library(calibrar)
library(doSNOW)

# tous les variables d’environnement sont définies dans le script PBS

setwd(Sys.getenv("PBS_O_WORKDIR"))

corresp <- data.table::fread("InputsMed/df_fleet_fdi_isis.csv", data.table = FALSE)
matchZoneGSA <- data.table::fread("InputsMed/MatchZonePopGSA.csv", data.table = FALSE)

###############################################################################

runModel <- function(par) { #"par" provided by the calibrate function

	# get back the capturability parameters (par) and write them in the format the model is able to read (csv, semantic)
	textLines <- readLines("q.csv")
	for (i in seq_along(par)) {
		id <- 2 + i
		tmp <- strsplit(textLines[id], ";")[[1]]
		textLines[id] <- paste0(tmp[1], ";", par[i])
	}
	write(textLines, file = "q.csv")

	# launch ISIS from R, cf le script PBS
	system2("myIsis", stdout = "debugISIS.log", stderr = "debugISIS.log")

	# read the model outputs and tranform them into a list ==========

	# catch at age
	export1 <- Sys.getenv("SIMUL_PATH") %>%
		file.path(., "resultExports/Captures_AgeStructure_pop_3pop.csv.gz")
		read.table(file = ., header = TRUE, sep = ";") %>%
		mutate(names = paste0("gp", group, "_", year + 2018))

	# catch per fleet
	export2 <- Sys.getenv("SIMUL_PATH") %>%
		file.path(., "resultExports/CatchWeightSpFltYear.csv.gz") %>%
		read.table(file = ., header = TRUE, sep = ";") %>%
		filter(valSim != 0)

	export2d <- export2 %>%
		left_join(corresp) %>%
		left_join(matchZoneGSA) %>%
		mutate(value_gsa = valSim * prop) %>%
		group_by(yearQuarter, country_code, sub_region, super_length, super_gear) %>%
		summarise(value = sum(value_gsa)) %>%
		ungroup() %>%
		mutate(names = paste0(
			yearQuarter, country_code, sub_region, super_length, super_gear,
			collapse = "_"
		))

	# combine =======================================================
	export <- rbind(
		export1 %>% select(names, value),
		export2d %>% select(names, value)
	)
	output <- as.list(export$value)
	names(output) <- export$names
	return(output)
}

###############################################################################

# residual sum of squares function
userFunction <- function(obs, sim) ((obs - sim) / obs) ^ 2

# calibrationInfo
calib_set <- calibration_setup(
	file = "data_calib/calibrationInfo.csv",
	control = list(col_skip = 1) # pour compatibilité avec ancienne version
)

# Observed data
calib_obs <- calibration_data(setup = calib_set, path = "data_calib")

# creation of the objective function
obj_fn <- calibration_objFn(
	model = runModel, setup = calib_set, observed = calib_obs,
	aggFn = calibrar:::.weighted.sum
)

###############################################################################

cl <- makeCluster()
registerDoSNOW(cl)

invisible(clusterEvalQ(cl, { # cacher ces messages gênants
	library(calibrar)
	library(dplyr, warn.conflicts = FALSE)
}))

clusterExport(cl, c(
	"corresp", "matchZoneGSA", "userFunction",
	"obj_fn", "calib_set", "calib_obs"
))

###############################################################################

# control argument of the calibrate function
control <- list(
	maxgen = 500,
	master = "master",
	run = paste0("RUN_", Sys.getenv("SIMUL_NAME")),
	restart.file = paste0("results_", Sys.getenv("SIMUL_NAME")),
	REPORT = 1,
	parallel = TRUE,
	nCores = as.integer(Sys.getenv("NCPUS")),
	popsize = 50,
	trace = 3,
	convergence = 0.001
)
parinit <- c(1e-6, rep(1e-5, 5))

calib <- calibrate(
	par = parinit, fn = obj_fn, method = "default",
	lower = 0, upper = 1e-3, control = control
)

###############################################################################

stopCluster(cl)