Data.table: test.data.table() error when optional package reshape is not installed

[skanskan]

Hello.

I've just used test.data.table() to ckeck my data.table 1.10.5 installation and got this errors:

Running test id 168 Don't know how to automatically pick scale for object of type ITime. Defaulting to continuous.
Test 168 Error in seq.int(0, to0 - from, by) : 'to' cannot be NA, NaN or infinite
Running test id 168.1 Test 168.1 Error in seq.int(0, to0 - from, by) : 'to' cannot be NA, NaN or infinite
Running test id 486 Tests 487 and 488 not run. If required call library(reshape) first.
Running test id 1383.2 Tests 1383.3 not run. If required call library(reshape) first.
Error in eval(expr, envir, enclos) :
2 errors out of 5737 (lastID=1759, endian==little, sizeof(long double)==16, sizeof(pointer)==8) in inst/tests/tests.Rraw on Fri Apr 07 14:17:57 2017. Search tests.Rraw for test numbers: 168, 168.1.

I have reshape2 installed instead of reshape, maybe the test should move to the newer packages.

[mattdowle]

Ok will try. It was a bit of a nightmare to test reshape and reshape2 interaction. The tests detach both packages and then reattach them in a particular order so that one is above the other in the search() path. Now that time has passed and we no longer import either, I'm not sure the best thing to do: yes maybe we should just remove reshape from Suggests and its tests.

[MichaelChirico]

the tests at least guarantee consistency with reshape2 operations, as long
as that's still a priority should probably keep the tests in.

On Apr 7, 2017 9:24 PM, "Matt Dowle" [e-mail] wrote:

Ok will try. It was a bit of a nightmare to test reshape and reshape2
interaction. The tests detach both packages and then reattach them in a
particular order so that one is above the other in the search() path. Now
that time has passed and we no longer import either, I'm not sure the best
thing to do: yes maybe we should just remove reshape from Suggests and its
tests.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
https://github.com/Rdatatable/data.table/issues/2105#issuecomment-292685418,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AHQQdSjMWqvFvcHEGistvhwVZ4AAAKBQks5rtuHcgaJpZM4M21QB
.

[skanskan]

What about the other errors?
Running test id 168 Don't know how to automatically pick scale for object of type ITime. Defaulting to continuous.
Test 168 Error in seq.int(0, to0 - from, by) : 'to' cannot be NA, NaN or infinite
Running test id 168.1 Test 168.1 Error in seq.int(0, to0 - from, by) : 'to' cannot be NA, NaN or infinite

[MichaelChirico]

actually, only tests 168 and 168.1 are showing errors. the others are just
messages produced along the way

On Apr 7, 2017 9:36 PM, "skanskan" [e-mail] wrote:

What about the other errors?
Running test id 168 Don't know how to automatically pick scale for object
of type ITime. Defaulting to continuous.
Test 168 Error in seq.int(0, to0 - from, by) : 'to' cannot be NA, NaN or
infinite
Running test id 168.1 Test 168.1 Error in seq.int(0, to0 - from, by) :
'to' cannot be NA, NaN or infinite

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/Rdatatable/data.table/issues/2105#issuecomment-292686116,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AHQQdfYwWgcGItwPs6lcXVRuqcCb6GK9ks5rtuSGgaJpZM4M21QB
.

[skanskan]

But are this errors caused by my R+data.table installation or settings?
Or it's a problem with this data.table version?
Or a problem with the test script?
Can they affect my daily operation?

[MichaelChirico]

look at the file inst/tests/tests.Rraw and search for those test #s

On Apr 8, 2017 7:23 AM, "skanskan" [e-mail] wrote:

But are this errors caused by my R+data.table installation or settings?
Can they affect my daily operation?

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/Rdatatable/data.table/issues/2105#issuecomment-292711818,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AHQQdc0ZEwwxh0zAH0UKoiXH0K2IddlFks5rt25JgaJpZM4M21QB
.

[mattdowle]

These test fails (168 and 168.1) were caused by a ggplot2 update, iirc. If not, the problem has been fixed some time ago I think. Is test.data.table() ok for you now @skanskan ?
Also, yes, I'm going to remove reshape Suggests as part of #2671.

[skanskan]

My last run:

Running test id 167      Test 167 ran without errors but failed check that x equals y:
> x = names(print(ggplot(DT, aes(b, f)) + geom_point()))[c(1, 3)] 
First 6 of 2 (type 'character'): [1] "data"   "scales"
> y = c("data", "plot") 
First 6 of 2 (type 'character'): [1] "data" "plot"
1 string mismatch
Running test id 167.2      Test 167.2 ran without errors but failed check that x equals y:
> x = names(print(ggplot(DT) + geom_hex(aes(b, f)) + facet_wrap(~grp)))[c(1,      3)] 
First 6 of 2 (type 'character'): [1] "data"   "scales"
> y = c("data", "plot") 
First 6 of 2 (type 'character'): [1] "data" "plot"
1 string mismatch
Running test id 168      Don't know how to automatically pick scale for object of type ITime. Defaulting to continuous.
Don't know how to automatically pick scale for object of type ITime. Defaulting to continuous.
Running test id 486      Tests 487 and 488 not run. If required call library(reshape) first.
Running test id 896.4225      Tests 897-899 not run. If required call library(bit64) first.
Running test id 1372.2      Tests 1372.3+ not run. If required call library(GenomicRanges) first.
Running test id 1383.2      Tests 1383.3 not run. If required call library(reshape) first.
Tests 1383.4 not run. If required call library(caret) first.
Running test id 1626.68      Tests 1626.[69-76] not run. If required call library(bit64) first.
Running test id 1729.14      Tests 1730 & 1731 not run. If required call library(bit64) first.
Running test id 1750.34      Test 1750.[41-48] not run. If required call library(bit64) first.
Running test id 1751.2      Test 1752 not run. If required call library(nanotime) first.
Error in eval(exprs[i], envir) : 
  2 errors out of 7437 (lastID=1880.2, endian==little, sizeof(long double)==16, sizeof(pointer)==8) in inst/tests/tests.Rraw on Wed Mar 14 00:58:14 2018. Search tests.Rraw for test numbers: 167, 167.2.

[HughParsonage]

@mattdowle the fix is still yet to be merged in #2545 I think.

[mattdowle]

Ah yes. This error is a different more recent error, caused by a recent change in ggplot2. It's a bit of a silly test really. PR #2671 removes ggplot2 from Suggest and these tests. The fixed ggplot test thanks to Hugh needs going into the new (to be created) compatibility package, then.

[HughParsonage]

Do you want me to modify PR #2545 at all? Or just leave till you get time?
On Wed, 14 Mar 2018 at 11:54 am, Matt Dowle [e-mail]
wrote:

Ah yes. This error is a different more recent error, caused by a recent
change in ggplot2. It's a bit of a silly test really. PR #2671
https://github.com/Rdatatable/data.table/pull/2671 removes ggplot2 from
Suggest and these tests. The fixed ggplot test thanks to Hugh needs going
into the new (to be created) compatibility package, then.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
https://github.com/Rdatatable/data.table/issues/2105#issuecomment-372868701,
or mute the thread
https://github.com/notifications/unsubscribe-auth/AHvGDKCWtMJ37V9GZZMRJPhxhuLSQ585ks5teGpJgaJpZM4M21QB
.

[mattdowle]

Nothing needed thanks. Yeah I'll get to #2545 when I'm wrapping up fread. There's other fread tasks in that area so I'll try and do them at same time.