def pubmed_adapter(search=None, id=None):
'''
Sample queries:
#curl "http://localhost:8880/pubmed?"
curl "http://localhost:8880/pubmed?search=stem+cells"
curl "http://localhost:8880/pubmed?id=19358275"
'''
#FIXME: How do we handle no search or id param? Just serve up the latest entries? Or error as below?
#assert_(not(search and id), msg="You must specify the 'search' or 'id' query parameter is mandatory.")
if search:
#search = first_item(search)
#reldate: only search for last N days
#query = urllib.urlencode({'db' : NCBI_DB, 'term': query, 'reldate': '60', 'datetype': 'edat', 'retmax': DEFAULT_MAX_RESULTS, 'usehistory': 'y'})
query = urllib.urlencode({'term': search, 'db' : NCBI_DB, 'datetype': 'edat', 'retmax': DEFAULT_MAX_RESULTS, 'usehistory': 'y'})
search_url = NCBI_SEARCH_PATTERN + query
logger.debug("Term search URL: " + search_url)
doc = bindery.parse(search_url, standalone=True)
search_terms = search
ids = ( unicode(i) for i in doc.eSearchResult.IdList.Id )
ids = ','.join(ids)
self_link = '/pubmed?search='+search
else:
#ids = first_item(id)
#fulltext = fulltext[0] if fulltext else u'no'
#if fulltext == 'yes':
search_terms = ids
self_link = '/pubmed?id='+ids
query = urllib.urlencode({'db' : NCBI_DB, 'id': ids, 'retmode': 'xml'})
search_url = NCBI_ARTICLE_ACCESS_PATTERN + query
logger.debug("ID search URL: " + search_url)
alt_link = search_url
doc = bindery.parse(search_url, standalone=True, model=PUBMED_MODEL)
#doc = bindery.parse(open('/Users/uche/tmp/efetch.fcgi.html'), standalone=True, model=PUBMED_MODEL)
metadata, first_id = metadata_dict(generate_metadata(doc))
return atom_results(doc, metadata, self_link, alt_link, search_terms)
def moin2atomentries(wikibase, outputdir, rewrite, pattern):
wikibase_len = len(rewrite)
if pattern: pattern = re.compile(pattern)
#print (wikibase, outputdir, rewrite)
req = urllib2.Request(wikibase, headers={'Accept': RDF_IMT})
with closing(urllib2.urlopen(req)) as resp:
feed = bindery.parse(resp)
for item in feed.RDF.channel.items.Seq.li:
uri = split_fragment(item.resource)[0]
relative = uri[wikibase_len:]
print >> sys.stderr, uri, relative
if pattern and not pattern.match(relative):
continue
if rewrite:
uri = uri.replace(rewrite, wikibase)
req = urllib2.Request(uri, headers={'Accept': DOCBOOK_IMT})
with closing(urllib2.urlopen(req)) as resp:
page = bindery.parse(resp)
entrydate = dateparse(unicode(page.article.articleinfo.revhistory.revision.date))
if entrydate.tzinfo == None: entrydate = entrydate.replace(tzinfo=DEFAULT_TZ)
output = os.path.join(outputdir, OUTPUTPATTERN%pathsegment(relative))
if os.access(output, os.R_OK):
lastrev = dateparse(unicode(bindery.parse(output).entry.updated))
if lastrev.tzinfo == None: lastrev = lastrev.replace(tzinfo=DEFAULT_TZ)
if (entrydate == lastrev):
print >> sys.stderr, 'Not updated. Skipped...'
