def test_with_code_which_result_is_sequence(t):
results = sxq.execute_all(
"data(/user/name)",
"<user><name>Taro</name><name>Jiro</name></user>")
expects = ["Taro", "Jiro"]
for index, expected in enumerate(expects):
t.assertEqual(results[index], expected)
pass
pass
def test_with_code_which_result_is_sequence(t):
results = sxq.execute_all(
"data(/user/name)",
"<user><name>Taro</name><name>Jiro</name></user>")
expects = ["Taro", "Jiro"]
for index, expected in enumerate(expects):
t.assertEqual(results[index], expected)
pass
pass
def find_news(self, method='xquery', search_text='', not_in=False):
import sys
reload(sys)
sys.setdefaultencoding('utf8')
news = []
if method == '':
method = 'xquery'
for source in self.rss_urls.keys():
r = requests.get(self.rss_urls[source])
rss = r.content
if method == 'xquery':
query_s = """for $i in //item
where contains(lower-case($i/title), lower-case('""" + search_text + """'))
return <new>{$i/title, $i/pubDate, $i/link }</new>
"""
if not_in:
query_s = """for $i in //item
where not(contains(lower-case($i/title), lower-case('""" + search_text + """')))
return <new>{$i/title, $i/pubDate, $i/link }</new>
"""
news_list = sxq.execute_all(query_s, rss)
for n_i in news_list:
i = etree.fromstring(n_i)
title = i.xpath("./title")[0]
pubDate = i.xpath("./pubDate")[0]
link = i.xpath("./link")[0]
n = New(title=title.text,
link=link.text,
pubdate=pubDate.text)
news.append(n)
elif method == 'regexp':
h = HTMLParser.HTMLParser()
rss = h.unescape(rss)
pattern = ur'<item>(.*?)</item>'
regex = re.compile(pattern,
re.DOTALL + re.UNICODE + re.IGNORECASE)
for match in regex.finditer(rss):
item = match.group(1)
pattern = ur'<title>((.*?)' + search_text + '(.*?))</title>'
if not_in:
pattern = ur'<title>((.(?<!' + search_text + '))*?)</title>'
regex = re.compile(pattern,
re.DOTALL + re.UNICODE + re.IGNORECASE)
matching = regex.search(item)
if matching:
title = matching.group(1)
n = New(title=title)
news.append(n)
return news
def filtro_xquery(request):
if request.method != 'GET' or 'q'not in request.GET:
return HttpResponseBadRequest()
keyword = request.GET['q'].encode('ascii', 'xmlcharrefreplace').lower()
query = build_query(keyword)
items = []
for xml in get_feeds_xml():
results = simplexquery.execute_all(query, resolver=Resolver(xml))
if results:
items.extend(results)
return HttpResponse(items, mimetype='text/html')
def find_news(self, method='xquery', search_text='', not_in= False):
import sys
reload(sys)
sys.setdefaultencoding('utf8')
news=[]
if method=='':
method = 'xquery'
for source in self.rss_urls.keys():
r = requests.get(self.rss_urls[source])
rss = r.content
if method == 'xquery':
query_s="""for $i in //item
where contains(lower-case($i/title), lower-case('"""+ search_text+ """'))
return <new>{$i/title, $i/pubDate, $i/link }</new>
"""
if not_in:
query_s="""for $i in //item
where not(contains(lower-case($i/title), lower-case('"""+ search_text+ """')))
return <new>{$i/title, $i/pubDate, $i/link }</new>
"""
news_list= sxq.execute_all(query_s, rss);
for n_i in news_list:
i = etree.fromstring(n_i)
title = i.xpath("./title")[0]
pubDate = i.xpath("./pubDate")[0]
link = i.xpath("./link")[0]
n= New(title= title.text, link= link.text, pubdate=pubDate.text)
news.append(n)
elif method == 'regexp':
h = HTMLParser.HTMLParser()
rss= h.unescape(rss)
pattern = ur'<item>(.*?)</item>'
regex = re.compile(pattern, re.DOTALL + re.UNICODE + re.IGNORECASE)
for match in regex.finditer(rss):
item= match.group(1)
pattern = ur'<title>((.*?)'+ search_text+'(.*?))</title>'
if not_in:
pattern = ur'<title>((.(?<!' + search_text+ '))*?)</title>'
regex = re.compile(pattern, re.DOTALL + re.UNICODE + re.IGNORECASE)
matching= regex.search(item)
if matching:
title= matching.group(1)
n= New(title= title)
news.append(n)
return news
def test_with_resolver_returns_non_unicode_type(t):
t.assertRaises(
TypeError, lambda: sxq.execute_all('doc("foo.xml")/name',
resolver=lambda uri: 0))
pass
def test_with_resolver_cannot_accept_just_one_arg(t):
t.assertRaises(
TypeError, lambda: sxq.execute_all('doc("foo.xml")/name',
resolver=lambda: "abc"))
pass
def test_with_non_callable_resolver_type(t):
t.assertRaises(TypeError, lambda: sxq.execute_all("/user", "<_/>", 0))
pass
def test_with_non_unicode_context_type(t):
t.assertRaises(TypeError, lambda: sxq.execute_all("/user", 0))
pass
def test_with_non_unicode_xquery_type(t):
t.assertRaises(TypeError, lambda: sxq.execute_all(0))
pass
def test_with_non_xml_resolver_result(t):
t.assertRaises(
ValueError,
lambda: sxq.execute_all('doc("foo.xml")/name',
resolver=lambda uri: "non xml"))
pass
def test_with_non_unicode_xquery_type(t):
t.assertRaises(
TypeError,
lambda : sxq.execute_all(0))
pass
def test_with_resolver_cannot_accept_just_one_arg(t):
t.assertRaises(
TypeError,
lambda : sxq.execute_all('doc("foo.xml")/name',
resolver=lambda : "abc"))
pass
###xquery='xquery version "1.0"; for $noticia in doc("output1.xml")/feed/entry return data($noticia/title)'
