Publishing Large Feature Services

[JosiahParry]

C stack usage error encountered here. Probably from JSON conversion. We need a more memory efficient way to handle this. For massive services, we shuold probablytake the first n and publish. Then use the rest to add.

res <- arc_read("https://services.arcgis.com/P3ePLMYs2RVChkJx/arcgis/rest/services/USA_ZIP_Code_Points_analysis/FeatureServer/0", n_max = 1e10)

n <- nrow(res)
r <- 1000

to_pub <- res[1:(n-r),]
to_save <- res[((n-r)+1):n,]

set_arc_token(auth_user())

# There's a C stack limit we're reaching here when publishing
# probably from it being toooo big 
publish_layer(to_pub, "USA Zip Code Points")