Ran into a few more snags, just checking if people...
# configs
p
Ran into a few more snags, just checking if people are already aware of these things or if I should create some github issues? • The test profile of pipelines seems to take precedence over anything specified in institutional profiles, no matter the order in which the profiles are provided. •
* --max_time: string [7.days] does not match pattern ^(\d+\.?\s*(s|m|h|d|day)\s*)+$ (7.days)
errors occur when specifying duration as
#.days
, even though this seems to be a valid duration type according to the [nextflow docs](https://www.nextflow.io/docs/latest/script.html#duration). • Related:
max_time = 7.day
and
max_time = '7.day'
are reported differently by the terminal output at the start of a pipeline (
7.day
versus `7d`respectively). Quoted vs unquoted memory directives behave similarly. I haven't been able to determine whether they also behave differently, i.e. if they are both correctly parsed by the
check_max
function, but I fear they might? I tried checking logs and reports, but couldn't figure out what I had to look for. For what it's worth, I've seen both quoted and unquoted memory/time durations being used in various nf-core docs and configs, so I suspect that someone would have noticed if one didn't work as expected?
p
1. Feature, not a bug 2. We are stricter in nf-core when possible, due to wanting to validate with JSON schema via the launch website 3. All are valid. Nextflow does some automatic typing, plus the function tries to cast what it's given into a Nextflow time unit.
p
All right, then I can stop worrying about these šŸ˜‰ Thanks for all the help you (and several others) have been providing me here, it's very much appreciated!
šŸ‘šŸ» 1
PR submitted! šŸŽŠ
šŸŽ‰ 1