Initial commit

6 files changed, 1125 insertions, 0 deletions

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..c9b568f
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,2 @@
+*.pyc
+*.swp
diff --git a/S3.py b/S3.py
new file mode 100644
index 0000000..37682f3
--- /dev/null
+++ b/S3.py
@@ -0,0 +1,617 @@
+#!/usr/bin/env python
+
+#  This software code is made available "AS IS" without warranties of any
+#  kind.  You may copy, display, modify and redistribute the software
+#  code either by itself or as incorporated into your code; provided that
+#  you do not remove any proprietary notices.  Your use of this software
+#  code is at your own risk and you waive any claim against Amazon
+#  Digital Services, Inc. or its affiliates with respect to your use of
+#  this software code. (c) 2006-2007 Amazon Digital Services, Inc. or its
+#  affiliates.
+
+import base64
+import hmac
+import httplib
+import re
+import sha
+import sys
+import time
+import urllib
+import urlparse
+import xml.sax
+
+DEFAULT_HOST = 's3.amazonaws.com'
+PORTS_BY_SECURITY = { True: 443, False: 80 }
+METADATA_PREFIX = 'x-amz-meta-'
+AMAZON_HEADER_PREFIX = 'x-amz-'
+
+# generates the aws canonical string for the given parameters
+def canonical_string(method, bucket="", key="", query_args={}, headers={}, expires=None):
+    interesting_headers = {}
+    for header_key in headers:
+        lk = header_key.lower()
+        if lk in ['content-md5', 'content-type', 'date'] or lk.startswith(AMAZON_HEADER_PREFIX):
+            interesting_headers[lk] = headers[header_key].strip()
+
+    # these keys get empty strings if they don't exist
+    if not interesting_headers.has_key('content-type'):
+        interesting_headers['content-type'] = ''
+    if not interesting_headers.has_key('content-md5'):
+        interesting_headers['content-md5'] = ''
+
+    # just in case someone used this.  it's not necessary in this lib.
+    if interesting_headers.has_key('x-amz-date'):
+        interesting_headers['date'] = ''
+
+    # if you're using expires for query string auth, then it trumps date
+    # (and x-amz-date)
+    if expires:
+        interesting_headers['date'] = str(expires)
+
+    sorted_header_keys = interesting_headers.keys()
+    sorted_header_keys.sort()
+
+    buf = "%s\n" % method
+    for header_key in sorted_header_keys:
+        if header_key.startswith(AMAZON_HEADER_PREFIX):
+            buf += "%s:%s\n" % (header_key, interesting_headers[header_key])
+        else:
+            buf += "%s\n" % interesting_headers[header_key]
+
+    # append the bucket if it exists
+    if bucket != "":
+        buf += "/%s" % bucket
+
+    # add the key.  even if it doesn't exist, add the slash
+    buf += "/%s" % urllib.quote_plus(key.encode('utf-8'))
+
+    # handle special query string arguments
+
+    if query_args.has_key("acl"):
+        buf += "?acl"
+    elif query_args.has_key("torrent"):
+        buf += "?torrent"
+    elif query_args.has_key("logging"):
+        buf += "?logging"
+    elif query_args.has_key("location"):
+        buf += "?location"
+
+    return buf
+
+# computes the base64'ed hmac-sha hash of the canonical string and the secret
+# access key, optionally urlencoding the result
+def encode(aws_secret_access_key, str, urlencode=False):
+    b64_hmac = base64.encodestring(hmac.new(aws_secret_access_key, str, sha).digest()).strip()
+    if urlencode:
+        return urllib.quote_plus(b64_hmac)
+    else:
+        return b64_hmac
+
+def merge_meta(headers, metadata):
+    final_headers = headers.copy()
+    for k in metadata.keys():
+        final_headers[METADATA_PREFIX + k] = metadata[k]
+
+    return final_headers
+
+# builds the query arg string
+def query_args_hash_to_string(query_args):
+    query_string = ""
+    pairs = []
+    for k, v in query_args.items():
+        piece = k
+        if v != None:
+            piece += "=%s" % urllib.quote_plus(str(v).encode('utf-8'))
+        pairs.append(piece)
+
+    return '&'.join(pairs)
+
+
+class CallingFormat:
+    PATH = 1
+    SUBDOMAIN = 2
+    VANITY = 3
+
+    def build_url_base(protocol, server, port, bucket, calling_format):
+        url_base = '%s://' % protocol
+
+        if bucket == '':
+            url_base += server
+        elif calling_format == CallingFormat.SUBDOMAIN:
+            url_base += "%s.%s" % (bucket, server)
+        elif calling_format == CallingFormat.VANITY:
+            url_base += bucket
+        else:
+            url_base += server
+
+        url_base += ":%s" % port
+
+        if (bucket != '') and (calling_format == CallingFormat.PATH):
+            url_base += "/%s" % bucket
+
+        return url_base
+
+    build_url_base = staticmethod(build_url_base)
+
+
+
+class Location:
+    DEFAULT = None
+    EU = 'EU'
+
+
+
+class AWSAuthConnection:
+    def __init__(self, aws_access_key_id, aws_secret_access_key, is_secure=True,
+            server=DEFAULT_HOST, port=None, calling_format=CallingFormat.SUBDOMAIN):
+
+        if not port:
+            port = PORTS_BY_SECURITY[is_secure]
+
+        self.aws_access_key_id = aws_access_key_id
+        self.aws_secret_access_key = aws_secret_access_key
+        self.is_secure = is_secure
+        self.server = server
+        self.port = port
+        self.calling_format = calling_format
+
+    def create_bucket(self, bucket, headers={}):
