different CRS output depending on downstream_route_measure value

[sebdalgarno]

The output of the following

• downstream_route_measure = 100
• blue_line_key = 356308001

https://features.hillcrestgeo.ca/fwa/functions/postgisftw.fwa_locatealong/items.json?blue_line_key=356308001&downstream_route_measure=100

is in lon/lat (although EPSG isn't in the json as below)

{"type":"FeatureCollection","features":[{"type":"Feature","geometry":{"type":"Point","coordinates":[-122.014985692,49.167195259,9.797745916]},"properties":{}}],"numberReturned":1,"timeStamp":"2022-06-30T16:03:54-07:00","links":[{"href":"https://features.hillcrestgeo.ca/fwa/collections/postgisftw.fwa_locatealong/items","rel":"self","type":"application/json","title":"This document as JSON"},{"href":"https://features.hillcrestgeo.ca/fwa/collections/postgisftw.fwa_locatealong/items.html","rel":"alternate","type":"text/html","title":"This document as HTML"}]}

Whereas with different downstream_route_measure:

• downstream_route_measure = 10000
• blue_line_key = 356308001

https://features.hillcrestgeo.ca/fwa/functions/postgisftw.fwa_locatealong/items.html?blue_line_key=356308001&downstream_route_measure=10000

is bc albers EPSG 3005

{"type":"FeatureCollection","features":[{"type":"Feature","geometry":{"type":"Point","crs":{"type":"name","properties":{"name":"EPSG:3005"}},"coordinates":[1295576.091236034,467912.302100455,10.5]},"properties":{}}],"numberReturned":1,"timeStamp":"2022-06-16T17:01:17-07:00","links":[{"href":"https://features.hillcrestgeo.ca/fwa/collections/postgisftw.fwa_locatealong/items","rel":"self","type":"application/json","title":"This document as JSON"},{"href":"https://features.hillcrestgeo.ca/fwa/collections/postgisftw.fwa_locatealong/items.html","rel":"alternate","type":"text/html","title":"This document as HTML"}]}

@smnorris any idea what is going on here?

[smnorris]

The request is hitting the cache - this one looks fine. https://features.hillcrestgeo.ca/fwa/functions/postgisftw.fwa_locatealong/items.html?blue_line_key=356308001&downstream_route_measure=10000&nocache=true

I'll clear out the cache!

[smnorris]

Cache is clear, hopefully all works fine now.

[sebdalgarno]

awesome thanks! all good now. Does that clear the cache for all previous function/collection results? (i.e. not just fwa_locatealong)

[smnorris]

Yes, it is a general cache for every request to the web server. I'm not sure how useful it is for featureserv but the cache is really nice when browsing maps created with tileserv.

[sebdalgarno]

ok great - clearing it should help troubleshoot a couple other remaining issues on our end. Yes I can definitely see how useful caching can be - especially for large requests that are done a lot. I can always request that you clear it if we are running into issues (i.e. after API changes as just occurred).

[smnorris]

Note that I haven't yet upgraded the db. The switchover should be seamless but I'll be sure to give you a heads up and clear the cache when I do make the switch.

Not sure if performance will be noticeably better but for some functions there is potential for good speedups.

[sebdalgarno]

OK great thanks. Also sorry I'm just catching up on the previous discussion between you and Joe on setting nocache=TRUE. I've added another issue to modify our tests so that they aren't cached. That will limit the need to request that you clear the cache in future.