def test_extract(element, expr, expect, build_first):
extractor = JSONExtractor(expr)
assert not extractor.built
if build_first:
extractor.build()
assert extractor.built
assert expect == extractor.extract(element)
assert extractor.built
def test_extract_first_without_default(element, expr, build_first):
extractor = JSONExtractor(expr)
if build_first:
extractor.build()
with pytest.raises(ExtractError) as catch:
extractor.extract_first(element)
exc = catch.value
assert len(exc.extractors) == 1
assert exc.extractors[0] is extractor
assert exc.element is element
def test_invalid_css_selector_expr(element, expr, by):
extractor = JSONExtractor(expr)
with pytest.raises(ExprError) as catch:
if by == "build":
extractor.build()
elif by == "extract":
extractor.extract(element)
exc = catch.value
assert exc.extractor is extractor
if (data_extractor.json.json_extractor_backend is
data_extractor.json.JSONPathExtractor):
# JSONExtractor implementated by 'jsonpath-extractor'
# only raise SyntaxError in build method
assert isinstance(exc.exc, SyntaxError)
else:
from jsonpath_rw.lexer import JsonPathLexerError
assert isinstance(exc.exc, (JsonPathLexerError, Exception))
assert re.match(r"ExprError with .+? raised by .+? extracting", str(exc))
def test_extract_first(element, expr, expect, build_first):
extractor = JSONExtractor(expr)
if build_first:
extractor.build()
assert expect == extractor.extract_first(element, default="default")
