feat(name): corruption update

There was a bug in the corruption name: Any calls in parallel yeilded
non-deterministic results due to a side effect. In refactoring the code
to fix that, I changed it so that we can a new kind of corription.

Yes, we should do something much better.

pynpc/name_corruptor.py

@@ -19,23 +19,96 @@ def __init__(self, patterns: list[str]) -> None:
         self.patterns = patterns
         self.cursor = len(patterns)
 
-    def corrupt_once(self, name: str) -> str:
+    def corrupt(self, name: str, size: int = 3) -> list[str]:
         """Corrupts a name."""
-        starting_cursor = self.cursor
-        cursor = (starting_cursor + 1) % len(self.patterns)
-        while cursor != starting_cursor:
-            # grab a pattern like ("d", "t")
-            (pattern, replacement) = self.patterns[cursor]  # type: ignore[misc]
-
-            # replace it - eg "david".replace("d", "t") => "tavit"
-            new_name = name.replace(pattern, replacement)  # type: ignore[has-type]
-            if new_name != name:
-                # if the name changed, we're done
-                self.cursor = cursor
-                return new_name
-
-            # if not, keep going with the next pattern
-            cursor = (cursor + 1) % len(self.patterns)
-
-        # if we get here, we didn't find any patterns that worked
-        return name
+        names = []
+        start = self.cursor
+        for _ in range(size):
+            starting_cursor = start
+            cursor = (starting_cursor + 1) % len(self.patterns)
+            while True:
+                (pattern, replacement) = self.patterns[cursor]  # type: ignore[misc]
+                new_name = name.replace(pattern, replacement)  # type: ignore[has-type]
+                if new_name != name:
+                    start = cursor
+                    names.append(new_name)
+                    name = new_name
+                    break
+                cursor = (cursor + 1) % len(self.patterns)
+                if cursor == (starting_cursor - 1):
+                    break
+            names.append(name)
+        return sorted(set(names))
+
+
+if __name__ == "__main__":  # pragma: no cover
+    """This is test code.
+
+    As in, that's what we used to see if we were happy with the
+    corruption. It's super academic!
+    """
+    from pathlib import Path
+
+    import orjson
+    from rich import print as lpr
+
+    data = Path(Path(__file__).resolve().parent.parent, "pynpc", "data", "name-corruption-pattern.json")
+    patterns = parse_patterns(orjson.loads(data.read_text()))
+    x = NameCorruptor(patterns)
+
+    names = (
+        "agatha",
+        "aldwin",
+        "althea",
+        "anselm",
+        "armin",
+        "bartholomew",
+        "berengar",
+        "clarice",
+        "constance",
+        "dierk",
+        "eadric",
+        "edward",
+        "eldrida",
+        "elfric",
+        "erna",
+        "eustace",
+        "felicity",
+        "finnegan",
+        "giselle",
+        "gerald",
+        "godric",
+        "gunther",
+        "hadrian",
+        "heloise",
+        "isolde",
+        "ivor",
+        "jocelyn",
+        "lancelot",
+        "lysandra",
+        "magnus",
+        "melisande",
+        "merrick",
+        "osborn",
+        "philomena",
+        "reginald",
+        "rowena",
+        "sabine",
+        "seraphina",
+        "sigfrid",
+        "tiberius",
+        "ulf",
+        "urien",
+        "vespera",
+        "wendel",
+        "wilfred",
+        "winifred",
+        "xenia",
+        "ysabel",
+        "zephyr",
+        "zinnia",
+        "zuriel",
+        "zygmund",
+    )
+    for name in names:
+        lpr(f'("{name}", "{" ".join(x.corrupt(name, 4))}"),')

pynpc/npc.py

@@ -115,18 +115,13 @@ def __init__(self, localisation: list[str] | None = None, what: str = "fantasy")
 
