Python domhelpers.gatherTextNodes函数代码示例

    def test_proxySearch(self):
        """
        When the user searches for a proxy, the results are displayed in a
        table, in a form that will allow them to submit it to add new read or
        write proxies.
        """
        self.expectSomeRecords()
        self.resource.getResourceById = partial(FakePrincipalResource, self)
        document = yield self.renderPage(dict(resourceId=["qux"],
                                              proxySearch=["bob"]))

        # Form is filled out with existing input.
        self.assertEquals(
            document.getElementById("txt_proxySearch").getAttribute("value"),
            "bob"
        )
        proxyAddForm = document.getElementById("frm_proxyAdd")
        [proxyResultsTable] = getElementsByTagName(proxyAddForm, "table")

        rows = getElementsByTagName(proxyResultsTable, 'tr')
        self.assertEquals(len(rows), 3)
        firstRowCells = getElementsByTagName(rows[1], 'td')
        self.assertEquals(
            [gatherTextNodes(cell) for cell in firstRowCells[1:]],
            ["User", "bob", "[email protected], [email protected]", ""]
        )
        self.assertNotIn(
            "No matches found for proxy resource bob",
            gatherTextNodes(document)
        )

    def test_selectResourceById(self):
        """
        When a resource is selected by a 'resourceId' parameter,
        """
        self.resource.getResourceById = partial(FakePrincipalResource, self)
        document = yield self.renderPage(dict(resourceId=["qux"]))
        [detailsTitle] = getElementsByTagName(document, 'h3')
        detailString = gatherTextNodes(detailsTitle)
        self.assertEquals(detailString,
                          "Resource Details: Hello Fake Resource: 'qux'")
        hiddenResourceId = document.getElementById(
            "hdn_resourceId").getAttribute("value")
        self.assertEquals(hiddenResourceId, "qux")

        autoScheduleMenu = document.getElementById("sel_autoSchedule")

        # Now, some assertions about features that are covered in other tests
        # which should be turned _off_ here since we're not asking for them.

        # Not an auto-schedule resource; there should be no auto-schedule menu.
        self.assertIdentical(autoScheduleMenu, None)
        # No resource search present; we shouldn't be performing the search.
        self.assertNotIn("No matches found for resource",
                         gatherTextNodes(document))
        self.assertIdentical(document.getElementById("tab_searchResults"), None)
        # I'm not attempting to retrieve a property, there's nothing to fail to
        # parse.
        self.assertNotIn("Unable to parse property", gatherTextNodes(document))
        # I'm not searching for proxies, so there shouldn't be any results.
        self.assertNotIn("No matches found for proxy resource",
                         gatherTextNodes(document))

    def test_resourceSearch(self):
        """
        Searching for resources should result in an HTML table resource search.
        """
        self.expectSomeRecords()
        document = yield self.renderPage(dict(resourceSearch=["bob"]))

        # Form is filled out with existing input.
        self.assertEquals(
            document.getElementById("txt_resourceSearch").getAttribute("value"),
            "bob"
        )
        tables = getElementsByTagName(document, "table")
        # search results are the first table
        rows = getElementsByTagName(tables[0], 'tr')
        self.assertEquals(len(rows), 3)
        firstRowCells = getElementsByTagName(rows[1], 'td')
        self.assertEquals(
            [gatherTextNodes(cell) for cell in firstRowCells[1:]],
            ["Bob Bobson", "User", "bob", "boblogin",
             "[email protected], [email protected]"]
        )
        [resourceLink] = getElementsByTagName(
            firstRowCells[0], 'a')
        self.assertEquals(
            resourceLink.getAttribute("href"),
            "/admin/?resourceId=users:bob"
        )
        self.assertEquals(gatherTextNodes(resourceLink), "select")
        self.assertNotIn(
            "No matches found for resource bob",
            gatherTextNodes(document)
        )

 def test_textEntitiesDecoded(self):
     """
     Minidom does decode entities in text nodes.
     """
     doc5_xml = '<x>Souffl&amp;</x>'
     doc5 = self.dom.parseString(doc5_xml)
     actual = domhelpers.gatherTextNodes(doc5)
     expected = 'Souffl&'
     self.assertEqual(actual, expected)
     actual = domhelpers.gatherTextNodes(doc5.documentElement)
     self.assertEqual(actual, expected)

