Merge pull request #112 from strictdoc-project/stanislaw/develop

helpers: lxml: lxml_convert_to/from_reqif_ns_xhtml_string() helper

Author: stanislaw

reqif/helpers/lxml.py

@@ -1,7 +1,9 @@
+from copy import deepcopy
 from itertools import chain
 
 from lxml import etree
 from lxml.etree import tostring
+from lxml.html import fragment_fromstring
 
 
 def dump_xml_node(node):
@@ -41,10 +43,15 @@ def my_escape_title(string: str) -> str:
 # when the etree.tostring(...) method is used:
 # <reqif-xhtml:div xmlns:reqif-xhtml="http://www.w3.org/1999/xhtml">--</reqif-xhtml:div>  # noqa: E501
 # FIXME: Would be great to find a better solution for this.
-def stringify_namespaced_children(node) -> str:
+def stringify_namespaced_children(node, namespace_tag=None) -> str:
+    if namespace_tag is None:
+        nskey = next(iter(node.nsmap.keys()))
+    else:
+        nskey = namespace_tag
+
     def _stringify_reqif_ns_node(node):
         assert node is not None
-        nskey = next(iter(node.nsmap.keys()))
+
         output = ""
         node_no_ns_tag = etree.QName(node).localname
         output += f"<{nskey}:{node_no_ns_tag}"
@@ -90,10 +97,37 @@ def stringify_children(node):
     )
 
 
+def lxml_convert_to_reqif_ns_xhtml_string(string, reqif_xhtml=True) -> str:
+    namespace_tag = "reqif-xhtml" if reqif_xhtml else "xhtml"
+    node = fragment_fromstring(string, create_parent="NOT-USED")
+    return stringify_namespaced_children(node, namespace_tag=namespace_tag)
+
+
+def lxml_convert_from_reqif_ns_xhtml_string(lxml_node) -> str:
+    lxml_node_deep_copy = deepcopy(lxml_node)
+    lxml_strip_namespace_from_xml(lxml_node_deep_copy, full=True)
+    return tostring(
+        lxml_node_deep_copy, encoding=str, pretty_print=True
+    ).rstrip()
+
+
 def is_self_closed_tag(xml):
     # The tag cannot be closed if it has children or has a non-None text.
     if len(xml.getchildren()) > 0:
         return False
     if xml.text is not None:
         return False
     return True
+
+
+def lxml_strip_namespace_from_xml(root_xml, full=False):
+    for elem in root_xml.getiterator():
+        # Remove an XML namespace URI in the element's name but keep the
+        # namespaces in the HTML content as found in the
+        # <ATTRIBUTE-VALUE-XHTML> of ReqIF XML.
+        if not full and "http://www.w3.org/1999/xhtml" in elem.tag:
+            continue
+        elem.tag = etree.QName(elem).localname
+    # Remove unused namespace declarations
+    etree.cleanup_namespaces(root_xml)
+    return root_xml

reqif/helpers/string/__init__.py

@@ -6,6 +6,7 @@
 from lxml import etree
 from lxml.etree import DocInfo
 
+from reqif.helpers.lxml import lxml_strip_namespace_from_xml
 from reqif.models.error_handling import (
     ReqIFMissingTagException,
     ReqIFSchemaError,
@@ -100,7 +101,7 @@ def _parse_reqif(xml_reqif) -> ReqIFBundle:
         schema_namespace = namespace_info.get("xsi")
 
         xml_reqif_nons = (
-            ReqIFParser._strip_namespace_from_xml(copy.deepcopy(xml_reqif))
+            lxml_strip_namespace_from_xml(copy.deepcopy(xml_reqif))
             if namespace is not None
             else xml_reqif
         )
@@ -317,16 +318,3 @@ def _parse_reqif_content(
             spec_relation_groups=spec_relation_groups,
         )
         return reqif_content, lookup, exceptions
-
-    @staticmethod
-    def _strip_namespace_from_xml(root_xml):
-        for elem in root_xml.getiterator():
-            # Remove an XML namespace URI in the element's name but keep the
-            # namespaces in the HTML content as found in the
-            # <ATTRIBUTE-VALUE-XHTML> of ReqIF XML.
-            if "http://www.w3.org/1999/xhtml" in elem.tag:
-                continue
-            elem.tag = etree.QName(elem).localname
-        # Remove unused namespace declarations
-        etree.cleanup_namespaces(root_xml)
-        return root_xml

reqif/helpers/string.py reqif/helpers/string/xhtml_indent.pyreqif/helpers/string.py renamed to reqif/helpers/string/xhtml_indent.py

