SciCrunch / sparc-curation / 591
2%

DEFAULT BRANCH: master
Ran
Jobs 1
Files 34
Run time 3s
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
591

push

travis-ci

tgbugs 
manifest file metadata and xml files now in the json dataset export

will likely need to break "path_metadata" into its own file

temporary caching of individual dataset exports by writing to disk
from the worker process, much faster to load them back for debugging
using -j1 since there is nasty performance degradation with joblib

Tabular detect potential cases where csv/tsv don't match the separator

pipes XmlFilePipeline implemented, not the best, but it will do for now
also quite a few updates to PipelineExtras for all of this

https://github.com/SciCrunch/sparc-curation/issues/54 seems to have been
caused by a lable/label spelling error, but that bug a recent addition
and thus is unlikely to be the cause in the current production export,
so I am not going to close the issue yet

manifest and xml file record processing is currently handled in
derives, this seems silly, but will have to wait for sparcur.simple to
resolve the issue I suspect, scaffolds are also processed in derived

Path project_relative_path and dataset_relative_path

_jsonMetadata add error if path does not exist, this is by far the
easiest way to do the validation of manifest files

schemas fixed bug where @context @base seemed to be added twice, but
it was actually because sc.base_context was being mutated in place, so
we deepcopy it now, also fixed bug where manifest_records was required
at the top level of a manifest file instead of inside of content, for
the mbf metadata schema changed extracted to contents for consistency

upgraded ManifestFileSchema to be like the other not-export versions

bump augpathlib dep for find_cach_root broken symlink fix

have to make sure to set BlackfynnCache._anchor in joblib processes
now since we actually make use of it, amazingly up to this point we
didn't actually hit any codepaths in the pipelines that made use of
the anchor, probably because everything was internal to a single
dataset, we will need to make sure that these kinds of things ... (continued)

1982 of 10079 relevant lines covered (19.66%)

0.2 hits per line

Jobs
ID Job ID Ran Files Coverage
2 591.2 (SCIGRAPH_API=https://scicrunch.org/api/1/sparc-scigraph SCICRUNCH_API_KEY=[secure]) 0
19.66
 Travis Job 591.2
Source Files on build 591
Detailed source file information is not available for this build.