continue
print >> sys.stderr, 'Writing to ', output
with open(output, 'w') as output:
handle_page(uri, page, outputdir, relative, output)
return
def test_xbel_2(self):
#BM1 = 'http://hg.4suite.org/amara/trunk/raw-file/bb6c40828b2d/demo/7days/bm1.xbel'
#BM2 = 'http://hg.4suite.org/amara/trunk/raw-file/bb6c40828b2d/demo/7days/bm2.xbel'
doc1 = bindery.parse(FILE('bm1.xbel'))
doc2 = bindery.parse(FILE('bm2.xbel'))
merge(doc1.xbel, doc2.xbel)
normalize_whitespace(doc1)
output = doc1.xml_encode(lookup("xml-indent")) + '\n'
self.assertEqual(output, open(FILE('merged-2.xbel')).read())
def test_simple_atom_entry():
'''Basic ns fixup upon mutation'''
doc = bindery.parse(ATOMENTRY1)
s = cStringIO.StringIO()
xml_print(doc, stream=s)
out = s.getvalue()
#self.assertEqual(out, ATOMENTRY1)
diff = treecompare.xml_diff(out, ATOMENTRY1)
diff = '\n'.join(diff)
assert not diff, "Expected=%r, returned=%r diff=%r" % (ATOMENTRY1, out, diff)
#Make sure we can parse the result
doc2 = bindery.parse(out)
def test_combined(self):
#ATOM1 = 'http://zepheira.com/news/atom/entries/'
#ATOM2 = 'http://ma.gnolia.com/atom/full/people/Uche'
ATOM1 = FILE('zepheira_atom.xml') #local download for testing
ATOM2 = FILE('magnolia_uche.xml') #local download for testing
output = cStringIO.StringIO()
combined_output = open(FILE('entries_combined.txt')).read() #local file for testing
doc1 = bindery.parse(ATOM1)
doc2 = bindery.parse(ATOM2)
combined = itertools.chain(*[doc.feed.entry for doc in (doc1, doc2)])
for node in sorted(combined, key=operator.attrgetter('updated')):
print >> output, node.title
self.assertEqual(output.getvalue(), combined_output)
def test_simple_atom_entry():
'''Basic ns fixup upon mutation'''
doc = bindery.parse(ATOMENTRY1)
s = cStringIO.StringIO()
xml_print(doc, stream=s)
out = s.getvalue()
#self.assertEqual(out, ATOMENTRY1)
diff = treecompare.xml_diff(out, ATOMENTRY1)
diff = '\n'.join(diff)
assert not diff, "Expected=%r, returned=%r diff=%r" % (ATOMENTRY1, out,
diff)
#Make sure we can parse the result
doc2 = bindery.parse(out)
def aggregate_entries(envelope, entries):
'''
envelope - input source of atom feed document to enclose entries
if it has any entries, the new ones are appended
entries - sequence of entry input sources
'''
envelope_doc = bindery.parse(envelope, model=FEED_MODEL)
entrydocs = [ bindery.parse(entry, model=ENTRY_MODEL) for entry in entries ]
#for entry in sorted(entrydocs, key=lambda x: attrgetter('updated')):
for entry_doc in sorted(entrydocs, key=lambda x: str(x.entry.updated), reverse=True):
envelope_doc.feed.xml_append(entry_doc.entry)
metadata = generate_metadata(envelope_doc)
return envelope_doc, metadata
def parse_landing_zone(ips_dict, cluster_config):
""" Returns the first landing zone found.
return (public_ip, directory)"""
cluster_config = cluster_config.strip().encode('utf-8')
if not ips_dict:
raise ClusterError('Cannot parse landing zone. ' +
'No ip translation table. ' +
'Cluster may not be ready yet.')
if not cluster_config:
raise ClusterError('Cannot parse landing zone. ' +
'Config is empty. ' +
'Cluster may not be ready yet.')
doc = bindery.parse(cluster_config.strip())
# Create node to ip address translation table
nodes = {}
for e in doc.Environment.Hardware.Computer:
nodes[e.name] = [e.netAddress]
Software = doc.Environment.Software
# attributes - build, buildSet, computer, description, directory, name
instances = Software.DropZone
[nodes[instance.computer].append('Drop Zone') for instance in instances]
computer = instances[0].computer
private_ip = nodes[computer][0]
directory = instances[0].directory
public_ip = ips_dict[private_ip]
return (public_ip, directory)
def moincms(wikibase, outputdir, pattern):
if pattern: pattern = re.compile(pattern)
#print (wikibase, outputdir, rewrite)
req = urllib2.Request(wikibase, headers={'Accept': RDF_IMT})
resp = urllib2.urlopen(req)
original_wiki_base = dict(resp.info())[ORIG_BASE_HEADER]
feed = bindery.parse(resp)
process_list = []
for item in feed.RDF.channel.items.Seq.li:
uri = split_fragment(item.resource)[0]
#print >> sys.stderr, (uri, str(item.resource), split_fragment(item.resource))
#Deal with the wrapped URI
if original_wiki_base:
#print >> sys.stderr, (uri, original_wiki_base.rstrip('/')+'/')
relative = relativize(uri,
original_wiki_base.rstrip('/') +
'/').lstrip('/')
uri = absolutize(relative, wikibase)
#print >> sys.stderr, (uri, relative)
if pattern and not pattern.match(relative):
continue
n = node.factory(uri, relative, outputdir)
if n.up_to_date():
pass
#print >> sys.stderr, 'Up to date. Skipped...'
else:
process_list.append(n)
#Process nodes needing update according to priority
for n in sorted(process_list, key=attrgetter('PRIORITY'), reverse=True):
#print >> sys.stderr, "processing ", n.rest_uri
n.render()
return
def read_info(fname):
all = bindery.parse(fname.read()).vector
info = [v for v in all.xml_children if v.xml_type == "element"]
def setImplements(fcns):
for fcn in fcns:
fcn.name = str(fcn.nameStr)
fcn.impl = getattr(fcn, "implName", None)
fcn.impl = (str(fcn.impl) if fcn.impl else fcn.impl)
fcn.line_idx = int(str(fcn.lineNum)) - 1
fcn.is_generator = str(fcn.isGenerator) == "true"
for fcn in fcns:
fcn.is_toplevel = getattr(fcn, "is_toplevel", False) or bool(
fcn.impl)
if fcn.impl:
try:
get_singleton(v for v in fcns
if v.name == fcn.impl).is_toplevel = True
except GetSingletonEmptyException:
global couldntfindexception
couldntfindexception = True
print("couldn't find function %s for file %s" %
(fcn.impl, fcn.srcFile))
return fcns
return dict(list(info).equiv_classes(
lambda a: Path(str(a.srcFile)))).map_values(setImplements)
def scanXMLDocument(self, path):
i=0
self.XML_W = lookup("xml")
self.doc = bindery.parse(path)
self.control.SetText(open(path).read())
self.control.EmptyUndoBuffer()
self.control.Colourise(0, -1)
# line numbers in the margin
self.control.SetMarginType(1, wx.stc.STC_MARGIN_NUMBER)
self.control.SetMarginWidth(1, 25)
#self.control.BeginTextColour((255,0,0))
#self.control.SetValue(self.doc.xml_encode(self.XML_W))
self.Sentences = []
for i in range (0, self.control.GetNumberOfLines()):
Line = self.control.GetLineText(i)
Line.strip()
#To remove spaces in the line
Line = re.sub(r"\s+", "", Line, flags=re.UNICODE)
if (Line[0]=='<') & (Line[1]=='S') & (Line[2]=='i'):
"""
Everytime we encouter '<Si' at the beginning
of a line, we append the Sentence Tab with
the line number
"""
self.Sentences.append(i)
def testDelChildElement7(self):
DOC = "<a><b>spam</b><b>eggs</b></a>"
EXPECTED = '<a><b>spam</b></a>'
doc = bindery.parse(DOC)
del doc.a[u'b'][1]
self.compare_output(doc, XMLDECL+EXPECTED)
return
def test_update_bindery(self):
doc = bindery.parse(self.XML)
#Add a new text node to a (--> last child)
doc.a.xml_append(u'New Content')
self.assertEqual(doc.a.xml_children[-1].xml_value, u'New Content')
new_elem = doc.xml_element_factory(None, u'spam')
doc.a.xml_append(new_elem)
self.assertEqual(doc.a.xml_children[-1], new_elem)
new_text = amara.tree.text(u'New Content')
doc.a.xml_insert(1, new_text)
self.assertEqual(doc.a.xml_children[1], new_text)
#Remove the last b child from a
num_kids = len(doc.a.xml_children)
#e1 = doc.a.b[-1].e
b1 = doc.a.b[1]
b1.xml_parent.xml_remove(b1)
self.assertEqual(len(doc.a.xml_children), num_kids-1)
doc = bindery.nodes.entity_base()
#doc.xml_clear() #Remove all children from a
doc.xml_append_fragment(self.XML)
check_bindery(self, doc)
return
def __init__(self, egdoc):
from amara import bindery
self.model_document = bindery.parse(egdoc)
self.model_document.xml_model.prefixes = top_namespaces(
self.model_document)
self.setup_model()
return
def testSetChildElement5(self):
DOC = "<a><b>spam</b><b>spam</b></a>"
EXPECTED = '<a><b>spam</b><b>eggs</b></a>'
doc = bindery.parse(DOC)
doc.a.b[1] = u"eggs"
self.compare_output(doc, XMLDECL+EXPECTED)
return
def xml_append_fragment(self, frag):
from amara.bindery import parse
doc = parse(frag)
for child in doc.xml_children:
self.xml_append(child)
return
def testDelChildElement4(self):
DOC = "<a><b>spam</b><b>spam</b></a>"
doc = bindery.parse(DOC)
def edit():
del doc.a.b[2]
self.assertRaises(IndexError, edit)
return
def testSetAttribute7(self):
DOC = '<a><b>spam</b></a>'
EXPECTED = '<a foo="bar"><b>spam</b></a>'
doc = bindery.parse(DOC)
doc.a.xml_attributes[u"foo"] = u"bar"
self.compare_output(doc, XMLDECL+EXPECTED)
return
def testDelAttribute1(self):
DOC = '<a b="spam"><b>spam</b></a>'
EXPECTED = '<a><b>spam</b></a>'
doc = bindery.parse(DOC)
del doc.a.b
self.compare_output(doc, XMLDECL+EXPECTED)
return
def testSetAttribute2(self):
DOC = '<a b="spam"><b>spam</b></a>'
EXPECTED = '<a b="eggs"><b>spam</b></a>'
doc = bindery.parse(DOC)
doc.a.xml_attributes[u'b'] = u"eggs"
self.compare_output(doc, XMLDECL+EXPECTED)
return
def OnSave(self,e):
