VP and VPTS object use 0/1 rather than false/true for gap

[peterdesmet]

VPTS CSV defines gap as a boolean, with true and false values. In ODIM bird profile, gap is an integer, with 1 and 0 values.

bioRad converts gap from incoming data (via read_vpfiles() or read_vpts()) to 0 and 1, but I would argue to follow the VPTS CSV specs and use false and true. Note that we already have day as another boolean field.

library(bioRad)
library(dplyr)
# # https://aloftdata.s3-eu-west-1.amazonaws.com/baltrad/hdf5/dkste/2023/09/03/dkste_vp_20230903T000000Z_0xf0040b.h5
# vp <- read_vpfiles("dkste_vp_20230903T000000Z_0xf0040b.h5")
# vp %>% as.data.frame() %>% head() %>% select(gap)

vpts <- read_vpts("https://aloftdata.s3-eu-west-1.amazonaws.com/baltrad/daily/dkste/2023/dkste_vpts_20230903.csv")
vpts %>% as.data.frame() %>% head() %>% select(gap)
#>   gap
#> 1   0
#> 2   0
#> 3   0
#> 4   0
#> 5   1
#> 6   1

^{Created on 2023-09-12 with reprex v2.0.2}

Note that e.g. the column day uses boolean values already.

[bart1]

boolean seems more logical to me

[adokter]

Seems we have lost the boolean type along the way, since the example object example_vpts still has gap as a boolean. So let's restore it to what it was.

[adokter]

function vol2birdR::vol2bird() does write a boolean in VPTS CSV format, but it stores gap as a float in .h5 format. In both cases bioRad function read_vpts() returns 1/0:

pvolfile <- system.file("extdata", "volume.h5", package = "bioRad")
vpout_csv=tempfile(fileext=".csv")
vpout_h5=tempfile(fileext=".h5")
vol2birdR::vol2bird(pvolfile, vpfile=vpout_csv)
vol2birdR::vol2bird(pvolfile, vpfile=vpout_h5)
vpts_h5=read_vpts(vpout_h5)
vpts_csv=read_vpts(vpout_csv)

# 0/1 output, not boolean:
vpts_csv$data$gap 

# 0/1 output, not boolean:
vpts_h5$data$gap

[adokter]

fixes are needed to both read_vpts_csv() and read_vpts_hdf5() (which calls read_vpfiles()), assigning to @iskandari who worked on these recently