+        return Response(self._make_request('PUT', bucket, '', {}, headers))
+
+    def create_located_bucket(self, bucket, location=Location.DEFAULT, headers={}):
+        if location == Location.DEFAULT:
+            body = ""
+        else:
+            body = "<CreateBucketConstraint><LocationConstraint>" + \
+                   location + \
+                   "</LocationConstraint></CreateBucketConstraint>"
+        return Response(self._make_request('PUT', bucket, '', {}, headers, body))
+
+    def check_bucket_exists(self, bucket):
+        return self._make_request('HEAD', bucket, '', {}, {})
+
+    def list_bucket(self, bucket, options={}, headers={}):
+        return ListBucketResponse(self._make_request('GET', bucket, '', options, headers))
+
+    def delete_bucket(self, bucket, headers={}):
+        return Response(self._make_request('DELETE', bucket, '', {}, headers))
+
+    def put(self, bucket, key, object, headers={}):
+        if not isinstance(object, S3Object):
+            object = S3Object(object)
+
+        return Response(
+                self._make_request(
+                    'PUT',
+                    bucket,
+                    key,
+                    {},
+                    headers,
+                    object.data,
+                    object.metadata))
+
+    def get(self, bucket, key, headers={}):
+        return GetResponse(
+                self._make_request('GET', bucket, key, {}, headers))
+
+    def delete(self, bucket, key, headers={}):
+        return Response(
+                self._make_request('DELETE', bucket, key, {}, headers))
+
+    def get_bucket_logging(self, bucket, headers={}):
+        return GetResponse(self._make_request('GET', bucket, '', { 'logging': None }, headers))
+
+    def put_bucket_logging(self, bucket, logging_xml_doc, headers={}):
+        return Response(self._make_request('PUT', bucket, '', { 'logging': None }, headers, logging_xml_doc))
+
+    def get_bucket_acl(self, bucket, headers={}):
+        return self.get_acl(bucket, '', headers)
+
+    def get_acl(self, bucket, key, headers={}):
+        return GetResponse(
+                self._make_request('GET', bucket, key, { 'acl': None }, headers))
+
+    def put_bucket_acl(self, bucket, acl_xml_document, headers={}):
+        return self.put_acl(bucket, '', acl_xml_document, headers)
+
+    def put_acl(self, bucket, key, acl_xml_document, headers={}):
+        return Response(
+                self._make_request(
+                    'PUT',
+                    bucket,
+                    key,
+                    { 'acl': None },
+                    headers,
+                    acl_xml_document))
+
+    def list_all_my_buckets(self, headers={}):
+        return ListAllMyBucketsResponse(self._make_request('GET', '', '', {}, headers))
+
+    def get_bucket_location(self, bucket):
+        return LocationResponse(self._make_request('GET', bucket, '', {'location' : None}))
+
+    # end public methods
+
+    def _make_request(self, method, bucket='', key='', query_args={}, headers={}, data='', metadata={}):
+
+        server = ''
+        if bucket == '':
+            server = self.server
+        elif self.calling_format == CallingFormat.SUBDOMAIN:
+            server = "%s.%s" % (bucket, self.server)
+        elif self.calling_format == CallingFormat.VANITY:
+            server = bucket
+        else:
+            server = self.server
+
+        path = ''
+
+        if (bucket != '') and (self.calling_format == CallingFormat.PATH):
+            path += "/%s" % bucket
+
+        # add the slash after the bucket regardless
+        # the key will be appended if it is non-empty
+        path += "/%s" % urllib.quote_plus(key.encode('utf-8'))
+
+
+        # build the path_argument string
+        # add the ? in all cases since 
+        # signature and credentials follow path args
+        if len(query_args):
+            path += "?" + query_args_hash_to_string(query_args)
+
+        is_secure = self.is_secure
+        host = "%s:%d" % (server, self.port)
+        while True:
+            if (is_secure):
+                connection = httplib.HTTPSConnection(host)
+            else:
+                connection = httplib.HTTPConnection(host)
+
+            final_headers = merge_meta(headers, metadata);
+            # add auth header
+            self._add_aws_auth_header(final_headers, method, bucket, key, query_args)
+
+            connection.request(method, path, data, final_headers)
+            resp = connection.getresponse()
+            if resp.status < 300 or resp.status >= 400:
+                return resp
+            # handle redirect
+            location = resp.getheader('location')
+            if not location:
+                return resp
+            # (close connection)
+            resp.read()
+            scheme, host, path, params, query, fragment \
+                    = urlparse.urlparse(location)
+            if scheme == "http":    is_secure = True
+            elif scheme == "https": is_secure = False
+            else: raise invalidURL("Not http/https: " + location)
+            if query: path += "?" + query
+            # retry with redirect
+
+    def _add_aws_auth_header(self, headers, method, bucket, key, query_args):
+        if not headers.has_key('Date'):
+            headers['Date'] = time.strftime("%a, %d %b %Y %X GMT", time.gmtime())
+
