Skip to content
GitLab
Select Git revision
  • master default
Find file BlameHistoryPermalink
extract.R 7.73 KiB
1 
# Usefull library
2 
library(tools)
3 
library(dplyr)
4
5
6 
# General information on station
7 
iStatut = c('0'='inconnu',
8 
            '1'='station avec signification hydrologique',
9 
            '2'='station sans signification hydrologique',
10 
            '3'="station d'essai")
11
12 
iFinalite = c('0'='inconnue',
13 
              '1'="hydrométrie générale",
14 
              '2'='alerte de crue',
15 
              '3'="hydrométrie générale et alerte de crue",
16 
              '4'="gestion d'ouvrage",
17 
              '5'='police des eaux',
18 
              '6'="suivi d'étiage",
19 
              '7'='bassin expérimental',
20 
              '8'='drainage')
21
22 
iType = c('0'='inconnu',
23 
          '1'='une échelle',
24 
          '2'='deux échelles, station mère',
25 
          '3'='deux échelles, station fille',
26 
          '4'='débits mesurés',
27 
          '5'='virtuelle')
28
29 
iInfluence = c('0'='inconnue',
30 
               '1'='nulle ou faible',
31 
               '2'='en étiage seulement',
32 
               '3'='forte en toute saison')
33
34 
iDebit = c('0'='reconstitué',
35 
           '1'="réel (prise en compte de l'eau rajoutée ou retirée du bassin selon aménagements)",
36 
           '2'='naturel')
37
38 
iQBE = c('0'='qualité basses eaux inconnue',
39 
         '1'='qualité basses eaux bonne',
40 
         '2'='qualité basses eaux douteuse')
41
42 
iQME = c('0'='qualité moyennes eaux inconnue',
43 
         '1'='qualité moyennes eaux bonne',
44 
         '2'='qualité moyennes eaux douteuse')
45
46 
iQHE = c('0'='qualité hautes eaux inconnue',
47 
         '1'='qualité hautes eaux bonne',
48 
         '2'='qualité hautes eaux douteuse')
49
50
51 
# Extraction of information
52 
extract_info = function (data_path, filedir, filename, verbose=TRUE) {
53
54 
    # Convert the filename in vector
55 
    filename = c(filename)
56
57 
    # If the filename is 'all' or regroup more than one filename
58 
    if (all(filename == 'all') | length(filename) > 1) {
59
60 
        # If the filename is 'all'
61 
        if (all(filename == 'all')) {
62 
            # Create a filelist to store all the filename
63 
            filelist = c()
64 
            # Get all the filename in the data directory selected
65 
            filelist_tmp = list.files(file.path(data_path,
66 
                                                filedir))
67
68 
            # For all the filename in the directory selected
69 
            for (f in filelist_tmp) {
70 
                # If the filename extention is 'txt'
7172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140
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_info = 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_info = rbind(df_info, extract_info(data_path, filedir, f)) } # Set the rownames by default (to avoid strange numbering) rownames(df_info) = NULL return (df_info) } # Get the filename from the vector filename = filename[1] # Print information if asked if (verbose) { print(paste("extraction of info for file :", filename)) } # Get the file path to the data file_path = file.path(data_path, filedir, filename) # Extract all the header infotxt = c(readLines(file_path, n=41)) # Create a tibble with all the information needed df_info = tibble(code=trimws(substr(infotxt[11], 38, nchar(infotxt[11]))), nom=trimws(substr(infotxt[12], 39, nchar(infotxt[12]))), territoire=trimws(substr(infotxt[13], 39, nchar(infotxt[13]))), L93X=as.numeric(substr(infotxt[16], 38, 50)), L93Y=as.numeric(substr(infotxt[16], 52, 63)), surface=as.numeric(substr(infotxt[19], 38, 50)), statut=iStatut[trimws(substr(infotxt[26], 38, 50))], finalite=iFinalite[trimws(substr(infotxt[26], 52, 56))], type=iType[trimws(substr(infotxt[26], 58, 58))], influence=iInfluence[trimws(substr(infotxt[26], 60, 60))], debit=iDebit[trimws(substr(infotxt[26], 62, 62))], QBE=iQBE[trimws(substr(infotxt[26], 72, 72))], QME=iQME[trimws(substr(infotxt[26], 74, 74))], QHE=iQHE[trimws(substr(infotxt[26], 76, 76))], file_path=file_path ) return (df_info) } # Example # df_info = extract_info( # "/home/louis/Documents/bouleau/INRAE/CDD_stationnarite/data", # '',
141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210
# c('H5920011_HYDRO_QJM.txt', 'K4470010_HYDRO_QJM.txt')) # Extraction of data extract_data = function (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(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, extract_data(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 information if asked if (verbose) { print(paste("extraction of data for file :", filename)) } # Get the file path to the data file_path = file.path(data_path, filedir, filename) df_data = read.table(file_path, header=TRUE, na.strings=c(' -99', ' -99.000'), sep=';', skip=41)[,1:2] # Extract all the information for the station
211212213214215216217218219220221222223224225226227228
df_info = extract_info(data_path, filedir, filename, verbose=FALSE) # Get the code of the station code = df_info$code # Create a tibble with the date as Date class and the code of the station df_data = tibble(Date=as.Date(as.character(df_data$Date), format="%Y%m%d"), df_data[-1], code=code) return (df_data) } # Example # df_data = extract_data( # "/home/louis/Documents/bouleau/INRAE/CDD_stationnarite/data", # '', # c('H5920011_HYDRO_QJM.txt', 'K4470010_HYDRO_QJM.txt'))