# Save away the edited text
# Open the file, do an RU sure check for an overwrite!
# Grab the content to be saved
itcontains = self.control.GetValue().encode("utf-8")
# Checking if the XML is correct
try:
self.doc = bindery.parse(itcontains)
except bindery.Error as e:
line = e.lineNumber
dial = wx.MessageDialog(None, 'Error with XML parsing at line '+str(line-1)+' \nCannot save document, please correct the error', 'Error',
wx.OK | wx.ICON_ERROR)
if dial.ShowModal() == wx.ID_OK:
dial.Destroy()
return
dlg = wx.FileDialog(self, "Choose a file", self.dirname, self.filename, "*.*", \
wx.SAVE | wx.OVERWRITE_PROMPT)
if dlg.ShowModal() == wx.ID_OK:
# Open the file for write, write, close
self.filename=dlg.GetFilename()
self.dirname=dlg.GetDirectory()
filehandle=open(os.path.join(self.dirname, self.filename),'w')
filehandle.write(itcontains)
filehandle.close()
# Get rid of the dialog to keep things tidy
dlg.Destroy()
def testSetChildElement6(self):
DOC = "<a><b>spam</b><b>spam</b></a>"
doc = bindery.parse(DOC)
def edit():
doc.a.b[2] = u"eggs"
self.assertRaises(IndexError, edit)
return
def test_update_bindery(self):
doc = bindery.parse(self.XML)
#Add a new text node to a (--> last child)
doc.a.xml_append(u'New Content')
self.assertEqual(doc.a.xml_children[-1].xml_value, u'New Content')
new_elem = doc.xml_element_factory(None, u'spam')
doc.a.xml_append(new_elem)
self.assertEqual(doc.a.xml_children[-1], new_elem)
new_text = amara.tree.text(u'New Content')
doc.a.xml_insert(1, new_text)
self.assertEqual(doc.a.xml_children[1], new_text)
#Remove the last b child from a
num_kids = len(doc.a.xml_children)
#e1 = doc.a.b[-1].e
b1 = doc.a.b[1]
b1.xml_parent.xml_remove(b1)
self.assertEqual(len(doc.a.xml_children), num_kids - 1)
doc = bindery.nodes.entity_base()
#doc.xml_clear() #Remove all children from a
doc.xml_append_fragment(self.XML)
check_bindery(self, doc)
return
def factory(rest_uri, moin_link=None, opener=None):
opener = opener or urllib2.build_opener()
logger.debug('rest_uri: ' + rest_uri)
req = urllib2.Request(rest_uri, headers={'Accept': DOCBOOK_IMT})
resp = opener.open(req)
doc = bindery.parse(resp, standalone=True, model=MOIN_DOCBOOK_MODEL)
original_wiki_base = dict(resp.info())[ORIG_BASE_HEADER]
#self.original_wiki_base = dict(resp.info())[ORIG_BASE_HEADER]
#amara.xml_print(self.content_cache)
metadata, first_id = metadata_dict(generate_metadata(doc))
metadata = metadata[first_id]
akara_type = U(metadata[u'ak-type'])
logger.debug('Type: ' + akara_type)
try:
#Older Moin CMS resource types are implemented by registration to the global node.NODES
cls = node.NODES[akara_type]
except KeyError:
#Newer Moin CMS resource types are implemented by discovery of a URL,
#to which a POST request executes the desired action
return node.ENDPOINTS and (rest_uri, akara_type,
node.ENDPOINTS[akara_type], doc,
metadata, original_wiki_base)
else:
instance = cls(rest_uri,
moin_link,
opener,
cache=(doc, metadata, original_wiki_base))
return instance
def addAudioSample(self, start, end):
i = self.control.GetInsertionPoint()
j = self.control.LineFromPosition(i)
section = self.getSectionNumber(j)
line= self.control.GetFirstVisibleLine()
try:
self.doc = bindery.parse(self.control.GetValue().encode("utf-8"))
except bindery.Error as e:
print e.lineNumber
if section != 0:
try:
self.doc.TEXT.S[section-1].AUDIO.start = self.SimplifyAudioLimit(start)
self.doc.TEXT.S[section-1].AUDIO.end = self.SimplifyAudioLimit(end)
except AttributeError:
new_elem = self.doc.xml_element_factory(None, u"AUDIO")
