tests/test-atomic.R

In diffobj: Diffs for R Objects

NAME <- "atomic"
source(file.path('_helper', 'init.R'))

# - Basic Tests

all.equal(as.character(diffPrint(chr.1, chr.2)), rdsf(100))
all.equal(
  as.character(diffPrint(chr.1, chr.2, mode="unified")), rdsf(200)
)
all.equal(
  as.character(diffPrint(chr.1, chr.2, mode="context")), rdsf(400)
)
all.equal(
  as.character(diffPrint(chr.1[2:3], chr.2[2], mode="sidebyside")), rdsf(500)
)
# Check that `extra` works

all.equal(
  as.character(
    diffPrint(chr.1, chr.2, mode="unified", extra=list(quote=FALSE))
  ),
  rdsf(600)
)
# make sure blanks line up correctly

all.equal(
  as.character(diffPrint(chr.3, chr.4)), rdsf(700)
)
all.equal(
  as.character(diffPrint(chr.3, chr.4, mode="unified")), rdsf(800)
)

# - Word wrap in atomic

A <- A.1 <- B <- c(letters, LETTERS)
B[15] <- "Alabama"
A.1[5] <- "Ee"
C <- A[-15]
D <- C
E <- B[-45]

# Test simple changes to vectors; at 80 columns removing 1:8 corresponds to
# row deletion

all.equal(as.character(diffPrint(A[-(1:8)], A)), rdsf(900))
all.equal(as.character(diffPrint(A, A[-(1:8)])), rdsf(1000))

all.equal(as.character(diffPrint(A[-1], A[-2])), rdsf(1100))

# Replace single word

all.equal(as.character(diffPrint(A, B)), rdsf(1200))
all.equal(as.character(diffPrint(B, A)), rdsf(1250))

# Make sure turning off word.diff also turns of unwrapping, but that we can
# turn off unwrapping without turning off word diff

all.equal(
  as.character(diffPrint(A, B, word.diff=FALSE)), rdsf(1300)
)
all.equal(
  as.character(diffPrint(A, B, unwrap.atomic=FALSE)), rdsf(1400)
)
# Different wrap frequency and removed words that span lines

all.equal(
  as.character(diffPrint(A, A.1[-(13:18)])), rdsf(1425)
)
# Removing words

all.equal(as.character(diffPrint(C, B)), rdsf(1450))

# Two hunks

all.equal(as.character(diffPrint(D, E)), rdsf(1500))
all.equal(as.character(diffPrint(E, D)), rdsf(1600))

# Vignette example

state.abb.2 <- state.abb
state.abb.2[38] <- "Pennsylvania"

all.equal(
  as.character(diffPrint(state.abb, state.abb.2)), rdsf(1700)
)
# Number corner cases

all.equal(as.character(diffPrint(1:100, 2:101)), rdsf(1800))
all.equal(as.character(diffPrint(2:101, 1:100)), rdsf(1900))
all.equal(
  as.character(diffPrint(2:101, (1:100)[-9])), rdsf(2000)
)
all.equal(
  as.character(diffPrint((2:101)[-98], (1:100)[-9])), rdsf(2100)
)
# This is one of those that a better in-hunk align algorithm would benefit

int.1 <- int.2 <- 1:100
int.2[c(8, 20, 60)] <- 99
int.2 <- c(50:1, int.2)
all.equal(as.character(diffPrint(int.1, int.2)), rdsf(2200))

# - with names
rand.chrs <- do.call(paste0, expand.grid(LETTERS, LETTERS))
F <- F1 <- F2 <- (2:105)[-98]
G <- G2 <- G3 <- G4 <- G5 <- (1:100)[-9]
nm.1 <- rand.chrs[seq_along(F)]
nm.2 <- rand.chrs[seq_along(G)]
names(F1) <- names(F2) <- nm.1
names(G3) <- names(G2) <- names(G3) <-  names(G4) <- names(G5) <- nm.2
names(G3)[c(5, 25, 60)] <- c("XXXXX", rand.chrs[c(300, 350)])
names(G4)[c(5, 25, 60)] <- names(G5)[c(5, 25, 60)] <-
  c("XX", rand.chrs[c(300, 350)])
attr(F2, "blah") <- 1:5
attr(G5, "blah") <- 3:8

all.equal(as.character(diffPrint(F, G)), rdsf(2300))
all.equal(as.character(diffPrint(F1, G2)), rdsf(2400))

# Challenging case b/c the names wrap with values, and you have to pick one or
# the other to match when the wrap frequencies are different

all.equal(as.character(diffPrint(F1, G3)), rdsf(2500))
all.equal(as.character(diffPrint(F1, G4)), rdsf(2520))

# Attributes

all.equal(as.character(diffPrint(F2, G5)), rdsf(2530))
all.equal(as.character(diffPrint(F1, G5)), rdsf(2540))

# - Original tests
set.seed(2)
w1 <- sample(
  c(
  "carrot", "cat", "cake", "eat", "rabbit", "holes", "the", "a", "pasta",
  "boom", "noon", "sky", "hat", "blah", "paris", "dog", "snake"
  ), 25, replace=TRUE
)
w4 <- w3 <- w2 <- w1
w2[sample(seq_along(w1), 5)] <- LETTERS[1:5]
w3 <- w1[8:15]
w4 <- c(w1[1:5], toupper(w1[1:5]), w1[6:15], toupper(w1[1:5]))

all.equal(as.character(diffPrint(w1, w2)), rdsf(2600))
all.equal(as.character(diffPrint(w1, w3)), rdsf(2700))
all.equal(as.character(diffPrint(w1, w4)), rdsf(2800))

# - Simple word diffs
a <- c("a", "b", "c", "d")
b <- c("b", "c", "d", "e")
all.equal(as.character(diffPrint(a, b)), rdsf(2900))

a <- c("x", "a", "b", "c", "d", "z")
b <- c("x", "b", "c", "d", "e", "z")
all.equal(as.character(diffPrint(a, b)), rdsf(3000))

a <- c("x", "a", "b", "c", "d", "z")
b <- c("z", "b", "c", "d", "e", "x")
all.equal(as.character(diffPrint(a, b)), rdsf(3100))

# - Alignment edge cases
all.equal(
  as.character(diffPrint(20:50, 30:62)), rdsf(3200)
)
# below is off; should be aligning matching context line, part of the problem
# might be that we're doing the realignment without thinking about what the
# other hunk has.
#
# Possible encode each line as hunk#:diff/mix/cont

# all.equal(
#   as.character(diffPrint(20:50, 35:62)), rdsf(3300)
# )

# another interesting example where the existing algo seems to lead to a
# reasonable outcome

all.equal(
  as.character(diffPrint(c(1:24,35:45), c(1:8, 17:45))), rdsf(3400)
)