+        c_string = canonical_string(method, bucket, key, query_args, headers)
+        headers['Authorization'] = \
+            "AWS %s:%s" % (self.aws_access_key_id, encode(self.aws_secret_access_key, c_string))
+
+
+class QueryStringAuthGenerator:
+    # by default, expire in 1 minute
+    DEFAULT_EXPIRES_IN = 60
+
+    def __init__(self, aws_access_key_id, aws_secret_access_key, is_secure=True,
+                 server=DEFAULT_HOST, port=None, calling_format=CallingFormat.SUBDOMAIN):
+
+        if not port:
+            port = PORTS_BY_SECURITY[is_secure]
+
+        self.aws_access_key_id = aws_access_key_id
+        self.aws_secret_access_key = aws_secret_access_key
+        if (is_secure):
+            self.protocol = 'https'
+        else:
+            self.protocol = 'http'
+
+        self.is_secure = is_secure
+        self.server = server
+        self.port = port
+        self.calling_format = calling_format
+        self.__expires_in = QueryStringAuthGenerator.DEFAULT_EXPIRES_IN
+        self.__expires = None
+
+        # for backwards compatibility with older versions
+        self.server_name = "%s:%s" % (self.server, self.port)
+
+    def set_expires_in(self, expires_in):
+        self.__expires_in = expires_in
+        self.__expires = None
+
+    def set_expires(self, expires):
+        self.__expires = expires
+        self.__expires_in = None
+
+    def create_bucket(self, bucket, headers={}):
+        return self.generate_url('PUT', bucket, '', {}, headers)
+
+    def list_bucket(self, bucket, options={}, headers={}):
+        return self.generate_url('GET', bucket, '', options, headers)
+
+    def delete_bucket(self, bucket, headers={}):
+        return self.generate_url('DELETE', bucket, '', {}, headers)
+
+    def put(self, bucket, key, object, headers={}):
+        if not isinstance(object, S3Object):
+            object = S3Object(object)
+
+        return self.generate_url(
+                'PUT',
+                bucket,
+                key,
+                {},
+                merge_meta(headers, object.metadata))
+
+    def get(self, bucket, key, headers={}):
+        return self.generate_url('GET', bucket, key, {}, headers)
+
+    def delete(self, bucket, key, headers={}):
+        return self.generate_url('DELETE', bucket, key, {}, headers)
+
+    def get_bucket_logging(self, bucket, headers={}):
+        return self.generate_url('GET', bucket, '', { 'logging': None }, headers)
+
+    def put_bucket_logging(self, bucket, logging_xml_doc, headers={}):
+        return self.generate_url('PUT', bucket, '', { 'logging': None }, headers)
+
+    def get_bucket_acl(self, bucket, headers={}):
+        return self.get_acl(bucket, '', headers)
+
+    def get_acl(self, bucket, key='', headers={}):
+        return self.generate_url('GET', bucket, key, { 'acl': None }, headers)
+
+    def put_bucket_acl(self, bucket, acl_xml_document, headers={}):
+        return self.put_acl(bucket, '', acl_xml_document, headers)
+
+    # don't really care what the doc is here.
+    def put_acl(self, bucket, key, acl_xml_document, headers={}):
+        return self.generate_url('PUT', bucket, key, { 'acl': None }, headers)
+
+    def list_all_my_buckets(self, headers={}):
+        return self.generate_url('GET', '', '', {}, headers)
+
+    def make_bare_url(self, bucket, key=''):
+        full_url = self.generate_url(self, bucket, key)
+        return full_url[:full_url.index('?')]
+
+    def generate_url(self, method, bucket='', key='', query_args={}, headers={}):
+        expires = 0
+        if self.__expires_in != None:
+            expires = int(time.time() + self.__expires_in)
+        elif self.__expires != None:
+            expires = int(self.__expires)
+        else:
+            raise "Invalid expires state"
+
+        canonical_str = canonical_string(method, bucket, key, query_args, headers, expires)
+        encoded_canonical = encode(self.aws_secret_access_key, canonical_str)
+
+        url = CallingFormat.build_url_base(self.protocol, self.server, self.port, bucket, self.calling_format)
+
+        url += "/%s" % urllib.quote_plus(key.encode('utf-8'))
+
+        query_args['Signature'] = encoded_canonical
+        query_args['Expires'] = expires
+        query_args['AWSAccessKeyId'] = self.aws_access_key_id
+
+        url += "?%s" % query_args_hash_to_string(query_args)
+
+        return url
+
+
+class S3Object:
+    def __init__(self, data, metadata={}):
+        self.data = data
+        self.metadata = metadata
+
+class Owner:
+    def __init__(self, id='', display_name=''):
+        self.id = id
+        self.display_name = display_name
+
+class ListEntry:
+    def __init__(self, key='', last_modified=None, etag='', size=0, storage_class='', owner=None):
+        self.key = key
+        self.last_modified = last_modified
+        self.etag = etag
+        self.size = size
+        self.storage_class = storage_class
+        self.owner = owner
+
+class CommonPrefixEntry:
+    def __init(self, prefix=''):
+        self.prefix = prefix
+
+class Bucket:
+    def __init__(self, name='', creation_date=''):
+        self.name = name
+        self.creation_date = creation_date
+
+class Response:
+    def __init__(self, http_response):
+        self.http_response = http_response
+        # you have to do this read, even if you don't expect a body.
+        # otherwise, the next request fails.
