all passwords have been reset to "mb"'.
# * It disables the IP lookup admin tool.
# sub DB_STAGING_SERVER_SANITIZED { 1 }

# Testing features enable "Accept edit" and "Reject edit" links on edits,
# this should only be enabled on staging servers. Also, this enables non-admin
# users to edit user permissions.
# sub DB_STAGING_TESTING_FEATURES { my $self = shift; $self->DB_STAGING_SERVER }

# SSL_REDIRECTS_ENABLED should be set to 1 on production.  It enables
# the "RequireSSL" attribute on Catalyst actions, which will redirect
# users to the SSL version of a Catalyst action (and redirect back to
# http:// after the action is complete, though only if the user
# started there).  If set to 0 no SSL redirects will be done, which is
# suitable for local or development deployments.
# sub SSL_REDIRECTS_ENABLED { 0 }

################################################################################
# Documentation Server Settings
################################################################################
# sub WIKITRANS_SERVER     { 'wiki.musicbrainz.org' }

# The path to MediaWiki's api.php file. This is required to automatically
# determine which documentation pages need to be updated in the
# transclusion table.
# sub WIKITRANS_SERVER_API { 'wiki.musicbrainz.org/api.php' }

# To enable documentation search on your server, create your own Google Custom
# Search engine and enter its ID as the value of GOOGLE_CUSTOM_SEARCH.
# Alternatively, if you're okay with the search results pointing to
# the musicbrainz.org server, you can use '006539527923176875863:xsv3chs2ovc'.
# sub GOOGLE_CUSTOM_SEARCH { '' }

################################################################################
# Cache Settings
################################################################################

# PLUGIN_CACHE_OPTIONS are the options configured for Plugin::Cache.  $c->cache
# is provided by Plugin::Cache, and is required for HTTP Digest authentication
# in the webservice (Catalyst::Authentication::Credential::HTTP).
# sub PLUGIN_CACHE_OPTIONS {
#     my $self = shift;
#     return {
#         class => 'MusicBrainz::Server::CacheWrapper::Redis',
#         server => '127.0.0.1:6379',
#         namespace => $self->CACHE_NAMESPACE . 'Catalyst:',
#     };
# }

# The caching options here relate to object caching in Redis - such as for
# artists, releases, etc. in order to speed up queries. See below if you want
# to disable caching.
# sub CACHE_MANAGER_OPTIONS {
#     my $self = shift;
#     my %CACHE_MANAGER_OPTIONS = (
#         profiles => {
#             external => {
#                 class => 'MusicBrainz::Server::CacheWrapper::Redis',
#                 options => {
#                     server => '127.0.0.1:6379',
#                     namespace => $self->CACHE_NAMESPACE,
#                 },
#             },
#         },
#         default_profile => 'external',
#     );
#
#     return \%CACHE_MANAGER_OPTIONS
# }
#
# Alternatively, you may disable caching by pointing to Cache::Null:
# sub CACHE_MANAGER_OPTIONS {
#     my $self = shift;
#     my %CACHE_MANAGER_OPTIONS = (
#         profiles => {
#             null => {
#                 class => 'Cache::Null',
#                 wrapped => 1,
#             },
#         },
#         default_profile => 'null',
#     );
#     return \%CACHE_MANAGER_OPTIONS;
# }

# Sets the TTL for entities stored in Redis, in seconds. A value of 0
# indicates that no expiration is set.
# sub ENTITY_CACHE_TTL { 0 }

################################################################################
# Sessions (advanced)
################################################################################

# The session store holds user login sessions. Session::Store::MusicBrainz
# uses DATASTORE_REDIS_ARGS to connect to and store sessions in Redis.

# sub SESSION_STORE { 'Session::Store::MusicBrainz' }
# sub SESSION_STORE_ARGS { return {} }
# sub SESSION_EXPIRE { return 36000; } # 10 hours

# Redis by default has 16 numbered databases available, of which DB 0
# is the default.  Here you can configure which of these databases are
# used by musicbrainz-server.
#
# test_database will be completely erased on each test run, so make
# sure it doesn't point at any production data you may have in your
# redis server.

# sub DATASTORE_REDIS_ARGS {
#     my $self = shift;
#     return {
#         database => 0,
#         namespace => $self->CACHE_NAMESPACE,
#         server => '127.0.0.1:6379',
#         test_database => 1,
#     };
# }

################################################################################
# Session cookies
################################################################################

# How long (in seconds) a web/rdf session can go "idle" before being timed out
# sub WEB_SESSION_SECONDS_TO_LIVE { 3600 * 3 }

# The cookie name to use
# sub SESSION_COOKIE { 'AF_SID' }
# The domain into which the session cookie is written
# sub SESSION_DOMAIN { undef }

################################################################################
# Other Settings
################################################################################

# Set this value to something true (e.g. 1) to set the server to read-only.
# To date, this option is widely ignored in the code; don't be surprised if you
# set it to true and find that writes are still possible.
# sub DB_READ_ONLY { 0 }

