update authors

Author: AAoritz

DESCRIPTION

@@ -10,11 +10,11 @@ Authors@R: c(
            comment = "Josep reviewed the package for rOpenSci, see https://github.com/ropensci/software-review/issues/623#issuecomment-1951446662"),
     person("Le Meur", "Nolwenn", role = "rev",
            comment = "Nolwenn reviewed the package for rOpenSci, see https://github.com/ropensci/software-review/issues/623#issuecomment-1961501137"),
-    person("Filipe", "Batista e Silva", role = "ctb",
+    person("Filipe", "Batista e Silva", role = "aut",
            comment = c(ORCID = "0000-0002-8752-6464")),
-    person("Carmelo", "Attardo", role = "ctb"),
-    person("Cristian", "Pigaiani", role = "ctb"),
-    person("Michele", "Tucci", role = "ctb")
+    person("Carmelo", "Attardo", role = "aut"),
+    person("Cristian", "Pigaiani", role = "aut"),
+    person("Michele", "Tucci", role = "aut")
   )
 Description: Motivated by changing administrative boundaries over time,
     the 'nuts' package can convert European regional data with NUTS codes

data-raw/.RData

@@ -0,0 +1,73 @@
+*net install github, from("https://haghish.github.io/github/")
+*github install haghish/rcall
+
+
+// Run R code directly
+rcall: print("Hello from R")
+
+// Import data
+rcall: library(nuts)
+rcall: data(manure)
+rcall: manure
+rcall: summary(manure)
+
+rcall: write_dta(manure, "manure.dta")
+
+rcall: library(dplyr)
+rcall: library(haven)
+rcall: setwd("/home/moritz/dev/nuts/data-raw/")
+
+
+
+
+library(nuts)
+library(dplyr)
+library(haven)
+setwd("/home/moritz/dev/nuts/data-raw/");
+data(manure)
+
+
+
+
+cd /home/moritz/dev/nuts/data-raw/
+sysuse auto, clear
+describe, full
+save myauto.dta, replace
+rsource, terminator(END_OF_R)
+library(haven);
+setwd("/home/moritz/dev/nuts/data-raw/");
+rauto<-read_dta("myauto.dta", convert.f=TRUE);
+rauto;
+attributes(rauto);
+q();
+END_OF_R
+
+
+
+# Load packages
+library(nuts)
+library(dplyr)
+library(stringr)
+
+# Loading and subsetting Eurostat data
+data(patents, package = "nuts")
+
+pat_n2 <- patents %>% filter(nchar(geo) == 4) # NUTS-2 values
+
+pat_n2_mhab_12_no <- pat_n2 %>%
+  filter(unit == "P_MHAB") %>% # Patents per one million inhabitants
+  filter(time == 2012) %>% # 2012
+  filter(str_detect(geo, "^NO")) %>%  # Norway
+  dplyr::select(-unit)
+
+# Classifying the Data
+pat_classified <- nuts_classify(
+  data = pat_n2_mhab_12_no,
+  nuts_code = "geo"
+  )
+  
+  
+pat_classified <- nuts_classify(
+  data = pat_n2,
+  nuts_code = "geo"
+  )

data-raw/run_ropensci_nuts_in.do

@@ -0,0 +1,115 @@
+# NUTS R PACKAGE IN PYTHON
+#============================
+# Use the rOpenSci R package nuts for the conversion of European NUTS codes in Python with the help of this gist.
+# More information on the package can be found here: https://docs.ropensci.org/nuts/
+#
+# [Remark] It may be easier to run this in a conda environment created with:  
+# conda create -n r-env r-base r-essentials
+# conda activate r-env
+# pip install rpy2
+
+# Import modules
+import rpy2.robjects as ro
+from rpy2.robjects.packages import importr
+from rpy2.robjects import pandas2ri
+
+# Load nuts package
+nuts = importr('nuts')
+
+# Load required packages 
+ro.r('''
+library(nuts)
+library(dplyr)
+''')
+
+# Load some Eurostat data that ships with the nuts package
+ro.r('''
+data(manure)
+''')
+manure_df = ro.r('manure')
+
+# Convert to pandas 
+with ro.conversion.localconverter(ro.default_converter + pandas2ri.converter):
+    manure_df = ro.conversion.rpy2py(manure_df)
+
+# Let's only look at NUTS 3 codes (when number of digits in NUTS codes is 5)
+manure_df = manure_df[manure_df['geo'].str.len() == 5]
+
+# The indicator I07A_EQ_Y measures the number of manure storage holdings
+manure_df = manure_df[manure_df['indic_ag'] == 'I07A_EQ_Y']
+manure_df = manure_df.drop(['indic_ag'], axis=1)
+
+# Focus on the year 2010
+manure_df = manure_df[manure_df['time'] == 2010]
+print(manure_df)
+
+
+# 1. DETECT NUTS VERSIONS
+#--------------------------
+# Let's explore how NUTS versions can be found in this Eurostat dataset
+
+# Pass your pandas data frame to R for the conversion
+with ro.conversion.localconverter(ro.default_converter + pandas2ri.converter):
+    ro.globalenv['manure_df'] = manure_df
+
+# Convert manure holdings in dataset to level of NUTS codes 2021 version
+results = ro.r('''
+nuts_classify(nuts_code = "geo", data = manure_df) 
+''')
+
+# Convert result back to pandas 
+with ro.conversion.localconverter(ro.default_converter + pandas2ri.converter):
+    manure_classified_df = ro.conversion.rpy2py(results)
+
+# 1. This is the original data set augmented with three variables: 
+# from_version = detected NUTS version(s)
+# from_level = detected NUTS level
+# country = detected country
+print(manure_classified_df[0])
+
+# 2. Breakdown of the overlap with each NUTS version
+print(manure_classified_df[1])
+
+# 3. Missing NUTS codes 
+print(manure_classified_df[2])
+
+
+# 2. CONVERT NUTS VERSIONS
+#--------------------------
+# Let's convert manure holdings in dataset to level of NUTS codes 2021 version
+results = ro.r('''
+manure_df %>%
+  # We first classify the NUTS codes 
+  nuts_classify(nuts_code = "geo") %>%
+  # And then convert
+  nuts_convert_version(to_version = '2021',
+                       weight = 'pop11',
+                       variables = c('values' = 'absolute'))
+''')
+
+# Convert result back to pandas 
+with ro.conversion.localconverter(ro.default_converter + pandas2ri.converter):
+    manure_converted_df = ro.conversion.rpy2py(results)
+
+# Voila, the converted data
+print(manure_converted_df)
+
+
+# 3. AGGREGATE NUTS LEVELS
+#--------------------------
+# Let's aggregate NUTS 3 to NUTS 2
+results = ro.r('''
+manure_df %>%
+  # We first classify the NUTS codes 
+  nuts_classify(nuts_code = "geo") %>%
+  # And aggregate
+  nuts_aggregate(to_level = 2, 
+                 variables = c('values' = 'absolute'))
+''')
+
+# Convert result back to pandas 
+with ro.conversion.localconverter(ro.default_converter + pandas2ri.converter):
+    manure_aggregated_df = ro.conversion.rpy2py(results)
+
+# The data aggregated to nuts level 2
+print(manure_aggregated_df)