PacificBiosciences / HiPhase

Small variant, structural variant, and short tandem repeat phasing tool for PacBio HiFi reads
Other
70 stars 4 forks source link

[Suggestion] reducing messages to STDOUT to speed up the utility #32

Closed davidyuyuan closed 6 months ago

davidyuyuan commented 7 months ago

HiPhase works great. I'd like to suggest to move the warning messages such as the one below to debug. This would reduce a large amount of messages to STDOUT. The log can be less cluttered and the execution can be faster without major changes.

[2024-04-15T18:17:31.161Z WARN hiphase::writers::ordered_vcf_writer] Received 'error seeking to "chrUn_KI270512v1":0 in indexed file', while seeking to chrUn_KI270512v1:0-18446744073709551615 in vcf #0, likely no variants present

holtjma commented 7 months ago

Hello,

While this would reduce log clutter, it may also mask out legitimate errors in the underlying htslib crate/library that HiPhase is using. Currently, the errors from rust_htslib are passed through as warnings in this context, allowing users like yourself to identify issues with the input files that would be otherwise hidden without the warning. We could potentially add an option to disable the messages, but we would still leave these warnings on by default.

I wouldn't worry about run-time of the output, it is extremely insignificant relative to the rest of the phasing process.

Matt

p.s. glad HiPhase is working for you!

davidyuyuan commented 7 months ago

Thanks, Matt. An option to turn it off would be great.

holtjma commented 6 months ago

In the hundred of samples we've run internally, we haven't found anything significant that those messages were capturing that were not caught elsewhere in the code. As a result, we elected to just downgrade them to DEBUG level.

This should be resolved in v1.4.1, thanks for the suggestion!