Skip to content
GitLab
Select Git revision
  • master default
Find file BlameHistoryPermalink
extractNV.R 5.80 KiB
1 
# Usefull library
2 
library(tools)
3 
library(dplyr)
4
5
6 
# Extraction of metadata
7 
extractNVlist_meta = function (computer_data_path, filedir, listdir, listname, verbose=TRUE) {
8
9 
    # Print metadata if asked
10 
    if (verbose) {
11 
        print(paste("extraction of NV meta for file :", listname))
12 
    }
13
14 
    # Get the file path to the list of station
15 
    list_path = file.path(computer_data_path, listdir, listname)
16
17 
    if (file.exists(list_path) & substr(list_path, nchar(list_path), nchar(list_path)) != '/') {
18
19 
        # Extract the data as a data frame
20 
        df_meta = read.table(list_path,
21 
                             header=TRUE)
22
23 
        # Create a filelist to store all the filename
24 
        codelist = c()
25
26 
        dir_path = file.path(computer_data_path, filedir)
27
28 
        # Get all the filename in the data directory selected
29 
        filelist_tmp = list.files(dir_path)
30
31 
        # For all the filename in the directory selected
32 
        for (f in filelist_tmp) {
33 
            # If the filename extention is 'txt'
34 
            if (file_ext(f) == 'txt') {
35 
                # Store the filename in the filelist
36 
                codelist = c(codelist, gsub('.txt', '', f))
37 
            }
38 
        }
39
40 
        exist = df_meta$CODE %in% codelist
41 
        missing = codelist[!(codelist %in% df_meta$CODE)]
42
43 
        # Display stations for which it misses metadata
44 
        print(paste('missing station meta for :', missing))
45
46 
        df_meta = df_meta[exist,]
47
48 
        # Create a tibble with all the metadata needed
49 
        df_meta =
50 
            tibble(code=as.character(df_meta$CODE),
51 
                   nom=as.character(df_meta$NOM),
52 
                   L93X=df_meta$X_L2E,
53 
                   L93Y=df_meta$Y_L2E,
54 
                   surface_km2=df_meta$S_km2,
55 
                   altitude_m=df_meta$Alt,
56 
                   file_path=file.path(dir_path,
57 
                                       paste(df_meta$CODE, '.txt', sep='')),
58 
                   source='NV'
59 
                   )
60
61 
        df_meta = bind_rows(df_meta,
62 
                            data.frame(code=missing,
63 
                                       file_path=file.path(dir_path,
64 
                                                           paste(missing,
65 
                                                                 '.txt',
66 
                                                                 sep=''))))
67
68 
        df_meta = df_meta[order(df_meta$code),]
69
70 
    } else {
7172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140
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))
141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179
} # 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'))