     def _get_name(self, name: str, sz: int = 3) -> str:
         """Get corruptions variations from a name."""
+        rlog.debug("Generating name", name=name, sz=sz)
         try:
-            first, last = name.split(" ")
+            first, _ = name.split(" ")
         except ValueError as e:
             rlog.error("Name is not in the expected format: 'first last'", name=name, error=e)
             first = name
-            last = ""
-        generated = [first]
-        for _ in range(sz):
-            _next = generated[-1]
-            corrupted = self._corruptor.corrupt_once(_next)
-            generated.append(corrupted)
-        return f"{first} (" + " ".join(generated[1:]) + ") " + last
+        return "(" + " ".join(self._corruptor.corrupt(first, sz)) + ") "
 
     def reading(self) -> Reading:
         """Return either upwards or revesed tarot cards draw."""
@@ -141,12 +136,15 @@ def generate(self) -> None:
         self.name_fem = person.full_name(gender=Gender.FEMALE)
         if self.name_fem != transliterate(self.name_fem):
             self.name_fem += " —  " + transliterate(self.name_fem)
+        self.name_fem += f" →  {self._get_name(transliterate(self.name_fem), 3)}"
         self.name_mal = person.full_name(gender=Gender.MALE)
         if self.name_mal != transliterate(self.name_mal):
             self.name_mal += " —  " + transliterate(self.name_mal)
+        self.name_mal += f" →  {self._get_name(transliterate(self.name_mal), 3)}"
         self.name_non = person.full_name()
         if self.name_non != transliterate(self.name_non):
             self.name_non += " —  " + transliterate(self.name_non)
+        self.name_non += f" →  {self._get_name(transliterate(self.name_non), 3)}"
         _arc = self._resources["archetypes"].get_value()
         self.nature = Trait(_arc["name"], _arc["description"])
         self.demeanour = self.nature

tests/test_name_corruptor.py

@@ -26,85 +26,75 @@ def test_parse_multiple_sequences():
     assert parse_patterns(sut) == expected
 
 
-def test_corrupt_once():
-    corruptor = NameCorruptor(parse_patterns([["th", "ff"]]))
-    assert corruptor.corrupt_once("agatha") == "agaffa"
-
-
 @pytest.mark.parametrize(
     ("name", "expected"),
     [
-        ("agatha", "agatha agadda agata agata agata"),
-        ("aldwin", "aldwin altwin altwen altwun altwum"),
-        ("althea", "althea alddea altea altea altea"),
-        ("anselm", "anselm antzelm intzelm aintzelm aiwantzelm"),
-        ("armin", "armin armen armun armum armium"),
-        (
-            "bartholomew",
-            "bartholomew barddolomew bartolomew partolomew fartolomew",
-        ),
-        ("berengar", "berengar baerengar baerungar baerumgar baeriumgar"),
-        ("clarice", "clarice claricke clarike clarijke clarihjke"),
-        (
-            "constance",
-            "constance konstance khonstance khonstince khonstaince",
-        ),
-        ("dierk", "dierk dieyrk tieyrk tiheyrk tyeyrk"),
-        ("eadric", "eadric eadrick eadrik eadrijk eatrijk"),
-        ("edward", "edward edvard etvard echard echart"),
-        ("eldrida", "eldrida eltrita eltrihta eltryta eltryta"),
-        ("elfric", "elfric elfrick elfrik elfrijk elvrijk"),
-        ("erna", "erna erne aerne aerne aerne"),
-        ("eustace", "eustace eustache eusteiche eusteeche eusteeckhe"),
-        ("felicity", "felicity velicity vhelicity vhelihcihty vhelycyty"),
-        ("finnegan", "finnegan vinnegan vhinnegan vhennegan vhunnegan"),
-        ("giselle", "giselle gitzelle gitzella gitzellya gihtzellya"),
-        ("gerald", "gerald gaerald gaeralt gairalt gaidralt"),
-        ("godric", "godric godrick godrik godrijk gotrijk"),
-        ("gunther", "gunther gundder gunter guntaer gumtaer"),
-        ("hadrian", "hadrian hatrian hatrihan hatryan hatryin"),
-        ("heloise", "heloise heloihse heloyse heloize heloite"),
-        ("isolde", "isolde isolte ihsolte ysolte izolte"),
-        ("ivor", "ivor ivhor ihvhor yvhor yvhhor"),
-        ("jocelyn", "jocelyn jocielyn joselyn jotzelyn jotzelyn"),
-        ("lancelot", "lancelot lancielot lanselot lantzelot lantzelod"),
-        ("lysandra", "lysandra lyzandra lytsandra lytzahndra lytzahntra"),
-        ("magnus", "magnus magnaes magnees magnees magnees"),
-        (
-            "melisande",
-            "melisande melizande melitsande melitzahnde melitzahnte",
-        ),
-        ("merrick", "merrick merrickk merrikk merrijkk maerrijkk"),
-        ("osborn", "osborn osporn osforn osvorn osvhorn"),
-        ("philomena", "philomena ffilomena filomena vilomena vhilomena"),
-        ("reginald", "reginald regineld reginelt regenelt regunelt"),
-        ("rowena", "rowena rowene rowune rowume rowiume"),
-        ("sabine", "sabine zabine tsabine tzahbine tzahbene"),
-        ("seraphina", "seraphina seraffina serafina seravina seravhina"),
-        ("sigfrid", "sigfrid sigvrid sigvhrid sigvhrit sihgvhriht"),
-        ("tiberius", "tiberius tibaerius tihbaerihus tybaeryus typaeryus"),
-        ("ulf", "ulf ulv ulvh ulvh ulvh"),
-        ("urien", "urien urieyn uriheyn uryeyn uryeyn"),
-        ("vespera", "vespera vhespera vhespaera vhesfaera vhesfaira"),
-        ("wendel", "wendel wentel wuntel wumtel wiumtel"),
-        ("wilfred", "wilfred wilvred wilvhred wilvhret wihlvhret"),
-        ("winifred", "winifred winivred winivhred winivhret wenivhret"),
-        ("xenia", "xenia xunia xumia xiumia xihumiha"),
-        ("ysabel", "ysabel yzabel ytsabel ytzahbel ytzahpel"),
-        ("zephyr", "zephyr zeffyr zefyr zevyr zevhyr"),
-        ("zinnia", "zinnia zennia zunnia zumnia ziumnia"),
-        ("zuriel", "zuriel zurieyl zuriheyl zuryeyl tzuryeyl"),
-        ("zygmund", "zygmund zygmunt zygmumt zygmiumt zygmihumt"),
+        ("agatha", "agadda agata"),
+        ("aldwin", "altwen altwin altwum altwun"),
+        ("althea", "alddea altea"),
+        ("anselm", "aintzelm aiwantzelm antzelm intzelm"),
+        ("armin", "armen armium armum armun"),
