If you wish to have weeWX store a new observation in the archive there are two key steps you need to take. First you need to get the observation 'into' weewx and second you need to get weewx to store the value in the archive. Getting the observation 'into' weeWX typically involves some sort of driver adding the observation to loop packets or archive records. From what I can understand of your posts so far, you appear to be using a modified version of the fileparse driver to do this. That is what Matthew meant by "you already have the observations".You can see this by running weeWX directly, you shoudl see you abs appearing in loop packets and/or archive records.
The second step, having weeWX save the observation, requires some action with a database. WeeWX stores archive data whenever an archive record is produced. Once any conversions, corrections etc have been done weeWX looks at each observation in the archive record and each field in the in-use schema and stores archive record observations that have a matching database field in the database; so if there is an observation 'outTemp' in the archive record and there is a field 'outTemp' in the archive table schema, the 'outTemp' value from the archive record will be saved in the 'outTemp' field in the archive table, and so on. If there is an observation called, say, 'SOC' in the archive record but there is no 'SOC' field in the in-use schema then the 'SOC' value in the archive record is effectively discarded, it will not be saved to archive. So one of the approaches to storing additional observations in the archive is to 'extend the default schema' to include the new observation. The beauty of this approach is that it is fairly simple and does not really involve writing any code to manipulate the database, you put a few lines of code in a file extend the schema, a change of setting in weewx.conf and then a bit of database shuffling with wee_database. You now have a database whose archive table includes fields for the new observation(s). WeewWX takes care of getting data into the new field(s) as part of its normal operation. This is what is meant by Matthew when he said 'just extend the schema'. The section of the Customization Guide you need is the section Matthew provided as a link in response to your original post. You will defintely need to follow steps 1 and 2, changing files/obs/fields to suit your system. Step 3 may not be required, especially if using SQLite, but steps 4 onwards will be required. Gary On Sunday, 29 January 2017 12:56:22 UTC+10, Robert Mantel wrote: > > Ok, so there is no soc.py because I thought that extending the schema > meant just adding the schema extension to the filelparse.py and then > reconfiguring the database. I didn't understand from mwall's comments what > exactly was meant by "just extend the schema you already have the > observations"...I can gather the soc value from fileparse.py so I'm not > sure if I need an soc.py? I guess what I'm asking is do I need a new > service for just gathering the soc? If someone can clarify what section of > the customization guide I need to do I'll go from there. I can pretty much > gather anything from my Particle P1 (basically and arduino with wifi) front > end and feed it into fileparse, with the right sensors attached. > > On Friday, 27 January 2017 17:46:08 UTC-5, gjr80 wrote: >> >> Hi, >> >> Think we will need to see the contents of weewx.conf, the contents of >> soc.py and what is the location of soc.py. >> >> Gary >> >> -- 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.
