def test_associate_tag_is_written_to_url(self):
tag = "ABC12345"
api = API("XXX", "XXX", "de", associate_tag=tag)
url = api._build_url(Operation="ItemSearch", SearchIndex="Books")
qs = parse_qs(urlparse(url)[4])
assert qs["AssociateTag"][0] == tag
def test_associate_tag_is_written_to_url(self):
tag = 'ABC12345'
api = API('XXX', 'XXX', 'de', associate_tag=tag)
url = api._build_url(Operation='ItemSearch', SearchIndex='Books')
qs = parse_qs(urlparse(url)[4])
assert qs['AssociateTag'][0] == tag
def test_associate_tag_is_written_to_url(self):
tag = 'ABC12345'
api = API('XXX', 'XXX', 'de', associate_tag=tag)
url = api._build_url(Operation='ItemSearch', SearchIndex='Books')
qs = parse_qs(urlparse(url)[4])
assert qs['AssociateTag'][0] == tag
def __init__(self, access_key_id, secret_access_key, locale, cachedir=DEFAULT_CACHE_DIR, **kwargs):
"""
:param cachedir: Path to directory containing cached responses.
"""
API.__init__(self, access_key_id, secret_access_key, locale, **kwargs)
self.cache = cachedir
if self.cache and not os.path.isdir(self.cache):
os.mkdir(self.cache)
def pytest_funcarg__api(request):
server = request.getfuncargvalue("server")
api = API("XXX", "XXX", "uk") # locale does not matter here!
api.host = "%s:%s" % server.server_address
if hasattr(request, "param"): # for tests generated by pytest_generate_tests
api.VERSION = request.param["version"]
api.REQUESTS_PER_SECOND = 10000 # just for here!
return api
def pytest_funcarg__api(request):
server = request.getfuncargvalue('server')
api = API('XXX', 'XXX', 'uk') # locale does not matter here!
api.host = '%s:%s' % server.server_address
if hasattr(request, 'param'): # for tests generated by pytest_generate_tests
api.VERSION = request.param['version']
api.REQUESTS_PER_SECOND = 10000 # just for here!
return api
def __init__(self, *args, **kwargs):
"""
:param cachedir: Path to directory containing cached responses.
"""
self.cache = kwargs.pop("cachedir", DEFAULT_CACHE_DIR)
API.__init__(self, *args, **kwargs)
if self.cache and not os.path.isdir(self.cache):
os.mkdir(self.cache)
def create_cart(asin, quantity=1):
api = API(settings.AMAZON_AWS_KEY, settings.AMAZON_SECRET_KEY,
settings.AMAZON_API_LOCALE, settings.AMAZON_ASSOCIATE_TAG)
cart = api.cart_create({asin: quantity})
try:
return unicode(cart.Cart.PurchaseURL)
except ValueError, InvalidCartItem:
raise ValidationError()
def __init__(self, *args, **kwargs):
"""
:param cachedir: Path to directory containing cached responses.
"""
self.cache = kwargs.pop('cachedir', DEFAULT_CACHE_DIR)
self.cachetime = kwargs.pop('cachetime', False) # i.e. indefinite
API.__init__(self, *args, **kwargs)
if self.cache and not os.path.isdir(self.cache):
os.mkdir(self.cache)
def __init__(self, access_key_id, secret_access_key, locale,
cachedir=DEFAULT_CACHE_DIR, **kwargs):
"""
:param cachedir: Path to directory containing cached responses.
"""
API.__init__(self, access_key_id, secret_access_key, locale, **kwargs)
self.cache = cachedir
if self.cache and not os.path.isdir(self.cache):
os.mkdir(self.cache)
def __init__(self, *args, **kwargs):
"""
:param cachedir: Path to directory containing cached responses.
"""
self.cache = kwargs.pop('cachedir', DEFAULT_CACHE_DIR)
self.cachetime = kwargs.pop('cachetime', False) # i.e. indefinite
API.__init__(self, *args, **kwargs)
if self.cache and not os.path.isdir(self.cache):
os.mkdir(self.cache)
def pytest_funcarg__api(request):
server = request.getfuncargvalue('server')
api = API('XXX', 'XXX', 'uk') # locale does not matter here!
api.host = '%s:%s' % server.server_address
if hasattr(request,
'param'): # for tests generated by pytest_generate_tests
api.VERSION = request.param['version']
api.REQUESTS_PER_SECOND = 10000 # just for here!
return api
def create_cart(asin, quantity=1):
api = API(
settings.AMAZON_AWS_KEY,
settings.AMAZON_SECRET_KEY,
settings.AMAZON_API_LOCALE,
settings.AMAZON_ASSOCIATE_TAG)
cart = api.cart_create({asin: quantity})
try:
return unicode(cart.Cart.PurchaseURL)
except ValueError, InvalidCartItem:
raise ValidationError()
def getAmazonContent(query):
api = API(AWS_KEY, SECRET_KEY, 'uk', associate_tag='giracake-21', processor=minidom_response_parser)
node = api.item_search('All', Keywords=query, ResponseGroup="Medium")
products = node.getElementsByTagName('Item')
data = []
for product in products:
#print product.toprettyxml()
summaryNodes = product.getElementsByTagName('OfferSummary')
if len(summaryNodes) > 0:
for summaryNode in summaryNodes:
priceNodes = summaryNode.getElementsByTagName('LowestNewPrice')
for priceNode in priceNodes:
formattedPrice = priceNode.getElementsByTagName('FormattedPrice')
price = formattedPrice[0].childNodes[0].nodeValue
imageSetsNodes = product.getElementsByTagName('ImageSets')
if len(imageSetsNodes) > 0:
for imageSetsNode in imageSetsNodes:
imageSetNodes = imageSetsNode.getElementsByTagName('ImageSet')
for imageSetNode in imageSetNodes:
if imageSetNode.attributes["Category"].value != "primary":
continue
mediumImage = imageSetNode.getElementsByTagName('LargeImage')
url = mediumImage[0].getElementsByTagName('URL')[0]
image = url.childNodes[0].nodeValue
prod = {}
prod['url'] = getText(product.getElementsByTagName('DetailPageURL')[0].childNodes)
prod['title'] = getText(product.getElementsByTagName('Title')[0].childNodes)
prod['img'] = image
prod['price'] = price
#imagesets = getText(product.getElementsByTagName('MediumImage'))[0].childNodes
#print imagesets
#imageset = getText(imagesets.getElementsByTagName('ImageSet'))[0]
data += [prod]
return data
def test_init_with_config_file_and_parameters(self, configfiles):
configfiles.add_file(
"""
[Credentials]
access_key = ABCDEFGH12345
secret_key = abcdegf43""", '~/.amazon-product-api')
API(locale='de')
def _fetch(self, url):
"""
Retrieves XML response from Amazon. In case of a timeout, it will try
:const:`~RetryAPI.TRIES`` times before raising an error.
