def validate_xml(xml, xsl_fname='fgdc', as_dataframe=False):
"""
Parameters
----------
xml : lxml document
or
filename
or
string containing xml representation
xsl_fname : str (optional)
can be one of:
'fgdc' - uses the standard fgdc schema
../resources/FGDC/fgdc-std-001-1998-annotated.xsd
'bdp' = use the Biological Data profile schema,
../resources/FGDC/BDPfgdc-std-001-1998-annotated.xsd
full file path to another local schema.
if not specified defaults to 'fgdc'
Returns
-------
list of tuples
(
"""
if xsl_fname.lower() == 'fgdc':
xsl_fname = utils.get_resource_path('fgdc/fgdc-std-001-1998-annotated.xsd')
elif xsl_fname.lower() == 'bdp':
xsl_fname = utils.get_resource_path('fgdc/BDPfgdc-std-001-1998-annotated.xsd')
else:
xsl_fname = xsl_fname
xmlschema_doc = etree.parse(xsl_fname)
xmlschema = etree.XMLSchema(xmlschema_doc)
xml_str = xml_utils.node_to_string(xml_utils.xml_document_loader(xml))
tree = etree.ElementTree(etree.fromstring(xml_str))
if xmlschema.validate(tree):
return []
# tree = etree.ElementTree(xml)
line_lookup = dict([(e.sourceline, tree.getpath(e)) for e in tree.xpath('.//*')])
line_lookup[tree.getroot().sourceline] = tree.getpath(tree.getroot())
errors = []
for error in xmlschema.error_log:
try:
errors.append((line_lookup[error.line][1:],
clean_error_message(error.message), error.line))
except KeyError:
errors.append(('Unknown', clean_error_message(error.message),
error.line))
if as_dataframe:
cols = ['xpath', 'message', 'line number']
return pd.DataFrame.from_records(errors, columns=cols)
else:
return errors
def validate_xml(xml, xsl_fname="fgdc", as_dataframe=False):
"""
Parameters
----------
xml : lxml document
or
filename
or
string containing xml representation
xsl_fname : str (optional)
can be one of:
'fgdc' - uses the standard fgdc schema
../resources/FGDC/fgdc-std-001-1998-annotated.xsd
'bdp' = use the Biological Data profile schema,
../resources/FGDC/BDPfgdc-std-001-1998-annotated.xsd
full file path to another local schema.
if not specified defaults to 'fgdc'
as_dataframe : bool
used to specify return format (list of tuples or dataframe)
Returns
-------
list of tuples
(xpath, error message, line number)
or
pandas dataframe
"""
if xsl_fname.lower() == "fgdc":
xsl_fname = utils.get_resource_path(FGDC_XSD_NAME)
elif xsl_fname.lower() == "bdp":
xsl_fname = utils.get_resource_path(BDP_XSD_NAME)
else:
xsl_fname = xsl_fname
xmlschema = xml_utils.load_schema(xsl_fname)
xml_doc = xml_utils.xml_document_loader(xml)
xml_str = xml_utils.node_to_string(xml_doc)
tree_node = xml_utils.string_to_node(xml_str.encode("utf-8"))
lxml._etree._ElementTree(tree_node)
errors = []
srcciteas = []
src_xpath = "dataqual/lineage/srcinfo/srccitea"
src_nodes = tree_node.xpath(src_xpath)
for i, src in enumerate(src_nodes):
srcciteas.append(src.text)
if src.text is None:
if len(src_nodes) == 1:
errors.append((
"metadata/" + src_xpath,
"source citation abbreviation cannot be empty",
1,
))
else:
xpath = "metadata/dataqual/lineage/srcinfo[{}]/srccitea"
errors.append((
xpath.format(i + 1),
"source citation abbreviation cannot be empty",
1,
))
procstep_xpath = "dataqual/lineage/procstep"
procstep_nodes = tree_node.xpath(procstep_xpath)
for proc_i, proc in enumerate(procstep_nodes):
srcprod_nodes = proc.xpath("srcprod")
for srcprod_i, srcprod in enumerate(srcprod_nodes):
srcciteas.append(srcprod.text)
if srcprod.text is None:
error_xpath = procstep_xpath
if len(procstep_nodes) > 1:
error_xpath += "[{}]".format(proc_i + 1)
error_xpath += "/srcprod"
if len(srcprod_nodes) > 1:
error_xpath += "[{}]".format(proc_i + 1)
errors.append((
"metadata/" + error_xpath,
"source produced abbreviation cannot be empty",
1,
))
srcused_xpath = "dataqual/lineage/procstep/srcused"
srcused_nodes = tree_node.xpath(srcused_xpath)
for i, src in enumerate(srcused_nodes):
if src.text not in srcciteas:
if len(srcused_nodes) == 1:
errors.append((
"metadata/" + srcused_xpath,
"Source Used Citation Abbreviation {} "
"not found in Source inputs "
"used".format(src.text),
1,
))
else:
xpath = "metadata/dataqual/lineage/procstep[{}]/srcused"
errors.append((
xpath.format(i + 1),
"Source Used Citation Abbreviation {} "
"not found in Source inputs "
"used".format(src.text),
1,
))
if xmlschema.validate(tree_node) and not errors:
return []
line_lookup = dict([(e.sourceline, tree_node.getroottree().getpath(e))
for e in tree_node.xpath(".//*")])
sourceline = tree_node.sourceline
line_lookup[sourceline] = tree_node.getroottree().getpath(tree_node)
fgdc_lookup = get_fgdc_lookup()
for error in xmlschema.error_log:
error_msg = clean_error_message(error.message, fgdc_lookup)
try:
errors.append((line_lookup[error.line][1:], error_msg, error.line))
except KeyError:
errors.append(("Unknown", error_msg, error.line))
errors = list(OrderedDict.fromkeys(errors))
if as_dataframe:
cols = ["xpath", "message", "line number"]
return pd.DataFrame.from_records(errors, columns=cols)
else:
return errors