optimize detection of xml-content

document.py

@@ -8,6 +8,7 @@
 import os.path
 import json
 import copy
+import facturx
 from lxml import etree
 from datetime import datetime, date
 from decimal import Decimal
@@ -22,6 +23,8 @@ from trytond.protocols.jsonrpc import JSONEncoder
 
 
 xml_types = [
+    (['xsd', 'Factur-X_1.07.2_MINIMUM', 'Factur-X_1.07.2_MINIMUM.xsd'],
+        'Factur-X minimum', 'facturx_extended'),
     (['xsd', 'Factur-X_1.07.2_EXTENDED', 'Factur-X_1.07.2_EXTENDED.xsd'],
         'Factur-X extended', 'facturx_extended'),
     (['xsd', 'Factur-X_1.07.2_EN16931', 'Factur-X_1.07.2_EN16931.xsd'],
@@ -30,8 +33,6 @@ xml_types = [
         'Factur-X basic', ''),
     (['xsd', 'Factur-X_1.07.2_BASICWL', 'Factur-X_1.07.2_BASICWL.xsd'],
         'Factur-X basic-wl', ''),
-    (['xsd', '/Factur-X_1.07.2_MINIMUM', 'Factur-X_1.07.2_MINIMUM.xsd'],
-        'Factur-X minimum', ''),
     (['xsd', 'CII D22B XSD', 'CrossIndustryInvoice_100pD22B.xsd'],
         'CrossIndustryInvoice D22', ''),
     (['xsd', 'os-UBL-2.1', 'xsd/maindoc', 'UBL-Invoice-2.1.xsd'],
@@ -53,19 +54,29 @@ class Incoming(metaclass=PoolMeta):
     def default_company(cls):
         return Transaction().context.get('company')
 
+    @classmethod
+    def process(cls, documents, with_children=False):
+        """ check if content can be converted
+
+        Args:
+            documents (list): records of model document.incoming
+            with_children (bool, optional): convert sub-documents.
+                Defaults to False.
+        """
+        for document in documents:
+            cls._readxml_check_content(document)
+        super().process(documents, with_children)
+
     def _process_supplier_invoice(self):
         """ try to detect content of 'data', read values
         """
         invoice = super()._process_supplier_invoice()
 
-        self.xsd_type = None
         if self.mime_type == 'application/xml':
             # detect xml-content
             xml_info = self._facturx_detect_content()
             if xml_info:
                 (xsd_type, funcname, xmltree) = xml_info
-                self.xsd_type = xsd_type
-
                 xml_read_func = getattr(self, '_readxml_%s' % funcname, None)
                 if not xml_read_func:
                     raise UserError(gettext(
@@ -74,7 +85,9 @@ class Incoming(metaclass=PoolMeta):
                 # read xml data, write to 'self.parsed_data'
                 xml_read_func(xmltree)
                 # update invoice with imported data
-                invoice = self._readxml_update_invoice(invoice)
+                if self.parsed_data:
+                    invoice = self._readxml_update_invoice(invoice)
+        # raise ValueError('stop')
         return invoice
 
     def _readxml_xpath(self, tags):
@@ -797,7 +810,7 @@ class Incoming(metaclass=PoolMeta):
                 ('ram:BatchID', 'lot'),
                 ('ram:BrandName', 'brand_name'),
                 ('ram:ModelName', 'model_name'),
-                ('ram:OriginTradeCountry', 'trade_country')
+                (['ram:OriginTradeCountry', 'ram:ID'], 'trade_country')
                 ]:
             result[key] = self._readxml_getvalue(xmldata, xpath_line + [
                 'ram:SpecifiedTradeProduct', xkey])
@@ -809,8 +822,8 @@ class Incoming(metaclass=PoolMeta):
                 'ram:ApplicableProductCharacteristic'], [
                 ('ram:TypeCode', 'code'),
                 ('ram:Description', 'description'),
-                ('ram:ValueMeasure', 'uom'),
+                ('ram:ValueMeasure', 'value', Decimal),
-                ('ram:ValueY', 'value')])
+                ('ram:Value', 'uom')])
 
         # classification of product
         result['classification'] = self._readxml_read_listdata(
@@ -958,7 +971,29 @@ class Incoming(metaclass=PoolMeta):
                 msg='cannot convert %(val)s' % {
                     'val': date_string}))
 
-    def _facturx_detect_content(self):
+    @classmethod
+    def _readxml_check_content(cls, document):
+        """ try to detect content, fire exception if fail
+
+        Args:
+            document (record): model document.incoming
+        """
+        xml_data = etree.fromstring(document.data)
+        try:
+            fx_flavour = facturx.get_flavor(xml_data)
+            fx_level = facturx.get_level(xml_data)
+            document._facturx_detect_content(fx_flavour, fx_level)
+            document.save()
+        except Exception as e1:
+            raise UserError(gettext(
+                'document_incoming_invoice_xml.msg_convert_error',
+                msg='%(name)s [%(flavour)s|%(level)s] %(msg)s' % {
+                    'name': document.name or '-',
+                    'flavour': fx_flavour,
+                    'level': fx_level,
+                    'msg': str(e1)}))
+
+    def _facturx_detect_content(self, flavour=None, level=None):
         """ check xml-data against xsd of XRechnung and FacturX,
             begin with extended, goto minimal, then xrechnung
 
@@ -969,13 +1004,23 @@ class Incoming(metaclass=PoolMeta):
         xml_data = etree.fromstring(self.data)
 
         for xsdpath, xsdtype, funcname in xml_types:
+            if flavour == 'factur-x' and level:
+                if not (xsdtype.lower().startswith('factur-x') and
+                        xsdtype.lower().endswith(level)):
+                    # skip check if xml-content is already known
+                    continue
+
             fname = os.path.join(*[os.path.split(__file__)[0]] + xsdpath)
             schema = etree.XMLSchema(etree.parse(fname))
             try:
                 schema.assertValid(xml_data)
-            except etree.DocumentInvalid:
+                self.xsd_type = xsdtype
+                return (xsdtype, funcname, xml_data)
+            except etree.DocumentInvalid as e1:
+                # fire exception for knows xml-content
+                if flavour == 'factur-x' and level:
+                    raise e1
                 pass
-            return (xsdtype, funcname, xml_data)
         return None
 
 # end Incoming