"""
attempts = 0
delay = self.DELAY
while True:
try:
attempts += 1
return API._fetch(self, url)
except urllib2.URLError, e:
# if a timeout occurred
# wait for some time before trying again
reason = getattr(e, 'reason', None)
if isinstance(reason,
socket.timeout) and attempts < self.TRIES:
time.sleep(delay)
delay *= self.BACKOFF
continue
# otherwise reraise the original error
raise
def _fetch(self, url):
"""
Retrieves XML response from Amazon. In case of a timeout, it will try
:const:`~RetryAPI.TRIES`` times before raising an error.
"""
attempts = 0
delay = self.DELAY
while True:
try:
attempts += 1
return API._fetch(self, url)
except URLError:
e = sys.exc_info()[1] # Python 2/3 compatible
# if a timeout occurred
# wait for some time before trying again
reason = getattr(e, 'reason', None)
if isinstance(reason, socket.timeout) and attempts < self.TRIES:
time.sleep(delay)
delay *= self.BACKOFF
continue
# otherwise reraise the original error
raise
def test_init_with_config_file(self, configfiles):
configfiles.add_file(
"""
[Credentials]
access_key = ABCDEFGH12345
secret_key = abcdegf43
locale = de""", '~/.amazon-product-api')
api = API()
def __init__(self, locale='de'):
gtk.Window.__init__(self, gtk.WINDOW_TOPLEVEL)
self.set_title("BrowseNode Explorer")
self.set_size_request(400, 200)
self.connect("delete_event", self.on_delete)
self.locale = locale
self.api = API(locale=self.locale)
# create a TreeStore with one string column to use as the model
self.treestore = gtk.TreeStore(int, str)
# create the TreeView using treestore
self.treeview = gtk.TreeView(self.treestore)
# add column id
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('id', renderer, text=0)
self.treeview.append_column(column)
# add column name
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('name', renderer, text=1)
column.set_sort_column_id(1) # Allow sorting on the column
self.treeview.append_column(column)
# make it clickable
self.treeview.add_events(gtk.gdk.BUTTON_PRESS_MASK)
self.treeview.connect('button_press_event', self.on_tree_click)
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.add(self.treeview)
self.add(scrolled)
self.show_all()
# populate with root nodes
# but avoid duplicated node ids
node_ids = set(NODE_IDS[self.locale].values())
for name, id in NODE_IDS[self.locale].items():
if id in node_ids:
self.treestore.append(None, [id, name])
node_ids.remove(id)
def search():
form = SearchForm()
if form.validate_on_submit():
referrer = form.referrer.data
query = form.query.data
results = model.session.query(model.Library).filter(
model.Library.product_desc.like('%' + query + '%')).all()
if not results:
#redirect to new page with amazon search results
return redirect(
url_for('amazon_bottlenose2', query=query, referrer=referrer))
else:
for i in results:
similar_products = None
if i.product.asin:
asin = i.product.asin
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
try:
similar_root = api.similarity_lookup(asin,
ResponseGroup='Large')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = similar_root.nsmap.get(None, '')
similar_products = similar_root.xpath(
'//aws:Items/aws:Item', namespaces={'aws': nspace})
except:
similar_products = None
#render page with search results
if referrer == 'dashboard':
return render_template('results.html',
results=results,
similar_products=similar_products)
else:
form = AddProductForm()
return render_template('add_product_results.html',
results=results,
similar_products=similar_products,
form=form)
# else:
# flash("Invalid Search")
return render_template("search.html", form=form)
def amazon_search():
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
result = api.item_search('All',
ResponseGroup='Large', AssociateTag='boitba-20', Keywords='tent marmot', ItemPage=2)
total_results = result.results
#item_search returns pages not items
# extract paging information
#total_results = result.results
#Stotal_pages = len(result) # or result.pages
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
return render_template("amazon_search.html", node = result, total_results=total_results)
def amazon(query):
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
similar_root = api.similarity_lookup('B0058U6DQC', ResponseGroup='Large')
product_root = api.item_lookup('B0058U6DQC', ResponseGroup='Large')
product_root = api.item_search(title='unicorn', ResponseGroup='Large')
more_products = api.item_search('Books', Publisher='Galileo Press')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = similar_root.nsmap.get(None, '')
similar_products = similar_root.xpath('//aws:Items/aws:Item',
namespaces={'aws' : nspace})
# more_products = product_root.xpath('//aws:Items/aws:Item',
# namespaces={'aws' : nspace})
return render_template("amazon.html", similar_products=similar_products, more_products = more_products, query=query)
def setup_class(cls):
"""
Collect all XML files stored.