@@ -0,0 +1,17 @@
+from lxml.etree import fromstring
+
+from reqif.helpers.lxml import (
+    lxml_convert_from_reqif_ns_xhtml_string,
+)
+
+
+def test__01__() -> None:
+    spec_type_string = """\
+<xhtml:div xmlns:xhtml="http://www.w3.org/1999/xhtml">Some<xhtml:span>combination</xhtml:span>of<xhtml:b>tags</xhtml:b></xhtml:div>\
+"""  # noqa: E501
+    expected_spec_type_string = """\
+<div>Some<span>combination</span>of<b>tags</b></div>\
+"""
+    lxml_node = fromstring(spec_type_string)
+    xhtml_string = lxml_convert_from_reqif_ns_xhtml_string(lxml_node=lxml_node)
+    assert xhtml_string == expected_spec_type_string

reqif/parser.py

@@ -0,0 +1,48 @@
+from reqif.helpers.lxml import (
+    lxml_convert_to_reqif_ns_xhtml_string,
+)
+
+
+def test__reqif_convert_to_ns_xhtml_string__1__xhtml_ns() -> None:
+    spec_type_string = """\
+<div>Some<span>combination</span>of<b>tags</b></div>\
+"""
+    expected_spec_type_string = """\
+<xhtml:div>Some<xhtml:span>combination</xhtml:span>of<xhtml:b>tags</xhtml:b></xhtml:div>\
+"""
+
+    ns_xhtml_string = lxml_convert_to_reqif_ns_xhtml_string(
+        spec_type_string, reqif_xhtml=False
+    )
+
+    assert ns_xhtml_string == expected_spec_type_string
+
+
+def test__reqif_convert_to_ns_xhtml_string__2__reqif_xhtml_ns() -> None:
+    spec_type_string = """\
+<div>Some<span>combination</span>of<b>tags</b></div>\
+"""
+    expected_spec_type_string = """\
+<reqif-xhtml:div>Some<reqif-xhtml:span>combination</reqif-xhtml:span>of<reqif-xhtml:b>tags</reqif-xhtml:b></reqif-xhtml:div>\
+"""
+
+    ns_xhtml_string = lxml_convert_to_reqif_ns_xhtml_string(
+        spec_type_string, reqif_xhtml=True
+    )
+
+    assert ns_xhtml_string == expected_spec_type_string
+
+
+def test__reqif_convert_to_ns_xhtml_string__3__string_without_tags() -> None:
+    spec_type_string = """\
+Some combination of words. Multi string.\
+"""
+    expected_spec_type_string = """\
+Some combination of words. Multi string.\
+"""
+
+    ns_xhtml_string = lxml_convert_to_reqif_ns_xhtml_string(
+        spec_type_string, reqif_xhtml=False
+    )
+
+    assert ns_xhtml_string == expected_spec_type_string

tests/unit/reqif/helpers/lxml/__init__.py

@@ -1,4 +1,4 @@
-from reqif.helpers.string import (
+from reqif.helpers.string.xhtml_indent import (
     reqif_indent_xhtml_string,
     reqif_unindent_xhtml_string,
 )

tests/unit/reqif/helpers/test_lxml.py …sts/unit/reqif/helpers/lxml/test_lxml.pytests/unit/reqif/helpers/test_lxml.py renamed to tests/unit/reqif/helpers/lxml/test_lxml.py tests/unit/reqif/helpers/test_lxml.py renamed to tests/unit/reqif/helpers/lxml/test_lxml.py

@@ -1,4 +1,4 @@
-from reqif.helpers.string import reqif_indent_xhtml_string
+from reqif.helpers.string.xhtml_indent import reqif_indent_xhtml_string
 from reqif.parser import ReqIFParser
 from reqif.unparser import ReqIFUnparser