def xslt(req, *opts):
"""
Transform the working document using an XSLT file.
:param req: The request
:param opts: Options (unused)
:return: the transformation result
Apply an XSLT stylesheet to the working document. The xslt pipe takes a set of keyword arguments. The only required
argument is 'stylesheet' which identifies the xslt resource. This is looked up either in the package or as a
user-supplied file. The rest of the keyword arguments are made available as string parameters to the XSLT transform.
**Examples**
.. code-block:: yaml
- xslt:
sylesheet: foo.xsl
x: foo
y: bar
"""
stylesheet = req.args.get('stylesheet', None)
if stylesheet is None:
raise PipeException("xslt requires stylesheet")
if req.t is None:
raise PipeException("Your plumbing is missing a select statement.")
params = dict((k, "\'%s\'" % v) for (k, v) in req.args.items())
del params['stylesheet']
ot = xslt_transform(req.t, stylesheet, params)
#log.debug(ot)
return ot
def xslt(req, *opts):
"""
Transform the working document using an XSLT file.
:param req: The request
:param opts: Options (unused)
:return: the transformation result
Apply an XSLT stylesheet to the working document. The xslt pipe takes a set of keyword arguments. The only required
argument is 'stylesheet' which identifies the xslt resource. This is looked up either in the package or as a
user-supplied file. The rest of the keyword arguments are made available as string parameters to the XSLT transform.
**Examples**
.. code-block:: yaml
- xslt:
sylesheet: foo.xsl
x: foo
y: bar
"""
if req.t is None:
raise PipeException("Your plumbing is missing a select statement.")
stylesheet = req.args.get('stylesheet', None)
if stylesheet is None:
raise PipeException("xslt requires stylesheet")
params = dict((k, "\'%s\'" % v) for (k, v) in req.args.items())
del params['stylesheet']
try:
return xslt_transform(req.t, stylesheet, params)
# log.debug(ot)
except Exception as ex:
traceback.print_exc(ex)
raise ex
def request(self, **kwargs):
stats['MD Requests'] += 1
pfx = kwargs.get('pfx', None)
path = kwargs.get('path', None)
content_type = kwargs.get('content_type', None)
log.debug("request pfx=%s, path=%s, content_type=%s" % (pfx, path, content_type))
def escape(m):
st = m.group(0)
if st == '<':
return '<'
if st == '>':
return '>'
return st
def _d(x):
if x is None or len(x) == 0:
return None, None
if x.startswith("{base64}"):
x = x[8:].decode('base64')
if '.' in x:
(p, sep, ext) = x.rpartition('.')
return p, ext
else:
return x, None
_ctypes = {'xml': 'application/xml',
'json': 'application/json',
'htm': 'text/html',
'html': 'text/html',
'ds': 'text/html',
's': 'application/json'}
alias = None
if pfx:
alias = pfx
pfx = self.aliases.get(alias, None)
if pfx is None:
raise NotFound()
path, ext = _d(path)
if pfx and path:
q = "{%s}%s" % (pfx, path)
else:
q = path
logging.debug("request %s %s" % (path, ext))
log.debug(cherrypy.request.headers)
accept = {}
if content_type is None:
if ext is not None and ext in _ctypes:
accept = {_ctypes[ext]: True}
else:
accept = MDServer.MediaAccept()
if ext is not None:
path = "%s.%s" % (path, ext)
else:
accept = {content_type: True}
with self.lock.readlock:
if ext == 'ds':
pdict = dict()
pdict['http'] = cherrypy.request
entityID = kwargs.get('entityID', None)
if entityID is None:
raise HTTPError(400, "400 Bad Request - missing entityID")
pdict['sp'] = self.md.sha1_id(entityID)
pdict['ret'] = kwargs.get('return', None)
if not path:
pdict['search'] = "/search/"
else:
pdict['search'] = "%s.s" % path
if pdict['ret'] is None:
raise HTTPError(400, "400 Bad Request - Missing 'return' parameter")
pdict['returnIDParam'] = kwargs.get('returnIDParam', 'entityID')
cherrypy.response.headers['Content-Type'] = 'text/html'
pdict['style'] = '/static/css/style.css' # TODO figure out how to sensibly set this per request
return template("ds.html").render(**pdict)
elif ext == 's':
paged = bool(kwargs.get('paged', False))
query = kwargs.get('query', None)
page = kwargs.get('page', 0)
page_limit = kwargs.get('page_limit', 10)
entity_filter = kwargs.get('entity_filter', None)
cherrypy.response.headers['Content-Type'] = 'application/json'
if paged:
res, more, total = self.md.search(query,
path=q,
page=int(page),
page_limit=int(page_limit),
entity_filter=entity_filter)
log.debug(dumps({'entities': res, 'more': more, 'total': total}))
return dumps({'entities': res, 'more': more, 'total': total})
else:
return dumps(self.md.search(query, path=q, entity_filter=entity_filter))
elif accept.get('text/html'):
if not q:
if pfx:
title = pfx
else:
title = "Metadata By Attributes"
return template("index.html").render(http=cherrypy.request,
md=self.md,
alias=alias,
aliases=self.aliases,
title=title)
else:
entities = self.md.lookup(q)
if not entities:
raise NotFound()
if len(entities) > 1:
return template("metadata.html").render(http=cherrypy.request, md=self.md, entities=entities)
else:
entity = entities[0]
t = html.fragment_fromstring(unicode(xslt_transform(entity, "entity2html.xsl")))
for c_elt in t.findall(".//code[@role='entity']"):
c_txt = dumptree(entity, pretty_print=True, xml_declaration=False).decode("utf-8")
p = c_elt.getparent()
p.remove(c_elt)
if p.text is not None:
p.text += c_txt # re.sub(".",escape,c_txt)
else:
p.text = c_txt # re.sub(".",escape,c_txt)
xml = dumptree(t, xml_declaration=False).decode('utf-8')
return template("basic.html").render(http=cherrypy.request, content=xml)
else:
for p in self.plumbings:
state = {'request': True,
'headers': {'Content-Type': 'text/xml'},
'accept': accept,
'url': cherrypy.url(relative=False),
'select': q,
'stats': {}}
r = p.process(self.md, state=state)
if r is not None:
cache_ttl = state.get('cache', 0)
log.debug("caching for %d seconds" % cache_ttl)
caching.expires(secs=cache_ttl)
for k, v in state.get('headers', {}).iteritems():
cherrypy.response.headers[k] = v
return r
raise NotFound()
def request(self, **kwargs):
"""The main request processor. This code implements all rendering of metadata.
"""
stats['MD Requests'] += 1
if not self.ready:
raise HTTPError(503, _("Service Unavailable (repository loading)"))
pfx = kwargs.get('pfx', None)
path = kwargs.get('path', None)
content_type = kwargs.get('content_type', None)
log.debug("MDServer pfx=%s, path=%s, content_type=%s" % (pfx, path, content_type))
def _d(x, do_split=True):
if x is not None:
x = x.strip()
log.debug("_d(%s,%s)" % (x, do_split))
if x is None or len(x) == 0:
return None, None
if x.startswith("{base64}"):
x = x[8:].decode('base64')
if do_split and '.' in x:
(pth, dot, extn) = x.rpartition('.')
assert (dot == '.')
if extn in _ctypes:
return pth, extn
return x, None
_ctypes = {'xml': 'application/xml',
'json': 'application/json',
'htm': 'text/html',
'html': 'text/html',
'ds': 'text/html',
's': 'application/json'}
alias = None
if pfx:
alias = pfx
pfx = self.aliases.get(alias, None)
if pfx is None:
raise NotFound()
path, ext = _d(path, content_type is None)
if pfx and path:
q = "{%s}%s" % (pfx, path)
path = "/%s/%s" % (alias, path)
else:
q = path
if ext is not None:
log.debug("request path: %s.%s, headers: %s" % (path, ext, cherrypy.request.headers))
else:
log.debug("request path: %s, headers: %s" % (path, cherrypy.request.headers))
accept = {}
if content_type is None:
if ext is not None and ext in _ctypes:
accept = {_ctypes[ext]: True}
else:
accept = MDServer.MediaAccept()
if ext is not None:
path = "%s.%s" % (path, ext)
else:
accept = {content_type: True}
with self.lock.readlock:
if ext == 'ds':
pdict = dict()
entity_id = kwargs.get('entityID', None)
if entity_id is None:
raise HTTPError(400, _("400 Bad Request - missing entityID"))
pdict['sp'] = self.md.sha1_id(entity_id)
e = self.md.store.lookup(entity_id)
if e is None or len(e) == 0:
raise HTTPError(404)
if len(e) > 1:
raise HTTPError(400, _("400 Bad Request - multiple matches for") + " %s" % entity_id)
pdict['entity'] = self.md.simple_summary(e[0])
if not path:
pdict['search'] = "/search/"
pdict['list'] = "/role/idp.json"
else:
pdict['search'] = "%s.s" % path
pdict['list'] = "%s.json" % path
cherrypy.response.headers['Content-Type'] = 'text/html'
return render_template("ds.html", **pdict)
elif ext == 's':
paged = bool(kwargs.get('paged', False))
query = kwargs.get('query', None)
page = kwargs.get('page', 0)
page_limit = kwargs.get('page_limit', 10)
entity_filter = kwargs.get('entity_filter', None)
related = kwargs.get('related', None)
cherrypy.response.headers['Content-Type'] = 'application/json'
if query is None:
log.debug("empty query - creating one")
query = [cherrypy.request.remote.ip]
referrer = cherrypy.request.headers.get('referrer', None)
if referrer is not None:
log.debug("including referrer: %s" % referrer)
url = urlparse.urlparse(referrer)
host = url.netloc
if ':' in url.netloc:
(host, port) = url.netloc.split(':')
for host_part in host.rstrip(self.psl.get_public_suffix(host)).split('.'):
if host_part is not None and len(host_part) > 0:
query.append(host_part)
log.debug("created query: %s" % ",".join(query))
if paged:
res, more, total = self.md.search(query,
path=q,
page=int(page),
page_limit=int(page_limit),
entity_filter=entity_filter,
related=related)
# log.debug(dumps({'entities': res, 'more': more, 'total': total}))
return dumps({'entities': res, 'more': more, 'total': total})
else:
return dumps(self.md.search(query,
path=q,
entity_filter=entity_filter,
related=related))
elif accept.get('text/html'):
if not q:
if pfx:
title = pfx
else:
title = _("Metadata By Attributes")
return render_template("index.html",
md=self.md,
alias=alias,
aliases=self.aliases,
title=title)
else:
entities = self.md.lookup(q)
if not entities:
raise NotFound()
if len(entities) > 1:
return render_template("metadata.html",
md=self.md,
subheading=q,
entities=entities)
else:
entity = entities[0]
t = html.fragment_fromstring(unicode(xslt_transform(entity, "entity2html.xsl")))
for c_elt in t.findall(".//code[@role='entity']"):
c_txt = dumptree(entity)
parser = etree.XMLParser(remove_blank_text=True)
src = StringIO(c_txt)
tree = etree.parse(src, parser)
c_txt = dumptree(tree, pretty_print=True, xml_declaration=False).decode("utf-8")
p = c_elt.getparent()
p.remove(c_elt)
if p.text is not None:
p.text += c_txt
else:
p.text = c_txt
xml = dumptree(t, xml_declaration=False).decode('utf-8')
return render_template("entity.html",
headline=self.md.display(entity).strip(),
subheading=entity.get('entityID'),
entity_id=entity.get('entityID'),
content=xml)
else:
for p in self.plumbings:
state = {'request': True,
'headers': {'Content-Type': 'text/xml'},
'accept': accept,
'url': cherrypy.url(relative=False),
'select': q,
'path': path,
'stats': {}}
r = p.process(self.md, state=state)
if r is not None:
cache_ttl = state.get('cache', 0)
log.debug("caching for %d seconds" % cache_ttl)
for k, v in state.get('headers', {}).iteritems():
cherrypy.response.headers[k] = v
caching.expires(secs=cache_ttl)
return r
raise NotFound()
def request(self, **kwargs):
"""The main request processor. This code implements all rendering of metadata.
"""
stats['MD Requests'] += 1
if not self.ready:
raise HTTPError(503, _("Service Unavailable (repository loading)"))
pfx = kwargs.get('pfx', None)
path = kwargs.get('path', None)
content_type = kwargs.get('content_type', None)
log.debug("MDServer pfx=%s, path=%s, content_type=%s" % (pfx, path, content_type))
def _d(x, do_split=True):
if x is not None:
x = x.strip()
log.debug("_d(%s,%s)" % (x, do_split))
if x is None or len(x) == 0:
return None, None
if x.startswith("{base64}"):
x = x[8:].decode('base64')
if do_split and '.' in x:
(pth, dot, extn) = x.rpartition('.')
assert (dot == '.')
if extn in _ctypes:
return pth, extn
return x, None
_ctypes = {'xml': 'application/xml',
'json': 'application/json',
'htm': 'text/html',
'html': 'text/html',
'ds': 'text/html',
's': 'application/json'}
alias = None
if pfx:
alias = pfx
pfx = self.aliases.get(alias, None)
if pfx is None:
raise NotFound()
path, ext = _d(path, content_type is None)
if pfx and path:
q = "{%s}%s" % (pfx, path)
path = "/%s/%s" % (alias, path)
else:
q = path
if ext is not None:
log.debug("request path: %s.%s, headers: %s" % (path, ext, cherrypy.request.headers))
else:
log.debug("request path: %s, headers: %s" % (path, cherrypy.request.headers))
accept = {}
if content_type is None:
if ext is not None and ext in _ctypes:
accept = {_ctypes[ext]: True}
else:
accept = MDServer.MediaAccept()
if ext is not None:
path = "%s.%s" % (path, ext)
else:
accept = {content_type: True}
with self.lock.readlock:
if ext == 'ds':
pdict = dict()
entity_id = kwargs.get('entityID', None)
if entity_id is None:
raise HTTPError(400, _("400 Bad Request - missing entityID"))
pdict['sp'] = self.md.sha1_id(entity_id)
e = self.md.store.lookup(entity_id)
if e is None or len(e) == 0:
raise HTTPError(404)
if len(e) > 1:
raise HTTPError(400, _("400 Bad Request - multiple matches for") + " %s" % entity_id)
pdict['entity'] = self.md.simple_summary(e[0])
if not path:
pdict['search'] = "/search/"
pdict['list'] = "/role/idp.json"
else:
pdict['search'] = "%s.s" % path
pdict['list'] = "%s.json" % path
cherrypy.response.headers['Content-Type'] = 'text/html'
return render_template("ds.html", **pdict)
elif ext == 's':
paged = bool(kwargs.get('paged', False))
query = kwargs.get('query', None)
page = kwargs.get('page', 0)
page_limit = kwargs.get('page_limit', 10)
entity_filter = kwargs.get('entity_filter', None)
related = kwargs.get('related', None)
cherrypy.response.headers['Content-Type'] = 'application/json'
if query is None:
log.debug("empty query - creating one")
query = [cherrypy.request.remote.ip]
referrer = cherrypy.request.headers.get('referrer', None)
if referrer is not None:
log.debug("including referrer: %s" % referrer)
url = urlparse.urlparse(referrer)
host = url.netloc
if ':' in url.netloc:
(host, port) = url.netloc.split(':')
for host_part in host.rstrip(self.psl.get_public_suffix(host)).split('.'):
if host_part is not None and len(host_part) > 0:
query.append(host_part)
log.debug("created query: %s" % ",".join(query))
if paged:
res, more, total = self.md.search(query,
path=q,
page=int(page),
page_limit=int(page_limit),
entity_filter=entity_filter,
related=related)
# log.debug(dumps({'entities': res, 'more': more, 'total': total}))
return dumps({'entities': res, 'more': more, 'total': total})
else:
return dumps(self.md.search(query,
path=q,
entity_filter=entity_filter,
related=related))
elif accept.get('text/html'):
if not q:
if pfx:
title = pfx
else:
title = _("Metadata By Attributes")
return render_template("index.html",
md=self.md,
alias=alias,
aliases=self.aliases,
title=title)
else:
entities = self.md.lookup(q)
if not entities:
raise NotFound()
if len(entities) > 1:
return render_template("metadata.html",
md=self.md,
subheading=q,
entities=entities)
else:
entity = entities[0]
t = html.fragment_fromstring(unicode(xslt_transform(entity, "entity2html.xsl")))
for c_elt in t.findall(".//code[@role='entity']"):
c_txt = dumptree(entity)
parser = etree.XMLParser(remove_blank_text=True)
src = StringIO(c_txt)
tree = etree.parse(src, parser)
c_txt = dumptree(tree, pretty_print=True, xml_declaration=False).decode("utf-8")
p = c_elt.getparent()
p.remove(c_elt)
if p.text is not None:
p.text += c_txt
else:
p.text = c_txt
xml = dumptree(t, xml_declaration=False).decode('utf-8')
return render_template("entity.html",
headline=self.md.display(entity).strip(),
subheading=entity.get('entityID'),
entity_id=entity.get('entityID'),
content=xml)
else:
for p in self.plumbings:
state = {'request': True,
'headers': {'Content-Type': 'text/xml'},
'accept': accept,
'url': cherrypy.url(relative=False),
'select': q,
'path': path,
'stats': {}}
r = p.process(self.md, state=state)
if r is not None:
cache_ttl = state.get('cache', 0)
log.debug("caching for %d seconds" % cache_ttl)
for k, v in state.get('headers', {}).iteritems():
cherrypy.response.headers[k] = v
caching.expires(secs=cache_ttl)
return r
raise NotFound()
def request(self, **kwargs):
"""The main request processor. This code implements all rendering of metadata.
"""
stats['MD Requests'] += 1
pfx = kwargs.get('pfx', None)
path = kwargs.get('path', None)
content_type = kwargs.get('content_type', None)
def escape(m):
st = m.group(0)
if st == '<':
return '<'
if st == '>':
return '>'
return st
def _d(x):
if x is None or len(x) == 0:
return None, None
if x.startswith("{base64}"):
x = x[8:].decode('base64')
if '.' in x:
(pth, sep, extn) = x.rpartition('.')
return pth, extn
else:
return x, None
_ctypes = {'xml': 'application/xml',
'json': 'application/json',
'htm': 'text/html',
'html': 'text/html',
'ds': 'text/html',
's': 'application/json'}
alias = None
if pfx:
alias = pfx
pfx = self.aliases.get(alias, None)
if pfx is None:
raise NotFound()
path, ext = _d(path)
if pfx and path:
q = "{%s}%s" % (pfx, path)
else:
q = path
log.debug("request path: %s, ext: %s, headers: %s" % (path, ext, cherrypy.request.headers))
accept = {}
if content_type is None:
if ext is not None and ext in _ctypes:
accept = {_ctypes[ext]: True}
else:
accept = MDServer.MediaAccept()
if ext is not None:
path = "%s.%s" % (path, ext)
else:
accept = {content_type: True}
with self.lock.readlock:
if ext == 'ds':
pdict = dict()
entity_id = kwargs.get('entityID', None)
if entity_id is None:
raise HTTPError(400, "400 Bad Request - missing entityID")
pdict['sp'] = self.md.sha1_id(entity_id)
pdict['ret'] = kwargs.get('return', None)
if not path:
pdict['search'] = "/search/"
else:
pdict['search'] = "%s.s" % path
if pdict['ret'] is None:
raise HTTPError(400, "400 Bad Request - Missing 'return' parameter")
pdict['returnIDParam'] = kwargs.get('returnIDParam', 'entityID')
cherrypy.response.headers['Content-Type'] = 'text/html'
return render_template("ds.html", **pdict)
elif ext == 's':
paged = bool(kwargs.get('paged', False))
query = kwargs.get('query', None)
page = kwargs.get('page', 0)
page_limit = kwargs.get('page_limit', 10)
entity_filter = kwargs.get('entity_filter', None)
cherrypy.response.headers['Content-Type'] = 'application/json'
if query is None:
log.debug("empty query - creating one")
query = [cherrypy.request.remote.ipgit]
referrer = cherrypy.request.headers.get('referrer', None)
if referrer is not None:
log.debug("including referrer: %s" % referrer)
url = urlparse.urlparse(referrer)
host = url.netloc
if ':' in url.netloc:
(host, port) = url.netloc.split(':')
for host_part in host.rstrip(self.psl.get_public_suffix(host)).split('.'):
if host_part is not None and len(host_part) > 0:
query.append(host_part)
log.debug("created query: %s" % ",".join(query))
if paged:
res, more, total = self.md.search(query,
path=q,
page=int(page),
page_limit=int(page_limit),
entity_filter=entity_filter)
#log.debug(dumps({'entities': res, 'more': more, 'total': total}))
return dumps({'entities': res, 'more': more, 'total': total})
else:
return dumps(self.md.search(query,
path=q,
entity_filter=entity_filter))
elif accept.get('text/html'):
if not q:
if pfx:
title = pfx
else:
title = _("Metadata By Attributes")
return render_template("index.html",
md=self.md,
alias=alias,
aliases=self.aliases,
title=title)
else:
entities = self.md.lookup(q)
if not entities:
raise NotFound()
if len(entities) > 1:
return render_template("metadata.html",
md=self.md,
entities=entities)
else:
entity = entities[0]
t = html.fragment_fromstring(unicode(xslt_transform(entity, "entity2html.xsl")))
for c_elt in t.findall(".//code[@role='entity']"):
c_txt = dumptree(entity, pretty_print=True, xml_declaration=False).decode("utf-8")
p = c_elt.getparent()
p.remove(c_elt)
if p.text is not None:
p.text += c_txt # re.sub(".",escape,c_txt)
else:
p.text = c_txt # re.sub(".",escape,c_txt)
xml = dumptree(t, xml_declaration=False).decode('utf-8')
return render_template("basic.html", content=xml)
else:
for p in self.plumbings:
state = {'request': True,
'headers': {'Content-Type': 'text/xml'},
'accept': accept,
'url': cherrypy.url(relative=False),
'select': q,
'path': path,
'stats': {}}
r = p.process(self.md, state=state)
if r is not None:
cache_ttl = state.get('cache', 0)
log.debug("caching for %d seconds" % cache_ttl)
for k, v in state.get('headers', {}).iteritems():
cherrypy.response.headers[k] = v
caching.expires(secs=cache_ttl)
return r
raise NotFound()