sf 0.9 submission to CRAN

[edzer]

See here for revdep checks. I went through all of them, and will resubmit a version that addresses point 2.

1. Warning: replacing previous import ‘sf::st_make_valid’ by ‘lwgeom::st_make_valid’ when loading ‘bdl’

   st_make_valid is being moved from lwgeom to sf; packages importing
   both will now generate this warning on startup. This will go away
   automatically when I can submit lwgeom after sf, with lwgeom no
   longer exporting this method.

   Packages affected: bdl, EmissV, RGISTools, tmap

2. problems now solved in sf;

   • st_sample has become a generic, but didn't export a sfg method
   • st_make_valid strips crs

   Packages affected: cartography, dssd, eSDM, GADMTools, transformr

3. have been informed:

• compstatr too strict crs test slu-openGIS/compstatr#19
• eixport: caused by silicate; reported crs structure is going to change hypertidy/silicate#104
• exactextractr: submitted sf will break exactextractr on CRAN isciences/exactextractr#24
• geobr: upcoming sf will break geobr on CRAN ipeaGIT/geobr#129
• geosample: by email
• geouy: upcoming sf will break geouy RichDeto/geouy#3
• jpndistrict: upcoming sf breaks jpndistrict on CRAN uribo/jpndistrict#30
• hereR: upcoming sf breaks hereR on CRAN munterfi/hereR#36
• macleish: upcoming sf breaks macleish on CRAN beanumber/macleish#26
• mapview: upcoming sf breaks mapview on CRAN mapview#271
• nhdR: upcoming sf breaks nhdR on CRAN jsta/nhdR#62
• fasterize: Upcoming sf breaks fasterize ecohealthalliance/fasterize#31
• oceanis: upcoming sf breaks oceanis on CRAN insee-psar-at/oceanis-package#2
• raceland: upcoming sf breaks raceland on CRAN Nowosad/raceland#5
• rFIA: email to [email protected]
• RFmerge: upcoming sf breaks RFmerge on CRAN hzambran/RFmerge#1
• rmapshaper: upcoming sf breaks rmapshaper on CRAN ateucher/rmapshaper#103
• sen2r: https://github.com/ranghetti/sen2r/issues/260
• slga: upcoming sf will break slga obrl-soil/slga#1
• spsurvey: email to [email protected]
• stars: author is me
• tanaka: vulnerability to PROJ6/GDAL3 riatelab/tanaka#1
• tmaptools: upcoming sf breaks tmaptools r-tmap/tmaptools#20

4. False positives:

• cdcfluview: curl error, happens at random places, also with current CRAN version of sf;
• gtfs2gps: network error
• NLMR: caused by fasterize, reported there
• rasterDT: caused by fasterize, reported there
• reproducible: caused by fasterize, reported there

[edzer]

Reverse dependency checks of second submission: https://win-builder.r-project.org/incoming_pretest/sf_0.9-0_20200321_121257/reverseDependencies/summary.txt

[edzer]

(As expected: cartography, dssd, eSDM, GADMTools, transformr are now resolved)

[rafapereirabr]

Hi @edzer . Do you have a planed date to submit the new version of sf to CRAN?

[edzer]

Yes, it is in the submission process, and could appear any time soon now.

[edzer]

Third submission: on PROJ 7.0.0 tests would fail; 5187565 removes the unload_gdal() / load_gdal() test; potential issue raised here: OSGeo/gdal#2344