Pull in fwa_upstreamtrace

[joethorley]

https://features.hillcrestgeo.ca/fwa/functions/postgisftw.fwa_upstreamtrace/items.html?start_blue_line_key=354153927&start_measure=10000

may just need wrapping in fwatlasbc

[joethorley]

Seems fine so far but there may be edge cases that I haven't thought of - supply blue line key and measure and all stream upstream is returned, cutting the stream at the provided point. If the provided measure is within the tolerance to the endpoint, the source geom is returned (or nothing, if at the endpoint). (edited) 


Very cool - is it ready to pull into the R packages?


I think so. I haven't added it to the repo yet but it is live on the server and good to go.
There is no functionality to provide a watershed code like you were originally asking for in the GH issue, let me know if that is needed.


blk is better I think


there is just one nuance between the results and the source data in the streams table - any record with null local code will not be returned (because we can't tell where it is positioned)
this is a data error, apparently those are being cleaned up but it is very slow because there are so many
And of course unconnected segments won't come through either.