config.R

library(nhdplusTools)
library(dplyr)
library(sf)
library(hyRefactor)
library(tidyr)
library(hyfabric)

if(nchar(Sys.getenv("RPU_CODE")) == 3) {
  rpu_code <- Sys.getenv("RPU_CODE")
}

if(exists("rpu_code") && !exists("VPU")) {
  VPU <- substr(rpu_code, 1, 2)
  
  if(rpu_code %in% c("10a", "10b", "10c", "10d")) {
    VPU <- "10L"
  }
  
  if(rpu_code %in% c("10e", "10f", "10g", "10h", "10i")) {
    VPU <- "10U"
  }
  
  if(rpu_code == "03g") {
    VPU = "08"
  }
}

# NHDPlus V2 Region
if(!exists("VPU") && !nchar(VPU <- Sys.getenv("HYDREG")) > 1) 
  VPU <- "01"

if(!exists("rpu_code") && !nchar(rpu_code <- Sys.getenv("RPU_CODE")) == 3) 
  rpu_code <- "01a"

if(!nchar(out_gpkg <- Sys.getenv("OUT_GPKG")) > 5) 
  out_gpkg <- file.path("cache", paste0(rpu_code, ".gpkg"))

if(!nchar(cache_split <- Sys.getenv("CACHE_SPLIT")) > 5) 
  cache_split <- file.path("temp", paste0(rpu_code, ".rda"))

if(!nchar(lookup_table_file <- Sys.getenv("LOOKUP_TABLE")) > 5) 
  lookup_table_file <- file.path("cache", paste0(rpu_code, "_lookup.csv"))

options(scipen = 9999)

# Source Data
crs <- 5070
data_paths <- jsonlite::read_json(file.path("cache", "data_paths.json"))
huc12_xWalk <- file.path(data_paths$nhdplus_dir, "CrosswalkTable_NHDplus_HU12.csv")
HUC12 <- file.path(data_paths$nhdplus_dir, "HUC12.rds")

# Defined at various points
raw_wbd <- paste0("HU12_", VPU) # Raw HUC12
gage_loc <- paste0("GageLoc_", VPU) # HI GageLoc
TE <- paste0("TE_", VPU) # Thermo-electric 
NID <- paste0("NID_", VPU) # Raw NID points
gage_events <- paste0("gage_events_", VPU) # gages with flowline index information for possible splitting

# Defined during NHD_Navigate.Rmd
nav_gpkg <- file.path("cache", paste0("GF_", VPU,".gpkg"))
nhd_flowline <- "nhd_flowline"
nhd_catchment <- "nhd_catchment"
xwalk_layer <- paste0("HUC12_nhd_", VPU) # HUC12 - nhdcat crosswalk, built in Nav for VPU 20
nav_poi_layer <- paste0("pois_", VPU) # Rolling Nav POI layer added to/modified througout nav workflow
WBs_layer <-  paste0("WB_", VPU) # Waterbodies within VPU
poi_moved <- paste0("POIs_mv_", VPU) # POIs moved from original COMID assignment
inc_pois <- paste0("inc_POIs_", VPU) # layer of POIs used to derive first cut of segments
nsegments_layer <- paste0("nsegment_", VPU) # Minimally-sufficient network dissolved by POI_ID
pois_all <- paste0("POIs_", VPU) # All POIs binded together
poi_xWalk <- paste0("poi_xWalk_", VPU) # POIs that changed COMIDS during the navigate part of the workflow
final_poi_layer <- paste0("final_POIS_", VPU)
pois_merge <- paste0("merPOIs_", VPU) # All POIs binded together

# Defined during Refactor
rpu <- paste0("rpu_", rpu_code)
split_meters <- 10000
combine_meters <- 1000
min_da_km <- 20
nhd_outlet <- "nhd_outlet"
nhd_catchment <- "nhd_catchment"
full_catchment_layer <- "full_catchment"
refactored_layer <- "collapsed"
reconciled_layer <- "reconciled"
gf_outlets_layer <- "gf_outlets"
divide_layer <- "divides"
refactor_table <- "refactor_lookup"
agg_fline_layer <- "agg_fline"
agg_cats_layer <- "agg_cats"
outlets_layer <- "outlets"
lookup_VPU <- paste0(VPU, "_lookup")

# Defined during Nondend
ND_gpkg <- file.path("cache", paste0("NHD_", VPU,".gpkg"))


# Making an "events" script to address "events" information
#     i.e. gages, NID, TE Plants, waterbodies
gage_gpkg <- "cache/Gage_info.gpkg"
gage_cand <- "Potential_Gages"
VPU_gage_Table <- file.path("cache", paste0("R", VPU, "_Gages.csv"))
CONUS_gage_Table <- "data/gages_MDA.rds"

# Defined during NonDend.Rmd
fin_gpkg <- file.path("cache", paste0(VPU, "_final.gpkg"))
lookup_miss <- paste0("lookup_missing_", VPU)
missing_cats <- paste0("miss_cats_", VPU)
protoHRUs <- paste0("poi_cats_", VPU)
cat_boundary <- paste0("boundaries_", VPU)

gf_gpkg <- nav_gpkg
poi_layer <- paste0("POIs_", VPU)



# parallel factor for catchment reconciliation.
para_reconcile <- 2
para_split_flines <- 2

data_paths <- jsonlite::read_json(file.path("cache", "data_paths.json"))
nhdplus_path(data_paths$nhdplus_gdb)
staged_nhd <- stage_national_data(nhdplus_data = data_paths$nhdplus_gdb, 
                                  output_path = data_paths$nhdplus_dir)

# Output Layers

nhd_outlet <- "nhd_outlet"
nhd_catchment <- "nhd_catchment"
full_catchment <- "full_catchment"
refactored_layer <- "collapsed"
reconciled_layer <- "reconciled"
gf_outlets_layer <- "gf_outlets"
divide_layer <- "divides"
refactor_table <- "refactor_lookup"
agg_fline_layer <- "agg_fline"
agg_cats_layer <- "agg_cats"
outlets_layer <- "outlets"
lookup_VPU <- paste0(VPU, "_lookup")

# Manually verified
crs_out <- st_crs('PROJCS["NAD83 / Conus Albers",GEOGCS["NAD83",DATUM["North American Datum 1983",SPHEROID["GRS 1980",6378137.0,298.257222101,AUTHORITY["EPSG","7019"]],TOWGS84[1.0,1.0,-1.0,0.0,0.0,0.0,0.0],AUTHORITY["EPSG","6269"]],PRIMEM["Greenwich",0.0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.017453292519943295],AXIS["Geodetic longitude",EAST],AXIS["Geodetic latitude",NORTH],AUTHORITY["EPSG","4269"]],PROJECTION["Albers Equal Area",AUTHORITY["EPSG","9822"]],PARAMETER["central_meridian",-96.0],PARAMETER["latitude_of_origin",23.0],PARAMETER["standard_parallel_1",29.5],PARAMETER["false_easting",0.0],PARAMETER["false_northing",0.0],PARAMETER["standard_parallel_2",45.5],UNIT["m",1.0],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","5070"]]')