diff --git a/plotting/panel.R b/plotting/panel.R
index ba163204dbdc2e337b6aa46402188d5132f23941..fdd8723559f5f407d87535b78623c683677c6800 100644
--- a/plotting/panel.R
+++ b/plotting/panel.R
@@ -35,6 +35,7 @@ panel = function (df_data, df_info, figdir, filedir, is_sqrt=FALSE) {
print(paste("Plotting for sation :", code))
df_data_code = df_data[df_data$code==code,]
+ # df_data_code_NoNA = df_data_code[!is.na(df_data_code$Qm3s),]
# Plot
p = ggplot(df_data_code, aes(x=Date, y=Qm3s)) +
diff --git a/processing/analyse.R b/processing/analyse.R
index b7441a6443d7a9a3c3687a8c264a4de90094e428..d5fcbe7feec437b3eeeb9bf5c832a1976726ac31 100644
--- a/processing/analyse.R
+++ b/processing/analyse.R
@@ -1,5 +1,5 @@
# Usefull library
-# library(StatsAnalysisTrend)
+library(StatsAnalysisTrend)
library(dplyr)
@@ -57,3 +57,11 @@ get_lacune = function (df_data, df_info) {
return (df_lac)
}
+
+
+
+
+
+
+
+
diff --git a/processing/extractBH.R b/processing/extractBH.R
index 2c841d4c5f4f5ad87858825c7662819d3540d477..09ea390447fabbe5d2efc24fd513c338ce601981 100644
--- a/processing/extractBH.R
+++ b/processing/extractBH.R
@@ -164,7 +164,7 @@ extractBH_info = function (computer_data_path, filedir, filename, verbose=TRUE)
file_path = file.path(computer_data_path, filedir, filename)
- if (file.exists(file_path)) {
+ if (file.exists(file_path) & substr(file_path, nchar(file_path), nchar(file_path)) != '/') {
# Extract all the header
infotxt = c(readLines(file_path, n=41))
@@ -185,7 +185,8 @@ extractBH_info = function (computer_data_path, filedir, filename, verbose=TRUE)
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
+ file_path=file_path,
+ source='BH'
)
return (df_info)
@@ -263,7 +264,7 @@ extractBH_data = function (computer_data_path, filedir, filename, verbose=TRUE)
# Get the file path to the data
file_path = file.path(computer_data_path, filedir, filename)
- if (file.exists(file_path)) {
+ 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,
diff --git a/processing/extractNV.R b/processing/extractNV.R
index 72abcd0d479bcca31dba1215bcdbd8837d603e1b..8d1f94ee3b110d281e17f1d97a87639e8ee89ae5 100644
--- a/processing/extractNV.R
+++ b/processing/extractNV.R
@@ -11,54 +11,62 @@ extractNVlist_info = function (computer_data_path, filedir, listdir, listname, v
print(paste("extraction of NV info for file :", listname))
}
- # Get the file path to the data
- list_path = file.path(computer_data_path, listdir, listname)
-
- # Extract the data as a data frame
- df_info = read.table(list_path,
- header=TRUE)
-
- # 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)
+ # Get the file path to the list of station
+ list_path = file.path(computer_data_path, listdir, listname)
- # 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))
- }
- }
+ if (file.exists(list_path) & substr(list_path, nchar(list_path), nchar(list_path)) != '/') {
- exist = df_info$CODE %in% codelist
- missing = codelist[!(codelist %in% df_info$CODE)]
- print(paste('missing station info for :', missing))
-
- df_info = df_info[exist,]
-
- # Create a tibble with all the information needed
- df_info =
- tibble(code=as.character(df_info$CODE),
