Merge pull request #6959 from markotoplak/1header-fix

[FIX] 1 line headers: only allow single-letter types and flags
biolab · Dec 20, 2024 · 9fb3969 · 9fb3969
2 parents f70085b + f75a884
commit 9fb3969
Show file tree

Hide file tree

Showing 3 changed files with 42 additions and 11 deletions.
diff --git a/Orange/data/io_base.py b/Orange/data/io_base.py
@@ -169,17 +169,27 @@ def _header1(cls, headers: List[List[str]]) -> Tuple[List, List, List]:
              e.g. d#sex,c#age,cC#IQ
         """
 
-        def is_flag(x):
-            return bool(cls._type_from_flag([x])[0] and
-                        _RE_TYPES.match(cls._type_from_flag([x])[0]) or
-                        Flags.RE_ALL.match(cls._flag_from_flag([x])[0]))
-
-        flags, names = zip(*[i.split(cls.HEADER1_FLAG_SEP, 1)
-                             if cls.HEADER1_FLAG_SEP in i and
-                             is_flag(i.split(cls.HEADER1_FLAG_SEP)[0])
-                             else ('', i)
-                             for i in headers[0]])
-        names = list(names)
+        roles = "".join([f for f in Flags.ALL.values() if len(f) == 1])  # cimw
+        types = "".join([t for t in flatten(getattr(vartype, 'TYPE_HEADERS')
+                                            for vartype in Variable.registry.values())
+                         if len(t) == 1]).upper()  # CNDST
+
+        res = ('^((?P<flags>'
+               f'[{roles}{types}]|'
+               f'([{roles}][{types}])|'
+               f'([{types}][{roles}])'
+               ')#)?(?P<name>.*)')
+
+        header1_re = re.compile(res)
+
+        flags = []
+        names = []
+        for i in headers[0]:
+            m = header1_re.match(i)
+            f, n = m.group("flags", "name")
+            flags.append('' if f is None else f)
+            names.append(n)
+
         return names, cls._type_from_flag(flags), cls._flag_from_flag(flags)
 
     @classmethod

diff --git a/Orange/data/tests/test_io_base.py b/Orange/data/tests/test_io_base.py
@@ -70,6 +70,18 @@ def test_get_header_data_1_flags2(self):
         self.assertListEqual(types, types_)
         self.assertListEqual(flags, flags_)
 
+    def test_get_header_data_1_hashes(self):
+        names, types, flags = _TableHeader.create_header_data(
+            [["Some long text#and here", "vd#Invalid spec", "C#Valid spec",
+              "m#Meta", "cD#Discrete class", "Si#Ignored string"]])
+        names_ = ["Some long text#and here", "vd#Invalid spec", "Valid spec",
+                  "Meta", "Discrete class", "Ignored string"]
+        types_ = ["", "", "c", "", "d", "s"]
+        flags_ = ["", "", "", "m", "c", "i"]
+        self.assertListEqual(names, names_)
+        self.assertListEqual(types, types_)
+        self.assertListEqual(flags, flags_)
+
     def test_get_header_data_3(self):
         names, types, flags = _TableHeader.create_header_data(self.header3[:3])
         self.assertListEqual(names, ["a", "b", "c", "d", "w", "e", "f", "g"])

diff --git a/i18n/si/msgs.jaml b/i18n/si/msgs.jaml
@@ -1155,6 +1155,15 @@ data/io_base.py:
     ^\s*( |{}|)*\s*$: false
     class `_TableHeader`:
         '#': false
+        def `_header1`:
+            TYPE_HEADERS: false
+            ^((?P<flags>: false
+            [{roles}{types}]|: false
+            ([{roles}][{types}])|: false
+            ([{types}][{roles}]): false
+            )#)?(?P<name>.*): false
+            flags: false
+            name: false
     class `_TableBuilder`:
         def `__init__`:
             'Feature ': false