Robert Coup

@SpaceIm can you explain what the **specific** issue(s) it is causing for Conan? If it's a case of adding geosop back then I'm fine with it, but are there other...

> since it was the only issue with add_subdirectory(). All other logic based on PROJECT_IS_TOP_LEVEL can be removed I think. No, CMAKE_SOURCE_DIR/CMAKE_BINARY_DIR are real problems when using GEOS as a...

Deployments have had a rolling restart accessible for a while now, but statefulsets & daemonsets still don't.

@robe2 thanks! Can easily make those changes. > Your tests also don't seem to guard against running on PROJ < 6.1, but that might be okay. Is this the [best...

> Is the "forward" parameter really needed ? > I mean: can a pipeline not always be expressed differently, > to be backward ? I'm thinking of a function that,...

> > But if I get my wish -- all of this will be unnecessary, cause we'll drop support for Proj < 6.1 so no ifdef will be needed -...

> could in theory be used to implement a ST_ReverseTransformPipeline() Feel to me like `ST_TransformPipeline()` is already extra-for-experts territory: PostGIS has got a _long_ way with `ST_Transform()`, and the people...

@robe2 @strk I've made those changes now. I went with adding `ST_InverseTransformPipeline()` instead of the `forward` parameter. Feedback welcome 😃

ok, so `make check` seems to work locally (and exits with 0). But looking closer: ``` Loading PostGIS into 'postgis_reg-3.4' failed (Error encountered loading /Users/rcoup/code/postgis/regress/00-regress-install/share/contrib/postgis/postgis_comments.sql: /tmp/pgis_reg/regress_log) ``` ``` $ cat...

@strk thanks. I re-added a similar error locally & `make check` fails (returns 2) now. 👍