"""
# TODO: Skip tests if no XML files are found?
cls.test_files = [
os.path.join(XML_TEST_DIR, dir, f) for dir in TESTABLE_API_VERSIONS
for f in os.listdir(os.path.join(XML_TEST_DIR, dir))
if f.lower().endswith('.xml')
]
cls.api = API(cls.ACCESS_KEY, cls.SECRET_KEY, 'us')
def amazon_search():
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
result = api.item_search('All',
ResponseGroup='Large',
AssociateTag='boitba-20',
Keywords='tent marmot',
ItemPage=2)
total_results = result.results
#item_search returns pages not items
# extract paging information
#total_results = result.results
#Stotal_pages = len(result) # or result.pages
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
return render_template("amazon_search.html",
node=result,
total_results=total_results)
def search():
form = SearchForm()
if form.validate_on_submit():
referrer = form.referrer.data
query = form.query.data
results = model.session.query(model.Library).filter(model.Library.product_desc.like('%'+ query + '%')).all()
if not results:
#redirect to new page with amazon search results
return redirect(url_for('amazon_bottlenose2', query=query, referrer=referrer))
else:
for i in results:
similar_products = None
if i.product.asin:
asin = i.product.asin
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
try:
similar_root = api.similarity_lookup(asin, ResponseGroup='Large')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = similar_root.nsmap.get(None, '')
similar_products = similar_root.xpath('//aws:Items/aws:Item',
namespaces={'aws' : nspace})
except:
similar_products = None
#render page with search results
if referrer == 'dashboard':
return render_template('results.html', results=results, similar_products=similar_products)
else:
form = AddProductForm()
return render_template('add_product_results.html', results=results, similar_products=similar_products, form=form)
# else:
# flash("Invalid Search")
return render_template("search.html", form=form)
def __init__(self, locale='de'):
gtk.Window.__init__(self, gtk.WINDOW_TOPLEVEL)
self.set_title("BrowseNode Explorer")
self.set_size_request(400, 200)
self.connect("delete_event", self.on_delete)
self.locale = locale
self.api = API(AWS_KEY, SECRET_KEY, self.locale)
# create a TreeStore with one string column to use as the model
self.treestore = gtk.TreeStore(int, str)
# create the TreeView using treestore
self.treeview = gtk.TreeView(self.treestore)
# add column id
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('id', renderer, text=0)
self.treeview.append_column(column)
# add column name
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('name', renderer, text=1)
column.set_sort_column_id(1) # Allow sorting on the column
self.treeview.append_column(column)
# make it clickable
self.treeview.add_events(gtk.gdk.BUTTON_PRESS_MASK)
self.treeview.connect('button_press_event', self.on_tree_click)
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.add(self.treeview)
self.add(scrolled)
self.show_all()
# populate with root nodes
# but avoid duplicated node ids
node_ids = set(NODE_IDS[self.locale].values())
for name, id in NODE_IDS[self.locale].items():
if id in node_ids:
self.treestore.append(None, [id, name])
node_ids.remove(id)
def _fetch(self, url):
if self.cache:
path = os.path.join(self.cache, "%s.xml" % self.get_hash(url))
# if response was fetched previously, use that one
if os.path.isfile(path):
return open(path)
# fetch original response from Amazon
resp = API._fetch(self, url)
if self.cache:
fp = open(path, "w+")
fp.write(etree.tostring(etree.parse(resp), pretty_print=True))
fp.seek(0)
return fp
return resp
def _fetch(self, url):
if self.cache:
path = os.path.join(self.cache, '%s.xml' % self.get_hash(url))
# if response was fetched previously, use that one
if os.path.isfile(path):
return open(path)
# fetch original response from Amazon
resp = API._fetch(self, url)
if self.cache:
fp = open(path, 'w+')
fp.write(etree.tostring(etree.parse(resp), pretty_print=True))
fp.seek(0)
return fp
return resp
def _fetch(self, url):
if self.cache:
path = os.path.join(self.cache, '%s.xml' % self.get_hash(url))
# if response was fetched previously, use that one
if os.path.isfile(path):
return open(path)
# fetch original response from Amazon
resp = API._fetch(self, url)
if self.cache:
fp = open(path, 'w+')
node = xml.dom.minidom.parseString(resp.read())
fp.write(node.toprettyxml())
#from lxml import etree
#fp.write(etree.tostring(etree.parse(resp), pretty_print=True))
fp.seek(0)
return fp
return resp
def amazon(query):
api = API(AWS_KEY, SECRET_KEY, 'us', ASSOC_TAG)
similar_root = api.similarity_lookup('B0058U6DQC', ResponseGroup='Large')
product_root = api.item_lookup('B0058U6DQC', ResponseGroup='Large')
product_root = api.item_search(title='unicorn', ResponseGroup='Large')
more_products = api.item_search('Books', Publisher='Galileo Press')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = similar_root.nsmap.get(None, '')
similar_products = similar_root.xpath('//aws:Items/aws:Item',
namespaces={'aws': nspace})
# more_products = product_root.xpath('//aws:Items/aws:Item',
# namespaces={'aws' : nspace})
return render_template("amazon.html",
similar_products=similar_products,
more_products=more_products,
query=query)
def __init__(self, *args, **kwargs):
API.__init__(self, *args, **kwargs)
self.batch_mode = False
"""
Find similar items to "Small Favor: A Novel of the Dresden Files"
(ASIN 0451462009).
"""
from amazonproduct.api import API
from config import AWS_KEY, SECRET_KEY
if __name__ == '__main__':
api = API(AWS_KEY, SECRET_KEY, 'us')
root = api.similarity_lookup('0451462009')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = root.nsmap.get(None, '')
books = root.xpath('//aws:Items/aws:Item',
namespaces={'aws' : nspace})
for book in books:
print 'ASIN %-10s' % book.ASIN,
print unicode(book.ItemAttributes.Author), ':',
print unicode(book.ItemAttributes.Title)
class SoupProcessor(BaseProcessor):
"""
Custom response parser using BeautifulSoup to parse the returned XML.
"""
def parse(self, fp):
soup = BeautifulSoup.BeautifulSoup(fp.read())
# parse errors
for error in soup.findAll('error'):
code = error.find('code').text
msg = error.find('message').text
raise AWSError(code, msg)
return soup
if __name__ == '__main__':
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='us', processor=SoupProcessor())
result = api.item_lookup('0718155157')
print result
# ...
# now do something with it!
class BrowseNodeExplorer (gtk.Window):
"""
Gtk explorer for Amazon BrowseNodes.
"""
def on_delete(self, widget, event, data=None):
# closes the window and quits.
gtk.main_quit()
return False
def on_tree_click(self, widget, event, data=None):
# if double click
if event.type == gtk.gdk._2BUTTON_PRESS:
# get data from highlighted selection
treeselection = self.treeview.get_selection()
model, iter = treeselection.get_selected()
name_of_data = self.treestore.get_value(iter, 0)
# and fetch selected node
self.fetch_nodes(name_of_data)
def __init__(self, locale='de'):
gtk.Window.__init__(self, gtk.WINDOW_TOPLEVEL)
self.set_title("BrowseNode Explorer")
self.set_size_request(400, 200)
self.connect("delete_event", self.on_delete)
self.locale = locale
self.api = API(AWS_KEY, SECRET_KEY, self.locale)
# create a TreeStore with one string column to use as the model
self.treestore = gtk.TreeStore(int, str)
# create the TreeView using treestore
self.treeview = gtk.TreeView(self.treestore)
# add column id
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('id', renderer, text=0)
self.treeview.append_column(column)
# add column name
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('name', renderer, text=1)
column.set_sort_column_id(1) # Allow sorting on the column
self.treeview.append_column(column)
# make it clickable
self.treeview.add_events(gtk.gdk.BUTTON_PRESS_MASK)
self.treeview.connect('button_press_event', self.on_tree_click)
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.add(self.treeview)
self.add(scrolled)
self.show_all()
# populate with root nodes
# but avoid duplicated node ids
node_ids = set(NODE_IDS[self.locale].values())
for name, id in NODE_IDS[self.locale].items():
if id in node_ids:
self.treestore.append(None, [id, name])
node_ids.remove(id)
def _find_row(self, node_id):
def match_func(row, data):
# data is a tuple containing column number, key
column, key = data
return row[column] == key
def search(rows, func, data):
if not rows: return None
for row in rows:
if func(row, data):
return row
result = search(row.iterchildren(), func, data)
if result: return result
return None
return search(self.treestore, match_func, (0, node_id))
def fetch_nodes(self, node_id):
"""
Fetches a BrowseNode from Amazon.
"""
# fetch matching row from treestore
row = self._find_row(node_id)
# fetch Amazon data
node = self.api.browse_node_lookup(node_id).BrowseNodes.BrowseNode
id = node.BrowseNodeId.pyval
name = node.Name.pyval
is_root = hasattr(node, 'IsCategoryRoot') and node.IsCategoryRoot.pyval == 1
#~ from lxml import etree
#~ print etree.tostring(node, pretty_print=True)
#try:
# parents = dict((parent.BrowseNodeId.pyval, parent.Name.pyval)
# for parent in node.Ancestors.BrowseNode)
#except AttributeError:
# parents = {}
#
#piter = None
#for parent_id, parent_name in parents.items():
# piter = self.treestore.append(None, [parent_id, parent_name])
#
#iter = self.treestore.append(piter, [id, name])
# replace node name
if is_root:
row[1] = node.Ancestors.BrowseNode.Name.text
try:
children = dict((child.BrowseNodeId.pyval, child.Name.pyval)
for child in node.Children.BrowseNode)
except AttributeError:
children = {}
for child_id, child_name in children.items():
self.treestore.append(row.iter, [child_id, child_name])
# expand nodes of just added
self.treeview.expand_row(tuple(row.path), True)
def main(self):
gtk.main()
def __init__(self, aws_access_key, aws_secret_key, affiliate_key):
self.api = API(aws_access_key, aws_secret_key, 'us', affiliate_key)
def pytest_funcarg__api(request):
"""
Initialises API for each test call (formerly done with ``setup_method()``).
"""
server = request.getfuncargvalue('server')
url_reg = re.compile(r'^http://(?P<host>[\w\-\.]+)(?P<path>/onca/xml.*)$')
# the following parameters are injected by pytest_generate_tests
locale = request.param.locale
version = request.param.version
xml_response = request.param.xml_response
processor = TESTABLE_PROCESSORS[request.param.processor]
api = API(locale=locale, processor=processor)
api.VERSION = version
api.REQUESTS_PER_SECOND = 10000 # just for here!
def counter(fnc):
"""
Wrapper function for ``_fetch`` which
1. keeps track of the times has been called and adjusts the path to the
corresponding XML response
2. Fetches any response that has not been cached from the live servers
"""
api._count = 0
def wrapped(url):
api._count += 1
path = xml_response
if api._count > 1:
root, ext = os.path.splitext(path)
path = '%s-%i%s' % (root, api._count, ext)
try:
if request.config.option.fetch == 'all':
raise ResponseRequired
try:
content = open(path, 'r').read()
# If the XML response has been previously fetched compare
# request arguments in order to see if there are any changes
cached_params = utils.arguments_from_cached_xml(content)
current_params = utils.arguments_from_url(url)
if cached_params != current_params:
raise ArgumentMismatch
except IOError:
if request.config.option.fetch in ('outdated', 'missing'):
raise ResponseRequired
raise pytest.skip('No cached XML response found!')
except ArgumentMismatch:
if request.config.option.fetch == 'outdated':
raise ResponseRequired
msg = ('Cached arguments in %s differ from the ones '
'currently tested against!\ncached=%r\ncurrent=%r' %
(path, cached_params, current_params))
return pytest.skip(msg)
except AttributeError:
# XML for error messages have no Argument elements!
pass
except ResponseRequired:
# fetch XML via urllib2 rather than directly via
# lxml.etree.parse() to avoid, for instance, problems with HTTP
# 403 errors
resp = requests.get(url, headers={'User-Agent': USER_AGENT})
root = lxml.etree.fromstring(resp.text)
# overwrite sensitive information in XML document.
for arg in root.xpath('//aws:Argument',
namespaces={'aws': root.nsmap[None]}):
if arg.get('Name') in ('Signature', 'AWSAccessKeyId',
'AssociateTag'):
arg.set('Value', 'X' * 15)
content = lxml.etree.tostring(root, pretty_print=True)
# complain loudly about missing credentials
# UNLESS it was actually on purpose!
if (six.b('MissingClientTokenId') in content
and getattr(request.function, 'refetch', True)):
raise pytest.fail(
'Cannot fetch XML response without credentials!')
if not os.path.exists(os.path.dirname(path)):
os.mkdir(os.path.dirname(path))
open(path, 'wb').write(content)
# We simply exchange the real host with the local one now!
# Note: Although strictly speaking it does not matter which URL is
# called exactly, to appeal to one's sense of correctness, let's
# keep at least the correct path!
url = url_reg.sub(r'%s\g<path>' % server.url, url)
server.serve_content(content)
return fnc(url)
return wrapped
api._fetch = counter(api._fetch)
return api
def __init__(self, *args, **kwargs):
API.__init__(self, *args, **kwargs)
self.batch_mode = False
# -*- coding: utf-8 -*-
from amazonproduct.api import API
from amazonproduct.errors import AWSError
from amazonproduct.processors import BaseProcessor
import BeautifulSoup
class SoupProcessor(BaseProcessor):
def parse(self, fp):
soup = BeautifulSoup.BeautifulSoup(fp.read())
for error in soup.findAll('error'):
code = error.find('code').text
msg = error.find('message').text
raise AWSError(code, msg)
return soup
if __name__ == '__main__':
api = API(locale='jp', processor=SoupProcessor())
result = api.item_lookup('B00LCL7A3G')
print result
class BrowseNodeExplorer(gtk.Window):
"""
Gtk explorer for Amazon BrowseNodes.
"""
def on_delete(self, widget, event, data=None):
# closes the window and quits.
gtk.main_quit()
return False
def on_tree_click(self, widget, event, data=None):
# if double click
if event.type == gtk.gdk._2BUTTON_PRESS:
# get data from highlighted selection
treeselection = self.treeview.get_selection()
model, iter = treeselection.get_selected()
name_of_data = self.treestore.get_value(iter, 0)
# and fetch selected node
self.fetch_nodes(name_of_data)
def __init__(self, locale='de'):
gtk.Window.__init__(self, gtk.WINDOW_TOPLEVEL)
self.set_title("BrowseNode Explorer")
self.set_size_request(400, 200)
self.connect("delete_event", self.on_delete)
self.locale = locale
self.api = API(locale=self.locale)
# create a TreeStore with one string column to use as the model
self.treestore = gtk.TreeStore(int, str)
# create the TreeView using treestore
self.treeview = gtk.TreeView(self.treestore)
# add column id
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('id', renderer, text=0)
self.treeview.append_column(column)
# add column name
renderer = gtk.CellRendererText()
column = gtk.TreeViewColumn('name', renderer, text=1)
column.set_sort_column_id(1) # Allow sorting on the column
self.treeview.append_column(column)
# make it clickable
self.treeview.add_events(gtk.gdk.BUTTON_PRESS_MASK)
self.treeview.connect('button_press_event', self.on_tree_click)
scrolled = gtk.ScrolledWindow()
scrolled.set_policy(gtk.POLICY_AUTOMATIC, gtk.POLICY_AUTOMATIC)
scrolled.add(self.treeview)
self.add(scrolled)
self.show_all()
# populate with root nodes
# but avoid duplicated node ids
node_ids = set(NODE_IDS[self.locale].values())
for name, id in NODE_IDS[self.locale].items():
if id in node_ids:
self.treestore.append(None, [id, name])
node_ids.remove(id)
def _find_row(self, node_id):
def match_func(row, data):
# data is a tuple containing column number, key
column, key = data
return row[column] == key
def search(rows, func, data):
if not rows: return None
for row in rows:
if func(row, data):
return row
result = search(row.iterchildren(), func, data)
if result: return result
return None
return search(self.treestore, match_func, (0, node_id))
def fetch_nodes(self, node_id):
"""
Fetches a BrowseNode from Amazon.
"""
# fetch matching row from treestore
row = self._find_row(node_id)
# fetch Amazon data
node = self.api.browse_node_lookup(node_id).BrowseNodes.BrowseNode
id = node.BrowseNodeId.pyval
name = node.Name.pyval
is_root = hasattr(node,
'IsCategoryRoot') and node.IsCategoryRoot.pyval == 1
#~ from lxml import etree
#~ print etree.tostring(node, pretty_print=True)
#try:
# parents = dict((parent.BrowseNodeId.pyval, parent.Name.pyval)
# for parent in node.Ancestors.BrowseNode)
#except AttributeError:
# parents = {}
#
#piter = None
#for parent_id, parent_name in parents.items():
# piter = self.treestore.append(None, [parent_id, parent_name])
#
#iter = self.treestore.append(piter, [id, name])
# replace node name
if is_root:
row[1] = node.Ancestors.BrowseNode.Name.text
try:
children = dict((child.BrowseNodeId.pyval, child.Name.pyval)
for child in node.Children.BrowseNode)
except AttributeError:
children = {}
for child_id, child_name in children.items():
self.treestore.append(row.iter, [child_id, child_name])
# expand nodes of just added
self.treeview.expand_row(tuple(row.path), True)
def main(self):
gtk.main()
from pymongo import Connection
from amazonproduct.api import API
from config import AWS_KEY, SECRET_KEY
if __name__ == '__main__':
item_attr = ['Brand', 'Manufacturer', 'Model', 'OperatingSystem', 'ProductGroup', 'ProductTypeName', 'ReleaseDate', 'Title']
item_img = ['SmallImage', 'MediumImage', 'LargeImage']
item_offer = ['SalesRank', 'TotalNew']
item_price = ['LowestNewPrice', 'LowestRefurbishedPrice', 'LowestUsedPrice']
img_attr = ['Height', 'URL', 'Width']
connection = Connection()
db = connection.linux_laptops
api = API(AWS_KEY, SECRET_KEY, 'us', associate_tag='notebux-20')
paginator = api.item_search('Electronics', Keywords='linux laptop', ResponseGroup='Medium')
for root in paginator:
nspace = root.nsmap.get(None, '')
items = root.xpath('//aws:Items/aws:Item', namespaces={'aws' : nspace})
for i in items:
doc = {}
try:
doc['ASIN'] = str(i.ASIN)
doc['URL'] = str(i.DetailPageURL)
# Images
doc['Images'] = {}
for a in item_img:
item_attr = [
'Brand', 'Manufacturer', 'Model', 'OperatingSystem', 'ProductGroup',
'ProductTypeName', 'ReleaseDate', 'Title'
]
item_img = ['SmallImage', 'MediumImage', 'LargeImage']
item_offer = ['SalesRank', 'TotalNew']
item_price = [
'LowestNewPrice', 'LowestRefurbishedPrice', 'LowestUsedPrice'
]
img_attr = ['Height', 'URL', 'Width']
connection = Connection()
db = connection.linux_laptops
api = API(AWS_KEY, SECRET_KEY, 'us', associate_tag='notebux-20')
paginator = api.item_search('Electronics',
Keywords='linux laptop',
ResponseGroup='Medium')
for root in paginator:
nspace = root.nsmap.get(None, '')
items = root.xpath('//aws:Items/aws:Item', namespaces={'aws': nspace})
for i in items:
doc = {}
try:
doc['ASIN'] = str(i.ASIN)
doc['URL'] = str(i.DetailPageURL)
# Images
def pytest_funcarg__api(request):
"""
Initialises API for each test call (formerly done with ``setup_method()``).
"""
server = request.getfuncargvalue('server')
url_reg = re.compile(r'^http://(?P<host>[\w\-\.]+)(?P<path>/onca/xml.*)$')
# the following parameters are injected by pytest_generate_tests
locale = request.param['locale']
version = request.param['version']
xml_response = request.param['xml_response']
processor = TESTABLE_PROCESSORS[request.param['processor']]
if isinstance(processor, type):
processor = processor()
api = API(locale=locale, processor=processor)
api.VERSION = version
api.REQUESTS_PER_SECOND = 10000 # just for here!
def counter(fnc):
"""
Wrapper function for ``_fetch`` which
1. keeps track of the times has been called and adjusts the path to the
corresponding XML response
2. Fetches any response that has not been cached from the live servers
"""
api._count = 0
def wrapped(url):
api._count += 1
path = xml_response
if api._count > 1:
root, ext = os.path.splitext(path)
path = '%s-%i%s' % (root, api._count, ext)
try:
if request.config.option.fetch == 'all':
raise ResponseRequired
try:
content = open(path, 'r').read()
# If the XML response has been previously fetched compare
# request arguments in order to see if there are any changes
cached_params = utils.arguments_from_cached_xml(content)
current_params = utils.arguments_from_url(url)
if cached_params != current_params:
raise ArgumentMismatch
except IOError:
if request.config.option.fetch in ('outdated', 'missing'):
raise ResponseRequired
raise pytest.skip('No cached XML response found!')
except ArgumentMismatch:
if request.config.option.fetch == 'outdated':
raise ResponseRequired
return pytest.skip('Cached arguments in %s differ from the '
'ones currently tested against!' % path)
#'\ncached=%r\ncurrent=%r' % (path,
#cached_params, current_params))
except AttributeError:
# XML for error messages have no Argument elements!
pass
except ResponseRequired:
# fetch XML via urllib2 rather than directly via
# lxml.etree.parse() to avoid, for instance, problems with HTTP
# 403 errors
try:
req = urllib2.Request(url, headers={'User-Agent': USER_AGENT})
xml = urllib2.urlopen(req).read()
except urllib2.HTTPError, e:
xml = e.read()
root = lxml.etree.fromstring(xml)
# overwrite sensitive information in XML document.
for arg in root.xpath('//aws:Argument',
namespaces={'aws': root.nsmap[None]}):
if arg.get('Name') in ('Signature', 'AWSAccessKeyId',
'AssociateTag'):
arg.set('Value', 'X'*15)
content = lxml.etree.tostring(root, pretty_print=True)
# complain loudly about missing credentials
# UNLESS it was actually on purpose!
if ('MissingClientTokenId' in content
and getattr(request.function, 'refetch', True)):
raise pytest.fail('Cannot fetch XML response without credentials!')
if not os.path.exists(os.path.dirname(path)):
os.mkdir(os.path.dirname(path))
open(path, 'wb').write(content)
# We simply exchange the real host with the local one now!
# Note: Although strictly speaking it does not matter which URL is
# called exactly, to appeal to one's sense of correctness, let's
# keep at least the correct path!
url = url_reg.sub(r'%s\g<path>' % server.url, url)
server.serve_content(content)
return fnc(url)
return wrapped
class AmazonProductSearch():
def __init__(self, aws_access_key, aws_secret_key, affiliate_key):
self.api = API(aws_access_key, aws_secret_key, 'us', affiliate_key)
def _create_product(self, result):
product = {}
product['ASIN'] = str(result.ASIN)
product['Amount'] = str(result.OfferSummary.LowestNewPrice.Amount)
product['Binding'] = str(result.ItemAttributes.Binding)
product['DetailPageURL'] = str(result.DetailPageURL)
return product
def item_search(self, title, actor, expected_running_time,
ResponseGroup='OfferFull, Medium', release_year=None):
'''
This method searches Amazon for DVD, Blu-ray, and/or Amazon Instant
Video listings of the movie that we are searching. Actor and running
time are included in the query to improve the accuracy of the results
returned by Amazon.
Args:
title: Title of the movie
actor: Top billing actor for the movie
expected_running_time: Total running time for the movie
ResponseGroup: See response group in amazon product search api
documentation
'''
try:
results = self.api.item_search('DVD', actor=actor, Keywords=title,
ResponseGroup=ResponseGroup)
except Exception as e:
print e
return []
rv = []
max_release_year_diff = 1
max_running_time_diff = 5
bindings_seen = {"DVD": False, "Blu-ray": False,
"Amazon Instant Video": False}
for result in results:
etree.tostring(result, pretty_print=True)
try:
if all(bindings_seen.values()):
return rv
result_binding = str(result.ItemAttributes.Binding)
result_release_year = int(str(result.ItemAttributes.ReleaseDate).split('-')[0])
result_running_time = int(result.ItemAttributes.RunningTime)
result_title = str(result.ItemAttributes.Title)
if fuzz.partial_ratio(title.lower(), result_title.lower()) < 100:
continue
if release_year is not None and abs(release_year - result_release_year) > max_release_year_diff:
continue
if abs(expected_running_time - result_running_time) > max_running_time_diff:
continue
if result_binding not in bindings_seen.keys():
continue
if bindings_seen.get(result_binding):
continue
rv.append(self._create_product(result))
bindings_seen[result_binding] = True
except Exception as e:
print e
continue
return rv
"""
Get all books published by "Galileo Press".
"""
from amazonproduct.api import API
import lxml
if __name__ == '__main__':
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='de')
result = api.item_search('Books', Publisher='Galileo Press',
ResponseGroup='Large')
# extract paging information
total_results = result.results
total_pages = len(result) # or result.pages
for book in result:
print 'page %d of %d' % (result.current, total_pages)
#~ from lxml import etree
#~ print etree.tostring(book, pretty_print=True)
print book.ASIN,
print unicode(book.ItemAttributes.Author), ':',
print unicode(book.ItemAttributes.Title),
aws_key = '1QTKKA77QPEB2MGMBYR2'
aws_secret = '643VohcwGOqct4C2AeLOHbmN4uR6Is2Gby8+4cX4'
result_ns = {
'xmlns': 'http://webservices.amazon.com/AWSECommerceService/2011-08-01'
}
def response_parser(fp):
dom = fns.parse_xml(fp.read())
return dom
client = API(aws_key,
aws_secret,
'us',
'springpartner-20',
processor=response_parser)
def get_items(response):
return response.xpath('//xmlns:Item', namespaces=result_ns)
def get_price(item):
"""
Returns the amazon price for an item.
"""
result = item.xpath('//xmlns:LowestNewPrice/xmlns:FormattedPrice',
namespaces=result_ns)
if len(result) > 0:
"""
Custom response parser using BeautifulSoup to parse the returned XML.
"""
def parse(self, fp):
soup = BeautifulSoup.BeautifulSoup(fp.read())
# parse errors
for error in soup.findAll('error'):
code = error.find('code').text
msg = error.find('message').text
raise AWSError(code, msg)
return soup
if __name__ == '__main__':
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='us', processor=SoupProcessor())
result = api.item_lookup('0718155157')
print result
# ...
# now do something with it!
# -*- coding: utf-8 -*-
from amazonproduct.api import API
from amazonproduct.errors import AWSError
from amazonproduct.processors import BaseProcessor
import BeautifulSoup
class SoupProcessor(BaseProcessor):
def parse(self, fp):
soup = BeautifulSoup.BeautifulStoneSoup(fp.read())
for error in soup.findAll('error'):
code = error.find('code').text
msg = error.find('message').text
raise AWSError(code, msg)
return soup
if __name__ == '__main__':
api = API(locale='jp', processor=SoupProcessor())
result = api.item_search('Apparel', Condition='All', Availability='Available', Keywords='Shirt')
print result
import sys
from amazonproduct.api import API
if __name__ == '__main__':
if len(sys.argv[1:]) == 0:
print __doc__
print 'Usage: %s ISBN' % sys.argv[0]
sys.exit(1)
for isbn in sys.argv[1:]:
isbn = isbn.replace('-', '')
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='us')
for root in api.item_lookup(isbn,
IdType='ISBN',
SearchIndex='Books',
ResponseGroup='EditorialReview'):
nspace = root.nsmap.get(None, '')
reviews = root.xpath('//aws:EditorialReview',
namespaces={'aws': nspace})
for review in reviews:
print unicode(review.Source)
print '-' * 40
print unicode(review.Content)
"""
Find similar items to "Small Favor: A Novel of the Dresden Files"
(ASIN 0451462009).
"""
from amazonproduct.api import API
if __name__ == '__main__':
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='us')
root = api.similarity_lookup('0451462009')
#~ from lxml import etree
#~ print etree.tostring(root, pretty_print=True)
nspace = root.nsmap.get(None, '')
books = root.xpath('//aws:Items/aws:Item', namespaces={'aws': nspace})
for book in books:
print 'ASIN %-10s' % book.ASIN,
print unicode(book.ItemAttributes.Author), ':',
print unicode(book.ItemAttributes.Title)
Get all editorial reviews for books with the specified ISBNs.
