braindead downloader

gistfile1.py

import threading, sys
import urlparse, httplib

is_done = False

class Downloader(threading.Thread):
    def __init__(self, url, buf, n, semin, semout):
        self.buf = buf
        self.bufn = n
        self.semin = semin
        self.semout = semout
        scheme, netloc, path, query, fragment = urlparse.urlsplit(url)
        self.host, self.path = netloc, path
        threading.Thread.__init__(self)

    def run(self):
        h = httplib.HTTPConnection(self.host)
        h.request("GET", self.path)
        r = h.getresponse()

        global is_done
        i = 0
        while True:
            self.semin.acquire()
            self.buf[i] = r.read(1024)
            ## when to stop???
            if not self.buf[i]:
                is_done = True
                break
            i = (i+1) % self.bufn
            self.semout.release()
        h.close()

class Writer(threading.Thread):
    def __init__(self, dest, buf, n, semin, semout):
        self.dest = dest
        self.buf = buf
        self.bufn = n
        self.semin = semin
        self.semout = semout
        threading.Thread.__init__(self)

    def run(self):
        global is_done

        i = 0
        f = open(self.dest, "w")
        while True:
            if is_done:
                break
            self.semout.acquire()
            f.write(buf[i])
            i = (i+1) % self.bufn
            self.semin.release()
        f.close()

n = 100
buf = [0 for i in range(n)]
semin = threading.Semaphore(n)
semout = threading.Semaphore(0)

url="http://www.google.com"
dest = "tmp"
if len(sys.argv) > 1:
    url = sys.argv[1]
if len(sys.argv) > 2:
    dest = sys.argv[2]

dl = Downloader(url=url, buf=buf, n=n, semin=semin, semout=semout)
wr = Writer(dest=dest, buf=buf, n=n, semin=semin, semout=semout)
dl.start()
wr.start()

buf = [0 for i in range(n)] => buf = [0] * n

@lyxint, 差不多吧，写过就扔的代码，不需要搞微优化。

主要问题是如何判断已经下载完成。这个 is_done 全局变量会有 race condition。

最好别用0, 用None可以省空间

@lyxint 万恶之源，万恶之源啊