# Set this value to a message that you'd like to display to users when
# they attempt to write to your read-only database (not used if DB_READ_ONLY
# is false)
# sub DB_READ_ONLY_MESSAGE { <<EOF }
# This server is temporarily in read-only mode
# for database maintainance.
# EOF

# Amazon associate and developer ids
#my %amazon_store_associate_ids = (
#    'amazon.ca'         => 'music0b72-20',
#    'amazon.co.jp'    => 'musicbrainzjp-22',
#    'amazon.co.uk'    => 'music080d-21',
#    'amazon.com'    => 'musicbrainz0d-20',
#    'amazon.de'         => 'music059-21',
#    'amazon.fr'         => 'music083d-21',
#    'amazon.it'         => 'music084d-21',
#    'amazon.es'         => 'music02e-21',
#);

# sub AWS_ASSOCIATE_ID
# {
#     shift;
#     return keys %amazon_store_associate_ids if not @_;
#     return $amazon_store_associate_ids{$_[0]};
# }

# sub AWS_PRIVATE { '' }
# sub AWS_PUBLIC { '' }

# To enable use of reCAPTCHA:
# 1. make sure $ENV{'REMOTE_ADDR'} is the ip address of the visitor.
# 2. replace undef with your recaptcha keys:
# sub RECAPTCHA_PUBLIC_KEY { return undef }
# sub RECAPTCHA_PRIVATE_KEY { return undef }

# internet archive private/public keys (for coverartarchive.org).
# sub COVER_ART_ARCHIVE_ACCESS_KEY { };
# sub COVER_ART_ARCHIVE_SECRET_KEY { };
# sub COVER_ART_ARCHIVE_UPLOAD_PREFIXER { shift; sprintf('//%s.s3.us.archive.org/', shift) };
# sub COVER_ART_ARCHIVE_DOWNLOAD_PREFIX { '//coverartarchive.org' };

# Mapbox access token must be set to display area/place maps.
# sub MAPBOX_MAP_ID { 'mapbox/streets-v11' }
# sub MAPBOX_ACCESS_TOKEN { '' }

# Disallow OAuth2 requests over plain HTTP
# sub OAUTH2_ENFORCE_TLS { my $self = shift; !$self->DB_STAGING_SERVER }

# sub USE_ETAGS { 1 }

# sub CATALYST_DEBUG { 1 }

# If you are developing on MusicBrainz, you should set this to a true value
# This will turn off some optimizations (such as CSS/JS compression) to make
# developing and debuging easier
# sub DEVELOPMENT_SERVER { 1 }

# How long to wait before rechecking template files (undef uses the
# Template::Toolkit default)
# sub STAT_TTL { shift->DEVELOPMENT_SERVER() ? undef : 1200 }

# Please activate the officially approved languages here. Not every .po
# file is active because we might have fully translated languages which
# are not yet properly supported, like right-to-left languages
# sub MB_LANGUAGES {qw()}

# Should the site fall back to browser settings when trying to set a language
# (note: will still only use languages in MB_LANGUAGES)
# sub LANGUAGE_FALLBACK_TO_BROWSER{ 1 }

# Bugs can be sent to a Sentry instance (https://sentry.io) via these settings.
# The DSNs can be found on the project configuration page.
# sub SENTRY_DSN { undef }
# sub SENTRY_DSN_PUBLIC { undef }

# Configure which html validator should be used.  If you run tests
# often, you should probably run a local copy of the validator.  See
# http://about.validator.nu/#src for instructions.
# sub HTML_VALIDATOR { 'http://validator.w3.org/nu/?out=json' }
# local use example: sub HTML_VALIDATOR { 'http://localhost:8888?out=json' }

# Set to 1 if you're a developer and plan to run tests locally. Never
# enable in production.
# sub USE_SET_DATABASE_HEADER { 0 }

################################################################################
# Profiling
################################################################################
# Set these to >0 to enable profiling

# Log if a request in /ws takes more than x seconds
# sub PROFILE_WEB_SERVICE { 0 }

# Log if a request in / (not /ws) takes more than x seconds
# sub PROFILE_SITE { 0 }

# The maximum amount of time a process can be serving a single request. This
# function takes a Catalyst::Request as input, and should return the amount of time
# in seconds that it should take to respond to this request.
# If undef, the process is never killed.
# sub DETERMINE_MAX_REQUEST_TIME { undef }

# sub LOGGER_ARGUMENTS {
#     return (
#         outputs => [
#             [ 'Screen', min_level => 'debug', newline => 1 ],
#         ],
#     )
# }

1;

=head1 COPYRIGHT AND LICENSE

Copyright (C) 1998 Robert Kaye

This file is part of MusicBrainz, the open internet music database,
and is licensed under the GPL version 2, or (at your option) any
later version: http://www.gnu.org/licenses/gpl-2.0.txt

=cut