def test_overlapping_error():
"""Test that important values within the lose list cause an exception."""
results = [[Meta2()]]
lose = ["key2-3"]
with pytest.raises(OverlappingLoseAndImportantException) as e_info:
generate_metadata_dict(results, lose)
assert ("The given lose dict contains values that are marked as important"
in six.text_type(e_info.value))
def test_generate_metadata_dict():
"""Test generating metadata dict using the metadata objects."""
results = [[Meta1()], [Meta2()], [Meta3()]]
lose = ["value2-1"]
metadata_dict = generate_metadata_dict(results, lose)
assert metadata_dict == {
0: {
"index": 0,
"key1": "value1-1",
"key2": "value2",
"key3": "key2-3",
"key4": "importantvalue",
"mimetype": "mime",
"version": 1.0,
"stream_type": "binary"
},
1: {
"index": 1,
"key1": "value1",
"key2": "value2",
"mimetype": "anothermime",
"version": 2,
"stream_type": "audio"
}
}
def test_generate_metadata_dict(meta_class_fx, meta_classes, lose, valid_dict,
expected_conflicts):
"""Test generating metadata dict using the metadata objects.
Tests both a successful case and a case with conflicts in
metadata, both while filling import values and while merging the
results.
"""
results = []
for meta_class in meta_classes:
results.append([meta_class_fx(meta_class)])
(metadata_dict, conflicts) = generate_metadata_dict(results, lose)
if valid_dict:
assert metadata_dict == {
0: {
"index": 0,
"key1": "value1-1",
"key2": "value2",
"key3": "key2-3",
"key4": "importantvalue",
"mimetype": "mime",
"version": 1.0,
"stream_type": "binary"
},
1: {
"index": 1,
"key1": "value1",
"key2": "value2",
"mimetype": "anothermime",
"version": 2,
"stream_type": "audio"
}
}
assert conflicts == expected_conflicts
def scrape_file(self):
"""
No need to scrape anything, just merge already collected metadata.
"""
streams, conflicts = generate_metadata_dict(self._scraper_results,
LOSE)
self.streams = streams
for error_message in conflicts:
self._errors.append(error_message)
self._messages.append("Scraper results merged into streams")
def scrape(self, check_wellformed=True):
"""Scrape file and collect metadata.
:check_wellformed: True, full scraping; False, skip well-formed check.
"""
self.detect_filetype()
# File not found or MIME type could not be determined
if not self.mimetype:
self.streams = {}
return
self._params["mimetype_guess"] = self.mimetype
for scraper_class in iter_scrapers(mimetype=self.mimetype,
version=self.version,
check_wellformed=check_wellformed,
params=self._params):
scraper = scraper_class(self.filename, check_wellformed,
self._params)
self._scrape_file(scraper)
self.streams = generate_metadata_dict(self._scraper_results, LOSE)
self._check_utf8(check_wellformed)
self._check_mimetype_version()