Extract NV rm

processing/extractNV.R

-# Usefull library
-library(tools)
-library(dplyr)
-
-
-iRegHydro = c('D'='Affluents du Rhin',
-              'E'="Fleuves ctiers de l'Artois-Picardie",
-              'A'='Rhin',
-              'B'='Meuse',
-              'F'='Seine aval (Marne incluse)',
-              'G'='Fleuves ctiers haut normands',
-              'H'='Seine amont',
-              'I'='Fleuves ctiers bas normands',
-              'J'='Bretagne',
-              'K'='Loire',
-              'L'='Loire',
-              'M'='Loire',
-              'N'='Fleuves ctiers au sud de la Loire',
-              'O'='Garonne',
-              'P'='Dordogne',
-              'Q'='Adour',
-              'R'='Charente',
-              'S'="Fleuves ctiers de l'Adour-Garonne",
-              'U'='Sane',
-              'V'='Rhne',
-              'W'='Isre',
-              'X'='Durance',
-              'Y'='Fleuves ctiers du Rhne-Mditranne et Corse',
-              'Z'='les',
-              '1'='Guadeloupe',
-              '2'='Martinique',
-              '5'='Guyane',
-              '6'='Guyane',
-              '7'='Guyane',
-              '8'='Guyane',
-              '9'='Guyane',
-              '4'='Runion')
-
-
-# Extraction of metadata
-extractNVlist_meta = function (computer_data_path, filedir, listdir, listname, verbose=TRUE) {
-    
-    # Print metadata if asked
-    if (verbose) {
-        print(paste("extraction of NV meta for file :", listname))
-    }
-
-    # Get the file path to the list of station
-    list_path = file.path(computer_data_path, listdir, listname)
-
-    if (file.exists(list_path) & substr(list_path, nchar(list_path), nchar(list_path)) != '/') {
-
-        # Extract the data as a data frame
-        df_meta = read.table(list_path,
-                             header=TRUE,
-                             encoding='UTF-8'
-                             ) 
-
-        # Create a filelist to store all the filename
-        codelist = c()
-        
-        dir_path = file.path(computer_data_path, filedir)
-
-        # Get all the filename in the data directory selected
-        filelist_tmp = list.files(dir_path)
-
-        # For all the filename in the directory selected
-        for (f in filelist_tmp) {
-            # If the filename extention is 'txt'
-            if (file_ext(f) == 'txt') {
-                # Store the filename in the filelist
-                codelist = c(codelist, gsub('.txt', '', f)) 
-            }
-        }  
-
-        exist = df_meta$CODE %in% codelist
-        missing = codelist[!(codelist %in% df_meta$CODE)]
-        
-        # Display stations for which it misses metadata
-        print(paste('missing station meta for :', missing))
-        
-        df_meta = df_meta[exist,]
-
-        # Create a tibble with all the metadata needed
-        df_meta =
-            tibble(code=as.character(df_meta$CODE),
-                   nom=as.character(df_meta$NOM),
-                   L93X=df_meta$X_L2E,
-                   L93Y=df_meta$Y_L2E,
-                   surface_km2=df_meta$S_km2,
-                   altitude_m=df_meta$Alt,
-                   file_path=file.path(dir_path,
-                                       paste(df_meta$CODE, '.txt', sep='')),
-                   source='NV',
-                   )
-
-        df_meta = bind_rows(df_meta, 
-                            data.frame(code=missing,
-                                       file_path=file.path(dir_path,
-                                                           paste(missing,
-                                                                 '.txt',
-                                                                 sep=''))))
-
-        df_meta = df_meta[order(df_meta$code),]
-
-        df_meta$region_hydro = iRegHydro[substr(df_meta$code, 1, 1)]
-        
-    } else {
-        print(paste('filename', list_path, 'do not exist'))
-        return (NULL)
-    }
-
-    return (df_meta)
-}
-
-# Example
-# df_meta = extractNVlist_meta(
-#     "/home/louis/Documents/bouleau/INRAE/CDD_stationnarite/data",
-#     'France207',
-#     '',
-#     'liste_bv_principaux_global.txt')
-
-
-# Extraction of data
-extractNV_data = function (computer_data_path, filedir, filename, verbose=TRUE) {
-    
-    # Convert the filename in vector
-    filename = c(filename)
-
-    # If the filename is 'all' or regroup more than one filename
-    if (all(filename == 'all') | length(filename) > 1) {
-
-        # If the filename is 'all'
-        if (all(filename == 'all')) {
-            # Create a filelist to store all the filename
-            filelist = c()
-             # Get all the filename in the data directory selected
-            filelist_tmp = list.files(file.path(computer_data_path,
-                                                filedir))
-
-            # For all the filename in the directory selected
-            for (f in filelist_tmp) {
-                # If the filename extention is 'txt'
-                if (file_ext(f) == 'txt') {
-                    # Store the filename in the filelist
-                    filelist = c(filelist, f) 
-                }
-            }
-            # If the filename regroup more than one filename
-        } else if (length(filename > 1)) {
-             # The filelist correspond to the filename
-            filelist = filename
-        } 
-
-        # Create a blank data frame
-        df_data = data.frame()
-
-        # For all the file in the filelist
-        for (f in filelist) {
-
-            # Concatenate by raw data frames created by this function when filename correspond to only one filename
-            df_data = rbind(df_data,
-                            extractNV_data(computer_data_path, 
-                                         filedir, 
-                                         f))
-        }
-
-        # Set the rownames by default (to avoid strange numbering)
-        rownames(df_data) = NULL
-        return (df_data)
-    }
-
-    # Get the filename from the vector
-    filename = filename[1]
-    
-    # Print metadata if asked
-    if (verbose) {
-        print(paste("extraction of NV data for file :", filename))
-    }
-
-    # Get the file path to the data
-    file_path = file.path(computer_data_path, filedir, filename)
-    
-    if (file.exists(file_path) & substr(file_path, nchar(file_path), nchar(file_path)) != '/') {
-
-        # Extract the data as a data frame
-        df_data = read.table(file_path,
-                             header=FALSE,
-                             skip=1,
-                             na.strings=c('-1', '-99.000'))   
-
-        # Create a tibble with the date as Date class and the code of the station
-        date = paste(df_data[,1],
-                     df_data[,2],
-                     df_data[,3],
-                     sep='-')
-
-        df_data = tibble(Date=as.Date(as.character(date),
-                                      format="%Y-%m-%d"),
-                         Qm3s=df_data[,4],
-                         QCode=df_data[,5],
-                         code=gsub('.txt', '', filename))
-
-        return (df_data)
-
-    } else {
-        print(paste('filename', file_path, 'do not exist'))
-        return (NULL)
-    }
-}
-
-# Example
-# df_data = extractNV_data(
-    # "/home/louis/Documents/bouleau/INRAE/CDD_stationnarite/data",
-    # 'France207',
-    # c('O0015310.txt', 'Q0214010.txt', 'P0115020.txt'))