I have been experimenting with this exact idea. For now I am running
wee_reports. I wanted it to run the reports soon after an archive record,
so it is triggered by receiving a MQTT message from the computer collecting
the data.
But, based on Tom’s comments, I think it would be fun to convert it into a
driver.
Rich


On Thu, Mar 14, 2019 at 5:42 PM Thomas Keffer <[email protected]> wrote:

> Running WeeWX without generating reports is simple enough: just remove the
> service weewx.engine.StdReport from the list of services to be run (option
> report_services).
>
> Having the reports done by a separate program running on a VM is a little
> more complicated.
>
> The simplest option is to use wee_reports to do the report generation. You
> could either have a crontab kick it off, or use some other signaling
> mechanism to start wee_reports. This is completely satisfactory, except for
> one detail: wee_reports would not have access to any new types introduced
> by any extensions you might have. If you don't have any extensions that add
> new types, then it's not a problem.
>
> Another more ambitious option is to write a dirt simple driver whose job
> it is to watch the database, and notice new, incoming records. It would
> then retrieve the record and emit a NEW_ARCHIVE_RECORD event. Because this
> driver would not be emitting the normal LOOP packets, this would be an
> experiment and would require understanding how the engine works. It would
> be a "good project for the interested student," but you would be on your
> own figuring it out.
>
> Those are the choices that come off the top of my head.
>
> -tk
>
> On Thu, Mar 14, 2019 at 2:07 PM Your Name Here <[email protected]>
> wrote:
>
>> I want to have my Raspberry Pi 3+ no longer make the reports (although,
>> they could be piped to /dev/null if that isn't an option). Unless there are
>> specific driver updates, I can just keep it running without dealing with
>> upgrades.
>>
>> I want a VM to generate the web pages instead.
>>
>> This would allow me to try out various VM setups and update/upgrades
>> independently of the data gathering installation of weewx on the Pi.
>>
>> This is not the same as using the simulator because that uses the
>> simulated data for web page generation.
>>
>> The VM wouldn't have a driver, per se, and it doesn't need to modify the
>> database (I use MySQL in another VM). I don't want it modifying the
>> database (especially if I mess up and the simulator is running—I don't want
>> that mixed in with the good data).
>>
>> I have tried a few combinations that I thought would work (I commented
>> out process and archive services; I tried simulator and used a "grant
>> select" only user against the database). I haven't figured it out yet.
>>
>> How can I make this work?
>>
>> Thanks.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "weewx-user" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> For more options, visit https://groups.google.com/d/optout.
>>
> --
> You received this message because you are subscribed to the Google Groups
> "weewx-user" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to