I noticed few issues with 2022.12 update Scrape moved to integration.
Sadly, it did not offer to move existing scrape sensors configuration so I have to do it by hand.
As I have number of scraping sensors it is tedious to do them all using web interface. Is there yaml for scrape integration I could edit. Editing yaml would be much easier? I did try to search through config directories but did not found it.
It seems documentation on Scrape is not updated. It still refers to adding scrape sensors in configuration.yaml.
Strangely, when I added Scrape resource in integration and added sensor, it did not create Scrape device. It created only entities for each sensor. Then I added another resource in integration and sensor and now configuration for all is stiffed in Integration configuration. With few more several scrape sources it would become quite crowded place.
One would expect that each resource would be added as separate device and configured as such.
Then, I noticed that I misspelled name of a sensor in resource. Automatically created id is, naturally, also misspelled. There is no option to fix name or id. It seems I have to delete sensor and add it again from scratch (did not try it, so maybe that would not work).
I can go to Entities and change name and id there but name and id in integration stays misspelled. I would like to be able to fix that in integration. Is there a way?
Note for others that are yet to do this. Comment scrape sensor configuration in configuration.yaml first. Otherwise, names would be occupied so new sensors created in integration if named the same would get suffixes.
In old scrape there was option scan_interval to set intervals for values to be scraped. In Scrape integration it seems there is no such option. How can we control frequency of scraping events in Scrape integration?
In the GUI you can de-select the 10 minute auto-scan and then you update via an automation at your selected time of day / interval / weekday etc. by using the homeassistant.update_entity service.
That is actually a much better way to scrape certain things, since it does not keep running (like the scan_intervall) all day/night (or weekends) when you don’t need it - therefore putting much less stress on the scraped website - and reducing the chances of getting yourself banned
To be honest, I did not consider this before, but now - even when using the yaml solution - I would put a scan_interval of 24 hours (because I believe you can not disable it totally) and update my scrapes using an automation only when it really makes sense.