In [3]:
# header for 2018-1 kernel
from pyCHX.chx_packages import *
%matplotlib notebook
plt.rcParams.update({'figure.max_open_warning': 0})
plt.rcParams.update({ 'image.origin': 'lower' })
plt.rcParams.update({ 'image.interpolation': 'none' })
import pickle as cpk
#from pyCHX.chx_xpcs_xsvs_jupyter_V1 import *
#%run /home/yuzhang/pyCHX_link/pyCHX/chx_generic_functions.py
#%matplotlib inline
import papermill as pm
In [4]:
# import database -> should be hidden from user in same package....
import datetime
import pymongo
from tqdm import tqdm
from bson import ObjectId
import matplotlib.patches as mpatches
from IPython.display import clear_output
cli = pymongo.MongoClient('xf11id-ca')
samples_2 = cli.get_database('samples').get_collection('samples_2')
data_acquisition_collection = cli.get_database('samples').get_collection('data_acquisition_collection')
beamline_pos = cli.get_database('samples').get_collection('beamline_pos')
from databroker import Broker
db = Broker.named('temp') # for real applications, 'temp' would be 'chx'
print('available databases:')
print(cli.database_names())
print('\n available collection in database samples:')
print(cli.samples.collection_names())
In [ ]:
In [ ]:
In [36]:
def _chx_analysis_data( uid,
#template_pipeline = '/home/yuzhang/analysis/2019_1/commisionning/XPCS_Single_2019_V1_SQRange.ipynb',
template_pipeline = '/home/yuzhang/analysis/2019_1/petrash/XPCS_Single_2019_V2.ipynb',
outDir = '/home/yuzhang/analysis/2019_1/petrash/ResPipelines/',
):
''' YG. Octo 6, 2018, Compress a eiger data using papermail
Input:
uid: string, the uique data id
force compress: if True, will force to compress data no matter the data was compressed already
The default compress pipeline
template_pipeline: str, the filename of the template pipeline
outDir:str, the path for the output pipeline
Output:
save the current pipeline to outDir
'''
output_pipeline = outDir + template_pipeline.split('/')[-1] + '_%s.ipynb'%uid
pm.execute_notebook(
template_pipeline, output_pipeline,
parameters = dict( uid = uid, ),
kernel_name='python3', report_mode=True )
In [37]:
if True:
temp1 = data_acquisition_collection.find_one({'_id':'general_list'})['compression_completed']
temp2= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_completed']
#temp3= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_failed']
temp4= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_failed_userX']
s2 = set(temp2)
#s3 = set(temp3)
s4 = set(temp4)
#######################################
#######Get uids to be compressed#######
uids = [x for x in temp1 if x not in s2 and x not in s4]
print(uids)
In [39]:
#temp1[-10:]
In [35]:
#temp4[-10:]
In [40]:
#temp2[-10:]
In [5]:
#not_done=[x for x in temp1 if x not in s2 and x not in s4]
#not_done
In [42]:
temp1[-1], temp2[-1], temp4[-1]
Out[42]:
In [43]:
def get_masked_analysis_database( start_uid ):
'''Give the uid for the first running and get the masked database'''
temp1 = data_acquisition_collection.find_one({'_id':'general_list'})['compression_completed']
for i, t in enumerate(temp1):
if t == start_uid:
print(i,t)
start_id = i
masked = data_acquisition_collection.update_one(
{'_id':'general_list'},{'$set':{'analysis_failed_userX': temp1[:start_id] } })
In [44]:
start_fuid = '2494fe5e-6d1a-4c18-805a-57c6e7d34acc' #'25171c35-ce50-450b-85a0-ba9e116651e3'
get_masked_analysis_database( start_uid = start_fuid )
In [9]:
#data_acquisition_collection.update_one( {'_id':'general_list'},{'$set':{'analysis_failed_petrash': temp1[:797] } })
data_acquisition_collection.update_one( {'_id':'general_list'},{'$set':{'analysis_in_progress': [] } })
Out[9]:
In [10]:
data_acquisition_collection.find_one({'_id':'general_list'})['analysis_in_progress']
Out[10]:
In [ ]:
In [11]:
#temp1
In [12]:
#temp4.append(['2891d947-58b1-4db8-9b7e-c93f94259e78', 'f34aceea-dacb-439f-9e91-94bc37156354', '0639c0c3-7257-436f-9bfe-b964830e3823'] )
#data_acquisition_collection.update_one({'_id':'general_list'},{'$set':{'analysis_failed_Surita':tem}})
#data_acquisition_collection.update_one({'_id':'general_list'},{'$set':{'analysis_failed':[]}})
In [ ]:
uids
In [ ]:
# get list of uids
end_of_compression_key='none' # stops if only this key is left in compressed_uid_list, 'none': not looking for key, just for empty list timeout
empty_list_timeout= 3600 * 24 * 7 #[s] stops if compressed_uid_list is empty for x s
time_count=0
run_condition = True
while run_condition:
clear_output()
temp1 = data_acquisition_collection.find_one({'_id':'general_list'})['compression_completed']
temp2= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_completed']
temp3= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_failed_userX']
temp4 = data_acquisition_collection.find_one({'_id':'general_list'})['analysis_in_progress']
#data_acquisition_collection.find_one({'_id':'general_list'})['analysis_failed']
s2 = set(temp2)
s3 = set(temp3)
s4 = set( temp4 )
#######################################
#######Get uids to be compressed#######
uids = [x for x in temp1 if x not in s2 and x not in s3]
######################################
if uids: # list of uids is not empty
print('uid list for analysis is NOT empty, found '+str(len(uids))+' uids awaiting analysis.')
time_count=0
if end_of_compression_key != 'none' and uids[0] == end_of_compression_key: #looking for a stop key, next uid up IS the stop key
run_condition = False
print('Stop Key for analysis detected!')
else:
print('Doing data analysis for uid '+uids[0])
#######################################
#for ics in tqdm(range(100)):
# time.sleep(.35)
########################################
uid = uids[0]
if uid not in s4:
try:
temp4.append( uid )
data_acquisition_collection.update_one({'_id':'general_list'},
{'$set':{ 'analysis_in_progress': temp4}})
_chx_analysis_data( uid )
# update list of compressed uids:
temp2= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_completed']
temp2.append(uids[0])
data_acquisition_collection.update_one({'_id':'general_list'},
{'$set':{ 'analysis_completed': temp2}})
except:
temp3= data_acquisition_collection.find_one({'_id':'general_list'})['analysis_failed_userX']
temp3.append(uids[0])
data_acquisition_collection.update_one({'_id':'general_list'},{'$set':{ 'analysis_failed_userX': temp3}})
##remove this uid from uid in analysis_in_progress
temp4 = data_acquisition_collection.find_one({'_id':'general_list'})['analysis_in_progress']
tx = [u for u in temp4 if u!=uid ]
data_acquisition_collection.update_one( {'_id':'general_list'},
{'$set':{'analysis_in_progress': tx } })
else:
if time_count > empty_list_timeout:
print('uid list for analysis was empty for > '+str(empty_list_timeout)+'s -> stop looking for new uids')
run_condition = False
else:
time_count=time_count+5
print('list of uids for analysis is emtpy...going to look again in 5s.')
time.sleep(5)
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]: