Check if equations contain valid R code

library(allodb)
library(tidyverse)

## -- Attaching packages ------------------------------------------- tidyverse 1.2.1 --

## v ggplot2 2.2.1     v purrr   0.2.4
## v tibble  1.4.2     v dplyr   0.7.5
## v tidyr   0.8.1     v stringr 1.3.1
## v readr   1.1.1     v forcats 0.3.0

## -- Conflicts ---------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

Data.

data("equations", package = "allodb")
eqn <- equations$equation_allometry

Try to evaluate each equation and catch errors.

eval_toy <- function(text) {
  toy_value <- 10
  eval(parse(text = text), envir = list(dbh = toy_value))
}

contents <- eqn %>% 
  map(safely(eval_toy)) %>% 
  transpose()

ok <- map_lgl(contents$error, is.null) 

invalid <- tibble(
  equation_allometry = eqn[!ok], 
  messages = map_chr(contents$error[!ok], "message")
)

unique(invalid$messages)

## [1] "object 'dba' not found" "object 'BA' not found"

Issue with ‘dba’.

filter(invalid, grepl("dba", messages))

## # A tibble: 39 x 2
##    equation_allometry messages              
##    <chr>              <chr>                 
##  1 38.111*(dba^2.9)   object 'dba' not found
##  2 38.111*(dba^2.9)   object 'dba' not found
##  3 51.996*(dba^2.77)  object 'dba' not found
##  4 37.637*(dba^2.779) object 'dba' not found
##  5 43.992*(dba^2.86)  object 'dba' not found
##  6 43.992*(dba^2.86)  object 'dba' not found
##  7 43.992*(dba^2.86)  object 'dba' not found
##  8 29.615*(dba^3.243) object 'dba' not found
##  9 29.615*(dba^3.243) object 'dba' not found
## 10 29.615*(dba^3.243) object 'dba' not found
## # ... with 29 more rows

Proposed solution: #38 and #41.

Issue with ‘BA’.

filter(invalid, grepl("BA", messages))

## # A tibble: 1 x 2
##   equation_allometry messages             
##   <chr>              <chr>                
## 1 51.68+0.02*BA      object 'BA' not found

How should we deal with this?