gistfile1.txt

from msgspec import json, Struct, Meta
from typing import Annotated
from msgspec import Struct, Meta

import re

_VCF_REGEX = ".*\.vcf(\.(gz|lz4|zstd|bzip2))?$"

class AncestrySubmission(Struct):
  vcf_path: Annotated[str, Meta(pattern=_VCF_REGEX)]


test_json = '{"vcf_path": "test.vcf"}'

json.decode(test_json)
# Out[15]: {'vcf_path': 'test.vcf'}

json.decode(test_json, type=AncestrySubmission)
# Out[19]: AncestrySubmission(vcf_path='test.vcf')

bad_test_json = '{"vcf_path": "test.blah"}'

json.decode(bad_test_json, type=AncestrySubmission)
# ---------------------------------------------------------------------------
# ValidationError                           Traceback (most recent call last)
# Cell In[21], line 1
# ----> 1 json.decode(bad_test_json, type=AncestrySubmission)

# ValidationError: Expected `str` matching regex '.*\\.vcf(\\.(gz|lz4|zstd|bzip2))?$' - at `$.vcf_path`