+        self.body = http_response.read()
+        if http_response.status >= 300 and self.body:
+            self.message = self.body
+        else:
+            self.message = "%03d %s" % (http_response.status, http_response.reason)
+
+
+
+class ListBucketResponse(Response):
+    def __init__(self, http_response):
+        Response.__init__(self, http_response)
+        if http_response.status < 300:
+            handler = ListBucketHandler()
+            xml.sax.parseString(self.body, handler)
+            self.entries = handler.entries
+            self.common_prefixes = handler.common_prefixes
+            self.name = handler.name
+            self.marker = handler.marker
+            self.prefix = handler.prefix
+            self.is_truncated = handler.is_truncated
+            self.delimiter = handler.delimiter
+            self.max_keys = handler.max_keys
+            self.next_marker = handler.next_marker
+        else:
+            self.entries = []
+
+class ListAllMyBucketsResponse(Response):
+    def __init__(self, http_response):
+        Response.__init__(self, http_response)
+        if http_response.status < 300: 
+            handler = ListAllMyBucketsHandler()
+            xml.sax.parseString(self.body, handler)
+            self.entries = handler.entries
+        else:
+            self.entries = []
+
+class GetResponse(Response):
+    def __init__(self, http_response):
+        Response.__init__(self, http_response)
+        response_headers = http_response.msg   # older pythons don't have getheaders
+        metadata = self.get_aws_metadata(response_headers)
+        self.object = S3Object(self.body, metadata)
+
+    def get_aws_metadata(self, headers):
+        metadata = {}
+        for hkey in headers.keys():
+            if hkey.lower().startswith(METADATA_PREFIX):
+                metadata[hkey[len(METADATA_PREFIX):]] = headers[hkey]
+                del headers[hkey]
+
+        return metadata
+
+class LocationResponse(Response):
+    def __init__(self, http_response):
+        Response.__init__(self, http_response)
+        if http_response.status < 300: 
+            handler = LocationHandler()
+            xml.sax.parseString(self.body, handler)
+            self.location = handler.location
+
+class ListBucketHandler(xml.sax.ContentHandler):
+    def __init__(self):
+        self.entries = []
+        self.curr_entry = None
+        self.curr_text = ''
+        self.common_prefixes = []
+        self.curr_common_prefix = None
+        self.name = ''
+        self.marker = ''
+        self.prefix = ''
+        self.is_truncated = False
+        self.delimiter = ''
+        self.max_keys = 0
+        self.next_marker = ''
+        self.is_echoed_prefix_set = False
+
+    def startElement(self, name, attrs):
+        if name == 'Contents':
+            self.curr_entry = ListEntry()
+        elif name == 'Owner':
+            self.curr_entry.owner = Owner()
+        elif name == 'CommonPrefixes':
+            self.curr_common_prefix = CommonPrefixEntry()
+
+
+    def endElement(self, name):
+        if name == 'Contents':
+            self.entries.append(self.curr_entry)
+        elif name == 'CommonPrefixes':
+            self.common_prefixes.append(self.curr_common_prefix)
+        elif name == 'Key':
+            self.curr_entry.key = self.curr_text
+        elif name == 'LastModified':
+            self.curr_entry.last_modified = self.curr_text
+        elif name == 'ETag':
+            self.curr_entry.etag = self.curr_text
+        elif name == 'Size':
+            self.curr_entry.size = int(self.curr_text)
+        elif name == 'ID':
+            self.curr_entry.owner.id = self.curr_text
+        elif name == 'DisplayName':
+            self.curr_entry.owner.display_name = self.curr_text
+        elif name == 'StorageClass':
+            self.curr_entry.storage_class = self.curr_text
+        elif name == 'Name':
+            self.name = self.curr_text
+        elif name == 'Prefix' and self.is_echoed_prefix_set:
+            self.curr_common_prefix.prefix = self.curr_text
+        elif name == 'Prefix':
+            self.prefix = self.curr_text
+            self.is_echoed_prefix_set = True
+        elif name == 'Marker':
+            self.marker = self.curr_text
+        elif name == 'IsTruncated':
+            self.is_truncated = self.curr_text == 'true'
+        elif name == 'Delimiter':
+            self.delimiter = self.curr_text
+        elif name == 'MaxKeys':
+            self.max_keys = int(self.curr_text)
+        elif name == 'NextMarker':
+            self.next_marker = self.curr_text
+
+        self.curr_text = ''
+
+    def characters(self, content):
+        self.curr_text += content
+
+
+class ListAllMyBucketsHandler(xml.sax.ContentHandler):
+    def __init__(self):
+        self.entries = []
+        self.curr_entry = None
+        self.curr_text = ''
+
+    def startElement(self, name, attrs):
+        if name == 'Bucket':
+            self.curr_entry = Bucket()
+
+    def endElement(self, name):
+        if name == 'Name':
+            self.curr_entry.name = self.curr_text
+        elif name == 'CreationDate':
+            self.curr_entry.creation_date = self.curr_text
+        elif name == 'Bucket':
+            self.entries.append(self.curr_entry)
+
+    def characters(self, content):
+        self.curr_text = content
+
+
+class LocationHandler(xml.sax.ContentHandler):
+    def __init__(self):
+        self.location = None
+        self.state = 'init'
+
+    def startElement(self, name, attrs):
+        if self.state == 'init':
+            if name == 'LocationConstraint':
+                self.state = 'tag_location'
+                self.location = ''
+            else: self.state = 'bad'
+        else: self.state = 'bad'
+
+    def endElement(self, name):
+        if self.state == 'tag_location' and name == 'LocationConstraint':
+            self.state = 'done'
+        else: self.state = 'bad'
+
+    def characters(self, content):
+        if self.state == 'tag_location':
+            self.location += content
diff --git a/__init__.py b/__init__.py
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/__init__.py
diff --git a/django/__init__.py b/django/__init__.py
new file mode 100644
index 0000000..7027c19
--- /dev/null
+++ b/django/__init__.py
@@ -0,0 +1,121 @@
+from django.conf import settings
+from amazons3 import S3
+
+from django.core.files.storage import Storage
+
+class S3Error(Exception):
+    "Misc. S3 Service Error"
+    pass
+
+class S3Storage(Storage):
+    options = None
+
+    def __init__(self, options=None):
+        if not options:
+            options = settings.S3_SETTINGS
+            self.options = options
+        self.perm_tuple = (
+            'private',
+            'public-read',
+            'public-read-write',
+            'authenticated-read'
+        )
+        if self.options['default_perm'] not in self.perm_tuple:
+            self.options['default_perm'] = 'private'
+
+        self.connect()
+
+    def connect(self):
+        self.conn = S3.AWSAuthConnection(self.options['aws_key'], self.options['aws_secret_key'])
+
+        res = self.conn.check_bucket_exists(self.options['bucket'])
+
+        if res.status != 200:
+            res = self.conn.create_bucket(self.options['bucket'])
+            if res.http_response.status != 200:
+                raise S3Error, 'Unable to create bucket %s' % (self.options['bucket'])
+
+        return True
+
+    def exists(self, filename):
+        contents = self.conn.list_bucket(self.options['bucket'])
+        if filename in [f.key for f in contents.entries]:
+            return True
+        else:
+            return False
+
+    def size(self, filename):
+        contents = self.conn.list_bucket(self.options['bucket'])
+        for f in contents.entries:
+            if f.name == filename:
+                return f.size
+
+        return False
+
+    def url(self, filename):
+        server = self.options['bucket']
+        if not self.options['vanity_url']:
+            server += '.s3.amazonaws.com'
+        return 'http://' + server + '/' + filename
+
+
+    def _save(self, filename, content):
+        # a stupid hack
+        content = content.file
+        try:
+            data = content.read()
+        except IOError, err:
+            raise S3Error, 'Unable to read %s: %s' % (filename, err.strerror)
+
+        if not content.content_type:
+            import mimetypes
+            content_type = mimetypes.guess_type(filename)[0]
+            if content_type is None:
+                content_type = 'text/plain'
+        else:
+            content_type = content.content_type
+
+        perm = self.options['default_perm']
+
+        res = self.conn.put(
+            self.options['bucket'],
+            filename,
+            S3.S3Object(data),
+            {
+                'x-amz-acl': perm,
+                'Content-Type': content_type
+            }
+        )
+
+        if res.http_response.status != 200:
+            raise S3Error, 'Unable to upload file %s: Error code %s: %s' % (filename, self.options['bucket'], res.body)
+
+
+        content.filename = filename
+        content.url = self.url(filename)
+
+        return filename
+
+    def delete(self, filename):
+        res = self.conn.delete(self.options['bucket'], filename)
+        if res.http_response.status != 204:
+            pass
+            #raise S3Error, 'Unable to delete file %s' % (filename)
+
+        return (res.http_response.status == 204)
+
+    def path(self, filename):
+        raise NotImplementedError
+
+    def open(self, filename, mode):
+        from urllib import urlopen
+        return urlopen(self.url(filename))
+
+    def get_available_name(self, filename):
+        import os
+        basefilename = os.path.splitext(filename)
+        i = 1
+        while self.exists(filename):
+            filename = '%s-%d%s' % (basefilename[0], i, basefilename[1])
+
+        return filename
diff --git a/s3-driver.py b/s3-driver.py
new file mode 100644
index 0000000..29f700b
--- /dev/null
+++ b/s3-driver.py
@@ -0,0 +1,118 @@
+#!/usr/bin/env python
+
+#  This software code is made available "AS IS" without warranties of any
+#  kind.  You may copy, display, modify and redistribute the software
+#  code either by itself or as incorporated into your code; provided that
+#  you do not remove any proprietary notices.  Your use of this software
+#  code is at your own risk and you waive any claim against Amazon
+#  Digital Services, Inc. or its affiliates with respect to your use of
+#  this software code. (c) 2006-2007 Amazon Digital Services, Inc. or its
+#  affiliates.
+
+import S3
+import time
+import sys
+
+AWS_ACCESS_KEY_ID = '<INSERT YOUR AWS ACCESS KEY ID HERE>'
+AWS_SECRET_ACCESS_KEY = '<INSERT YOUR AWS SECRET ACCESS KEY HERE>'
+# remove these next two lines when you've updated your credentials.
+print "update s3-driver.py with your AWS credentials"
+sys.exit();
+
+# convert the bucket to lowercase for vanity domains
+# the bucket name must be lowercase since DNS is case-insensitive
+BUCKET_NAME = AWS_ACCESS_KEY_ID.lower() + '-test-bucket'
+KEY_NAME = 'test-key'
+
+conn = S3.AWSAuthConnection(AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY)
+generator = S3.QueryStringAuthGenerator(AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY)
+
+
+# Check if the bucket exists.  The high availability engineering of 
+# Amazon S3 is focused on get, put, list, and delete operations. 
+# Because bucket operations work against a centralized, global
+# resource space, it is not appropriate to make bucket create or
+# delete calls on the high availability code path of your application.
+# It is better to create or delete buckets in a separate initialization
+# or setup routine that you run less often.