new_elem.xml_attributes.setnode(new_elem.xml_attribute_factory(None, u"start", self.SimplifyAudioLimit(start)))
new_elem.xml_attributes.setnode(new_elem.xml_attribute_factory(None, u"end", self.SimplifyAudioLimit(end)))
self.doc.TEXT.S[section-1].xml_insert(0, new_elem)
self.control.SetValue(self.doc.xml_encode(self.XML_W))
i = self.control.PositionFromLine(j)
self.control.SetInsertionPoint(i)
self.control.ScrollToLine(line)
self.focus = section-1
def moincms(wikibase, outputdir, pattern):
if pattern: pattern = re.compile(pattern)
#print (wikibase, outputdir, rewrite)
req = urllib2.Request(wikibase, headers={'Accept': RDF_IMT})
resp = urllib2.urlopen(req)
original_wiki_base = dict(resp.info())[ORIG_BASE_HEADER]
feed = bindery.parse(resp)
process_list = []
for item in feed.RDF.channel.items.Seq.li:
uri = split_fragment(item.resource)[0]
#print >> sys.stderr, (uri, str(item.resource), split_fragment(item.resource))
#Deal with the wrapped URI
if original_wiki_base:
#print >> sys.stderr, (uri, original_wiki_base.rstrip('/')+'/')
relative = relativize(uri, original_wiki_base.rstrip('/')+'/').lstrip('/')
uri = absolutize(relative, wikibase)
#print >> sys.stderr, (uri, relative)
if pattern and not pattern.match(relative):
continue
n = node.factory(uri, relative, outputdir)
if n.up_to_date():
pass
#print >> sys.stderr, 'Up to date. Skipped...'
else:
process_list.append(n)
#Process nodes needing update according to priority
for n in sorted(process_list, key=attrgetter('PRIORITY'), reverse=True):
#print >> sys.stderr, "processing ", n.rest_uri
n.render()
return
def list_records(self, set="", resumption_token = ""):
'''
List records. Use either the resumption token or set id.
'''
if resumption_token:
params = {'verb' : 'ListRecords', 'resumptionToken': resumption_token}
else:
params = {'verb' : 'ListRecords', 'metadataPrefix': 'oai_dc', 'set': set}
qstr = urllib.urlencode(params)
url = self.root + '?' + qstr
self.logger.debug('OAI request URL: {0}'.format(url))
start_t = time.time()
resp, content = self.h.request(url)
retrieved_t = time.time()
self.logger.debug('Retrieved in {0}s'.format(retrieved_t - start_t))
doc = bindery.parse(url, model=OAI_LISTRECORDS_MODEL)
records, first_id = metadata_dict(generate_metadata(doc),
nesteddict=False)
for id_, props in records:
for k, v in props.iteritems():
props[k] = [ U(item) for item in v ]
if (doc.OAI_PMH.ListRecords is not None) and (doc.OAI_PMH.ListRecords.resumptionToken is not None):
resumption_token = U(doc.OAI_PMH.ListRecords.resumptionToken)
else:
resumption_token = ''
return {'records' : records, 'resumption_token' : resumption_token}
def testSetAttribute5(self):
DOC = '<a><b>spam</b></a>'
EXPECTED = '<a xmlns:ns="urn:bogus" ns:foo="bar"><b>spam</b></a>'
doc = bindery.parse(DOC)
doc.a.xmlns_attributes[u'ns']= u'urn:bogus'
doc.a.xml_attributes[u'ns:foo'] = u'bar'
self.compare_output(doc, XMLDECL+EXPECTED)
return
def testSetAttribute4(self):
DOC = '<a><b>spam</b></a>'
EXPECTED = '<a><b xml:lang="en">spam</b></a>'
doc = bindery.parse(DOC)
# doc.a.b.xml_set_attribute((u"xml:lang"), u"en")
doc.a.b.xml_attributes[u'xml:lang'] = u'en'
self.compare_output(doc, XMLDECL+EXPECTED)
return
def testSetAttribute3(self):
from xml.dom import Node
DOC = '<a b="spam"><b>spam</b></a>'
EXPECTED = '<a b="eggs"><b>spam</b></a>'
doc = bindery.parse(DOC)
doc.a[ATTRIBUTE_NODE, None, u'b'] = u'eggs'
self.compare_output(doc, XMLDECL+EXPECTED)
return
def render(self):
#Copy attachments to dir
req = urllib2.Request(self.rest_uri, headers={'Accept': ATTACHMENTS_IMT})
resp = urllib2.urlopen(req)
doc = bindery.parse(resp, model=ATTACHMENTS_MODEL)
for attachment in (doc.attachments.attachment or ()):
print attachment
return
def test_parse_with_file_path(self):
"""Parse with file path"""
fname = tempfile.mktemp('.xml')
fout = open(fname, 'w')
fout.write(MONTY_XML)
fout.close()
doc = parse(fname)
self.run_checks(doc)
def Xtest_parse_with_url(self):
doc = parse(TEST_URL)
#Minimal node testing
self.assertEqual(len(doc.xml_children), 1)
self.assertEqual(doc.xml_children[0].xml_type, tree.element.xml_type)
self.assertEqual(doc.xml_children[0].xml_qname, 'disclaimer')
self.assertEqual(doc.xml_children[0].xml_namespace, None)
self.assertEqual(doc.xml_children[0].xml_prefix, None,)
def test_xslt():
url = server() + "akara.xslt?" + urllib.urlencode({"@xslt": XSLT_URL})
req = urllib2.Request(url)
req.add_header("Content-Type", "text/xml")
response = urllib2.urlopen(req, XML_DATA)
doc = bindery.parse(response)
assert str(doc.html.head.title) == "Document Title", repr(str(doc.html.head.title))
def test_xpath1(self):
"""Test AVT"""
doc = bindery.parse(self.DOC)
self.assertEqual(doc.xml_select(u'ns:friends/ns:bff', prefixes=self.PREFIXES), u'Suzie Q')
self.assertEqual(
[ unicode(f.xml_select(u'concat(@rank, ": ", .)')) for f in doc.friends.bff ],
[u'1: Suzie Q', u'2: Betty Boost'])
return
def test_parse_with_file_path(self):
"""Parse with file path"""
fname = tempfile.mktemp('.xml')
fout = open(fname, 'w')
fout.write(MONTY_XML)
fout.close()
doc = parse(fname)
self.run_checks(doc)
def __init__(self, schdoc):
from amara import bindery
dispatcher.__init__(self)
self.model_document = bindery.parse(schdoc)
self.model_document.xml_model.prefixes = top_namespaces(self.model_document)
self.rules = []
self.setup_model()
return
def __init__(self, parent):
self._init_ctrls(parent)
# replace the file name below
fileName = 'Musica/crdo-NRU_F4_10_AGRICULTURAL_ACTIVITIES.xml'
file = os.path.join(os.getcwd(), fileName)
self.xmlDoc = bindery.parse(file)
self.styledTextCtrl1.AddTextUTF8(self.xmlDoc.xml(indent=u'yes'))
def ejsonize(isrc):
'''
Convert Atom syntax to a dictionary
Note: the conventions used are designed to simplify conversion to Exhibit JSON
(see: http://www.ibm.com/developerworks/web/library/wa-realweb6/ ; listing 3)
'''
doc = bindery.parse(isrc, model=FEED_MODEL)
def process_entry(e):
known_elements = [u'id', u'title', u'link', u'author', u'category', u'updated', u'content', u'summary']
data = {
u"id": unicode(e.id),
#XXX Shall we use title for label?
u"label": unicode(e.id),
u"type": u"Entry",
u"title": unicode(e.title),
u"link": first_item([ l.href for l in e.link if l.rel in [None, u"alternate"] ], []),
#Nested list comprehension to select the alternate link,
#then select the first result ([0]) and gets its href attribute
u"authors": [ unicode(a.name) for a in iter(e.author or []) ],
#Nested list comprehension to create a list of category values
u"categories": [ unicode(c.term) for c in iter(e.category or []) ],
u"updated": unicode(e.updated),
u"summary": unicode(e.summary),
}
if not data[u"categories"]: del data[u"categories"]
if e.summary is not None:
data[u"summary"] = unicode(e.summary)
if e.content is not None:
try:
data[u"content_src"] = unicode(e.content.src)
except AttributeError:
data[u"content_text"] = deserialize_text_construct(e.content)
for child in e.xml_elements:
if child.xml_namespace != ATOM_NAMESPACE and child.xml_local not in known_elements:
data[child.xml_local] = unicode(child)
return data
try:
doc_entries = iter(doc.feed.entry)
feedinfo = {
u"id": unicode(doc.feed.id),
#XXX Shall we use title for label?