    def check_80_columns(self, dom, filename):
        for node in domhelpers.findNodesNamed(dom, 'pre'):
            # the ps/pdf output is in a font that cuts off at 80 characters,
            # so this is enforced to make sure the interesting parts (which
            # are likely to be on the right-hand edge) stay on the printed
            # page.
            for line in domhelpers.gatherTextNodes(node, 1).split('\n'):
                if len(line.rstrip()) > 80:
                    self._reportError(filename, node, 
                                      'text wider than 80 columns in pre')
        for node in domhelpers.findNodesNamed(dom, 'a'):
            if node.getAttribute('class', '').endswith('listing'):
                try:
                    fn = os.path.dirname(filename) 
                    fn = os.path.join(fn, node.getAttribute('href'))
                    lines = open(fn,'r').readlines()
                except:
                    self._reportError(filename, node,
                                      'bad listing href: %r' %
                                      node.getAttribute('href'))
                    continue

                for line in lines:
                    if len(line.rstrip()) > 80:
                        self._reportError(filename, node, 
                                          'listing wider than 80 columns')

 def testComplexNotification(self):
     listNode = self.d.getElementById("theList")
     self.assert_(listNode, "Test %s failed" % outputNum)
     liNodes = domhelpers.getElementsByTagName(listNode, "li")
     self.assert_(liNodes, "DOM was not updated by notifying Widgets. Test %s" % outputNum)
     text = domhelpers.gatherTextNodes(liNodes[0])
     self.assert_(text.strip() == "test", "Wrong output: %s. Test %s" % (text, outputNum))

    def test_proxiesListing(self):
        """
        Resource principals will have their proxies listed in a table.
        """
        def fakeResourceById(request, resid):
            return FakePrincipalResource(self, request, resid,
                                         recordType="resources")

        self.resource.getResourceById = fakeResourceById
        document = yield self.renderPage(dict(resourceId=["qux"]))
        proxiesForm = document.getElementById("frm_proxies")
        [proxiesTable] = getElementsByTagName(proxiesForm, "table")
        rows = getElementsByTagName(proxiesTable, "tr")

        # header + 3 data rows (see FakePrincipalResource)
        self.assertEquals(len(rows), 4)
        firstRowCells = getElementsByTagName(rows[1], "td")
        # name, buttons, name, buttons
        self.assertEquals(len(firstRowCells), 4)
        lastRowCells = getElementsByTagName(rows[-1], "td")
        # name, buttons, blank space
        self.assertEquals(len(lastRowCells), 3)
        self.assertEquals(lastRowCells[-1].getAttribute("colspan"), "2")

        self.assertNotIn("This resource has no proxies.",
                         ''.join(gatherTextNodes(document)))

    def test_gatherTextNodes(self):
        doc1=microdom.parseString('<a>foo</a>')
        actual=domhelpers.gatherTextNodes(doc1)
        expected='foo'
        self.assertEqual(actual, expected)
        actual=domhelpers.gatherTextNodes(doc1.documentElement)
        self.assertEqual(actual, expected)

        doc2_xml='<a>a<b>b</b><c>c</c>def<g>g<h>h</h></g></a>'
        doc2=microdom.parseString(doc2_xml)
        actual=domhelpers.gatherTextNodes(doc2)
        expected='abcdefgh'
        self.assertEqual(actual, expected)
        actual=domhelpers.gatherTextNodes(doc2.documentElement)
        self.assertEqual(actual, expected)

        doc3_xml=('<a>a<b>b<d>d<g>g</g><h>h</h></d><e>e<i>i</i></e></b>' +
                  '<c>c<f>f<j>j</j></f></c></a>')
        doc3=microdom.parseString(doc3_xml)
        actual=domhelpers.gatherTextNodes(doc3)
        expected='abdgheicfj'
        self.assertEqual(actual, expected)
        actual=domhelpers.gatherTextNodes(doc3.documentElement)
        self.assertEqual(actual, expected)

        doc4_xml='''<html>
  <head>
  </head>
  <body>
    stuff
  </body>
</html>
'''
        doc4=microdom.parseString(doc4_xml)
        actual=domhelpers.gatherTextNodes(doc4)
        expected='\n    stuff\n  '
        assert actual==expected, 'expected %s, got %s' % (expected, actual)
        actual=domhelpers.gatherTextNodes(doc4.documentElement)
        self.assertEqual(actual, expected)
        
        doc5_xml='<x>Souffl&eacute;</x>'
        doc5=microdom.parseString(doc5_xml)
        actual=domhelpers.gatherTextNodes(doc5)
        expected='Souffl&eacute;'
        self.assertEqual(actual, expected)
        actual=domhelpers.gatherTextNodes(doc5.documentElement)
        self.assertEqual(actual, expected)

 def testUnEntities(self):
     s = """
             <HTML>
                 This HTML goes between Stupid <=CrAzY!=> Dumb.
             </HTML>
         """
     d = microdom.parseString(s, beExtremelyLenient=1)
     n = domhelpers.gatherTextNodes(d)
     self.assertNotEquals(n.find('>'), -1)

