def download_query(self,
query,
table_name,
save_to='data',
context='MyDB'):
"""Perform a query based on the context given"""
if context != 'MyDB':
jobid = cj.submitJob(query, context=context)
cj.waitForJob(jobid, verbose=True)
job_status = cj.getJobStatus(jobid)
if job_status['Success'] != 5:
raise CasJobsError(
'Error Performing the query, {}'.format(job_status))
# Now the results are safely saved in myDB
# You can go ahead and download by using a session
self.download_csv(table_name)
self._download_from_scidrive(table_name, save_to)
def testCasJobsSubmit(token):
tbl="mriiscplanck1_"+str(randint(0,1000000))
sql="""select top 10 galaxyId,snapnum,stellarmass into """+tbl+""" from MRIIscPlanck1"""
jobId= CasJobs.submitJob(sql,context="Henriques2015a",token=token)
print("jobId=",jobId)
jobDesc = CasJobs.waitForJob(jobId)
print( jobDesc)
def test_CasJobs_submitJob(self):
jobId = CasJobs.submitJob(sql=CasJobs_TestQuery + " into MyDB." +
CasJobs_TestTableName1,
context=CasJobs_TestDatabase)
jobDescription = CasJobs.waitForJob(jobId=jobId, verbose=True)
df = CasJobs.executeQuery(sql="DROP TABLE " + CasJobs_TestTableName1,
context="MyDB",
format="csv")
self.assertNotEqual(jobId, "")
def createQueries(username, password, amount, table_name, context):
CasJobs_Query = "SELECT top " + str(
amount
) + " statement into [myscratch:default]." + table_name + " FROM sdssweblogs.SqlLog WHERE CHARINDEX('FROM PhotoObjAll', statement) > 0 AND CHARINDEX('p.dec BETWEEN', statement) > 0 AND CHARINDEX('p.ra BETWEEN', statement) > 0 AND access='Skyserver.Search.SQL'"
login(username, password)
jobId = CasJobs.submitJob(sql=CasJobs_Query, context=context)
jobDescription = CasJobs.waitForJob(jobId=jobId, verbose=True)
print(
"Data has been saved on sciserver at context myscratch with tablename "
+ table_name)
def createData(username, password, dec_low, dec_high, ra_low, ra_high,
table_name, context, all_data):
CasJobs_Database = "DR16"
CasJobs_Query = "Select ra,dec into [myscratch:default]." + table_name + " from dr16.photoobjall "
if (not all_data):
CasJobs_Query += "where dec between " + str(dec_low) + " and " + str(
dec_high) + " and ra between " + str(ra_low) + " and " + str(
ra_high)
login(username, password)
jobId = CasJobs.submitJob(sql=CasJobs_Query, context=context)
jobDescription = CasJobs.waitForJob(jobId=jobId, verbose=True)
print(
"Data has been saved on sciserver at context myscratch with tablename "
+ table_name)
tname = 'test'
data = CasJobs.getTables('MyDB')
if data:
print(data)
print(
'There are already tables in your database under that name, removing them before doing new query...'
)
#CLEAN UP: delete table from myDB
SkyQuery.dropTable(tableName=tname, datasetName='myDB')
#long asynchronous job
print(
'submitting SQL job to SciServer... could take some time depending on query length and SciServer load...'
)
jobID = CasJobs.submitJob(query + "into myDB." + tname, "dr14")
CasJobs.waitForJob(jobID, verbose=True)
try:
#Download table from MYDB into Pandas dataframe. additional parms: , top=10
print('Attempting to download table...')
data_table = SkyQuery.getTable(tableName=tname, datasetName='MyDB')
print('Done! Table shape is: ' + str(data_table.shape))
#save df to disk in case of super long queries:
filename = 'test_query_table_1000'
print('Saving tables to disk as: ' + filename)
save_obj(data_table, filename)
except:
print(
'ERROR, No data found in your SciServer database, query may not have completed, check your SQL syntax?'