u"label": unicode(doc.feed.id),
u"type": u"Feed",
u"title": unicode(doc.feed.title),
u"link": first_item([ l.href for l in doc.feed.link if l.rel in [None, u"alternate"] ], []),
u"authors": [ unicode(a.name) for a in iter(doc.feed.author or []) ],
u"updated": unicode(doc.feed.updated),
}
except AttributeError:
try:
doc_entries = iter(doc.entry)
feedinfo = None
except AttributeError:
#FIXME L10N
raise ValueError("Does not appear to be a valid Atom file")
return [ process_entry(e) for e in doc_entries ]
def test_nasty_xml_1(self):
"""XML with 2 elements with same local name and different NS on same parent"""
doc = parse(NASTY_NS_XML1)
self.assertEqual(len(doc.top.xml_children), 5)
self.assertEqual(len(list(doc.top.monty)), 1)
self.assertEqual(len(list(doc.top.monty_)), 1)
self.assertEqual(doc.top.monty.xml_namespace, u"urn:bogus:a")
self.assertEqual(doc.top.monty_.xml_namespace, u"urn:bogus:b")
self.assertEqual(doc.top.monty.xml_following_sibling.xml_following_sibling, doc.top.monty_)
def testInsertBefore1(self):
DOC = "<a><b>eggs</b></a>"
EXPECTED = '<a><b>spam</b><b>eggs</b></a>'
doc = bindery.parse(DOC)
new = doc.xml_element_factory(None, u'b')
new.xml_append(doc.xml_text_factory(u'spam'))
doc.a.xml_insert(0, new)
self.compare_output(doc, XMLDECL+EXPECTED)
return
def test_xslt():
url = server() + "akara.xslt?" + urllib.urlencode({"@xslt": XSLT_URL})
req = urllib2.Request(url)
req.add_header("Content-Type", "text/xml")
response = urllib2.urlopen(req, XML_DATA)
doc = bindery.parse(response)
assert str(doc.html.head.title) == "Document Title", repr(
str(doc.html.head.title))
def render(self):
#Copy attachments to dir
req = urllib2.Request(self.rest_uri,
headers={'Accept': ATTACHMENTS_IMT})
resp = urllib2.urlopen(req)
doc = bindery.parse(resp, model=ATTACHMENTS_MODEL)
for attachment in (doc.attachments.attachment or ()):
print attachment
return
def test_xpath1(self):
"""Test AVT"""
doc = bindery.parse(self.DOC)
self.assertEqual(doc.xml_select(u'ns:friends/ns:bff'), u'Suzie Q')
self.assertEqual([
unicode(f.xml_select(u'concat(@rank, ": ", .)'))
for f in doc.friends.bff
], [u'1: Suzie Q', u'2: Betty Boost'])
return
def test_xpath(self):
doc = bindery.parse(self.MONTY_XML)
m = doc.monty
p1 = doc.monty.python
self.assertEqual(p1.xml_select(u'string(@spam)'), u'eggs')
for p, line in zip(doc.xml_select(u'//python'), self.lines_py):
output = cStringIO.StringIO()
xml_print(p, stream=output)
self.assertEqual(output.getvalue(), line)
def test_xpath(self):
doc = bindery.parse(self.MONTY_XML)
m = doc.monty
p1 = doc.monty.python
self.assertEqual(p1.xml_select(u'string(@spam)'), u'eggs')
for p, line in zip(doc.xml_select(u'//python'), self.lines_py):
output = cStringIO.StringIO()
xml_print(p, stream=output)
self.assertEqual(output.getvalue(), line)
def parseXMLDocument(self, path):
i=0
self.XML_W = lookup("xml")
try:
self.doc = bindery.parse(path)
except amara.lib.IriError:
self.doc = bindery.parse(path, standalone=True)
#self.control.SetText(open(path).read())
self.control.EmptyUndoBuffer()
self.control.Colourise(0, -1)
# line numbers in the margin
self.control.SetMarginType(1, wx.stc.STC_MARGIN_NUMBER)
self.control.SetMarginWidth(1, 25)
self.control.SetValue(self.doc.xml_encode(self.XML_W).decode("utf-8").replace("<S id=","\n <S id=").replace("<", u"\u2039").replace(">", u"\u203A"))
def testInsertAfter3(self):
DOC = "<a><b>spam</b><c>ham</c><c>pork</c></a>"
EXPECTED = "<a><b>spam</b><c>eggs</c><c>ham</c><c>pork</c></a>"
doc = bindery.parse(DOC)
