unittest fails #502

nunotexbsd · 2024-04-01T14:18:07Z

Tests fails with same error on 2.3.5 and 2.3.6.

Any clues?

Thanks

R version 4.3.3 (2024-02-29)
Platform: amd64-portbld-freebsd14.0 (64-bit)
Running under: FreeBSD 140amd64-main 14.0-RELEASE-p5 FreeBSD 14.0-RELEASE-p5 amd64

* checking tests ...
  Running 'testthat.R'
 ERROR
Running the tests in 'tests/testthat.R' failed.
Last 13 lines of output:

  `actual$X.a4..a4..a4..e5.2` is a character vector ('<a4><a4><a4><e5>', '<a4><a4><a4><e5>', '<a4><a4><a4><e5>')
  `expected$X.a4..a4..a4..e5.2` is absent

  `actual$<a4><a4><a4><e5>1` is absent
  `expected$<a4><a4><a4><e5>1` is an integer vector (1, 2, 3)

  `actual$<a4><a4><a4><e5>2` is absent
  `expected$<a4><a4><a4><e5>2` is a character vector ('\244\244\244\345', '\244\244\244\345', '\244\244\244\345')

  [ FAIL 4 | WARN 0 | SKIP 16 | PASS 670 ]
  Error: Test failures
  In addition: Warning message:
  call dbDisconnect() when finished working with a connection
  Execution halted
* DONE

Status: 1 ERROR, 2 WARNINGs
See
  '/wrkdirs/usr/ports/databases/R-cran-RSQLite/work/RSQLite.Rcheck/00check.log'
for details.

*** Error code 1

Failed test log
testthat.Rout.fail.log

The text was updated successfully, but these errors were encountered:

krlmlr · 2024-04-01T14:48:22Z

Thanks. What's your locale? Can you please try running in a UTF-8 locale?

nunotexbsd · 2024-04-01T16:15:45Z

I'm running tests inside poudriere jail (packager) and root have default C.UTF-8 locale.
Tests was run with this setup.

Should I try anything else?

root@140amd64-main:~ # locale
LANG=C.UTF-8
LC_CTYPE="C.UTF-8"
LC_COLLATE="C.UTF-8"
LC_TIME="C.UTF-8"
LC_NUMERIC="C.UTF-8"
LC_MONETARY="C.UTF-8"
LC_MESSAGES="C.UTF-8"
LC_ALL=

krlmlr · 2024-04-01T16:24:43Z

Thanks. This definitely looks like an encoding issue, but it's hard for me to reproduce. Is there a Docker file/image that helps with this?

nunotexbsd · 2024-04-01T16:47:12Z

Never used Docker but I found this link:
https://wiki.freebsd.org/Docker

Other possibility is to run a FreeBSD in a VM:
https://www.freebsd.org/where/

On my local account (outside jail), same UTF-8 locale, same error:

== Failed tests ================================================================
-- Failure ('test-encoding.R:61:3'): list the field of tables whose colnames are BIG5 encoded (#277) --
dbListFields(con, "a") (`actual`) not identical to colnames(df) (`expected`).

<snip>

krlmlr · 2024-04-01T16:55:04Z

Thanks, this is useful.

I wonder why this test is even run on your platform, it has:

  skip_on_os("linux")
  skip_on_os("mac")
  skip_on_os("solaris")

Basically, skip everywhere except on Windows. Can you please share your Sys.info() ?

