Advanced Config¶
Here, we'll break down all of the configuration options per section.
Main¶
Here's the main
section.
uuid
will be created automatically, you need not worry about this.cache_dir
is handy for keeping cached files out of the script directory. This is helpful for reinstalls as well as development.
[main]
# uuid =
# cache_dir =
Streaming¶
method
can be set toffmpeg
,vlc
ordirect
.bytes_per_read
determines how many bytes of the stream to read before sending the data to your client. Increasing this value may cause longer load times, and lowering it may effectstuttering
.origin_quality
can be set to high,medium,low for most variants. Variants that make use of m3u8 will Autoselect High for the direct method if not set. ffmpeg/vlc will determine the best stream on their own. Some Variants can allow alternative values.transcode_quality
works with ffmpeg/vlc to use fHDHR for handling quality instead of the origin. Valid settings include: heavy,mobile,internet720,internet480,internet360,internet240
[streaming]
# method = direct
# bytes_per_read = 1152000
# origin_quality = None
# transcode_quality = None
fhdhr¶
The fhdhr
contains all the configuration options for interfacing between this script and your media platform.
address
andport
are what we will allow the script to listen on.0.0.0.0
is the default, and will respond to all.discovery_address
may be helpful for making SSDP work properly. Ifaddress
is not0.0.0.0
, we will use that. If this is not set to a real IP, we won't run SSDP. SSDP is only really helpful for discovering in Plex/Emby. It's a wasted resource since you can manually add theip:port
of the script to Plex.tuner_count
is a limit of devices able to stream from the script. The default is 4.friendlyname
is to set the name that Plex sees the script as.reporting_*
are settings that show how the script projects itself as a hardware device.device_auth
andrequire_auth
are for an unimplemented Authentication feature.chanscan_on_start
Scans Origin for new channels at startup.
[fhdhr]
# address = 0.0.0.0
# discovery_address = 0.0.0.0
# port = 5004
# tuner_count = 4
# friendlyname = fHDHR
# reporting_firmware_name = fHDHR
# reporting_manufacturer = BoronDust
# reporting_model = fHDHR
# reporting_firmware_ver = 20201001
# reporting_tuner_type = Antenna
# device_auth = fHDHR
# require_auth = False
# chanscan_on_start = True
EPG¶
images
can be set toproxy
orpass
. If you chooseproxy
, images will be reverse proxied through fHDHR.method
which method(s) to use for xmltv generation.update_frequency
determines how often we check for new scheduling information. In Seconds.reverse_days
allows Blocks of EPG data to be created prior to the start of the EPG Source data.forward_days
allows Blocks of EPG data to be created after the end of the EPG Source data.block_size
in seconds, sets the default block size for data before, after and missing timeslots.xmltv_offset
allows the final xmltv file to have an offset for users with timezone issues.
[epg]
# images = pass
# method = None
# update_frequency = 43200
# reverse_days = 1
# forward_days = 7
# block_size = 1800
# xmltv_offset = +0000
ffmpeg¶
The ffmpeg
section includes:
path
. This is useful if ffmpeg is not in your systems PATH, or you want to manually specify.
[ffmpeg]
# path = ffmpeg
vlc¶
The vlc
section includes:
path
. This is useful if vlc is not in your systems PATH, or you want to manually specify.
[vlc]
# path = cvlc
Logging¶
level
determines the amount of logging you wish to see in the console, as well as to the logfile (stored in your cache directory).
[logging]
# level = WARNING
Database¶
- experiment with these settings at your own risk. We use sqlalchemy to provide database options, but we default to sqlite.
Note: this section needs improvement.
[database]
# type = sqlite
# driver = None
user = None
pass = None
host = None
port = None
name = None
RMG¶
# enabled = True
SSDP¶
# enabled = True
# max_age = 1800
# proto = ipv6
# iface = None
# multicast_address = None
Last update:
2021-02-04