Created
September 20, 2011 18:17
-
-
Save lrvick/1229850 to your computer and use it in GitHub Desktop.
Import remote gzipped Sqlite3 SQL file as local sqlite3 database
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#The goal is to emulate the following bash line properly in python: | |
#wget -O - "https://github.com/downloads/Tawlk/synt/sample_data.bz2" | bzcat | sqlite3 sample_data.db | |
import bz2 | |
import sqlite3 | |
import time | |
import urllib2 | |
import os | |
from cStringIO import StringIO | |
response = urllib2.urlopen('https://github.com/downloads/Tawlk/synt/sample_data.bz2') | |
total_bytes = int(response.info().getheader('Content-Length').strip()) | |
saved_bytes = 0 | |
start_time = time.time() | |
last_seconds = 0 | |
last_seconds_start = 0 | |
data_buffer = StringIO() | |
decompressor = bz2.BZ2Decompressor() | |
if not os.path.exists('~/.synt'): | |
os.makedirs('~/.synt/') | |
if os.path.exists('~/.synt/samples.db'): | |
os.remove('~/.synt/samples.db') | |
conn = sqlite3.connect('~/.synt/samples.db') | |
prcount=0 | |
def import_progress(): | |
global prcount, sql_lines | |
prcount +=20 | |
print "Processed %s of ~2000000 records" % (prcount) | |
return 0 | |
conn.set_progress_handler(import_progress,20) | |
while True: | |
seconds = (time.time() - start_time) | |
chunk = response.read(8192) | |
if not chunk: | |
break | |
saved_bytes += len(chunk) | |
data_buffer.write(decompressor.decompress(chunk)) | |
if seconds > 1: | |
percent = round((float(saved_bytes) / total_bytes)*100, 2) | |
speed = round((float(total_bytes / seconds ) / 1024),2) | |
speed_type = 'Kb/s' | |
if speed > 1000: | |
speed = round((float(total_bytes / seconds ) / 1048576),2) | |
speed_type = 'Mb/s' | |
if last_seconds >= 0.5: | |
last_seconds = 0 | |
last_seconds_start = time.time() | |
print("Downloaded %d of %d Mb, %s%s (%0.2f%%)\r" % (saved_bytes/1048576, total_bytes/1048576, speed, speed_type, percent)) | |
else: | |
last_seconds = (time.time() - last_seconds_start) | |
if saved_bytes == total_bytes: | |
print("Downloaded %d of %d Mb, %s%s (100%%)\r" % (saved_bytes/1048576, total_bytes/1048576, speed, speed_type)) | |
try: | |
conn.executescript(data_buffer.getvalue()) | |
except Exception, e: | |
print("Sqlite3 import failed with: %s" % e) | |
break |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment