Issues with nimble test
lucia-ramirez opened this issue · comments
Hello,
I was following the steps to download hts-nim, but I got the following error when I ran nimble test:
Executing task test in /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/hts.nimble
Hint: used config file '/software/teamtrynka/conda/ase/nim/config/nim.cfg' [Conf]
Hint: used config file '/software/teamtrynka/conda/ase/nim/config/config.nims' [Conf]
Hint: used config file '/lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/nim.cfg' [Conf]
Hint: 7880 lines; 0.338s; 6.992MiB peakmem; Debug build; proj: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all; out: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all [SuccessX]
Hint: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all [Exec]
could not import: sam_hdr_destroy
Error: execution of an external program failed: '/lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all '
stack trace: (most recent call last)
/tmp/nimblecache-1036732846/nimscriptapi_3424262110.nim(187, 16)
/lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/hts.nimble(20, 8) testTask
/software/teamtrynka/conda/ase/nim/lib/system/nimscript.nim(260, 7) exec
/software/teamtrynka/conda/ase/nim/lib/system/nimscript.nim(260, 7) Error: unhandled exception: FAILED: nim c -d:useSysAssert -d:useGcAssert --lineDir:on --debuginfo -r tests/all [OSError]
Error: Exception raised during nimble script execution
I am using htslib 1.12, nim 1.4.8 and nimble 0.13.1.
Thanks!
Hi Lucia,
this means that nim is finding an older version of htslib (< 1.10) when sam_hdr_destroy did not exist.
You can use something like: export LD_LIBRARY_PATH=/path/to/dir/
where dir is the directory that contains a libhts.so that's >= 1.10.
-Brent
Hello, I already added the path export LD_LIBRARY_PATH=/software/teamtrynka/conda/ase/lib/ (where /software/teamtrynka/conda/ase/lib/ contains libhts.a, libhts.so, libhts.so.1.12, libhts.so.3) but I am still getting the same error
Maybe you can create a new conda env and try again. This is definitely the case of nim (and the C compiler used by nim) finding an older htslib. If you update ld library path, you might make sure all the nim cache directories are cleared as that might not trigger a deep recompile since it doesn't see a file change. The caches will be in ~/.cache/nim and you can remove the entire directory.
It was indeed the cache - Thanks!
Sorry, I just realized that nimble test still drop some issues for bgzf-suite, vcf-suite and test files is this okay?
Executing task test in /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/hts.nimble
Hint: used config file '/software/teamtrynka/conda/ase/nim/config/nim.cfg' [Conf]
Hint: used config file '/software/teamtrynka/conda/ase/nim/config/config.nims' [Conf]
Hint: used config file '/lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/nim.cfg' [Conf]
Hint: 7880 lines; 0.897s; 6.988MiB peakmem; Debug build; proj: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all; out: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all [SuccessX]
Hint: /lustre/scratch123/hgi/mdt2/teams/trynka/lucia/bin/hts-nim/tests/all [Exec]
[Suite] flag test-suite
[OK] test flags
[Suite] flag cigar-suite
[OK] test op
[OK] new cigar
[Suite] flag hts-suite
[OK] test hts
[Suite] bgzf-suite
[OK] test-write
[OK] test-read
[OK] bgzf-iterator
[E::bgzf_read_block] Failed to read BGZF block data at offset 0 expected 24944 bytes; hread returned 982
[OK] bgzf-iterator-err
[OK] write-with-index
[OK] read-new-index
[OK] csi-iterator
[OK] bgzi-query from VCF rows
[Suite] fai-suite
[OK] fai-read
[OK] fai chrom_len
[OK] fai indexing
[OK] issue26
[OK] close
[Suite] aux-test
[OK] test aux
[OK] tag
[OK] drop tag
[Suite] vcf suite
[E::bcf_hdr_read] Input is not detected as bcf or vcf format
[OK] empty file rasies
[OK] filter
[OK] test writer
[OK] vcf write missing format values
[OK] iterating over info fields
[OK] that adding a new sample and setting values works.
[OK] format fields
[OK] test empty format
[OK] test format setting
[OK] add string to header
[OK] add info to header
[OK] add and set Flag
[OK] load index
[OK] remove info from header
[OK] add format to header
[OK] remove format from header
[OK] set qual
[OK] set id
[OK] new from string
[Suite] genotypes suite
[OK] test alts
[OK] unknown alts
[OK] contig names
[Suite] header record
[OK] info test
[OK] format test
[Suite] vcf alleles
[OK] update alleles with low-level setters works
[OK] update alleles with high-level setters works
[Suite] bug suite
[OK] csq reader
[OK] re-use same header issue 77
[Suite] speed tests
312078 in .. 5.220469364 seconds
[OK] standard format getter
[Suite] bam-suite
[OK] xam_index
[OK] test sa sam
[OK] test sa bam
[OK] cram writing
[OK] set aux
[OK] set qname
[OK] test from string
[Suite] stats suite
[OK] test fisher
0.09873714670538818
0.002221433773229274
0.002221433773229274
[OK] binomial
[Suite] bugs
[OK] vcf/bcf subset alts
[Suite] test files
[OK] that iteration works
[OK] file types
[E::hts_open_format] Failed to open file "tests/sadfasdfasdfasdf" : No such file or directory
[OK] that bad path gives reasonable error message
[OK] readLine
[Suite] Ensure the samples included in Readme.md works as intended
Success: Readme.md sample code between lines 39 and 62 seems OK.
Success: Readme.md sample code between lines 69 and 107 seems OK.
Success: Readme.md sample code between lines 112 and 119 seems OK.
[OK] Code extraction and running.
yes, that's as expected. those are checking the error handling in hts-nim and htslib also reports some messages.