Source code for yt.utilities.poster.encode

"""multipart/form-data encoding module

This module provides functions that facilitate encoding name/value pairs
as multipart/form-data suitable for a HTTP POST or PUT request.

multipart/form-data is the standard way to upload files over HTTP"""

__all__ = ['gen_boundary', 'encode_and_quote', 'MultipartParam',
        'encode_string', 'encode_file_header', 'get_body_size', 'get_headers',
        'multipart_encode']

try:
    import uuid
    def gen_boundary():
        """Returns a random string to use as the boundary for a message"""
        return uuid.uuid4().hex
except ImportError:
    import random, sha
[docs] def gen_boundary(): """Returns a random string to use as the boundary for a message""" bits = random.getrandbits(160) return sha.new(str(bits)).hexdigest()
import re, os, mimetypes from yt.extern.six.moves import urllib try: from email.header import Header except ImportError: # Python 2.4 from email.Header import Header
[docs]def encode_and_quote(data): """If ``data`` is unicode, return urllib.quote_plus(data.encode("utf-8")) otherwise return urllib.quote_plus(data)""" if data is None: return None if isinstance(data, unicode): data = data.encode("utf-8") return urllib.quote_plus(data)
def _strify(s): """If s is a unicode string, encode it to UTF-8 and return the results, otherwise return str(s), or None if s is None""" if s is None: return None if isinstance(s, unicode): return s.encode("utf-8") return str(s)
[docs]class MultipartParam(object): """Represents a single parameter in a multipart/form-data request ``name`` is the name of this parameter. If ``value`` is set, it must be a string or unicode object to use as the data for this parameter. If ``filename`` is set, it is what to say that this parameter's filename is. Note that this does not have to be the actual filename any local file. If ``filetype`` is set, it is used as the Content-Type for this parameter. If unset it defaults to "text/plain; charset=utf8" If ``filesize`` is set, it specifies the length of the file ``fileobj`` If ``fileobj`` is set, it must be a file-like object that supports .read(). Both ``value`` and ``fileobj`` must not be set, doing so will raise a ValueError assertion. If ``fileobj`` is set, and ``filesize`` is not specified, then the file's size will be determined first by stat'ing ``fileobj``'s file descriptor, and if that fails, by seeking to the end of the file, recording the current position as the size, and then by seeking back to the beginning of the file. ``cb`` is a callable which will be called from iter_encode with (self, current, total), representing the current parameter, current amount transferred, and the total size. """ def __init__(self, name, value=None, filename=None, filetype=None, filesize=None, fileobj=None, cb=None): self.name = Header(name).encode() self.value = _strify(value) if filename is None: self.filename = None else: if isinstance(filename, unicode): # Encode with XML entities self.filename = filename.encode("ascii", "xmlcharrefreplace") else: self.filename = str(filename) self.filename = self.filename.encode("string_escape").\ replace('"', '\\"') self.filetype = _strify(filetype) self.filesize = filesize self.fileobj = fileobj self.cb = cb if self.value is not None and self.fileobj is not None: raise ValueError("Only one of value or fileobj may be specified") if fileobj is not None and filesize is None: # Try and determine the file size try: self.filesize = os.fstat(fileobj.fileno()).st_size except (OSError, AttributeError): try: fileobj.seek(0, 2) self.filesize = fileobj.tell() fileobj.seek(0) except: raise ValueError("Could not determine filesize") def __lt__(self, other): attrs = ['name', 'value', 'filename', 'filetype', 'filesize', 'fileobj'] myattrs = [getattr(self, a) for a in attrs] oattrs = [getattr(other, a) for a in attrs] return myattrs < oattrs
[docs] def reset(self): if self.fileobj is not None: self.fileobj.seek(0) elif self.value is None: raise ValueError("Don't know how to reset this parameter")
@classmethod
[docs] def from_file(cls, paramname, filename): """Returns a new MultipartParam object constructed from the local file at ``filename``. ``filesize`` is determined by os.path.getsize(``filename``) ``filetype`` is determined by mimetypes.guess_type(``filename``)[0] ``filename`` is set to os.path.basename(``filename``) """ return cls(paramname, filename=os.path.basename(filename), filetype=mimetypes.guess_type(filename)[0], filesize=os.path.getsize(filename), fileobj=open(filename, "rb"))
@classmethod
[docs] def from_params(cls, params): """Returns a list of MultipartParam objects from a sequence of name, value pairs, MultipartParam instances, or from a mapping of names to values The values may be strings or file objects, or MultipartParam objects. MultipartParam object names must match the given names in the name,value pairs or mapping, if applicable.""" if hasattr(params, 'items'): params = params.items() retval = [] for item in params: if isinstance(item, cls): retval.append(item) continue name, value = item if isinstance(value, cls): assert value.name == name retval.append(value) continue if hasattr(value, 'read'): # Looks like a file object filename = getattr(value, 'name', None) if filename is not None: filetype = mimetypes.guess_type(filename)[0] else: filetype = None retval.append(cls(name=name, filename=filename, filetype=filetype, fileobj=value)) else: retval.append(cls(name, value)) return retval
[docs] def encode_hdr(self, boundary): """Returns the header of the encoding of this parameter""" boundary = encode_and_quote(boundary) headers = ["--%s" % boundary] if self.filename: disposition = 'form-data; name="%s"; filename="%s"' % (self.name, self.filename) else: disposition = 'form-data; name="%s"' % self.name headers.append("Content-Disposition: %s" % disposition) if self.filetype: filetype = self.filetype else: filetype = "text/plain; charset=utf-8" headers.append("Content-Type: %s" % filetype) headers.append("") headers.append("") return "\r\n".join(headers)