+if (conn.check_bucket_exists(BUCKET_NAME).status == 200):
+  print '----- bucket already exists! -----'
+else:
+  print '----- creating bucket -----'
+  print conn.create_located_bucket(BUCKET_NAME, S3.Location.DEFAULT).message
+  # to create an EU bucket
+  #print conn.create_located_bucket(BUCKET_NAME, S3.Location.EU).message
+
+print '----- bucket location -----'
+print conn.get_bucket_location(BUCKET_NAME).location
+
+print '----- listing bucket -----'
+print map(lambda x: x.key, conn.list_bucket(BUCKET_NAME).entries)
+
+print '----- putting object (with content type) -----'
+print conn.put(
+        BUCKET_NAME,
+        KEY_NAME,
+        S3.S3Object('this is a test'),
+        { 'Content-Type': 'text/plain' }).message
+
+print '----- listing bucket -----'
+print map(lambda x: x.key, conn.list_bucket(BUCKET_NAME).entries)
+
+print '----- getting object -----'
+print conn.get(BUCKET_NAME, KEY_NAME).object.data
+
+print '----- query string auth example -----'
+print "\nTry this url out in your browser (it will only be valid for 60 seconds).\n"
+generator.set_expires_in(60);
+url = generator.get(BUCKET_NAME, KEY_NAME)
+print url
+print '\npress enter> ',
+sys.stdin.readline()
+
+print "\nNow try just the url without the query string arguments.  it should fail.\n"
+print generator.make_bare_url(BUCKET_NAME, KEY_NAME)
+print '\npress enter> ',
+sys.stdin.readline()
+
+
+print '----- putting object with metadata and public read acl -----'
+print conn.put(
+    BUCKET_NAME,
+    KEY_NAME + '-public',
+    S3.S3Object('this is a publicly readable test'),
+    { 'x-amz-acl': 'public-read' , 'Content-Type': 'text/plain' }
+).message
+
+print '----- anonymous read test ----'
+print "\nYou should be able to try this in your browser\n"
+public_key = KEY_NAME + '-public'
+print generator.make_bare_url(BUCKET_NAME, public_key)
+print "\npress enter> ",
+sys.stdin.readline()
+
+print "----- getting object's acl -----"
+print conn.get_acl(BUCKET_NAME, KEY_NAME).object.data
+
+print "\n----- path style url example -----";
+print "Non-location-constrained buckets can also be specified as part of the url path.  (This was the original url style supported by S3.)\n";
+print "Try this url out in your browser (it will only be valid for 60 seconds).\n"
+generator.calling_format = S3.CallingFormat.PATH
+url = generator.get(BUCKET_NAME, KEY_NAME)
+print url
+print "\npress enter> ",
+sys.stdin.readline()
+
+print '----- deleting objects -----'
+print conn.delete(BUCKET_NAME, KEY_NAME).message
+print conn.delete(BUCKET_NAME, KEY_NAME + '-public').message
+
+print '----- listing bucket -----'
+print map(lambda x: x.key, conn.list_bucket(BUCKET_NAME).entries)
+
+print '----- listing all my buckets -----'
+print map(lambda x: x.name, conn.list_all_my_buckets().entries)
+
+print '----- deleting bucket ------'
+print conn.delete_bucket(BUCKET_NAME).message
+
diff --git a/s3-test.py b/s3-test.py
new file mode 100644
index 0000000..fbd8d9c
--- /dev/null
+++ b/s3-test.py
@@ -0,0 +1,267 @@
+#!/usr/bin/env python
+
+# This software code is made available "AS IS" without warranties of any
+# kind.  You may copy, display, modify and redistribute the software
+# code either by itself or as incorporated into your code; provided that
+# you do not remove any proprietary notices.  Your use of this software
+# code is at your own risk and you waive any claim against Amazon
+# Digital Services, Inc. or its affiliates with respect to your use of
+# this software code. (c) 2006-2007 Amazon Digital Services, Inc. or its
+# affiliates.
+
+import unittest
+import S3
+import httplib
+import sys
+
+AWS_ACCESS_KEY_ID = '<INSERT YOUR AWS ACCESS KEY ID HERE>'
+AWS_SECRET_ACCESS_KEY = '<INSERT YOUR AWS SECRET ACCESS KEY HERE>'
+# remove these next two lines when you've updated your credentials.
