gowachin/diver

Package to simulate scubadiving planification.

diver's Issues

Need an upper limit for ascent speed.

Despite being absurd, high ascent speed make consumption break.

A test for the asent speed in the dive function should be implemented to limit this value.
Maximum of 60 m/min could be right however I've met someone going faster in an accident...maybe 120 ?

Reproduction of error

conso_dive <- dive(20, 40, ascent_speed = 188888999)
tank <- tank(vol = 8.5, press = 140)
cons <- 20

conso(conso_dive, tank, cons)
#> Error in if (t1 == t2) { : valeur manquante là où TRUE / FALSE est requis

rules must be ranked by values or the while loop is infinite in conso !

Rules values in a tank can't be inversed, meaning that rule1 must be superior or equal to rule2.

Reproducing error example :

tank <- tank(vol = 12, press = 200, 
              rules = list(rules = c('mid' = 25, 'quarter' = 50), sys = '%'))
conso(dive(20, 40), tank, cons = 20)
# need to stop this because the while loop has gone wild.

A check in the tank function will be added ! But another check in conso or expand could be usefull later on.

Github actions failing

I need to upgrade the github aciton so that pkgdown and coverage can run nicely.
Also need to update the README !

https://community.rstudio.com/t/github-action-failure-with-rcmd-check-on-ubuntu-devel/129727

Minor errors with majoration

Majoration add minor time to successive dives because NA values in nitrogen table resulted in 0.82 residual nitrogen in later computation about time for a depth.

Consecutive dives time shouldn't use interval time !!!

Consecutive dives shouldn't compute the dive time with interval time !!
It's written in pdf from FFESSM in 1999 about MN90 tables. Need to check this and rewrite this function !

DiveR/R/DiveR.R

Line 182 in e70dd66

time <- dtime(dive1) + dive1$dtr + inter + time2

dtime work only on square dives

Curve dives (where depths and time are input as vector and not single value) can't be used properly by dtime function, meaning that it is not possible to compute ndive and summary them properly.

In the refactorisation of dive planning to add an attribute to direct different behavior of attributes functions.

Use {checkmate} to simplify the idiotproof

Idiotproof (checking function argument) is tedious and takes a lot of lines.

The package {checkmate} could be great to introduce but it also require to modify most of the tests of the package.
Could be usefull to add this feature for next release (along with deco algorithmes and {rcpp} usage)

Plot is sometimes missing information.

The text for deco stop is sometimes missing.

Ex : Missing 20' time mark above the second deco stop. A second dive with 19' depth time work.

d1 <- dive(40, 10)
d2 <- dive(30, 20)
ds <- ndive(d1, d2, inter = 60)
plot(ds)
summary(ds$dive2) # deco time is 20min

Should work like :

d2 <- dive(30, 19)
ds <- ndive(d1, d2, inter = 60)
plot(ds)
summary(ds$dive2) # deco time is 20min

It also could be nice to improve horizontal spacing for inter time and maj time.