nunotexbsd · 2024-04-01T17:03:54Z

> Sys.info()
                                                                                                                                               sysname
                                                                                                                                             "FreeBSD"
                                                                                                                                               release
                                                                                                                                        "15.0-CURRENT"
                                                                                                                                               version
"FreeBSD 15.0-CURRENT #89 main-n269067-125c4560bc70: Mon Apr  1 15:39:24 WEST 2024     [email protected]:/usr/obj/usr/src/amd64.amd64/sys/GENERIC-NODEBUG"
                                                                                                                                              nodename
                                                                                                                                            "leg.home"
                                                                                                                                               machine
                                                                                                                                               "amd64"
                                                                                                                                                 login
                                                                                                                                             "nunotex"
                                                                                                                                                  user
                                                                                                                                             "nunotex"
                                                                                                                                        effective_user
                                                                                                                                             "nunotex"

nunotexbsd · 2024-04-01T17:16:48Z

Searching for correct define for freebsd unix in https://stat.ethz.ch/R-manual/R-devel/library/base/html/Platform.html and maybe should I try skip_on_os("unix") ?

nunotexbsd · 2024-04-01T17:20:47Z

(...)
also .Platform seems the correct way:

> .Platform
$OS.type
[1] "unix"

$file.sep
[1] "/"

$dynlib.ext
[1] ".so"

$GUI
[1] "X11"

$endian
[1] "little"

$pkgType
[1] "source"

$path.sep
[1] ":"

$r_arch
[1] ""

krlmlr · 2024-04-01T17:22:05Z

This is what is used internally:

testthat:::system_os()
#> [1] "darwin"
testthat:::system_os
#> function () 
#> tolower(Sys.info()[["sysname"]])
#> <bytecode: 0x117a85400>
#> <environment: namespace:testthat>
testthat::skip_on_os
#> function (os, arch = NULL) 
#> {
#>     os <- match.arg(os, choices = c("windows", "mac", "linux", 
#>         "solaris"), several.ok = TRUE)
#>     msg <- switch(system_os(), windows = if ("windows" %in% os) "On Windows", 
#>         darwin = if ("mac" %in% os) "On Mac", linux = if ("linux" %in% 
#>             os) "On Linux", sunos = if ("solaris" %in% os) "On Solaris")
#>     if (!is.null(arch) && !is.null(msg)) {
#>         if (!is.character(arch)) {
#>             abort("`arch` must be a character vector")
#>         }
#>         if (system_arch() %in% arch) {
#>             msg <- paste(msg, system_arch())
#>         }
#>         else {
#>             msg <- NULL
#>         }
#>     }
#>     if (is.null(msg)) {
#>         invisible(TRUE)
#>     }
#>     else {
#>         skip(msg)
#>     }
#> }
#> <bytecode: 0x117b4f190>
#> <environment: namespace:testthat>

^{Created on 2024-04-01 with reprex v2.1.0}

So we'd either need a patch in testthat (which will take some time) or a workaround here.

Of all the patches possible, I could imagine a skip_on_os_other_than(), support for skip_on_os("bsd") or accepting the values from tolower(Sys.info()[["sysname"]]) in skip_on_os() . The workaround should likely use skip_if_not() .

nunotexbsd · 2024-04-01T17:47:07Z

> library(testthat)
> testthat:::system_os()
[1] "freebsd"

Paching test-encoding.R:

test_that("list the field of tables whose colnames are BIG5 encoded (#277)", {
  skip_on_os("linux")
  skip_on_os("mac")
  skip_on_os("solaris")
+  skip_on_os("freebsd")

test_that("write tables whose colnames or contents are BIG5 encoded (#277)", {
  skip_on_os("linux")
  skip_on_os("mac")
  skip_on_os("solaris")
+  skip_on_os("freebsd")

test result: (reduced from 4 to 2 fails)

* checking tests ...
  Running 'testthat.R'
 ERROR
Running the tests in 'tests/testthat.R' failed.
Last 13 lines of output:
   1. \-testthat::skip_on_os("freebsd") at test-encoding.R:40:3
   2.   \-base::match.arg(...)
  -- Error ('test-encoding.R:140:3'): write tables whose colnames or contents are BIG5 encoded (#277) --
  Error in `match.arg(os, choices = c("windows", "mac", "linux", "solaris"),
      several.ok = TRUE)`: 'arg' should be one of "windows", "mac", "linux", "solaris"
  Backtrace:
      x
   1. \-testthat::skip_on_os("freebsd") at test-encoding.R:140:3
   2.   \-base::match.arg(...)