+print "update s3-test.py with your AWS credentials"
+sys.exit();
+
+# for subdomains (bucket.s3.amazonaws.com),
+# the bucket name must be lowercase since DNS is case-insensitive
+BUCKET_NAME = "%s-test-bucket" % AWS_ACCESS_KEY_ID.lower();
+
+
+class TestAWSAuthConnection(unittest.TestCase):
+    def setUp(self):
+        self.conn = S3.AWSAuthConnection(AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY)
+
+    # test all operations for both regular and vanity domains
+    # regular: http://s3.amazonaws.com/bucket/key
+    # subdomain: http://bucket.s3.amazonaws.com/key
+    # testing pure vanity domains (http://<vanity domain>/key) is not covered here
+    # but is possible with some additional setup (set the server in @conn to your vanity domain)
+
+    def test_subdomain_default(self):
+        self.run_tests(S3.CallingFormat.SUBDOMAIN, S3.Location.DEFAULT)
+
+    def test_subdomain_eu(self):
+        self.run_tests(S3.CallingFormat.SUBDOMAIN, S3.Location.EU)
+
+    def test_path_default(self):
+        self.run_tests(S3.CallingFormat.PATH, S3.Location.DEFAULT)
+
+
+    def run_tests(self, calling_format, location):
+        self.conn.calling_format = calling_format
+
+        response = self.conn.create_located_bucket(BUCKET_NAME, location)
+        self.assertEquals(response.http_response.status, 200, 'create bucket')
+
+        response = self.conn.list_bucket(BUCKET_NAME)
+        self.assertEquals(response.http_response.status, 200, 'list bucket')
+        self.assertEquals(len(response.entries), 0, 'bucket is empty')
+
+        text = 'this is a test'
+        key = 'example.txt'
+
+        response = self.conn.put(BUCKET_NAME, key, text)
+        self.assertEquals(response.http_response.status, 200, 'put with a string argument')
+
+        response = \
+            self.conn.put(
+                    BUCKET_NAME,
+                    key,
+                    S3.S3Object(text, {'title': 'title'}),
+                    {'Content-Type': 'text/plain'})
+
+        self.assertEquals(response.http_response.status, 200, 'put with complex argument and headers')
+
+        response = self.conn.get(BUCKET_NAME, key)
+        self.assertEquals(response.http_response.status, 200, 'get object')
+        self.assertEquals(response.object.data, text, 'got right data')
+        self.assertEquals(response.object.metadata, { 'title': 'title' }, 'metadata is correct')
+        self.assertEquals(int(response.http_response.getheader('Content-Length')), len(text), 'got content-length header')
+
+        title_with_spaces = " \t  title with leading and trailing spaces     "
+        response = \
+            self.conn.put(
+                    BUCKET_NAME,
+                    key,
+                    S3.S3Object(text, {'title': title_with_spaces}),
+                    {'Content-Type': 'text/plain'})
+
+        self.assertEquals(response.http_response.status, 200, 'put with headers with spaces')
+
+        response = self.conn.get(BUCKET_NAME, key)
+        self.assertEquals(response.http_response.status, 200, 'get object')
+        self.assertEquals(
+                response.object.metadata,
+                { 'title': title_with_spaces.strip() },
+                'metadata with spaces is correct')
+
+        # delimited list tests
+        inner_key = 'test/inner.txt'
+        last_key = 'z-last-key.txt'
+        response = self.conn.put(BUCKET_NAME, inner_key, text)
+        self.assertEquals(response.http_response.status, 200, 'put inner key')
+
+        response = self.conn.put(BUCKET_NAME, last_key, text)
+        self.assertEquals(response.http_response.status, 200, 'put last key')
+
+        response = self.do_delimited_list(BUCKET_NAME, False, {'delimiter': '/'}, 2, 1, 'root list')
+
+        response = self.do_delimited_list(BUCKET_NAME, True, {'max-keys': 1, 'delimiter': '/'}, 1, 0, 'root list with max keys of 1', 'example.txt')
+
+        response = self.do_delimited_list(BUCKET_NAME, True, {'max-keys': 2, 'delimiter': '/'}, 1, 1, 'root list with max keys of 2, page 1', 'test/')
+
+        marker = response.next_marker
+
+        response = self.do_delimited_list(BUCKET_NAME, False, {'marker': marker, 'max-keys': 2, 'delimiter': '/'}, 1, 0, 'root list with max keys of 2, page 2')
+
+        response = self.do_delimited_list(BUCKET_NAME, False, {'prefix': 'test/', 'delimiter': '/'}, 1, 0, 'test/ list')
+
+        response = self.conn.delete(BUCKET_NAME, inner_key)
+        self.assertEquals(response.http_response.status, 204, 'delete %s' % inner_key)
+
+        response = self.conn.delete(BUCKET_NAME, last_key)
+        self.assertEquals(response.http_response.status, 204, 'delete %s' % last_key)
+
+
+        weird_key = '&=//%# ++++'
+
+        response = self.conn.put(BUCKET_NAME, weird_key, text)
+        self.assertEquals(response.http_response.status, 200, 'put weird key')
+
+        response = self.conn.get(BUCKET_NAME, weird_key)
+        self.assertEquals(response.http_response.status, 200, 'get weird key')
+
+        response = self.conn.get_acl(BUCKET_NAME, key)
+        self.assertEquals(response.http_response.status, 200, 'get acl')
+
+        acl = response.object.data
+
+        response = self.conn.put_acl(BUCKET_NAME, key, acl)
+        self.assertEquals(response.http_response.status, 200, 'put acl')
+
+        response = self.conn.get_bucket_acl(BUCKET_NAME)
+        self.assertEquals(response.http_response.status, 200, 'get bucket acl')
+
+        bucket_acl = response.object.data
+
+        response = self.conn.put_bucket_acl(BUCKET_NAME, bucket_acl)
+        self.assertEquals(response.http_response.status, 200, 'put bucket acl')
+
+        response = self.conn.get_bucket_acl(BUCKET_NAME)
+        self.assertEquals(response.http_response.status, 200, 'get bucket logging')
+
+        bucket_logging = response.object.data
+
+        response = self.conn.put_bucket_acl(BUCKET_NAME, bucket_logging)
