python post with multipart_form-data

 这是一个发送如同 enctype=”multipart/from-data” post格式 的web脚本，这种格式通常用来上传文件，但是通常也用来传一些大的post数据,这是因为它可以突破常规post(指为:application/x-www-form-urlencoded) 的大小限制.

  所以在碰到这类网站的时候，我们想要模拟浏览器发送请求，使用常规post方法 来传送数据显然是不行的，我曾经碰到过许多post这种类型的网站，例如 quora 例如 京东的论坛

这是一个从activestate搞来的实例， 相信能找到这篇文章的你能看懂：

import httplib, mimetypes
 
def post_multipart(host, selector, fields, files):
    """
    Post fields and files to an http host as multipart/form-data.
    fields is a sequence of (name, value) elements for regular form fields.
    files is a sequence of (name, filename, value) elements for data to be uploaded as files
    Return the server's response page.
    """
    content_type, body = encode_multipart_formdata(fields, files)
    h = httplib.HTTP(host)
    h.putrequest('POST', selector)
    h.putheader('content-type', content_type)
    h.putheader('content-length', str(len(body)))
    h.endheaders()
    h.send(body)
    errcode, errmsg, headers = h.getreply()
    return h.file.read()
 
def encode_multipart_formdata(fields, files):
    """
    fields is a sequence of (name, value) elements for regular form fields.
    files is a sequence of (name, filename, value) elements for data to be uploaded as files
    Return (content_type, body) ready for httplib.HTTP instance
    """
    BOUNDARY = '----------ThIs_Is_tHe_bouNdaRY_$'
    CRLF = '\r\n'
    L = []
    for (key, value) in fields:
        L.append('--' + BOUNDARY)
        L.append('Content-Disposition: form-data; name="%s"' % key)
        L.append('')
        L.append(value)
    for (key, filename, value) in files:
        L.append('--' + BOUNDARY)
        L.append('Content-Disposition: form-data; name="%s"; filename="%s"' % (key, filename))
        L.append('Content-Type: %s' % get_content_type(filename))
        L.append('')
        L.append(value)
    L.append('--' + BOUNDARY + '--')
    L.append('')
    body = CRLF.join(L)
    content_type = 'multipart/form-data; boundary=%s' % BOUNDARY
    return content_type, body
 
def get_content_type(filename):
    return mimetypes.guess_type(filename)[0] or 'application/octet-stream'