def sampleWordIDs(numSamples=1, wordLength=5):
# TODO: Implement limits:
# 1) data have same periodicity
# 2) data come from certain supertype
# 2) data come from certain subtype
# 2) data come from certain type
# 2) data come from certain data_master_ids
# 2) data come from differing data_master_ids (i.e. no way to have same data_master_id)
allIDs = EM_DBHelp.retrieve_AllWordSeriesIDs(db_connection, db_cursor)
words = []
for i in range(numSamples):
words.append(sample(allIDs, wordLength))
wordSeriesIDs = {ID for word in words for ID in word}
getFromDB = lambda (seriesID): EM_DBHelp.retrieve_WordSeriesMetaData(
db_connection, db_cursor, columnName, seriesID=seriesID
)
columnName = "dt_latest_word"
dates = [getFromDB(ID) for ID in wordSeriesIDs]
latestDate = min(dates)
columnName = "dt_earliest_word"
dates = [getFromDB(ID) for ID in wordSeriesIDs]
earliestDate = max(dates)
columnName = "int_data_master_id"
dataSeriesIDs = [getFromDB(ID) for ID in wordSeriesIDs]
return (words, wordSeriesIDs, dataSeriesIDs, earliestDate, latestDate)