)
#see tables in MyDB
def test_CasJobs_waitForJob(self):
jobId = CasJobs.submitJob(sql=CasJobs_TestQuery,
context=CasJobs_TestDatabase)
jobDescription = CasJobs.waitForJob(jobId=jobId, verbose=True)
self.assertGreaterEqual(jobDescription["Status"], 3)
# In[ ]: #execute a quick SQL query: df = CasJobs.executeQuery(sql=CasJobs_TestQuery, context=CasJobs_TestDatabase, format="pandas") print(df) # In[ ]: #submit a job, which inserts the query results into a table in the MyDB database context. #Wait until the job is done and get its status. jobId = CasJobs.submitJob(sql=CasJobs_TestQuery + " into MyDB." + CasJobs_TestTableName1, context="MyDB") jobDescription = CasJobs.waitForJob(jobId=jobId, verbose=False) print(jobId) print(jobDescription) # In[ ]: # drop or delete table in MyDB database context df = CasJobs.executeQuery(sql="DROP TABLE " + CasJobs_TestTableName1, context="MyDB", format="pandas") print(df) # In[ ]: #get job status
# Example of a longer query: get magnitudes and sizes (Petrosian radii) of one million galaxies
verylongquery = 'select top 10 objid, ra, dec \n'
verylongquery += 'u, g, r, i, z, err_u, err_g, err_r, err_i, err_z, petror90_r \n'
verylongquery += 'into mydb.' + bigtablename + '\n'
verylongquery += 'from galaxy\n'
verylongquery += 'where clean = 1'
print('Submitting query:\n', verylongquery)
print('\n')
thisjobid = CasJobs.submitJob(sql=verylongquery, context=this_context)
print('Job submitted with jobId = ', thisjobid)
print('\n')
waited = CasJobs.waitForJob(
jobId=thisjobid) # waited is a dummy variable; just print wait msg
jobDescription = CasJobs.getJobStatus(thisjobid)
print('\n')
print('Information about the job:')
#pprint(jobDescription)
jobDescriber(jobDescription)
# ## Thank you!
#
# Thanks for reviewing this SciServer example notebook. You can use this notebook as a template to develop your own notebooks, but please do so in a copy rather than in the original example notebook.
# As you begin to use any of our SciServer modules in your own notebooks, consult the SciServer scripting documentation at http://www.sciserver.org/docs/sciscript-python/SciServer.html (link opens in a new window).
#
# If you have questions, please email the SciServer helpdesk at [email protected].
sql_drop = """IF OBJECT_ID('spectrain') IS NOT NULL
DROP TABLE spectrain"""
CasJobs.executeQuery(sql=sql_drop, context='MYDB', format="pandas")
sql_get = """
select * from MYDB.NIPSspecphoto order by objid
{}
OFFSET {} ROWS FETCH NEXT 1407000 ROWS ONLY;""".format('', 1000000 * ii)
# ii+=1
df = CasJobs.executeQuery(sql='SELECT * FROM N15',
context='MYDB',
format="pandas")
jobid = CasJobs.submitJob(sql=sql_spec, context='DR13')
CasJobs.waitForJob(jobid)
df = CasJobs.executeQuery(sql=sql_spec, context='DR13', format="pandas")
df = CasJobs.executeQuery(sql='SELECT * FROM MYDB.spectrain',
context='MYDB',
format="pandas")
df.shape
df.to_csv('../data/sdss_stars/DR13/544k_spec_objid.csv', index=False)
'''
select class, subclass, u-g as ug, g-r as gr, r-i as ri, i-z as iz, u-r as ur, g-i as gi, r-z as rz, u-i as ui, g-z as gz, u-z as uz,
(u-15)/7 as u, (g-15)/7 as g, (i-15)/7 as i, (z-15)/7 as z, (r-15)/5 as r,
u-0.5*ext_u as u05, g-0.5*ext_g as g05, r-0.5*ext_r as r05, i-0.5*ext_i as i05, z-0.5*ext_z as z05,
u-0.25*ext_u as u25, g-0.25*ext_g as g25, r-0.25*ext_r as r25, i-0.25*ext_i as i25,z-0.25*ext_z as z25,