def textFromHtml(htmlText):
    """
    Convert html text into its text nodes, with extreme leniency.  If the input
    is unicode, keep it unicode.
    """
    d = microdom.parseString(htmlText, beExtremelyLenient=1)
    s = domhelpers.gatherTextNodes(d, joinWith=u" ")
    ## print '\n'.join('| ' + l for l in s.splitlines())
    return s

 def test_scrubTrustsH1(self):
     """
     Test that L{xquotient.scrubber.Scrubber} considers h1 to be a safe tag.
     Added because of #1895.
     """
     node = parseString("<h1>Foo</h1>").documentElement
     scrubbed = scrub(node)
     h1s = getElementsByTagName(scrubbed, 'h1')
     self.assertEquals(len(h1s), 1)
     self.assertEquals(gatherTextNodes(h1s[0]).strip(), "Foo")

    def test_gatherTextNodesDropsWhitespace(self):
        """
        Microdom discards whitespace-only text nodes, so L{gatherTextNodes}
        returns only the text from nodes which had non-whitespace characters.
        """
        doc4_xml = '''<html>
  <head>
  </head>
  <body>
    stuff
  </body>
</html>
'''
        doc4 = self.dom.parseString(doc4_xml)
        actual = domhelpers.gatherTextNodes(doc4)
        expected = '\n    stuff\n  '
        self.assertEqual(actual, expected)
        actual = domhelpers.gatherTextNodes(doc4.documentElement)
        self.assertEqual(actual, expected)

 def test_davProperty(self):
     """
     When a resource is selected by a resourceId parameter, and a DAV
     property is selected by the 'davPropertyName' parameter, that property
     will displayed.
     """
     self.resource.getResourceById = partial(FakePrincipalResource, self)
     document = yield self.renderPage(
         dict(resourceId=["qux"],
              davPropertyName=["DAV:#displayname"])
     )
     propertyName = document.getElementById('txt_davPropertyName')
     self.assertEquals(propertyName.getAttribute("value"),
                       "DAV:#displayname")
     propertyValue = DisplayName("The Name To Display").toxml()
     self.assertIn(cgi.escape(propertyValue),
                   gatherTextNodes(document))
     self.assertNotIn("Unable to parse property to read:",
                      gatherTextNodes(document))

    def scrub(self, node, filterCIDLinks=True):
        """
        Remove all potentially harmful elements from the node and
        return a wrapper node.

        For reasons (perhaps dubious) of performance, this mutates its
        input.
        """
        if node.nodeName == 'html':
            filler = body = lmx().div(_class="message-html")
            for c in node.childNodes:
                if c.nodeName == 'head':
                    for hc in c.childNodes:
                        if hc.nodeName == 'title':
                            body.div(_class="message-title").text(domhelpers.gatherTextNodes(hc))
                            break
                elif c.nodeName == 'body':
                    filler = body.div(_class='message-body')
                    break
        else:
            filler = body = lmx().div(_class="message-nohtml")
        for e in self.iternode(node):
            if getattr(e, 'clean', False):
                # If I have manually exploded this node, just forget about it.
                continue
            ennl = e.nodeName.lower()

            if filterCIDLinks and self._filterCIDLink(e):
                # we could replace these with a marker element, like we do
                # with dangerous tags, but i'm not sure there is a reason to
                e.parentNode.removeChild(e)

            if ennl in self._goodHtml:
                handler = getattr(self, '_handle_' + ennl, None)
                if handler is not None:
                    e = handler(e)
                newAttributes = {}
                oldAttributes = e.attributes
                e.attributes = newAttributes
                goodAttributes = self._goodHtml[ennl] + self._alwaysSafeAttributes
                for attr in goodAttributes:
                    if attr in oldAttributes:
                        newAttributes[attr] = oldAttributes[attr]
            else:
                e.attributes.clear()
                e.setTagName("div")
                e.setAttribute("class", "message-html-unknown")
                e.setAttribute("style", "display: none")
                div = Element('div')
                div.setAttribute('class', 'message-html-unknown-tag')
                div.appendChild(Text("Untrusted %s tag" % (ennl, )))
                e.childNodes.insert(0, div)
        filler.node.appendChild(node)
        return body.node

 def test_noProxiesListing(self):
     """
     When the selected resource principal has no proxies, the page should
     display a message saying so.
     """
     self.resource.getResourceById = partial(FakePrincipalResource, self,
                                             recordType='resources',
                                             hasProxies=False)
     document = yield self.renderPage(dict(resourceId=['qux']))
     self.assertIn("This resource has no proxies.",
                   ''.join(gatherTextNodes(document)))