Merge branch 'master' of github.com:data-cleaning/validatetools

Author: edwindj

NEWS.md

@@ -12,6 +12,8 @@ Romina Filippini and Simona Toti for reporting.
 
 * fix for `is_contradicted_by` and `detect_infeasible_rules` when the rules contained 
 if statements.
+
+* added extra simplification for conditional statements (issue #10)
 
 # validatetools 0.5.2
 

R/simplify_conditional.R

@@ -15,10 +15,16 @@ simplify_conditional <- function(x, ...){
   vals <- to_exprs(x)
   for (i in which(is_cond)){
     cond <- vals[[i]]
+    
+    cond <- simplify_dnf(cond, vals[-i])
+    vals[[i]] <- cond
+
     cond <- simplify_non_constraining(cond, vals)
     vals[[i]] <- cond
+    
     cond <- simplify_non_relaxing(cond, vals)
     vals[[i]] <- cond
+    
   }
   # TODO set meta data correctly for the resulting rule set
   do.call(validate::validator, vals)
@@ -49,3 +55,27 @@ simplify_non_constraining <- function(cond_expr, vals){
   }
   cond_expr
 }
+
+simplify_dnf <- function(cond_expr, vals){
+  clauses <- as_dnf(cond_expr)
+  rm <- integer()
+  for (i in seq_along(clauses)){
+    clause_neg <- invert_or_negate(clauses[[i]])
+    for (j in seq_along(clauses)){
+      
+      if (i == j || j %in% rm){
+        next
+      }
+      
+      test_rules <- do.call(validate::validator,c(clauses[j], clause_neg))
+      if (is_infeasible(test_rules, verbose=FALSE)){
+        rm <- c(rm, j)
+        next
+      }
+    }
+  }
+  if (length(rm)){
+    return(as.expression.dnf(clauses[-rm], as_if = TRUE)[[1]])
+  }
+  cond_expr
+}

README.Rmd

@@ -146,6 +146,16 @@ simplify_fixed_variables(rules)
 ### Simplifying conditional statements
 
 ```{r}
+# superfluous conditions
+rules <- validator(
+  r1 = if (age > 18) age <= 67,
+  r2 = if (income > 0 && income > 1000) job == TRUE 
+)
+# implies that age always is <= 67
+simplify_conditional(rules)
+
+
+
 # non-relaxing clause
 rules <- validator( 
   r1 = if (income > 0) age >= 16,

README.md

@@ -193,6 +193,19 @@ simplify_fixed_variables(rules)
 ### Simplifying conditional statements
 
 ``` r
+# superfluous conditions
+rules <- validator(
+  r1 = if (age > 18) age <= 67,
+  r2 = if (income > 0 && income > 1000) job == TRUE 
+)
+# implies that age always is <= 67
+simplify_conditional(rules)
+#> Object of class 'validator' with 2 elements:
+#>  r1: age <= 67
+#>  r2: income <= 1000 | (job == TRUE)
+
+
+
 # non-relaxing clause
 rules <- validator( 
   r1 = if (income > 0) age >= 16,

examples/simplify_conditional.R

@@ -1,5 +1,14 @@
 library(validate)
 
+# superfluous conditions
+rules <- validator(
+  r1 = if (x > 0) x < 1,
+  r2 = if (y > 0 && y > 1) z > 0
+)
+# r1 means that x > 0 is superfluous
+# r2 means that y > 0 is superfluous
+simplify_conditional(rules)
+
 # non-relaxing clause
 rules <- validator( r1 = if (x > 1) y > 3
                   , r2 = y < 2

man/simplify_conditional.Rd

@@ -113,3 +113,18 @@ test_that("a more complex if statement also works", {
   expect_equal(exprs_s$r2, exprs$r2)
 
 })
+
+test_that("simplify dnf works", {
+  rules <- validator(r1 = if (x > 0) x < 1)
+  rules_s <- simplify_conditional(rules)
+  
+  exprs_s <- to_exprs(rules_s)
+  expect_equal(exprs_s$r1, quote(x < 1))
+  
+  rules <- validator(r1 = if (x > 0 && x > 1) y > 0)
+  rules_s <- simplify_conditional(rules)
+  
+  exprs_s <- to_exprs(rules_s)
+  expect_equal(exprs_s$r1, quote(if (x > 1) y  >0))
+  
+})