config.yaml

# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#               Sample Newsreap Configuration File
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
# This is a sample Newsreap configuration file. When Newsreap is started up
# it will check the following locations for configuration files (in order):
#    ~/.config/newsreap/config.yaml
#    ~/newsreap/config.yaml
#    ~/.newsreap/config.yaml
#    /etc/newsreap/config.yaml
#    /etc/config.yaml
#
# So be sure to copy your configuration file in one of the paths identified
# above and everything will 'just work' assuming your configuration is
# correct! :)
#
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#   Global Configuration
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
global:
   # All of newsreap cataloging gets placed here
   base_dir: ~/.config/newsreap
   # Defines a working directory to download temporary content into
   work_dir: %{base_dir}/var

# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#   NNTP Server Definitions
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#
# You may define as many servers as you like; below identifies the default
# values. If you're happy with the default entries, you can omit them for
# simplicity if you like.
servers:
    # The NNTP Providers hostname and/or IP
  - host: myprovider
    # The NNTP Providers port (associated with the hostname)
    port: 563

    # Your login and password
    username: myusername
    password: mypassword

    # Use TLS when establishing a connection. This is directly tied to the
    # port you chose above.
    secure: True

    # Verify the ownership of the provides NNTP Server's certificate (if using
    # a secure connection) with the Certificate Authority. This is a good
    # option to enable to avoid a man-in-the-middle attack.
    verify_cert: False

    # Support compression if available;  if compression isn't available, we
    # automatically safely fall back to rfc3977.  Unless you're certain your
    # NNTP Provider doesn't support compression, there is no reason to change
    # this option (to rfc3977)
    iostream: gzip.rfc3977

    # Older NNTP Providers required you to select the Usenet group before
    # retrieving content from it. This isn't really a thing anymore. To be
    # backwards compatible, the option is here anyway.
    join_group: False

    # you either download your messages using BODY, or you download your
    # content using ARTICLE (this is all behind the scenes stuff). The option
    # is here for those who like to tweak their settings.
    use_body: False

    # You either use HEAD or you use STAT when fetching details about a post
    # Similar to the use_body above, there is really no reason to change this
    # value
    use_stat: True

# Define any number of servers you want
#  - host: my.other.provider
#    port: 563
#    username: myusername
#    password: mypassword
#    secure: True
#    verify_cert: False
#    iostream: gzip.rfc3977
#    join_group: False
#    use_body: False
#    use_stat: True

# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#   Processing (Before and After Downloading)
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
# The processing section defines specifics to how you want to handle i/o
# to and from the server.
processing:
   # Each thread you define here will spawn another connection to you're NNTP
   # Service Provider.  The more servers you spin, the more work/downloads you
   # can perform concurrently.  Make sure not to set this value higher then
   # the maximum number of connections your provider allows.
   threads: 3

   # This option is only used when indexing headers off of usenet for offline
   # searching/filtering. This defines the number of headers you want to scan
   # and process at a time. Setting this value between 25000 and 75000 seems
   # to be the sweet spot. Set it higher if you have lots of cpu power and
   # lower if you don't. Consider that each thread you defined above will
   # be processing this many headers concurrently with the others.  Setting
   # it to high doesn't allow you to distribute the load very well.
   header_batch_size: 25000

   # This should be the absolute path to a directory you've mapped to a
   # ramdisk. This is more of a Linux thing, but a ramdisk acts as a swapping
   # location when handling indexed results. e.g:
   #    mkdir /var/newsreap/swap
   #
   #    # Now use tmpfs to create a ramdisk.
   #    sudo mount -t tmpfs -o size=4G tmpfs /var/newsreap/swap
   #
   #    # Older systems may not have tmpfs, so you can use ramfs instead
   #    # ramfs does not have a physical limit and is specified as a starting
   #    # size but requires you to specify a value anyway. In the below
   #    # example, the 1G is not really referenced; you'll use all of your
   #    # ram until you have no more to offer
   #    sudo mount -t ramfs -o size=1G ramfs /var/newsreap/swap
   #
   #    If you used the examples identified above, you could set your
   #    ramdisk to be /var/newsreap/swap
   #
   #    Note: indexing will behave slower if the you specify a path that
   #          is not in memory.  In these cases, it's much better to leave
   #          this set to None.
   ramdisk: None

# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#   Posting to your NNTPServer
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
# If you plan on posting an article, this field just provides some templated
# information to make it easier to do so.
posting:
    # The poster is stamped on all posts you make.
    poster: 'Newsreap <reaper@newsreap.io>'

    # The default subject
    subject: '{{description}} (%Y-%m-%d)" - "{{filename}}" yEnc ({{index}}/{{count}})'

    # If posting binary content, what is the largest size you want your
    # article to be before it's broken up into segments. Most providers do not
    # allow you to exceed 760KB; so don't change this unless you know what
    # you're doing
    max_article_size: 760KB

    # Some archivers (such as rar, zip, 7z) can break up a large archive
    # into several smaller archived files
    # You can specify sizes like 25MB or 10MB, or just put in the byte
    # value if you want (no unit).
    #
    # The default is 'auto'; auto is a special keyword to automatically
    # calculate the size based on the content being archived.
    max_archive_size: auto

# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
#   Database Configuration
# -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
# The engine must be an SQLAlchemy URL.  They're pretty straight forward
# and SQLAlchemy allows us to support all of the major databases out there
# as well.
database:
   engine: 'sqlite://%{base_dir}/newsreap.db'
#   engine: 'sqlite:///:memory:'
#   engine: 'mysql+mysqldb://user:pass@hostname/newsreap'
#   engine: 'postgresql+psycopg2://user:pass@localhost/newsreap'