# Ejecución de consulta xquery.
###x=sxq.execute_all(xquery)
# TEST. Verificación que python entiende correctamente el uso de caracteres acentuados desde el código fuente.
# print([y.index('ó') for y in x])
# TEST. Verificación del tipo de dato devuelto en la consulta xquery
# print(type(x))
# print(x[0])
# Visualización de los contenidos de la lista de resultados devueltos por la consulta xquery
# Instrucción anterior que no interpretaba correctamente caracteres Unicode
# print([y.encode('utf-8') for y in x])
###for y in x : print(y)
# MUESTRA TODOS LOS TÍTULOS
xquery = 'xquery version "1.0"; for $noticia in doc("output1.xml")/feed/entry return data($noticia/title)'
x = sxq.execute_all(xquery)
print('BÚSQUEDA DE TODOS LOS TÍTULOS DE NOTICIAS\n')
for y in x:
print(y)
print('\n\nBÚSQUEDA POR ELEMENTO Y CONTENIDO\n')
tipoBusqueda = raw_input(
'Ingrese el Tipo de búsqueda: [T]-Título [D]-Descripción [C]-Categorías:\n'
)
palabra = raw_input('Escriba palabra a buscar:\n')
if tipoBusqueda == 'T':
textoBusqueda = 'TITULO'
elif tipoBusqueda == 'D':
textoBusqueda = 'DESCRIPCIÓN'
elif tipoBusqueda == 'C':
textoBusqueda = 'CATEGORÍA'
def test_with_non_callable_resolver_type(t):
t.assertRaises(
TypeError,
lambda : sxq.execute_all("/user", "<_/>", 0))
pass
def test_with_non_xml_context(t):
t.assertRaises(
ValueError,
lambda : sxq.execute_all("/user", "non xml"))
pass
def test_with_non_unicode_context_type(t):
t.assertRaises(
TypeError,
lambda : sxq.execute_all("/user", 0))
pass
def test_with_syntax_error_xquery(t):
t.assertRaises(
ValueError,
lambda : sxq.execute_all("/user'"))
pass
def test_with_non_xml_context(t):
t.assertRaises(ValueError, lambda: sxq.execute_all("/user", "non xml"))
pass
def test_with_resolver_returns_non_unicode_type(t):
t.assertRaises(
TypeError,
lambda : sxq.execute_all('doc("foo.xml")/name',
resolver=lambda uri: 0))
pass
def test_with_non_xml_resolver_result(t):
t.assertRaises(
ValueError,
lambda : sxq.execute_all('doc("foo.xml")/name',
resolver=lambda uri : "non xml"))
pass
class Resolver(object):
def __call__(self, uri):
print(uri)
return "<name>Jiro</name>"
pass
print(sxq.execute('doc("foo.xml")/name', resolver=Resolver()))
prompt("['execute' returns None if fails by anyway]")
print(sxq.execute("<user>"))
print(sxq.execute("<user>{string(/name)}</user>", "<name>Taro<name>"))
prompt("['execute_all' returns a list of multiple results]")
print(repr(sxq.execute_all("/user/name",
"<user><name>Taro</name><name>Jiro</name></user>")))
prompt("['execute_all' raise ValueError if xquery execution is failed]")
try: sxq.execute_all("/user'")
except ValueError: traceback.print_exc()
try: sxq.execute_all('doc("foo.xml")/name', resolver=lambda uri : "non xml")
except ValueError: traceback.print_exc()
prompt("['execute_all' raise TypeError if arg has invalid type]")
try: sxq.execute_all(0)
except TypeError: traceback.print_exc()
try: sxq.execute_all("", "", 0)
def test_with_syntax_error_xquery(t):
t.assertRaises(ValueError, lambda: sxq.execute_all("/user'"))
pass
