def test_merge(self):
p1 = self.person('Eve 1')
p2 = self.person('Eve 2')
merger = CromObjectMerger()
merger.merge(p1, p2)
referrers = p1.referred_to_by
self.assertEqual(len(referrers), 1)
def test_pipeline_sales(self):
'''
When dimensions get merged, the Unknown physical dimension classification (300055642)
gets dropped if there are any other classifications.
'''
h1 = vocab.Height(ident='', content=9.0)
h1.unit = vocab.instances.get('inches')
self.assertEqual({c._label for c in h1.classified_as}, {'Height'})
h2 = vocab.PhysicalDimension(ident='', content=9.0)
self.assertEqual({c._label for c in h2.classified_as}, {'Unknown physical dimension'})
merger = CromObjectMerger()
h = merger.merge(h1, h2)
self.assertEqual({c._label for c in h.classified_as}, {'Height'})
def merge_objects(self, objects):
r = JSONValueRewriter(self.prev_post_sales_map)
for k in list(objects.keys()):
data = objects[k]
updated = r.rewrite(data)
ident = updated['id']
if k != ident:
if ident in objects:
read = reader.Reader()
m = read.read(json.dumps(objects[ident]))
n = read.read(json.dumps(updated))
merger = CromObjectMerger()
m = merger.merge(m, n)
objects[ident] = json.loads(factory.toString(m, False))
else:
objects[ident] = updated
del (objects[k])
def _rewrite_output_files(files, r, update_filename, worker_id, total_workers,
kwargs):
i = 0
if not files:
return
print(
f'rewrite worker partition {worker_id} called with {len(files)} files [{files[0]} .. {files[-1]}]'
)
start = time.time()
rewritten_count = 0
processed_count = 0
ignore_errors = kwargs.get('ignore_errors', False)
for i, f in enumerate(files):
processed_count += 1
# print(f'{i} {f}', end="\r", flush=True)
with open(f) as data_file:
try:
bytes = data_file.read()
if 'content_filter_re' in kwargs:
filter_re = kwargs['content_filter_re']
if not re.search(filter_re, bytes):
pass
# print(f'skipping {f}')
continue
else:
pass
# print(f'processing {f}')
data = json.loads(bytes)
except json.decoder.JSONDecodeError:
sys.stderr.write(
f'Failed to load JSON during rewriting of {f}\n')
if ignore_errors:
continue
else:
raise
d = r.rewrite(data, file=f)
if update_filename:
newfile = filename_for(d, original_filename=f, **kwargs)
else:
newfile = f
if d == data and f == newfile:
# nothing changed; do not rewrite the file
continue
else:
pass
# print(f'*** rewrote data in {f} --> {newfile}')
if newfile != f:
if os.path.exists(newfile):
read = reader.Reader()
merger = CromObjectMerger()
with open(newfile, 'r') as fh:
content = fh.read()
try:
m = read.read(content)
n = read.read(d)
# print('========================= MERGING =========================')
# print('merging objects:')
# print(f'- {m}')
# print(f'- {n}')
merger.merge(m, n)
# except model.DataError as e:
except Exception as e:
print(
f'Exception caught while merging data from {newfile} ({str(e)}):'
)
print(d)
print(content)
if ignore_errors:
continue
else:
raise
data = factory.toString(m, False)
d = json.loads(data)
with open(newfile, 'w') as data_file:
rewritten_count += 1
json.dump(d, data_file, indent=2, ensure_ascii=False)
if newfile != f:
os.remove(f)
end = time.time()
elapsed = end - start
if rewritten_count:
print(
f'worker partition {worker_id}/{total_workers} finished with {rewritten_count}/{processed_count} files rewritten in %.1fs'
% (elapsed, ))
else:
print(
f'worker partition {worker_id}/{total_workers} finished in %.1fs' %
(elapsed, ))
with open(filename, 'r') as fh:
content = fh.read()
canon_file = None
canon_content = None
try:
m = read.read(content)
id = m.id
if id in seen:
canon_file = seen[id]
# print(f'*** {id} already seen in {canon_file} ; merging {filename}')
merger = CromObjectMerger()
with open(canon_file, 'r') as cfh:
canon_content = cfh.read()
n = read.read(canon_content)
try:
merger.merge(m, n)
except model.DataError as e:
print(
f'Exception caught while merging data from {newfile} ({str(e)}):'
)
print(d)
print(content)
raise
merged_data = factory.toString(m, False)
d = json.loads(merged_data)
with open(canon_file, 'w') as data_file:
json.dump(d,
data_file,
indent=2,
ensure_ascii=False)
os.remove(filename)