add UNICODE_CHARACTER_TO_ASCII and CHARACTER_TO_NAME tables

mathics_scanner/characters.py

@@ -165,6 +165,35 @@ def replace_box_unicode_with_ascii(input_string):
     NAMED_CHARACTERS_COLLECTION.get("unicode-to-wl-re", "")
 )
 
+# Unicode to ASCII
+
+CHARACTER_TO_NAME = {
+    char: rf"\[{name}]"
+    for name, char in NAMED_CHARACTERS_COLLECTION["named-characters"].items()
+}
+# TODO: add WL characters to CHARACTER_TO_NAME. For example, "\uF74C" in WMA is named as
+# \[DifferentialD]. Here we are using "\U0001D451" for that name, because is a character
+# we can print with standard fonts. The problem with this approach is that the map
+# would not be invertible anymore. 
+
+
+# This dictionary is used for the default encoding from Unicode/UTF-8 to ASCII
+
+UNICODE_CHARACTER_TO_ASCII = CHARACTER_TO_NAME.copy()
+UNICODE_CHARACTER_TO_ASCII.update(
+    {
+        ch: NAMED_CHARACTERS_COLLECTION["operator-to-ascii"][name]
+        for name, ch in NAMED_CHARACTERS_COLLECTION["operator-to-unicode"].items()
+        if name in NAMED_CHARACTERS_COLLECTION["operator-to-ascii"]
+    }
+)
+# TODO: add WL characters to UNICODE_CHARACTER_TO_ASCII. For example, "\uF74C" in WMA is named as
+# \[DifferentialD]. Here we are using "\U0001D451" for that name, because is a character
+# we can print with standard fonts. For the effects of this table, "\uF74C" should be mapped to
+# something that can be print as an ASCII string (probably, "d").
+
+
+
 
 # Deprecated
 def replace_wl_with_plain_text(wl_input: str, use_unicode=True) -> str:

mathics_scanner/data/named-characters.yml

@@ -813,7 +813,7 @@ CapitalDelta:
 
 CapitalDifferentialD:
   amslatex: '\mathbb{D}'
-  ascii: "d"
+  ascii: "D"
   esc-alias: DD
   has-unicode-inverse: true
   is-letter-like: true
@@ -2131,6 +2131,7 @@ DifferenceDelta:
 # \u2146
 DifferentialD:
   amslatex: '\,d'
+  ascii: "d"
   esc-alias: dd
   has-unicode-inverse: true
   # This can't be letter-like because it is used in derivatives as a function