- nom=as.character(df_info$NOM),
- L93X=df_info$X_L2E,
- L93Y=df_info$Y_L2E,
- surface_km2=df_info$S_km2,
- altitude_m=df_info$Alt,
- file_path=file.path(dir_path,
- paste(df_info$CODE, '.txt', sep=''))
- )
-
- df_info = bind_rows(df_info,
- data.frame(code=missing,
- file_path=file.path(dir_path,
- paste(missing,
- '.txt',
- sep=''))))
+ # Extract the data as a data frame
+ df_info = read.table(list_path,
+ header=TRUE)
+
+ # Create a filelist to store all the filename
+ codelist = c()
+
+ dir_path = file.path(computer_data_path, filedir)
+
+ # Get all the filename in the data directory selected
+ filelist_tmp = list.files(dir_path)
+
+ # For all the filename in the directory selected
+ for (f in filelist_tmp) {
+ # If the filename extention is 'txt'
+ if (file_ext(f) == 'txt') {
+ # Store the filename in the filelist
+ codelist = c(codelist, gsub('.txt', '', f))
+ }
+ }
+
+ exist = df_info$CODE %in% codelist
+ missing = codelist[!(codelist %in% df_info$CODE)]
+ print(paste('missing station info for :', missing))
+
+ df_info = df_info[exist,]
+
+ # Create a tibble with all the information needed
+ df_info =
+ tibble(code=as.character(df_info$CODE),
+ nom=as.character(df_info$NOM),
+ L93X=df_info$X_L2E,
+ L93Y=df_info$Y_L2E,
+ surface_km2=df_info$S_km2,
+ altitude_m=df_info$Alt,
+ file_path=file.path(dir_path,
+ paste(df_info$CODE, '.txt', sep='')),
+ source='NV'
+ )
+
+ df_info = bind_rows(df_info,
+ data.frame(code=missing,
+ file_path=file.path(dir_path,
+ paste(missing,
+ '.txt',
+ sep=''))))
+
+ } else {
+ print(paste('filename', list_path, 'do not exist'))
+ return (NULL)
+ }
return (df_info)
}
@@ -131,7 +139,7 @@ extractNV_data = function (computer_data_path, filedir, filename, verbose=TRUE)
# Get the file path to the data
file_path = file.path(computer_data_path, filedir, filename)
- if (file.exists(file_path)) {
+ 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,
diff --git a/processing/prepare.R b/processing/prepare.R
new file mode 100644
index 0000000000000000000000000000000000000000..d61ddedd0a50048cdb1db855b93481d226b18b1a
--- /dev/null
+++ b/processing/prepare.R
@@ -0,0 +1,13 @@
+
+
+prepare = function(df_data, df_info) {
+
+ df_data$Gcode = as.integer(factor(df_data$code))
+ df_data = tibble(Date=df_data$Date,
+ Qm3s=df_data$Qm3s,
+ Gcode=df_data$code)
+
+ df_info$Gcode = as.integer(factor(df_info$code))
+
+ return (list(data=df_data, info=df_info))
+}
diff --git a/script.R b/script.R
index 960d7266b9a576f7777575a9e93a1ab8fea1af07..8ea2ea69bea6dddfad8d5a0c091235bbaaa9ffbc 100644
--- a/script.R
+++ b/script.R
@@ -17,41 +17,45 @@ computer_work_path =
# Path to the directory where BH data is stored
BHfiledir =
# "test"
+ # ""
"BanqueHydro_Export2021"
## Manual selection ##
# Name of the file that will be analysed from the BH directory
BHfilename =
- FALSE
+ ""
# c("H5920011_HYDRO_QJM.txt", "K4470010_HYDRO_QJM.txt")
# "all"
## Or list selection ##
# Path to the list file of BH data that will be analysed
BHlistdir =
- # FALSE
""
BHlistname =
- # FALSE
+ # ""
"Liste-station_RRSE.docx"
### NIVALE ###
# Path to the directory where NV data is stored
NVfiledir =
- "France207"
+ ""
+ # "France207"