[docs] def encode(self, boundary): """Returns the string encoding of this parameter""" if self.value is None: value = self.fileobj.read() else: value = self.value if re.search("^--%s$" % re.escape(boundary), value, re.M): raise ValueError("boundary found in encoded string") return "%s%s\r\n" % (self.encode_hdr(boundary), value)
[docs] def iter_encode(self, boundary, blocksize=4096): """Yields the encoding of this parameter If self.fileobj is set, then blocks of ``blocksize`` bytes are read and yielded.""" total = self.get_size(boundary) current = 0 if self.value is not None: block = self.encode(boundary) current += len(block) yield block if self.cb: self.cb(self, current, total) else: block = self.encode_hdr(boundary) current += len(block) yield block if self.cb: self.cb(self, current, total) last_block = "" encoded_boundary = "--%s" % encode_and_quote(boundary) boundary_exp = re.compile("^%s$" % re.escape(encoded_boundary), re.M) while True: block = self.fileobj.read(blocksize) if not block: current += 2 yield "\r\n" if self.cb: self.cb(self, current, total) break last_block += block if boundary_exp.search(last_block): raise ValueError("boundary found in file data") last_block = last_block[-len(encoded_boundary)-2:] current += len(block) yield block if self.cb: self.cb(self, current, total)
[docs] def get_size(self, boundary): """Returns the size in bytes that this param will be when encoded with the given boundary.""" if self.filesize is not None: valuesize = self.filesize else: valuesize = len(self.value) return len(self.encode_hdr(boundary)) + 2 + valuesize
[docs]def encode_string(boundary, name, value): """Returns ``name`` and ``value`` encoded as a multipart/form-data variable. ``boundary`` is the boundary string used throughout a single request to separate variables.""" return MultipartParam(name, value).encode(boundary)
[docs]def encode_file_header(boundary, paramname, filesize, filename=None, filetype=None): """Returns the leading data for a multipart/form-data field that contains file data. ``boundary`` is the boundary string used throughout a single request to separate variables. ``paramname`` is the name of the variable in this request. ``filesize`` is the size of the file data. ``filename`` if specified is the filename to give to this field. This field is only useful to the server for determining the original filename. ``filetype`` if specified is the MIME type of this file. The actual file data should be sent after this header has been sent. """ return MultipartParam(paramname, filesize=filesize, filename=filename, filetype=filetype).encode_hdr(boundary)
[docs]def get_body_size(params, boundary): """Returns the number of bytes that the multipart/form-data encoding of ``params`` will be.""" size = sum(p.get_size(boundary) for p in MultipartParam.from_params(params)) return size + len(boundary) + 6
[docs]def get_headers(params, boundary): """Returns a dictionary with Content-Type and Content-Length headers for the multipart/form-data encoding of ``params``.""" headers = {} boundary = urllib.quote_plus(boundary) headers['Content-Type'] = "multipart/form-data; boundary=%s" % boundary headers['Content-Length'] = str(get_body_size(params, boundary)) return headers
class multipart_yielder: def __init__(self, params, boundary, cb): self.params = params self.boundary = boundary self.cb = cb self.i = 0 self.p = None self.param_iter = None self.current = 0 self.total = get_body_size(params, boundary) def __iter__(self): return self def next(self): """generator function to yield multipart/form-data representation of parameters""" if self.param_iter is not None: try: block = next(self.param_iter) self.current += len(block) if self.cb: self.cb(self.p, self.current, self.total) return block except StopIteration: self.p = None self.param_iter = None if self.i is None: raise StopIteration elif self.i >= len(self.params): self.param_iter = None self.p = None self.i = None block = "--%s--\r\n" % self.boundary self.current += len(block) if self.cb: self.cb(self.p, self.current, self.total) return block self.p = self.params[self.i] self.param_iter = self.p.iter_encode(self.boundary) self.i += 1 return next(self) def reset(self): self.i = 0 self.current = 0 for param in self.params: param.reset()
[docs]def multipart_encode(params, boundary=None, cb=None): """Encode ``params`` as multipart/form-data. ``params`` should be a sequence of (name, value) pairs or MultipartParam objects, or a mapping of names to values. Values are either strings parameter values, or file-like objects to use as the parameter value. The file-like objects must support .read() and either .fileno() or both .seek() and .tell(). If ``boundary`` is set, then it as used as the MIME boundary. Otherwise a randomly generated boundary will be used. In either case, if the boundary string appears in the parameter values a ValueError will be raised. If ``cb`` is set, it should be a callback which will get called as blocks of data are encoded. It will be called with (param, current, total), indicating the current parameter being encoded, the current amount encoded, and the total amount to encode. Returns a tuple of `datagen`, `headers`, where `datagen` is a generator that will yield blocks of data that make up the encoded parameters, and `headers` is a dictionary with the associated Content-Type and Content-Length headers. Examples: >>> datagen, headers = multipart_encode( [("key", "value1"), ("key", "value2")] ) >>> s = "".join(datagen) >>> assert "value2" in s and "value1" in s >>> p = MultipartParam("key", "value2") >>> datagen, headers = multipart_encode( [("key", "value1"), p] ) >>> s = "".join(datagen) >>> assert "value2" in s and "value1" in s >>> datagen, headers = multipart_encode( {"key": "value1"} ) >>> s = "".join(datagen) >>> assert "value2" not in s and "value1" in s """ if boundary is None: boundary = gen_boundary() else: boundary = urllib.quote_plus(boundary) headers = get_headers(params, boundary) params = MultipartParam.from_params(params) return multipart_yielder(params, boundary, cb), headers