Hello All,
I am from a small north east town in Victoria. On the bom website the forecast my product_id is made up from product derived from IDV10753 and IDV17003. How do i set this up as i single product_id? I have tried entering one or the other and it fails on both. The sensor works fine if i say put it IDV10450 (Melbourne).
Thanks
It won’t support that unfortunately.
Thanks anyway, I select a town that’s close by for now.
Yeah that’s the best approach as I suggested up above to someone else.
@DavidFW1960 and @BrendanMoran Thank you guys for the Bom forecast integration and the BOM weather card. I live in the sydney region and use the Prod ID of IDN10064 for the forecast however I live in the north western suburbs.
Upon looking at the IDN10064.xml the forecast for certain cities are also available within this xml.
E.g. The location of Penrith within the file provides the forecast for 7 days. While the remaining locations Liverpool, Parramatta etc. provide 1 day which will be the most accurate for any suburbs closest to those locations.
Is there a way to provide a mechanism to extract this data rather than using the Syd location forecast which is rather inaccurate for the different locations within Syd?
I will have a look at this when I am back home next week. Busy with family for a few days.
ok so yeah… seen this before. Liverpool looks like this:
It uses a combo of 2 IDNs for it’s data.
I have not looked at supporting that and it’s really above my hacking abilities at this point to do so.
I’d suggest picking the best station with a single IDN number for your location. Sorry.
Let me try to explain a bit more…
For Syd the file / IDN number which the forecast gets extracted from is IDN10064.xml which looks like
Data is extracted from the area type = “metropolitan” and “location” nodes where the description is “Sydney” which is essentially the weather for the city.
However Penrith also provides a 6 day forecast, but there is no way to access this info.
The rest of the nodes however only provides today’s forecast.
If there is a way to pass this secondary location (e.g. Penrith which can override the default which is description=“Sydney”) and extract the corresponding info one might get more accurate info if they are closer to that location. Obviously if there is no source data it would be good to default to the corresponding node where description =“Sydney”.
My Python skills are almost none existent hence thought I would reach out to you and see if you would be able to do this.
No need to explain I understand the problem.
It uses 2 different data sets with 2 IDN’s. I only use one IDN in the FTP component amd there is no easy way for me to allow 2 IDN’s.
Yes of course it’s possible but way above my Python skills as well - I’m just a hacker and hacked in a few minor changes from Brendans original.
Hey guys, seen what you have done here and it is excellent - i will look at integrating it into my system as it evolves and as i move more from my Node-red dashboard to HA.
Can i suggest it might be an idea to use Node-red as a middleman for this stuff ? It has excellent easy to use capabilties to enable extracting anything you want and to put it back out into a set format - so i was thinking as an evolution of this project - what about defining a middle layer/format that your system would accept for putting into HA. NR could then do the heavy lifting of bringing down the data and manipulating it into that format ?
regards
Craig
Why?
I don’t run NR, don’t intend to, don’t see any benefit over what is being done here.
My thought was that as you have said above you are only dabbling with python, and due to how the BOM works with the “stacked” forecasts for some regions that NR could be used to provide an easy to use graphical layer in the middle to enable people to massage the data into the format that you could set and expect each time, you could then spend your time managing the “front end” and middle layer and people could use NR (or any other tool) to put the data into the format that you are expecting.
Craig
Feel free to do that if you want to but I’m not touching NR.
OK fair enough - i will have to do a lot more playing before i understand how all this hangs together
Craig
I have tried to get the forecast working but have been getting nothing.
I managed to get the basic weather information using this guide.
I’m pretty new the HA so any help is appreciated.
Which product_id: did you use for the bom forecast platform?
Also, please post your forecast config and any errors in the log.
That guide is somewhat outdated for forecast. Best bet is to use https://github.com/DavidFW1960/bom_forecast which is better supported by the community
I’ve redone it using this guide, I’m getting errors on the uv_index and fire_danger
I’ve put these values in
my_bom_station: IDV60901.95867
my_bom_name: Melbourne
my_bom_product_id: IDV10450
Station from Vic/Scoresby observations
Product from Vic/Melbourne forecast
You need to post the yaml config so I can have a look. (The whole config)
Here’s my configuration.yaml
# Configure a default setup of Home Assistant (frontend, api, etc)
default_config:
homeassistant:
customize: !include customize.yaml
packages: !include_dir_named packages
frontend:
themes: !include_dir_merge_named themes
# Text to speech
tts:
- platform: google_translate
media_player:
- platform: lg_netcast
host: 192.168.1.116
access_token: 859067
name: Kitchen Remote
turn_on_action:
service: switch.turn_on
data:
entity_id: switch.tv_switch
camera:
- platform: bom
location: Melbourne
#https://github.com/bremor/bom_forecast
group: !include groups.yaml
automation: !include automations.yaml
script: !include scripts.yaml
scene: !include scenes.yaml
Secrets.Yaml
my_bom_station: IDV60901.95867
my_bom_name: Melbourne
my_bom_product_id: IDV10450
I have the stock weather.Yaml file in the packages folder.