Skip to content

Test Failures Found During R CMD Check #172

@Andrew-Acker125

Description

@Andrew-Acker125

Hello,
We've been running R CMD check on packages for our package registry manager PRISM (Home | PRISM) and noticed that visdat is failing during tests with 56 failures. A large amount of these failures are visual snapshot failures that seem to be related to a ggplot2 update. Some of these visual snapshots appear to be having the same problem as seen in issue #171.
The full error output:

checking tests Running testthat.R
Running the tests in tests/testthat.R failed.
Last 13 lines of output:
  'vis-value/vis-value-constant.svg', and 'vis-value/vis-value-vanilla.svg'
  Error:
  ! Test failures.
  Execution halted
  [ FAIL 56 | WARN 0 | SKIP 0 | PASS 53 ]
  Error:
  ! Test failures.
  Execution halted

Additional Error Info

1. Visual snapshot differences (33 failures)

  • test-test-vis-miss-one-col.R — 6 failures
  • test-vis-binary.R — 1 failure
  • test-vis-compare.R — 1 failure
  • test-vis-cor.R — 2 failures
  • test-vis-dat.R — 5 failures
  • test-vis-expect.R — 2 failures
  • test-vis-guess.R — 4 failures
  • test-vis-miss.R — 10 failures
  • test-vis-value.R — 2 failures

See attached zip file of expected vs actual for failures we could reproduce

visdat.zip

2. Error format snapshot changes (15 failures )

  • test-vis-binary.R — 1 failure
  • test-vis-compare.R — 4 failures
  • test-vis-cor.R — 2 failures
  • test-vis-dat.R — 2 failures
  • test-vis-expect.R — 1 failure
  • test-vis-guess.R — 2 failures
  • test-vis-large.R — 2 failures
  • test-vis-miss.R — 1 failure

All 15 follow the same pattern error display format changed from Error <rlang_error> to Condition Error in \fn(). Here is one example (test-vis-dat.R):
Expected:

Code
  vis_dat(AirPassengers)
Error <rlang_error>
  `vis_dat()` requires a <data.frame>
  the object I see has class(es):
  <ts>

Actual:

Code
  vis_dat(AirPassengers)
Condition
  Error in `test_if_dataframe()`:
  ! `vis_dat()` requires a <data.frame>
  the object I see has class(es):
  <ts>

3. Tibble print format changes (8 failures)

  • test-gather_funs.R — 2 failures
  • test-vis-cor.R — 2 failures
  • test-vis-dat.R — 2 failures
  • test-vis-miss.R — 2 failures
    All 8 follow the same pattern, format changed from # ... to # i. Here is one example (test-vis-dat.R):
    Expected:
     8     5     2 Solar.R  integer   8    
     9     5     2 Temp     integer   72   
    10     5     2 Wind     numeric   2    
   # ... with 755 more rows

Actual:

     8     5     2 Solar.R  integer   8    
     9     5     2 Temp     integer   72   
    10     5     2 Wind     numeric   2    
   # i 755 more rows

4. Spelling

spelling.Rout vs spelling.Rout.save comparison fails, the word "unqouted" in vis_cor.Rd:28 is now flagged as a potential spelling error.
Expected:

All Done!

Actual:

Potential spelling errors:
  WORD       FOUND IN
unqouted   vis_cor.Rd:28
If these are false positive, run `spelling::update_wordlist()`.All Done!

Happy to provide any more details or clarity if needed.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions