import cache
from infogami.utils import delegate
def _get_libraries(site=None):
"""Returns all the libraries each as a dict."""
if 'env' not in web.ctx:
delegate.fakeload()
site = site or web.ctx.site
keys = site.things(query={"type": "/type/library", "limit": 1000, "status": "approved"})
libraries = site.get_many(keys)
return [lib.dict() for lib in libraries]
# cache the result for an hour in memcache
_get_libraries_memoized = cache.memcache_memoize(_get_libraries, "inlibrary._get_libraries", timeout=60*60)
def get_libraries():
"""Returns all the libraries."""
libraries = _get_libraries_memoized()
return [web.ctx.site.new(doc['key'], doc) for doc in libraries]
def _get_library(libraries, ip):
"""Returns the library containing the given ip as dict.
"""
for lib in libraries:
if lib.has_ip(ip):
return lib
def get_library():
"""Returns library document if the IP of the current request is in the IP range of that library.
metadata = process_metadata_dict(item_json.get('metadata', {}))
if metadata:
# if any of the files is access restricted, consider it as
# an access-restricted item.
files = item_json.get('files', [])
metadata['access-restricted'] = any(f.get('private') == 'true' for f in files)
# remember the filenames to construct download links
metadata['_filenames'] = [f['name'] for f in files]
return metadata
def get_metadata(itemid):
item_json = get_item_json(itemid)
return extract_item_metadata(item_json)
get_metadata = cache.memcache_memoize(get_metadata, key_prefix='ia.get_metadata', timeout=5*60)
def process_metadata_dict(metadata):
"""Process metadata dict to make sure multi-valued fields like
collection and external-identifier are always lists.
The metadata API returns a list only if a field has more than one value
in _meta.xml. This puts burden on the application to handle both list and
non-list cases. This function makes sure the known multi-valued fields are
always lists.
"""
mutlivalued = set(['collection', 'external-identifier', 'isbn', 'subject', 'oclc-id'])
def process_item(k, v):
if k in mutlivalued and not isinstance(v, list):
v = [v]
elif k not in mutlivalued and isinstance(v, list):
stats.end()
return web.storage()
# archive.org returns html on internal errors.
# Checking for valid xml before trying to parse it.
if not metaxml.strip().startswith("<?xml"):
return web.storage()
try:
defaults = {"collection": [], "external-identifier": []}
return web.storage(xml2dict(metaxml, **defaults))
except Exception, e:
print >> web.debug, "Failed to parse metaxml for %s: %s" % (itemid, str(e))
return web.storage()
get_meta_xml = cache.memcache_memoize(get_meta_xml, key_prefix="ia.get_meta_xml", timeout=5*60)
def xml2dict(xml, **defaults):
"""Converts xml to python dictionary assuming that the xml is not nested.
To get some tag as a list/set, pass a keyword argument with list/set as value.
>>> xml2dict('<doc><x>1</x><x>2</x></doc>')
{'x': 2}
>>> xml2dict('<doc><x>1</x><x>2</x></doc>', x=[])
{'x': [1, 2]}
"""
d = defaults
dom = minidom.parseString(xml)
for node in dom.documentElement.childNodes:
