def search(self, environ, start_response):
"""WSGI method, called by the wsgi app for requests that matches
``searchendpoint``."""
queryparams = self._search_parse_query(environ['QUERY_STRING'])
res, pager = self._search_run_query(queryparams)
if pager['totalresults'] == 1:
title = "1 match"
else:
title = "%s matches" % pager['totalresults']
title += " for '%s'" % queryparams.get("q")
body = html.Body()
for r in res:
if not 'dcterms_title' in r or r['dcterms_title'] is None:
r['dcterms_title'] = r['uri']
if r.get('dcterms_identifier', False):
r['dcterms_title'] = r['dcterms_identifier'] + ": " + r[
'dcterms_title']
body.append(
html.Div([
html.H2([elements.Link(r['dcterms_title'], uri=r['uri'])]),
r.get('text', '')
], **{'class': 'hit'}))
pagerelem = self._search_render_pager(pager, queryparams,
environ['PATH_INFO'])
body.append(
html.Div([
html.P([
"Results %(firstresult)s-%(lastresult)s "
"of %(totalresults)s" % pager
]), pagerelem
], **{'class': 'pager'}))
data = self._transform(title, body, environ, template="xsl/search.xsl")
return self._return_response(data, start_response)
def search(self, environ, start_response):
"""WSGI method, called by the wsgi app for requests that matches
``searchendpoint``."""
queryparams = self._search_parse_query(environ['QUERY_STRING'])
# massage queryparams['issued'] if present, then restore it
y = None
if 'issued' in queryparams:
y = int(queryparams['issued'])
queryparams['issued'] = Between(datetime(y, 1, 1),
datetime(y, 12, 31, 23, 59, 59))
boost_types = [("sfs", 10)]
res, pager = self._search_run_query(queryparams,
boost_types=boost_types)
if y:
queryparams['issued'] = str(y)
if pager['totalresults'] == 1:
title = "1 träff"
else:
title = "%s träffar" % pager['totalresults']
title += " för '%s'" % queryparams.get("q")
body = html.Body()
if hasattr(res, 'aggregations'):
body.append(
self._search_render_facets(res.aggregations, queryparams,
environ))
for r in res:
if 'label' not in r:
label = r['uri']
elif isinstance(r['label'], list):
label = str(r['label']) # flattens any nested element
# structure, eg
# <p><strong><em>foo</em></strong></p>
# -> foo
else:
label = r['label']
rendered_hit = html.Div([
html.B([elements.Link(label, uri=r['uri'])], **
{'class': 'lead'})
], **{'class': 'hit'})
if r.get('text'):
rendered_hit.append(html.P([r.get('text', '')]))
if 'innerhits' in r:
for innerhit in r['innerhits']:
rendered_hit.append(self._search_render_innerhit(innerhit))
body.append(rendered_hit)
pagerelem = self._search_render_pager(pager, queryparams,
environ['PATH_INFO'])
body.append(
html.Div([
html.P([
"Träff %(firstresult)s-%(lastresult)s "
"av %(totalresults)s" % pager
]), pagerelem
], **{'class': 'pager'}))
data = self._transform(title, body, environ, template="xsl/search.xsl")
return self._return_response(data, start_response)
def exception(self, environ, start_response):
import traceback
from pprint import pformat
exc_type, exc_value, tb = sys.exc_info()
tblines = traceback.format_exception(exc_type, exc_value, tb)
tbstr = "\n".join(tblines)
# render the error
title = tblines[-1]
body = html.Body([
html.Div([
html.H1(self.exception_heading),
html.P([self.exception_description]),
html.H2("Traceback"),
html.Pre([tbstr]),
html.H2("Variables"),
html.Pre([
"request_uri: %s\nos.getcwd(): %s" %
(request_uri(environ), os.getcwd())
]),
html.H2("environ"),
html.Pre([pformat(environ)]),
html.H2("sys.path"),
html.Pre([pformat(sys.path)]),
html.H2("os.environ"),
html.Pre([pformat(dict(os.environ))])
])
])
msg = self._transform(title, body, environ)
return self._return_response(msg,
start_response,
status="500 Internal Server Error",
contenttype="text/html")
def _search_render_facets(self, facets, queryparams, environ):
facetgroups = []
commondata = self.repos[0].commondata
searchurl = request_uri(environ, include_query=False)
for facetresult in ('type', 'creator', 'issued'):
if facetresult in facets:
if facetresult in queryparams:
# the user has selected a value for this
# particular facet, we should not display all
# buckets (but offer a link to reset the value)
qpcopy = dict(queryparams)
del qpcopy[facetresult]
href = "%s?%s" % (searchurl, urlencode(qpcopy))
val = queryparams[facetresult]
if facetresult == "creator":
val = self.repos[0].lookup_label(val)
elif facetresult == "type":
val = self.repolabels.get(val, val)
lbl = "%s: %s" % (self.facetlabels.get(
facetresult, facetresult), val)
facetgroups.append(
html.LI([
lbl,
html.A(
"\xa0", **{
'href': href,
'class': 'glyphicon glyphicon-remove'
})
]))
else:
facetgroup = []
for bucket in facets[facetresult]['buckets']:
if facetresult == 'type':
lbl = self.repolabels.get(bucket['key'],
bucket['key'])
key = bucket['key']
elif facetresult == 'creator':
k = URIRef(bucket['key'])
pred = SKOS.altLabel if commondata.value(
k, SKOS.altLabel) else FOAF.name
lbl = commondata.value(k, pred)
key = bucket['key']
elif facetresult == "issued":
lbl = bucket["key_as_string"]
key = lbl
qpcopy = dict(queryparams)
qpcopy[facetresult] = key
href = "%s?%s" % (searchurl, urlencode(qpcopy))
facetgroup.append(
html.LI([
html.A("%s" % (lbl), **{'href': href}),
html.Span([str(bucket['doc_count'])],
**{'class': 'badge pull-right'})
]))
lbl = self.facetlabels.get(facetresult, facetresult)
facetgroups.append(
html.LI([html.P([lbl]),
html.UL(facetgroup)]))
return html.Div(facetgroups, **{'class': 'facets'})
def test_elements_from_soup(self):
from ferenda.elements import html
soup = BeautifulSoup(
"""<body>
<h1>Sample</h1>
<div class="main">
<img src="xyz.png"/>
<p>Some <b>text</b></p>
<dl>
<dt>Term 1</dt>
<dd>Definition 1</dd>
</dl>
</div>
<div id="foot">
<hr/>
<a href="/">home</a> - <a href="/about">about</a>
</div>
</body>""", "lxml")
body = html.elements_from_soup(soup.body)
# print("Body: \n%s" % serialize(body))
result = html.Body([
html.H1(["Sample"]),
html.Div([
html.Img(src="xyz.png"),
html.P(["Some ", html.B(["text"])]),
html.DL([html.DT(["Term 1"]),
html.DD(["Definition 1"])])
], **{"class": "main"}),
html.Div([
html.HR(),
html.A(["home"], href="/"), " - ",
html.A(["about"], href="/about")
],
id="foot")
])
self.maxDiff = 4096
self.assertEqual(serialize(body), serialize(result))
def handle_search(self, request, **values):
# return Response("<h1>Hello search: " + request.args.get("q") +" </h1>", mimetype="text/html")
res, pager = self._search_run_query(request.args)
if pager['totalresults'] == 1:
title = "1 match"
else:
title = "%s matches" % pager['totalresults']
title += " for '%s'" % request.args.get("q")
body = html.Body()
for r in res:
if not 'dcterms_title' in r or r['dcterms_title'] is None:
r['dcterms_title'] = r['uri']
if r.get('dcterms_identifier', False):
r['dcterms_title'] = r['dcterms_identifier'] + ": " + r[
'dcterms_title']
body.append(
html.Div([
html.H2([elements.Link(r['dcterms_title'], uri=r['uri'])]),
r.get('text', '')
], **{'class': 'hit'}))
pagerelem = self._search_render_pager(pager, dict(request.args),
request.path)
body.append(
html.Div([
html.P([
"Results %(firstresult)s-%(lastresult)s "
"of %(totalresults)s" % pager
]), pagerelem
], **{'class': 'pager'}))
data = self._transform(title,
body,
request.environ,
template="xsl/search.xsl")
return Response(data, mimetype="text/html")
def test_meta(self):
# test 3: use a mix of our own elements and html elements,
# with meta + uri attached to some nodes
g1 = Graph().parse(format='n3',
data="""
@prefix bibo: <http://purl.org/ontology/bibo/> .
