Failed test at t/Taxonomy.t

emollier commented 3 years ago

Greetings,

While investigating Debian Bug#983239, I tried running the test suite straight from origin/master and obtained the following output:

$ prove
t/EntrezGene.t ..... ok   
t/GenBank.t ........ ok     
t/GenPept.t ........ ok     
t/Query-Genbank.t .. ok     
t/RefSeq.t ......... ok   
t/Taxonomy.t ....... 43/? 
#   Failed test at t/Taxonomy.t line 102.
#          got: 'Actinomycetia'
#     expected: 'Actinobacteria'
t/Taxonomy.t ....... 69/? # Looks like you failed 1 test of 69.
t/Taxonomy.t ....... Dubious, test returned 1 (wstat 256, 0x100)
Failed 1/69 subtests 

Test Summary Report
-------------------
t/Taxonomy.t     (Wstat: 256 Tests: 69 Failed: 1)
  Failed test:  44
  Non-zero exit status: 1
Files=6, Tests=166, 152 wallclock secs ( 0.04 usr  0.01 sys +  1.88 cusr  0.23 csys =  2.16 CPU)
Result: FAIL

Is it possible the source NCBI database had some kind of shift recently?

Kind Regards, Étienne.

cjfields commented 3 years ago

Hi @emollier a recent pull request from @MrCurtis addressed this (see history above). This has been pushed out to CPAN in release 1.7.7. Thanks for reporting!

alexey-larionov commented 2 years ago

The error is still here in 1.7.7 on 10 May 2022

philgoetz commented 1 year ago

The error is still here in 1.7.7 on 10 May 2022

NCBI changed the database again, including:

Actinomycetia => Actinomycetes
homo sapiens no longer has the common name "man"; sorry to all you NLP programs out there
'Phygadeuon ovatus' formerly had no node, but now it's NCBI taxonomy node 2890685

Also, I changed the test for 'Trimorus ovatus' to 'Zaphod Beeblebrox', because you shouldn't test the functionality of reporting "No hit" for an actual species name, as it's likely to be added to the database eventually (as happened with P. ovatus).

I'll commit my fix here when I figure out how.

cjfields commented 1 year ago

We may want to consider how to future-proof these tests to not fail from remote database changes, though to be honest they don’t happen often.

On Jan 25, 2023, at 1:31 PM, Phil Goetz @.***> wrote:

The error is still here in 1.7.7 on 10 May 2022

NCBI changed the database again, including:

Actinomycetia => Actinomycetes
homo sapiens no longer has the common name "man"; sorry to all you NLP programs out there
'Phygadeuon ovatus' formerly had no node, but now it's NCBI taxonomy node 2890685

Also, I changed the test for 'Trimorus ovatus' to 'Zaphod Beeblebrox', because you shouldn't test the functionality of reporting "No hit" for an actual species name, as it's likely to be added to the database eventually (as happened with P. ovatus).

I'll commit my fix here when I figure out how.

— Reply to this email directly, view it on GitHubhttps://urldefense.com/v3/__https://github.com/bioperl/Bio-DB-NCBIHelper/issues/6*issuecomment-1404127195__;Iw!!DZ3fjg!9L4_DZ2w8fzrOkqjAYMxHBtQThoy88DJJXnQIrY_fwmEd6SxPKudWf5ggxL0ELO_y7GeoLu0Do3czwPH2iy1mEFnXr0Z$, or unsubscribehttps://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AAAMA4KFPLXODMIQEF73543WUF5RTANCNFSM4X7L4RCA__;!!DZ3fjg!9L4_DZ2w8fzrOkqjAYMxHBtQThoy88DJJXnQIrY_fwmEd6SxPKudWf5ggxL0ELO_y7GeoLu0Do3czwPH2iy1mK7FA0g3$. You are receiving this because you modified the open/close state.Message ID: @.***>

esiefker commented 10 months ago

I'm still seeing this bug a year and a half since it was closed. Is there a workaround? Or has this package just been impossible to install?


t/Taxonomy.t ........... 7/?
#   Failed test at t/Taxonomy.t line 48.
t/Taxonomy.t ........... 43/?
#   Failed test at t/Taxonomy.t line 102.
#          got: 'Actinomycetes'
#     expected: 'Actinomycetia'
t/Taxonomy.t ........... 68/?
#   Failed test 'Correct: No hit'
#   at t/Taxonomy.t line 181.
#          got: '2890685'
#     expected: 'No hit'
t/Taxonomy.t ........... 69/? # Looks like you failed 3 tests of 69.

cjfields commented 10 months ago

Hi @esiefker sorry, I haven't been seeing notifications for this repo so I'm glad you replied back. To be honest I'm happy it still works, even though the tests have been failing. Let me see what I can do to address this.

philgoetz commented 10 months ago

I committed my fixes, and got this message:

You’re making changes in a project you don’t have write access to. Submitting a change will write it to a new branch in your fork philgoetz/Bio-DB-NCBIHelper, so you can send a pull request.

Is that sufficient for someone with write privs to move my changes into the main branch?

carandraug commented 10 months ago

Is that sufficient for someone with write privs to move my changes into the main branch?

I think I found the patch on your github clone. I'll take care of merging it, give me a few minutes.

carandraug commented 10 months ago

@philgoetz I've split your fix into separate commits and pushed them.

I'm closing this issue (but someone else should test it and make a new release).

cjfields commented 10 months ago

@carandraug I'll set up tests and make a new release, thanks for the merge!

cjfields commented 10 months ago

@philgoetz @esiefker v1.7.8 has been pushed to CPAN. It will take a little while to show up, but should be ready pretty soon

bioperl / Bio-DB-NCBIHelper

Failed test at t/Taxonomy.t #6