Fix fn:unparsed-text* functions and update release info

  - Use codecs.getreader() for checking resource data stream

CHANGELOG.rst

@@ -2,6 +2,12 @@
 CHANGELOG
 *********
 
+`v4.2.1`_ (2024-02-10)
+======================
+* Fix dynamic context initialization with lxml a non-root element (issue #71)
+* Fix XP30+ function fn:function-lookup
+* Fix XP30+ fn:unparsed-text, fn:unparsed-text-lines and fn:unparsed-text-available
+
 `v4.2.0`_ (2024-02-03)
 ======================
 * Drop support for Python 3.7
@@ -443,3 +449,4 @@ CHANGELOG
 .. _v4.1.4: https://github.com/sissaschool/elementpath/compare/v4.1.3...v4.1.4
 .. _v4.1.5: https://github.com/sissaschool/elementpath/compare/v4.1.4...v4.1.5
 .. _v4.2.0: https://github.com/sissaschool/elementpath/compare/v4.1.5...v4.2.0
+.. _v4.2.1: https://github.com/sissaschool/elementpath/compare/v4.2.0...v4.2.1

README.rst

@@ -28,7 +28,7 @@ has it's own implementation of XPath 1.0.
 Installation and usage
 ======================
 
-You can install the package with *pip* in a Python 3.7+ environment::
+You can install the package with *pip* in a Python 3.8+ environment::
 
     pip install elementpath
 

elementpath/xpath30/_xpath30_functions.py

@@ -1268,33 +1268,29 @@ def evaluate_unparsed_text_functions(self, context=None):
         raise self.error('FOUT1190') from None
 
     if context is not None and uri in context.text_resources:
-        obj = context.text_resources[uri]
+        text = context.text_resources[uri]
     else:
         try:
             with urlopen(uri) as rp:
-                obj = rp.read()
-        except (ValueError, URLError) as err:
-            message = str(err)
-            if 'No such file' in message or \
-                    'unknown url type' in message or \
-                    'HTTP Error 404' in message or \
-                    'failure in name resolution' in message:
-                raise self.error('FOUT1170', message) from None
-            raise self.error('FOUT1190') from None
-        else:
-            if context is not None:
-                context.text_resources[uri] = obj
+                stream_reader = codecs.getreader(encoding)(rp)
+                text = stream_reader.read()
+        except URLError as err:
+            raise self.error('FOUT1170', err) from None
+        except ValueError as err:
+            if len(self) > 1:
+                raise self.error('FOUT1190', err) from None
 
-    try:
-        text = codecs.decode(obj, encoding)
-    except UnicodeDecodeError:
-        if len(self) > 1:
-            raise self.error('FOUT1190') from None
+            try:
+                with urlopen(uri) as rp:
+                    stream_reader = codecs.getreader('UTF-16')(rp)
+                    text = stream_reader.read()
+            except URLError as err:
+                raise self.error('FOUT1170', err) from None
+            except ValueError as err:
+                raise self.error('FOUT1190', err) from None
 
-        try:
-            text = codecs.decode(obj, 'UTF-16')
-        except UnicodeDecodeError:
-            raise self.error('FOUT1190') from None
+        if context is not None:
+            context.text_resources[uri] = text
 
     if not all(is_xml_codepoint(ord(s)) for s in text):
         raise self.error('FOUT1190')
@@ -1330,22 +1326,39 @@ def evaluate_unparsed_text_available_function(self, context=None):
 
     try:
         uri = self.get_absolute_uri(href)
+    except ValueError:
+        return False
+
+    try:
         codecs.lookup(encoding)
-        with urlopen(uri) as rp:
-            obj = rp.read()
-    except (ValueError, URLError, LookupError):
+    except LookupError:
         return False
 
     try:
-        return all(is_xml_codepoint(ord(s)) for s in codecs.decode(obj, encoding))
-    except UnicodeDecodeError:
+        with urlopen(uri) as rp:
+            stream_reader = codecs.getreader(encoding)(rp)
+            for line in stream_reader:
+                if any(not is_xml_codepoint(ord(s)) for s in line):
+                    return False
+    except URLError:
+        return False
+    except ValueError:
         if len(self) > 1:
             return False
+    else:
+        return True
 
-        try:
-            return all(is_xml_codepoint(ord(s)) for s in codecs.decode(obj, 'UTF-16'))
-        except UnicodeDecodeError:
-            return False
+    # Fallback auto-detection with utf-16
+    try:
+        with urlopen(uri) as rp:
+            stream_reader = codecs.getreader('UTF-16')(rp)
+            for line in stream_reader:
+                if any(not is_xml_codepoint(ord(s)) for s in line):
+                    return False
+    except (ValueError, URLError):
+        return False
+    else:
+        return True
 
 
 @method(function('environment-variable', nargs=1,

publiccode.yml

@@ -6,8 +6,8 @@ publiccodeYmlVersion: '0.2'
 name: elementpath
 url: 'https://github.com/sissaschool/elementpath'
 landingURL: 'https://github.com/sissaschool/elementpath'
-releaseDate: '2024-02-03'
-softwareVersion: v4.2.0
+releaseDate: '2024-02-10'
+softwareVersion: v4.2.1
 developmentStatus: stable
 platforms:
   - linux

tests/execute_w3c_tests.py

@@ -103,8 +103,6 @@
     # For XP30+
     'fn-root__K-NodeRootFunc-2',  # includes a XPath 3.0 fn:generate-id()
     'fn-codepoints-to-string__cbcl-codepoints-to-string-021',  # Too long ...
-    'fn-unparsed-text__fn-unparsed-text-038',  # Typo in filename
-    'fn-unparsed-text-lines__fn-unparsed-text-lines-038',  # Typo in filename
     'fn-serialize__serialize-xml-015b',  # Do not raise, attribute is good
     'fn-parse-xml-fragment__parse-xml-fragment-022-st',  # conflict with parse-xml-fragment-022
     'fn-for-each-pair__fn-for-each-pair-017',  # Requires PI and comments parsing