chore: import upstream snapshot with attribution

2026-07-13 13:25:10 +08:00
commit c397331b1e
3684 changed files with 990993 additions and 0 deletions
@@ -0,0 +1,57 @@
+import pynini
+from fun_text_processing.text_normalization.de.taggers.decimal import quantities
+from fun_text_processing.text_normalization.en.graph_utils import (
+    DAMO_NOT_QUOTE,
+    GraphFst,
+    delete_preserve_order,
+    insert_space,
+)
+from pynini.lib import pynutil
+
+
+class DecimalFst(GraphFst):
+    """
+    Finite state transducer for classifying decimal, e.g.
+        decimal { negative: "true" integer_part: "elf"  fractional_part: "vier null sechs" quantity: "billionen" } -> minus elf komma vier null sechs billionen
+        decimal { integer_part: "eins" quantity: "billion" } -> eins billion
+
+    """
+
+    def __init__(self, deterministic: bool = True):
+        super().__init__(name="decimal", kind="classify", deterministic=deterministic)
+
+        delete_space = pynutil.delete(" ")
+        self.optional_sign = pynini.closure(
+            pynini.cross('negative: "true"', "minus ") + delete_space, 0, 1
+        )
+        self.integer = (
+            pynutil.delete('integer_part: "')
+            + pynini.closure(DAMO_NOT_QUOTE, 1)
+            + pynutil.delete('"')
+        )
+        self.fractional_default = (
+            pynutil.delete('fractional_part: "')
+            + pynini.closure(DAMO_NOT_QUOTE, 1)
+            + pynutil.delete('"')
+        )
+
+        self.fractional = pynutil.insert(" komma ") + self.fractional_default
+
+        self.quantity = (
+            delete_space
+            + insert_space
+            + pynutil.delete('quantity: "')
+            + quantities
+            + pynutil.delete('"')
+        )
+        self.optional_quantity = pynini.closure(self.quantity, 0, 1)
+
+        graph = self.optional_sign + (
+            self.integer + self.quantity
+            | self.integer + delete_space + self.fractional + self.optional_quantity
+        )
+
+        self.numbers = graph
+        graph += delete_preserve_order
+        delete_tokens = self.delete_tokens(graph)
+        self.fst = delete_tokens.optimize()