  [ FAIL 2 | WARN 0 | SKIP 16 | PASS 670 ]
  Error: Test failures
  In addition: Warning message:
  call dbDisconnect() when finished working with a connection
  Execution halted
* DONE

krlmlr · 2024-04-01T17:50:11Z

The test failures indicate that skip_on_os("bsd") doesn't work as intended.

nunotexbsd · 2024-04-01T17:52:29Z

Error in match.arg(os, choices = c("windows", "mac", "linux", "solaris"),`

Is there a way to include freebsd in that list?

krlmlr · 2024-04-01T17:54:32Z

Yes, we'd need a patch in testthat. The alternative is a plain skip_if_not() call here, perhaps:

skip_if_not(.Platform$OS.type == "windows")

nunotexbsd · 2024-04-01T18:03:28Z

Cool, that works fine, tests run OK and it makes some kind of logic since this test is only for windows?

test_that("list the field of tables whose colnames are BIG5 encoded (#277)", {
-  skip_on_os("linux")
-  skip_on_os("mac")
-  skip_on_os("solaris")
+  skip_if_not(.Platform$OS.type == "windows")

test_that("write tables whose colnames or contents are BIG5 encoded (#277)", {
-  skip_on_os("linux")
-  skip_on_os("mac")
-  skip_on_os("solaris")
+  skip_if_not(.Platform$OS.type == "windows")

krlmlr · 2024-04-01T18:04:56Z

Agree. I'll keep looking for the "multiple skip_on_os()" anti-pattern in the future.

nunotexbsd · 2024-04-01T18:07:57Z

Nice, will this be fixed upstream or should I patch package manually?

I'm not the RSQLite FreeBSD port maintainer but my work depends on it for a set of new R ports.
( https://www.freshports.org/databases/R-cran-RSQLite/ )

krlmlr · 2024-04-01T19:23:30Z

Happy to review a PR here (do you mean this by "upstream"?), but it might take a while before this hits CRAN. Patching manually might be the best option right now.

nunotexbsd · 2024-04-01T19:33:42Z

Nice. By "upstream" I mean the project coders, "downstream" will be a FreeBSD port, e.g..
Ok, I will patch it manually then.

Thanks!

krlmlr · 2024-04-01T19:34:59Z

I'd appreciate if you could point me to the patch or submit a PR here.

nunotexbsd · 2024-04-01T19:52:20Z

0001-Use-a-better-logic-on-tests-testthat-test-encoding.R.patch.txt

Please correct commit msg if necessary.

I think a pull request will be simplier.
Hope that it is ok.

Thanks

Under FreeBSD test fails because it only skips linux, mac and solaris. Opimize logic and use this test only on windows. Fixes r-dbi/RSQLite#502

krlmlr added the help wanted label Apr 1, 2024

nunotexbsd mentioned this issue Apr 2, 2024

test: Fix test on FreeBSD #504

Merged

krlmlr closed this as completed in #504 Apr 7, 2024

bpvgoncalves pushed a commit to bpvgoncalves/RSQLCipher that referenced this issue Aug 31, 2024

Update test-encoding.R: Fix test on FreeBSD

e5fffbf

Under FreeBSD test fails because it only skips linux, mac and solaris. Opimize logic and use this test only on windows. Fixes r-dbi/RSQLite#502

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

unittest fails #502

unittest fails #502

nunotexbsd commented Apr 1, 2024 •

edited

Loading

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024 •

edited

Loading

unittest fails #502

unittest fails #502

Comments

nunotexbsd commented Apr 1, 2024 • edited Loading

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024

krlmlr commented Apr 1, 2024

nunotexbsd commented Apr 1, 2024 • edited Loading

nunotexbsd commented Apr 1, 2024 •

edited

Loading

nunotexbsd commented Apr 1, 2024 •

edited

Loading