"""
import sys
from amazonproduct.api import API
if __name__ == '__main__':
if len(sys.argv[1:]) == 0:
print __doc__
print 'Usage: %s ISBN' % sys.argv[0]
sys.exit(1)
for isbn in sys.argv[1:]:
isbn = isbn.replace('-', '')
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='us')
for root in api.item_lookup(isbn, IdType='ISBN',
SearchIndex='Books', ResponseGroup='EditorialReview'):
nspace = root.nsmap.get(None, '')
reviews = root.xpath('//aws:EditorialReview',
namespaces={'aws' : nspace})
for review in reviews:
print unicode(review.Source)
print '-' * 40
print unicode(review.Content)
def buy_items(item_list, quantity_list=None, category_list=None):
#Instantiate Amazon API object
amazon = API(locale='us')
#Default buy 1 of an item, search in 'All' general category
if quantity_list == None:
quantity_list = [1 for item in item_list]
if category_list == None:
category_list = ['All' for item in item_list]
item_and_quantity = dict()
item_names = dict()
item_prices = dict()
#Search for item in Amazon, get AWS product id
#Keep track of AWS product id the quantity to buy. Must be formatted as shown below
for i in range(len(item_list)):
#print item_list[i]
result = search_item_and_price(amazon, item_list[i], category_list[i])
if result != None:
item_and_quantity[result['aws_id']] = quantity_list[i]
#print type(quantity_list[i])
#Create remote cart of items to buy
cart = amazon.cart_create(item_and_quantity)
#Get cart information. ID and HMAC is used to track and reference the created
#cart. Purchase URL is what the user uses to purchase the desired items
cart_id = cart.Cart.CartId
cart_hmac = cart.Cart.HMAC
purchase_url = cart.Cart.PurchaseURL.text
#print type(purchase_url)
#Get contents of cart
cart_get = amazon.cart_get(cart_id, cart_hmac)
subtotal = cart_get.Cart.SubTotal.getchildren()[2].text
cart_contents = cart_get.Cart.CartItems.getchildren()[1:]
#print type(subtotal)
#pprint.pprint(item_and_quantity)
#Get the prices and names of items in the cart. Since adding items to the cart
#does not (seem to) let the programmer decide which offer listing to add, need
#to check what is in the cart. Need to muck around the Amazon Product Advertising
#API more, but man the documentation leaves much to be desired
for cart_item in cart_contents:
item_info = cart_item.getchildren()
aws_id = item_info[1].text
item_prices[aws_id] = item_info[7].getchildren()[-1].text
if type(item_info[4].text) == unicode:
item_names[aws_id] = item_info[4].text.encode('ascii', 'ignore')
else:
# Assuming it is a string type
item_names[aws_id] = item_info[4].text
#print item_names[aws_id]
#print type(item_names[aws_id])
#print type(item_prices[aws_id])
#pprint.pprint(item_and_quantity)
#pprint.pprint(item_names)
#pprint.pprint(item_prices)
#pprint.pprint(subtotal)
#print type(purchase_url)
return item_and_quantity, item_names, item_prices, subtotal, purchase_url
"""
Get all books published by "Galileo Press".
"""
from amazonproduct.api import API
import lxml
if __name__ == '__main__':
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale='de')
result = api.item_search('Books',
Publisher='Galileo Press',
ResponseGroup='Large')
# extract paging information
total_results = result.results
total_pages = len(result) # or result.pages
for book in result:
print 'page %d of %d' % (result.current, total_pages)
#~ from lxml import etree
#~ print etree.tostring(book, pretty_print=True)
print book.ASIN,
print unicode(book.ItemAttributes.Author), ':',
print unicode(book.ItemAttributes.Title),
from amazonproduct.api import API
from amazonproduct import errors
_here = os.path.dirname(os.path.abspath(__file__))
XSLT = os.path.join(_here, 'xml2json.xslt')
def jsonify(node):
xslt_root = etree.parse(XSLT)
transform = etree.XSLT(xslt_root)
result = transform(node)
return unicode(result)
if __name__ == '__main__':
api = API(locale='de')
collection = []
while True:
try:
ean = raw_input('EAN? ')
resp = api.item_lookup(ean, SearchIndex='All', IdType='EAN', ResponseGroup='Large')
items = resp.Items.Item
for item in items:
attrs = item.ItemAttributes
director = None
if hasattr(attrs, 'Director'):
if len(attrs.Director) > 1:
director = [p.text for p in attrs.Director]
else:
director = attrs.Director.text
def test_responsegroups_as_list(self):
api = API(locale='de')
url = api._build_url(ResponseGroup=['ItemAttributes', 'Images'])
qs = parse_qs(urlparse(url)[4])
assert qs['ResponseGroup'][0] == 'ItemAttributes,Images'
from xml.dom.minidom import parse
def minidom_response_parser(fp):
"""
Custom response parser using xml.dom.minidom.parse
instead of lxml.objectify.
"""
root = parse(fp)
# parse errors
for error in root.getElementsByTagName('Error'):
code = error.getElementsByTagName('Code')[0].firstChild.nodeValue
msg = error.getElementsByTagName('Message')[0].firstChild.nodeValue
raise AWSError(code, msg)
return root
if __name__ == '__main__':
api = API(AWS_KEY, SECRET_KEY, 'us',
processor=minidom_response_parser)
root = api.item_lookup('0718155157')
print root.toprettyxml()
# ...
# now do something with it!
help='ID is an Universal Product Code (UPC).')
parser.add_option('--sku', action='store_const', dest='id_type', const=SKU,
help='ID is an Stock Keeping Unit (SKU).')
parser.add_option('--locale', choices=HOSTS.keys(), dest='locale',
help='Amazon locale to use [default: %default].', default='de')
parser.add_option('-q', '--quiet', action='store_false', dest='verbose',
help='Suppress output.')
(options, ids) = parser.parse_args(sys.argv[1:])
if len(ids) == 0:
parser.error('No IDs specified!')
# Don't forget to create file ~/.amazon-product-api
# with your credentials (see docs for details)
api = API(locale=options.locale)
params = {
'ResponseGroup' : 'Images',
'SearchIndex' : 'All',
'IdType' : options.id_type,
}
# When IdType equals ASIN, SearchIndex cannot be present.
if options.id_type == ASIN:
del params['SearchIndex']
for id in ids:
id = id.replace('-', '')
