如何减少python中大列表使用的内存

Question

I'm writing a program, it works fine but when it loads the database(a 100MB text file) to a list it's memory usage becomes 700-800MB

Code used to load the file to a list:

database = []
db = open('database/db.hdb')
dbcontent = db.read()
for line in dbcontent.split('\n'):
    line = line.split(':')
    database.append(line)

Snipplet from db.hdb:

14200:917cb8a3d1d9eb24af6c5bcf3bf7e401:Trojan.Downloader-1420
7168:a105e2cc8148158cd048360eb847c7d0:Trojan.Downloader-1421
7168:c61ef67b5e7eef19ef732f55116742f6:Trojan.Downloader-1422
7168:851b6320148122104f50445ea2684c9f:Trojan.Downloader-1423
7168:ca128383c79a56d930eb4a7ff5026e31:Trojan.Downloader-1424
355204:4af89f8d219f94462cf2f8cb8eb4c6d7:Trojan.Bancos-2053
356984:2bfb53d76891059b79122e13d1537e4a:Trojan.Bancos-2054
363520:edbbdf497cda1ba79c06ea40673d963e:Trojan.Bancos-2055
367616:d85f719b032dbf39800d90ca881fd225:Trojan.Bancos-2056
370688:6cb572fd2452416dc4ea09e3ad917e66:Trojan.Bancos-2057
370688:ef34885677230061649d30ea66d7b0a1:Trojan.Bancos-2058
399360:8578b664706cfdc2f653680bac1b1b6e:Trojan.Bancos-2059
401408:de62af250b5a3e1ba1e9c517629383dd:Trojan.Bancos-2060
622592:8a236340c0a8c76343f6fb581314fadf:Trojan.Bancos-2061
622592:29f3499488ba1814c62fac3c2f3bda54:Trojan.Bancos-2062
622592:5d023bccf2ff097ccbc0ab0eab4a6ee7:Trojan.Bancos-2063
622592:3d6a25ed1f0e2001e72812ce1adf37d3:Trojan.Bancos-2064
622592:eaff242b601807e5805c189752d39124:Trojan.Bancos-2065
623104:8cd8e788d33cf40412d3346a525e4cce:Trojan.Bancos-2066
625152:25470d6895cb0e5c2e7181cb9a201ae0:Trojan.Bancos-2067
625152:436d574cef37b2e62d9b801b8fc2c4f1:Trojan.Bancos-2068
647168:51eb4e43f24cf511e6715cc8667babcd:Trojan.Bancos-2069

(The full file has ~1800000 lines)

How do I decrease the memory usage

Answer 1

You should use the file object as an iterator to reduce memory usage from file. You could then process database list in chunks rather than all together. For example:

results = []
database = []
for line in open("database/db.hdb"):
    line = line.split(':')
    #You could then manage database in chunks?
    database.append(line)
    if len(database) > MAX:
        #dosomething with database list so far to get result
        results.append(process_database(database))
        database = []
#do something now with individual results to make one result
combine_results(results)

Answer 2

As long as you don't need the complete file in memory, you could read one line at a time:

database = []
db = open('database/db.hdb')
line = db.readline()
while line:
    line = line.split(':')
    database.append(line)
    line = db.readline()

See here for details on file.readline()