+        self.assertEquals(response.http_response.status, 200, 'put bucket logging')
+
+        response = self.conn.list_bucket(BUCKET_NAME)
+        self.assertEquals(response.http_response.status, 200, 'list bucket')
+        entries = response.entries
+        self.assertEquals(len(entries), 2, 'got back right number of keys')
+        # depends on weird_key < key
+        self.assertEquals(entries[0].key, weird_key, 'first key is right')
+        self.assertEquals(entries[1].key, key, 'second key is right')
+
+        response = self.conn.list_bucket(BUCKET_NAME, {'max-keys': 1})
+        self.assertEquals(response.http_response.status, 200, 'list bucket with args')
+        self.assertEquals(len(response.entries), 1, 'got back right number of keys')
+
+        for entry in entries:
+            response = self.conn.delete(BUCKET_NAME, entry.key)
+            self.assertEquals(response.http_response.status, 204, 'delete %s' % entry.key)
+
+        response = self.conn.list_all_my_buckets()
+        self.assertEquals(response.http_response.status, 200, 'list all my buckets')
+        buckets = response.entries
+
+        response = self.conn.delete_bucket(BUCKET_NAME)
+        self.assertEquals(response.http_response.status, 204, 'delete bucket')
+
+        response = self.conn.list_all_my_buckets()
+        self.assertEquals(response.http_response.status, 200, 'list all my buckets again')
+
+        self.assertEquals(len(response.entries), len(buckets) - 1, 'bucket count is correct')
+
+    def verify_list_bucket_response(self, response, bucket, is_truncated, parameters, next_marker=''):
+        prefix = ''
+        marker = ''
+
+        if parameters.has_key('prefix'):
+            prefix = parameters['prefix']
+        if parameters.has_key('marker'):
+            marker = parameters['marker']
+
+        self.assertEquals(bucket, response.name, 'bucket name should match')
+        self.assertEquals(prefix, response.prefix, 'prefix should match')
+        self.assertEquals(marker, response.marker, 'marker should match')
+        if parameters.has_key('max-keys'):
+            self.assertEquals(parameters['max-keys'], response.max_keys, 'max-keys should match')
+        self.assertEquals(parameters['delimiter'], response.delimiter, 'delimiter should match')
+        self.assertEquals(is_truncated, response.is_truncated, 'is_truncated should match')
+        self.assertEquals(next_marker, response.next_marker, 'next_marker should match')
+
+    def do_delimited_list(self, bucket_name, is_truncated, parameters, regular_expected, common_expected, test_name, next_marker=''):
+        response = self.conn.list_bucket(bucket_name, parameters)
+        self.assertEquals(response.http_response.status, 200, test_name)
+        self.assertEquals(regular_expected, len(response.entries), 'right number of regular entries')
+        self.assertEquals(common_expected, len(response.common_prefixes), 'right number of common prefixes')
+
+        self.verify_list_bucket_response(response, bucket_name, is_truncated, parameters, next_marker)
+
+        return response
+
+class TestQueryStringAuthGenerator(unittest.TestCase):
+    def setUp(self):
+        self.generator = S3.QueryStringAuthGenerator(AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY)
+        if (self.generator.is_secure == True):
+            self.connection = httplib.HTTPSConnection(self.generator.server_name)
+        else:
+            self.connection = httplib.HTTPConnection(self.generator.server_name)
+
+    def check_url(self, url, method, status, message, data=''):
+        if (method == 'PUT'):
+            headers = { 'Content-Length': len(data) }
+            self.connection.request(method, url, data, headers)
+        else:
+            self.connection.request(method, url)
+
+        response = self.connection.getresponse()
+        self.assertEquals(response.status, status, message)
+
+        return response.read()
+
+    # test all operations for both regular and vanity domains
+    # regular: http://s3.amazonaws.com/bucket/key
+    # subdomain: http://bucket.s3.amazonaws.com/key
+    # testing pure vanity domains (http://<vanity domain>/key) is not covered here
+    # but is possible with some additional setup (set the server in @conn to your vanity domain)
+
+    def test_subdomain(self):
+        self.run_tests(S3.CallingFormat.SUBDOMAIN)
+
+    def test_path(self):
+        self.run_tests(S3.CallingFormat.PATH)
+
+    def run_tests(self, calling_format):
+        self.generator.calling_format = calling_format
+
+        key = 'test'
+
+        self.check_url(self.generator.create_bucket(BUCKET_NAME), 'PUT', 200, 'create_bucket')
+        self.check_url(self.generator.put(BUCKET_NAME, key, ''), 'PUT', 200, 'put object', 'test data')
+        self.check_url(self.generator.get(BUCKET_NAME, key), 'GET', 200, 'get object')
+        self.check_url(self.generator.list_bucket(BUCKET_NAME), 'GET', 200, 'list bucket')
+        self.check_url(self.generator.list_all_my_buckets(), 'GET', 200, 'list all my buckets')
+        acl = self.check_url(self.generator.get_acl(BUCKET_NAME, key), 'GET', 200, 'get acl')
+        self.check_url(self.generator.put_acl(BUCKET_NAME, key, acl), 'PUT', 200, 'put acl', acl)
+        bucket_acl = self.check_url(self.generator.get_bucket_acl(BUCKET_NAME), 'GET', 200, 'get bucket acl')
+        self.check_url(self.generator.put_bucket_acl(BUCKET_NAME, bucket_acl), 'PUT', 200, 'put bucket acl', bucket_acl)
+        bucket_logging = self.check_url(self.generator.get_bucket_logging(BUCKET_NAME), 'GET', 200, 'get bucket logging')
+        self.check_url(self.generator.put_bucket_logging(BUCKET_NAME, bucket_logging), 'PUT', 200, 'put bucket logging', bucket_logging)
+        self.check_url(self.generator.delete(BUCKET_NAME, key), 'DELETE', 204, 'delete object')
+        self.check_url(self.generator.delete_bucket(BUCKET_NAME), 'DELETE', 204, 'delete bucket')
+
+
+if __name__ == '__main__':
+    unittest.main()
+
+