# Name of the file that will be analysed from the NV directory
NVfilename =
- "all"
+ ""
+ # "all"
+
# Path to the list file of information about station that will be analysed
NVlistdir =
""
NVlistname =
- "liste_bv_principaux_global.txt"
+ ""
+ # "liste_bv_principaux_global.txt"
########################
@@ -63,6 +67,7 @@ setwd(computer_work_path)
# Sourcing R file
source('processing/extractBH.R')
source('processing/extractNV.R')
+source('processing/prepare.R')
source('processing/analyse.R')
source('plotting/panel.R')
@@ -86,7 +91,7 @@ print(paste('figdir :', figdir))
# BANQUE HYDRO #
# Get only the selected station from a list station file
-if (is.character(BHlistdir) & is.character(BHlistname) & is.character(BHfiledir)){
+if (BHlistname != ""){
df_selec = get_selection(computer_data_path,
BHlistdir,
BHlistname,
@@ -113,25 +118,52 @@ df_data_BH = extractBH_data(computer_data_path, BHfiledir, BHfilename)
# NIVALE #
# Extract information about selected stations
df_info_NV = extractNVlist_info(computer_data_path, NVfiledir, NVlistdir, NVlistname)
-### /!\ missing station info
+### /!\ missing station info ###
# Extract data about selected stations
df_data_NV = extractNV_data(computer_data_path, NVfiledir, NVfilename)
-
-common = levels(factor(df_info_NV[df_info_NV$code %in% df_info_BH$code,]$code))
-NVadd = levels(factor(df_info_NV[!(df_info_NV$code %in% df_info_BH$code),]$code))
-
-df_info_NVadd = df_info_NV[df_info_NV$code %in% NVadd,]
-df_info = full_join(df_info_BH, df_info_NVadd, by=c("code", "nom", "L93X", "L93Y", "surface_km2", "file_path"))
-
-df_data_NVadd = df_data_NV[df_data_NV$code %in% NVadd,]
-df_data = full_join(df_data_BH, df_data_NVadd, by=c("Date", "Qm3s", "code"))
+# JOIN #
+if (!is.null(df_data_NV) & !is.null(df_data_BH)) {
+
+ # Get the station in common
+ common = levels(factor(df_info_NV[df_info_NV$code %in% df_info_BH$code,]$code))
+ # Get the Nv station to add
+ NVadd = levels(factor(df_info_NV[!(df_info_NV$code %in% df_info_BH$code),]$code))
+
+ # Select only the NV info to add
+ df_info_NVadd = df_info_NV[df_info_NV$code %in% NVadd,]
+ # Join NV data to BH data
+ df_info = full_join(df_info_BH, df_info_NVadd, by=c("code", "nom", "L93X", "L93Y", "surface_km2", "file_path"))
+
+ # Select only the NV data to add
+ df_data_NVadd = df_data_NV[df_data_NV$code %in% NVadd,]
+ # Join NV info to BH info
+ df_data = full_join(df_data_BH, df_data_NVadd, by=c("Date", "Qm3s", "code"))
+
+} else if (is.null(df_data_NV) & !is.null(df_data_BH)) {
+ df_info = df_info_BH
+ df_data = df_data_BH
+} else if (!is.null(df_data_NV) & is.null(df_data_BH)) {
+ df_info = df_info_NV
+ df_data = df_data_NV
+} else {
+ stop('No data')
+}
+# TIME PANEL #
# Plot time panel of debit by stations
-# panel(df_data, df_info, figdir, filedir)
-# panel(df_data, df_info, figdir, filedir, is_sqrt=TRUE)
+# panel(df_data, df_info, figdir, "")
+# panel(df_data, df_info, figdir, "", is_sqrt=TRUE)
+### /!\ Removed 185 row(s) containing missing values (geom_path) -> remove NA ###
+
# Compute gap parameters for stations
# df_lac = get_lacune(df_data, df_info)
+
+df_list = prepare(df_data, df_info)
+
+
+
+