100 lines
3.4 KiB
Python
100 lines
3.4 KiB
Python
###########################################################################
|
|
# Copyright 2019 (C) Hui Lan <hui.lan@cantab.net>
|
|
# Written permission must be obtained from the author for commercial uses.
|
|
###########################################################################
|
|
|
|
|
|
# Purpose: dictionary & pickle as a simple means of database.
|
|
# Task: incorporate the functions into wordfreqCMD.py such that it will also show cumulative frequency.
|
|
# Note: unlike pick_idea.py, now the second item is not frequency, but a list of dates.
|
|
|
|
import pickle
|
|
from datetime import datetime
|
|
|
|
def lst2dict(lst, d):
|
|
'''
|
|
Store the information in list lst to dictionary d.
|
|
Now stores frequency count instead of dates list.
|
|
'''
|
|
for x in lst:
|
|
word = x[0]
|
|
if isinstance(x[1], list): # if it's a list of dates
|
|
count = len(x[1]) # convert to frequency
|
|
else:
|
|
count = x[1] # already a frequency
|
|
|
|
if not word in d:
|
|
d[word] = count
|
|
else:
|
|
d[word] += count
|
|
|
|
def deleteRecord(path,word):
|
|
with open(path, 'rb') as f:
|
|
db = pickle.load(f)
|
|
try:
|
|
db.pop(word)
|
|
except KeyError:
|
|
print("sorry")
|
|
with open(path, 'wb') as ff:
|
|
pickle.dump(db, ff)
|
|
|
|
def dict2lst(d):
|
|
if len(d) > 0:
|
|
keys = list(d.keys())
|
|
if isinstance(d[keys[0]], int):
|
|
return list(d.items()) # return (word, frequency) pairs directly
|
|
elif isinstance(d[keys[0]], list):
|
|
return [(k, len(v)) for k, v in d.items()] # convert date lists to counts
|
|
|
|
return []
|
|
|
|
def merge_frequency(lst1, lst2):
|
|
d = {}
|
|
lst2dict(lst1, d)
|
|
lst2dict(lst2, d)
|
|
return d
|
|
|
|
|
|
def load_record(pickle_fname):
|
|
f = open(pickle_fname, 'rb')
|
|
d = pickle.load(f)
|
|
f.close()
|
|
return d
|
|
|
|
|
|
def save_frequency_to_pickle(d, pickle_fname):
|
|
f = open(pickle_fname, 'wb')
|
|
d2 = {}
|
|
for k in d:
|
|
if not k in exclusion_lst and not k.isnumeric() and not len(k) < 2:
|
|
if isinstance(d[k], list):
|
|
d2[k] = len(d[k]) # store frequency count instead of dates list
|
|
else:
|
|
d2[k] = d[k]
|
|
pickle.dump(d2, f)
|
|
f.close()
|
|
|
|
|
|
exclusion_lst = ['one', 'no', 'has', 'had', 'do', 'that', 'have', 'by', 'not', 'but', 'we', 'this', 'my', 'him', 'so', 'or', 'as', 'are', 'it', 'from', 'with', 'be', 'can', 'for', 'an', 'if', 'who', 'whom', 'whose', 'which', 'the', 'to', 'a', 'of', 'and', 'you', 'i', 'he', 'she', 'they', 'me', 'was', 'were', 'is', 'in', 'at', 'on', 'their', 'his', 'her', 's', 'said', 'all', 'did', 'been', 'w']
|
|
|
|
if __name__ == '__main__':
|
|
# Test 1: Convert dates to frequencies
|
|
lst1 = [('apple',['201910251437', '201910251438']), ('banana',['201910251439'])]
|
|
d = {}
|
|
lst2dict(lst1, d)
|
|
print("Test 1 - Convert dates to frequencies:")
|
|
print(d) # Should show: {'apple': 2, 'banana': 1}
|
|
|
|
# Test 2: Save and load frequencies
|
|
save_frequency_to_pickle(d, 'frequency.p')
|
|
loaded_d = load_record('frequency.p')
|
|
print("\nTest 2 - Load saved frequencies:")
|
|
print(loaded_d) # Should match the previous output
|
|
|
|
# Test 3: Merge frequencies
|
|
lst2 = [('banana',['201910251439']), ('orange', ['201910251440', '201910251439'])]
|
|
lst1 = dict2lst(loaded_d)
|
|
merged_d = merge_frequency(lst2, lst1)
|
|
print("\nTest 3 - Merge frequencies:")
|
|
print(merged_d) # Should show banana with increased frequency
|