@prefix dcterms: <http://purl.org/dc/terms/> .
<http://localhost:8000/res/base/basefile#S1> a bibo:DocumentPart;
dcterms:title "First section";
bibo:chapter "1" .
""")
g2 = Graph().parse(format='n3',
data="""
@prefix bibo: <http://purl.org/ontology/bibo/> .
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix owl: <http://www.w3.org/2002/07/owl#> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
<http://localhost:8000/res/base/basefile#S2> a bibo:DocumentPart;
dcterms:title "Second section";
bibo:chapter "2";
dcterms:creator "Fred Bloggs"@en-GB;
dcterms:issued "2013-05-10"^^xsd:date;
owl:sameAs <http://example.org/s2> .
<http://example.org/s2> dcterms:title "Same same but different" .
<http://localhost:8000/res/base/unlrelated> dcterms:title "Unrelated document" .
""")
body = el.Body([
el.Heading(['Toplevel heading'], level=1),
html.P(['Introductory preamble']),
html.Div([
html.P(['Some text']),
el.Subsection([el.Paragraph(['More text'])],
ordinal='1.1',
title="First subsection")
],
uri='http://localhost:8000/res/base/basefile#S1',
meta=g1),
el.Section([el.Paragraph(['Even more text'])],
uri='http://localhost:8000/res/base/basefile#S2',
meta=g2)
])
want = """
<body xmlns="http://www.w3.org/1999/xhtml"
about="http://localhost:8000/res/base/basefile">
<h1>Toplevel heading</h1>
<p>Introductory preamble</p>
<div about="http://localhost:8000/res/base/basefile#S1"
content="First section"
property="dcterms:title"
typeof="bibo:DocumentPart">
<span href="http://localhost:8000/res/base/basefile"
rel="dcterms:isPartOf"/>
<span content="1"
property="bibo:chapter"
xml:lang=""/>
<p>Some text</p>
<div about="http://localhost:8000/res/base/basefile#S1.1"
content="First subsection"
property="dcterms:title"
typeof="bibo:DocumentPart"
class="subsection">
<span href="http://localhost:8000/res/base/basefile#S1"
rel="dcterms:isPartOf"/>
<span about="http://localhost:8000/res/base/basefile#S1.1"
content="1.1"
property="bibo:chapter"/>
<p>More text</p>
</div>
</div>
<div about="http://localhost:8000/res/base/basefile#S2"
class="section"
content="Second section"
property="dcterms:title"
typeof="bibo:DocumentPart">
<span href="http://localhost:8000/res/base/basefile"
rel="dcterms:isPartOf"/>
<span href="http://example.org/s2"
rel="owl:sameAs">
<span content="Same same but different"
property="dcterms:title"
xml:lang=""/>
</span>
<span content="2"
property="bibo:chapter"
xml:lang=""/>
<span content="2013-05-10"
property="dcterms:issued"
datatype="xsd:date"/>
<span content="Fred Bloggs"
property="dcterms:creator"
xml:lang="en-GB"/>
<p>Even more text</p>
</div>
</body>"""
self._test_asxhtml(want, body)