Remove specificity from default compare

followthemoney/types/address.py

@@ -42,7 +42,7 @@ def clean_text(
         return collapsed
 
     def _specificity(self, value: str) -> float:
-        return dampen(10, 60, value)
+        return dampen(10, 30, value)
 
     def node_id(self, value: str) -> Optional[str]:
         slug = slugify(value)

followthemoney/types/country.py

@@ -82,5 +82,8 @@ def clean_text(
     def country_hint(self, value: str) -> str:
         return value
 
+    def compare(self, left: str, right: str) -> float:
+        return 1.0 if left == right else 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"iso-3166-1:{value}")

followthemoney/types/email.py

@@ -79,5 +79,10 @@ def clean_text(
     # def country_hint(self, value)
     # TODO: do we want to use TLDs as country evidence?
 
+    def compare(self, left: str, right: str) -> float:
+        if left.lower() == right.lower():
+            return 1.0
+        return 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef("mailto:%s" % value.lower())

followthemoney/types/entity.py

@@ -64,6 +64,9 @@ def clean_text(
             return text
         return None
 
+    def compare(self, left: str, right: str) -> float:
+        return 1.0 if left == right else 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"entity:{value}")
 

followthemoney/types/gender.py

@@ -61,5 +61,8 @@ def clean_text(
             return None
         return code
 
+    def compare(self, left: str, right: str) -> float:
+        return 1.0 if left == right else 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"gender:{value}")

followthemoney/types/iban.py

@@ -47,6 +47,9 @@ def clean_text(
     def country_hint(self, value: str) -> str:
         return value[:2].lower()
 
+    def compare(self, left: str, right: str) -> float:
+        return 1.0 if left == right else 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(self.node_id(value))
 

followthemoney/types/ip.py

@@ -43,5 +43,8 @@ def clean_text(
         except ValueError:
             return None
 
+    def _specificity(self, value: str) -> float:
+        return 1.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"ip:{value}")

followthemoney/types/language.py

@@ -113,5 +113,8 @@ def clean_text(
             return None
         return code
 
+    def _specificity(self, value: str) -> float:
+        return 1.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"iso-639:{value}")

followthemoney/types/mimetype.py

@@ -37,6 +37,9 @@ def clean_text(
             return text
         return None
 
+    def _specificity(self, value: str) -> float:
+        return 1.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"urn:mimetype:{value}")
 

followthemoney/types/string.py

@@ -38,3 +38,6 @@ class HTMLType(StringType):
     label = _("HTML")
     plural = _("HTMLs")
     max_size = 30 * MEGABYTE
+
+    def compare(self, left: str, right: str) -> float:
+        return 0.0

followthemoney/types/topic.py

@@ -70,5 +70,8 @@ class TopicType(EnumType):
     def _locale_names(self, locale: Locale) -> EnumValues:
         return {k: gettext(v) for (k, v) in self._TOPICS.items()}
 
+    def compare(self, left: str, right: str) -> float:
+        return 1.0 if left == right else 0.0
+
     def rdf(self, value: str) -> Identifier:
         return URIRef(f"ftm:topic:{value}")

followthemoney/types/url.py

@@ -54,8 +54,10 @@ def clean_text(
             parsed = parsed._replace(path="/")
         return parsed.geturl()
 
-    def _specificity(self, value: str) -> float:
-        return dampen(10, 120, value)
+    def compare(self, left: str, right: str) -> float:
+        if left.lower() == right.lower():
+            return 1.0
+        return 0.0
 
     def rdf(self, value: str) -> Identifier:
         return URIRef(value)