def check_language(file):
''' test to make sure all not_in_new files crept into jl files becaues of order problem
because they are actually french (no french ids should be in the .jl file)
'''
not_in_new = pickle.load(open('not_in_new.pkl','rb'))
final = etree.parse(file).getroot()
for i,child in enumerate(final):
try:
formid = str(child.xpath("FORM[NAME='thisformid']/A/text()")[0]).lower()
if formid in not_in_new:
print common.language(child)
except IndexError:
print "SMALL NO FORM ID", child.xpath("FORM[NAME='thisformid']/A/text()")
#raise
print "Conlusion, all records have french primary ids, and thus must be removed"
def process_doubles(self, datafile):
tree = etree.parse(datafile)
root = tree.getroot()
for i,pair in enumerate(self.combined_elements(root)):
node_en = pair[0]
lang = common.language(node_en)
if lang == "English":
pass
#continue
# MAKE SURE It's english and that the order in pilot-matched.xml has not been broken
node_fr = pair[1]
package_en = Transform().process_node(i,node_en, "eng; CAN")
package_fr = Transform().process_node(i,node_fr,u"fra; CAN")
# Transfer French Data to English Package
for pack in package_fr['resources']:
if pack['format'] != "HTML":
package_en['resources'].append(pack)
if package_en['resources'] == []:
raise Exception
if not package_en['owner_org']:
print "############### NO ORGANIZATION ###########", package_en['id']
elif not package_en['title']:
print "############### NO TITLE ###########", package_en['id']
elif not package_en['id']:
"############ NO ID ###########",package_en['id']
else:
print i, "OK",package_en['id']
print package_en['title']
print package_en['title_fra']
#if i>80:sys.exit()
if package_fr['id'] == "dafb6413-5dab-45ca-bcd2-8c6ff4b67be5":
print "STOP"
self.outfile.write(json.dumps(package_en) + "\n")
def find_missing_bilingual(file1,file2):
cnt=Counter()
tree1 = etree.parse(file1)
tree2 = etree.parse(file2)
root1 = tree1.getroot()
root2 = tree2.getroot()
ids_bi=[]
ids_source=[]
for i,child in enumerate(root1):
try:
formid = str(child.xpath("FORM[NAME='thisformid']/A/text()")[0]).lower()
except:
print "SMALL NO FORM ID"
try:
lang=language(child)
if lang=="Bilingual":
ids_bi.append(formid)
except:
print "small NO Language"
for i,child in enumerate(root2):
try:
formid = str(child.xpath("FORM[NAME='thisformid']/A/text()")[0]).lower()
except:
print "BIG NO FORM ID"
print child.find("DC.TITLE").text
print child.xpath("FORM[NAME='title_en']/A/text()")
#print etree.tostring(child)
try:
lang=common.language(child)
if lang=="Bilingual":
ids_source.append(formid)
except:
print "small NO Language"
biset= set(ids_bi)
fullset = set(ids_source)
print biset.issubset(fullset)
diff = fullset.difference(biset)
print diff
def _parse_fields(self,node):
''' package fields '''
for ckan_name, pilot_name, field in schema_description.dataset_all_fields():
if pilot_name:
path = "FORM[NAME='%s']/A/text()"%pilot_name
element = node.xpath(path)
try:
# Deal with Pilot UUID CODES
value = element[0].strip()
if "|" in value:
split_value=value.split("|")[1]
self.fields[pilot_name] = field['choices_by_pilot_uuid'][split_value]['key']
else:
self.fields[pilot_name]=value
except IndexError:
self.fields[pilot_name]=""
else:
# ckan_name / field does not belong at PilotRecord level. Process in CanadaRecord
pass
''' Grab data that is not defined in schema '''
self.fields['language'] = common.language(node)
geo_lower_left = node.xpath("FORM[NAME='geo_lower_left']/A/text()")
geo_upper_right = node.xpath("FORM[NAME='geo_upper_right']/A/text()")
if geo_lower_left and geo_upper_right and geo_lower_left[0] != "N/A":
print "GEO ", geo_upper_right, geo_lower_left
try:
left,bottom = geo_lower_left[0].split(" ")
right, top = geo_upper_right[0].split(" ")
except ValueError:
''' To catch values that have a dash that should perhaps be a minus ['84 - 43'] ['41.5 - 141'] '''
left,bottom = geo_lower_left[0].replace(" - "," -").split(" -")
right, top = geo_upper_right[0].replace(" - "," -").split(" -")
coordinates = [[left, bottom], [left,top], [right, top], [right, bottom]]
self.fields['spatial']= {'type': 'Polygon', 'coordinates': coordinates}
''' resources'''
try:
for i in range(1,5):
url = node.xpath("FORM[NAME='dataset_link_en_%d']/A/text()" % i)
if url:
resource_dict = {}
resource_dict['url']=url[0]
if "http://data.gc.ca/commonwebsol/fileuploads/C/4/0/C4060F22-17EB-450D-9B5E-A1216E75DF47/Dictionnaire" in resource_dict['url']:
print "STOP"
# Force a language from parent
resource_dict['language'] = self.fields['language']
format = node.xpath("FORM[NAME='dataset_format_%d']/A/text()" % i)
size = node.xpath("FORM[NAME='dataset_size_%d']/A/text()" % i)
if format:resource_dict['format']=format[0].split("|")[1]
self.resources.append(PilotResource(resource_dict,'dataset_link_en_'))
else:
break
extras=['supplementary_documentation_en',
'supplementary_documentation_fr',
'data_dictionary_fr',
'dictionary_list:_en']
for extra in extras:
url= node.xpath("FORM[NAME='%s']/A/text()"% extra)
if url:
resource_dict = {}
resource_dict['url']=url[0]
self.resources.append(PilotResource(resource_dict,extra))
except:
raise