+        ("bartholomew", "barddolomew bartolomew fartolomew partolomew"),
+        ("berengar", "baerengar baeriumgar baerumgar baerungar"),
+        ("clarice", "claricke clarihjke clarijke clarike"),
+        ("constance", "khonstaince khonstance khonstince konstance"),
+        ("dierk", "dieyrk tieyrk tiheyrk tyeyrk"),
+        ("eadric", "eadrick eadrijk eadrik eatrijk"),
+        ("edward", "echard echart edvard etvard"),
+        ("eldrida", "eltrihta eltrita eltryta"),
+        ("elfric", "elfrick elfrijk elfrik elvrijk"),
+        ("erna", "aerne erne"),
+        ("eustace", "eustache eusteeche eusteeckhe eusteiche"),
+        ("felicity", "velicity vhelicity vhelihcihty vhelycyty"),
+        ("finnegan", "vhennegan vhinnegan vhunnegan vinnegan"),
+        ("giselle", "gihtzellya gitzella gitzelle gitzellya"),
+        ("gerald", "gaerald gaeralt gaidralt gairalt"),
+        ("godric", "godrick godrijk godrik gotrijk"),
+        ("gunther", "gumtaer gundder guntaer gunter"),
+        ("hadrian", "hatrian hatrihan hatryan hatryin"),
+        ("heloise", "heloihse heloite heloize heloyse"),
+        ("isolde", "ihsolte isolte izolte ysolte"),
+        ("ivor", "ihvhor ivhor yvhhor yvhor"),
+        ("jocelyn", "jocielyn joselyn jotzelyn"),
+        ("lancelot", "lancielot lanselot lantzelod lantzelot"),
+        ("lysandra", "lytsandra lytzahndra lytzahntra lyzandra"),
+        ("magnus", "magnaes magnees"),
+        ("melisande", "melitsande melitzahnde melitzahnte melizande"),
+        ("merrick", "maerrijkk merrickk merrijkk merrikk"),
+        ("osborn", "osforn osporn osvhorn osvorn"),
+        ("philomena", "ffilomena filomena vhilomena vilomena"),
+        ("reginald", "regenelt regineld reginelt regunelt"),
+        ("rowena", "rowene rowiume rowume rowune"),
+        ("sabine", "tsabine tzahbene tzahbine zabine"),
+        ("seraphina", "seraffina serafina seravhina seravina"),
+        ("sigfrid", "sigvhrid sigvhrit sigvrid sihgvhriht"),
+        ("tiberius", "tibaerius tihbaerihus tybaeryus typaeryus"),
+        ("ulf", "ulv ulvh"),
+        ("urien", "urieyn uriheyn uryeyn"),
+        ("vespera", "vhesfaera vhesfaira vhespaera vhespera"),
+        ("wendel", "wentel wiumtel wumtel wuntel"),
+        ("wilfred", "wihlvhret wilvhred wilvhret wilvred"),
+        ("winifred", "wenivhret winivhred winivhret winivred"),
+        ("xenia", "xihumiha xiumia xumia xunia"),
+        ("ysabel", "ytsabel ytzahbel ytzahpel yzabel"),
+        ("zephyr", "zeffyr zefyr zevhyr zevyr"),
+        ("zinnia", "zennia ziumnia zumnia zunnia"),
+        ("zuriel", "tzuryeyl zurieyl zuriheyl zuryeyl"),
+        ("zygmund", "zygmihumt zygmiumt zygmumt zygmunt"),
     ],
 )
 def test_corrupt_several(name, expected):
     data = Path(Path(__file__).resolve().parent.parent, "pynpc", "data", "name-corruption-pattern.json")
     patterns = parse_patterns(loads(data.read_text()))
 
     corruptor = NameCorruptor(patterns)
-    generated = [name]
-    for _ in range(len(expected.split(" ")) - 1):
-        _next = generated[-1]
-        corrupted = corruptor.corrupt_once(_next)
-        generated.append(corrupted)
-    assert " ".join(generated) == expected
+    generated = corruptor.corrupt(name, 4)
+    assert " ".join(generated) == expected, " ".join(generated)
+
+
+def test_latesha_hanging() -> None:
+    """Test that Latesha is hanging."""
+    data = Path(Path(__file__).resolve().parent.parent, "pynpc", "data", "name-corruption-pattern.json")
+    patterns = parse_patterns(loads(data.read_text()))
+    sut = NameCorruptor(patterns)
+    assert "Lataersha" in sut.corrupt("Latersha")

tests/test_npc.py

@@ -95,7 +95,10 @@ def test_resource_get_values(res) -> None:
 
 
 def test_get_name(random) -> None:
-    assert "fred" in random._get_name("fred")  # noqa: SLF001
+    sut = random._get_name("fred")  # noqa: SLF001
+    assert "vhred" in sut
+    assert "vhret" in sut
+    assert "vred" in sut
 
 
 @pytest.mark.parametrize(