def main():
rx = bd.BeeStringDict('RX', keysize=9, readonly=1)
rx_set = bd.BeeStringDict('RXset', keysize=9, readonly=0)
count = -1
empty = 0
keys = sorted(rx.keys())
print len(keys)
for key in keys:
line = rx.get(key, '')
if len(line) > 9:
subs = line.split('|')
codes = []
for sub in subs:
code = sub.split(':')[0]
if len(code) > 4:
codes.append(code)
#print codes
rx_set[key] = set(codes)
rx_set.commit()
else:
empty += 1
count += 1
if count % 10000 == 0:
print rich_string_wrap(num2comma2(count),'y',0,'k',0), \
rich_string_wrap(num2comma2(empty),'r',0,'k',0), ('%4.3f ' % (float(empty+0.5)/float(count+0.5)))
def process_pat_worker(in_queue, out_queue, plock, process_function, db_list,
**kwargs):
for db in db_list:
kwargs[db] = bd.BeeStringDict(db_list[db][0] + db,
keysize=db_list[db][1],
readonly=True)
#people_dx = bd.BeeStringDict('kanix_matches_F_Breast_Cancer.txt_demo_bd',
# keysize = 9, readonly=True)
#kwargs['person_dx_info'] = people_dx ### BADDD
#rx = bd.BeeStringDict(marketscan+'RX', keysize = 9, readonly=True)
#dx = bd.BeeStringDict(marketscan+'DX', keysize = 9, readonly=True)
#demo = bd.BeeStringDict(marketscan+'ID_demographics', keysize = 9, readonly=True)
#rxset = bd.BeeStringDict(marketscan+'RXSet', keysize = 9, readonly=True)
#dx = bd.BeeStringDict(marketscan+'DX', keysize = 9, readonly=True)
## iter = reads until get "None"
for (patline, index, tot_pat) in iter(in_queue.get, None):
if index % 50000 == 0:
plock.acquire()
print 'AT {:1.2f} %\n'.format(index / float(tot_pat) * 100)
sys.stdout.flush()
plock.release()
result = process_function(patline, **kwargs)
if not result is None:
out_queue.put(result)
out_queue.put(None)
plock.acquire()
print 'child: is in q empty? ' + str(in_queue.empty())
plock.release()