It is confusing, but the reason it's a "pull" (and not a push) request is because you're requesting the project manager to "pull" your changes into the main project.
To create a pull request, you have to first fork the repo, then make your changes within your fork. Once done and tested, go to the main project and request a new pull request. On Tuesday, August 23, 2016 at 11:00:49 AM UTC-4, Andrew Milner wrote: > > A major problem with that Tom - I do not quite understand this concept of > 'pull request' ie what exactly it is that I have to do!! Is there a simple > noddy guide to pulling weewx anywhere??? Always seems to me that it should > be 'push' request - ie I have something that I want to give but that could > be why I'm confuzzed!! > > > > On Tuesday, 23 August 2016 17:29:53 UTC+3, Tom Keffer wrote: > >> As always, a pull request is welcome. >> >> -tk >> >> On Mon, Aug 22, 2016 at 7:11 PM, Andrew Milner <[email protected]> >> wrote: >> >>> Just seen this thread, and this is something that somewhat annoyed me as >>> well, since all the extra digits cluttered up the database viewing. I >>> 'solved' (got around) it by specifying the precision in the column >>> definitions since MySQL allows this for float, real and double precision. >>> Couldn't the schema just be altered to specify the precision for all fields >>> and make weewx more human friendly? Most columns only require at most 2 >>> decimals, and many 0 or 1 for practical purposes, and weewx remains >>> oblivious to the column definition change. I changed the schema in mesowx >>> with no issues so it should work for weewx. >>> >>> >>> On Friday, 12 February 2016 15:11:38 UTC+2, Pat O'Brien wrote: >>> >>>> Thanks for the explanation. It all makes sense, especially laid out >>>> like that. Many areas in play with the observations, so I'm glad to see >>>> I'm >>>> doing it right! >>>> >>>> >>>> On Friday, February 12, 2016 at 8:03:53 AM UTC-5, mwall wrote: >>>>> >>>>> On Thursday, February 11, 2016 at 7:28:24 PM UTC-5, Pat O'Brien wrote: >>>>> >>>>> I was having a look in the database today, and noticed that a lot of >>>>>> the archive records and the archive_day_* tables have observations with >>>>>> 12 >>>>>> decimals. Seems to be a bit long. >>>>>> >>>>>> So I'm just checking to see if my driver should control the float >>>>>> length (maybe _packet['windSpeed'] = '{0:.2f}'.format( data[ >>>>>> "windSpeed"] ) this is untested) or is this supposed to be handled >>>>>> within weewx somewhere and I have it configured incorrectly? >>>>>> >>>>> >>>>> hi pat, >>>>> >>>>> you are doing it right - using float() is the appropriate way to >>>>> convert from string. >>>>> >>>>> precision (number of decimal places) shows up in a few places, >>>>> including: >>>>> >>>>> 1) how the database stores a REAL number. nothing to do here. it is >>>>> what it is. >>>>> >>>>> 2) how your database browser displays a REAL number. some database >>>>> browsing tools let you specify how many decimal places you want to see, >>>>> some even on a per-column basis. this does not affect the data - just >>>>> what >>>>> you see. >>>>> >>>>> 3) how weewx (in python) represents a float. nothing to do here. it >>>>> is what it is. >>>>> >>>>> 4) how weewx displays a float. this is an issue when writing a weewx >>>>> template. use the cheetah/python formatting to display the number of >>>>> decimal places you want to see. >>>>> >>>>> 5) how weewx uploads a float to wunderground, etc. this is only an >>>>> issue if you are writing an uploader - an extension that uploads data. >>>>> values are typically converted to a string then sent using http POST or >>>>> similar mechanism. use python formatting in your uploader to specify the >>>>> number of decimal places that the destination requires. >>>>> >>>>> m >>>>> >>>> >>