new = doc.xml_element_factory(None, u'c')
new.xml_append(doc.xml_text_factory(u'eggs'))
doc.a.xml_insert(doc.a.xml_index(doc.a.b) +1, new)
self.compare_output(doc, XMLDECL+EXPECTED)
return
def test_parse_with_stream(self):
"""Parse with stream"""
fname = tempfile.mktemp('.xml')
fout = open(fname, 'w')
fout.write(MONTY_XML)
fout.close()
fout = open(fname, 'r')
doc = parse(fout)
fout.close()
self.run_checks(doc)
def test_bindery(self):
doc = bindery.parse(self.MONTY_XML)
m = doc.monty
p1 = doc.monty.python #or m.python; p1 is just the first python element
self.assertEqual(p1.xml_attributes[(None, u'spam')], u'eggs')
self.assertEqual(p1.spam, u'eggs')
for p, line in zip(doc.monty.python, self.lines_py): #The loop will pick up both python elements
output = cStringIO.StringIO()
xml_print(p, stream=output)
self.assertEqual(output.getvalue(), line)
def testXml(self):
tmp = self.control.GetValue().encode("utf-8")
try:
self.doc = bindery.parse(tmp)
except bindery.Error as e:
line= e.lineNumber
res = self.errorDialog(line)
return res
return 1
def test_avt1(self):
"""Test AVT"""
doc = bindery.parse(self.DOC)
self.assertEqual(doc.xml_avt(u'Hello, {friends/bff}'), u'Hello, Suzie Q')
self.assertEqual(
[ f.xml_avt(u'Big up, {.}, rank {@rank}') for f in doc.friends.bff ],
[u'Big up, Suzie Q, rank 1', u'Big up, Betty Boost, rank 2'])
self.assertEqual(
[ f.xml_avt(u'{@rank}: {.}') for f in sorted(doc.friends.bff, key=attrgetter("rank")) ],
[u'1: Suzie Q', u'2: Betty Boost'])
return
def test_charsearch():
url = server() + "akara.unicode.search?q=DAGGER"
doc = bindery.parse(urllib2.urlopen(url))
names = set()
see_alsos = set()
for child in doc.xml_select(u"characters/character"):
names.add(child.name)
see_alsos.add(child.see_also)
assert names == set(["DAGGER", "DOUBLE DAGGER"]), names
assert see_alsos == set(
["http://www.fileformat.info/info/unicode/char/2020/index.htm",
"http://www.fileformat.info/info/unicode/char/2021/index.htm"]), see_alsos
def search(self, term):
qstr = urllib.urlencode({'verb' : 'GetRecord', 'metadataPrefix': 'oai_dc', 'identifier': dspace_id})
url = DSPACE_OAI_ENDPOINT + '?' + qstr
logger.debug('DSpace URL: ' + str(url))
#keywords = [ (k.strip(), JOVE_TAG) for k in unicode(row.xml_select(u'string(.//*[@class="keywords"])')).split(',') ]
doc = bindery.parse(url, model=OAI_MODEL)
#print >> sys.stderr, list(generate_metadata(doc))
resources, first_id = metadata_dict(generate_metadata(doc), nesteddict=False)
record = doc.OAI_PMH
resource = resources[first_id]
def test_namespace_free_xhtml3(self):
'namespace-free XHTML' + '...as XML with pretty print'
doc = self._build_namespace_free_xhtml()
s = cStringIO.StringIO()
xml_print(doc, stream=s, indent=True)
out = s.getvalue()
#self.assertEqual(out, ATOMENTRY1)
diff = treecompare.xml_diff(out, XHTML_EXPECTED_3, whitespace=False)
diff = '\n'.join(diff)
self.assertFalse(diff, msg=(None, diff))
#Make sure we can parse the result
doc2 = bindery.parse(out)
def test_avt1(self):
"""Test AVT"""
doc = bindery.parse(self.DOC)
#FIXME: try to work it so , prefixes= is not needed for this case (declaration in scope)
self.assertEqual(doc.xml_avt(u'Hello, {ns:friends/ns:bff}', prefixes=self.PREFIXES), u'Hello, Suzie Q')
self.assertEqual(
[ f.xml_avt(u'Big up, {.}, rank {@rank}') for f in doc.friends.bff ],
[u'Big up, Suzie Q, rank 1', u'Big up, Betty Boost, rank 2'])
self.assertEqual(
[ f.xml_avt(u'{@rank}: {.}') for f in sorted(doc.friends.bff, key=attrgetter("rank")) ],
[u'1: Suzie Q', u'2: Betty Boost'])
return
