[WIP] Make Glyphs production names available

glyphs-reader/data/update.py

@@ -17,7 +17,7 @@
 from lxml import etree
 from pathlib import Path
 from textwrap import dedent
-from typing import Iterable, Mapping, Optional, Tuple
+from typing import Iterable, Mapping, MutableSet, Optional, Tuple
 
 
 @dataclass(frozen=True)
@@ -26,6 +26,7 @@ class GlyphInfo:
     name: str
     category: str
     subcategory: Optional[str]
+    production_name: Optional[str]
 
 
 def codename(name: Optional[str]) -> Optional[str]:
@@ -48,6 +49,7 @@ def read_glyph_info(file: str) -> Tuple[GlyphInfo]:
             e.attrib["name"].strip(),
             codename(e.attrib["category"]),
             codename(e.attrib.get("subCategory", None)),
+            e.attrib.get("production", None),
         )
         if info.name not in by_name:
             by_name[info.name] = info
@@ -62,13 +64,16 @@ def read_glyph_info(file: str) -> Tuple[GlyphInfo]:
                 print(f'Ignoring alt name "{alt_name}", already taken')
                 continue
             by_name[alt_name] = dataclasses.replace(
-                by_name[e.attrib["name"]], name=alt_name, codepoint=None
+                by_name[e.attrib["name"]],
+                name=alt_name,
+                codepoint=None,
+                production_name=None,
             )
 
     return tuple(by_name.values())
 
 
-def minimal_names(names_taken: set(), names: Iterable[str]) -> Mapping[str, str]:
+def minimal_names(names_taken: MutableSet, names: Iterable[str]) -> Mapping[str, str]:
     counts = defaultdict(int)
     for name in names:
         counts[name] += 1
@@ -121,6 +126,22 @@ def main():
                 f"Multiple names are assigned 0x{codepoint:04x}, using the first one we saw"
             )
 
+    production_name_to_info_idx = {}
+    blank = 0
+    uprefix = 0
+    uniprefix = 0
+    interesting = 0
+    for i, gi in enumerate(glyph_infos):
+        if gi.production_name is None:
+            blank += 1
+            continue
+        if gi.production_name in production_name_to_info_idx:
+            print(
+                f"Multiple names are assigned {gi.production_name}, using the first one we saw"
+            )
+        else:
+            production_name_to_info_idx[gi.production_name] = i
+
     dest_file = Path(__file__).parent.parent / "src" / "glyphslib_data.rs"
 
     with open(dest_file, "w") as f:
@@ -131,7 +152,7 @@ def main():
         f.write(f"//! {len(glyph_infos)} glyph metadata records taken from glyphsLib\n")
         f.write("\n")
         f.write(
-            "use crate::glyphdata::{qr, q1, q2, q3, Category, QueryResult, Subcategory};\n"
+            "use crate::glyphdata::{qr, qru, qrv, qrc, q1, q1c, q2, q2c, q2u, q2v, q3, q3c, Category, QueryResult, Subcategory};\n"
         )
         f.write("\n")
 
@@ -150,24 +171,44 @@ def main():
 
             # map to shorthand
             if (None, None) == (gi.subcategory, gi.codepoint):
-                entry = f"q1({category})"
+                fn = "q1"
+                args = (category,)
             elif gi.subcategory is None:
                 # codepoint must not be
-                entry = f"q2({category}, 0x{gi.codepoint})"
+                fn = "q2"
+                args = (category, f"0x{gi.codepoint}")
             elif gi.codepoint is None:
                 # subcategory must not be
-                entry = f"q3({category}, {min_subcategories[gi.subcategory]})"
+                fn = "q3"
+                args = (category, min_subcategories[gi.subcategory])
             else:
                 # We must have all the things!
-                entry = f"qr({category}, {min_subcategories[gi.subcategory]}, 0x{gi.codepoint})"
+                fn = "qr"
+                args = (category, min_subcategories[gi.subcategory], f"0x{gi.codepoint}")
+
+            # Do we need to track a production name?
+            # If so modify which shorthand fn we call
+            if gi.production_name is not None:
+                use_custom = True
+                if gi.codepoint is not None:
+                    codepoint = int(gi.codepoint, 16)
+                    # Uni-prefix is most common so given the shortest prefix. See ProductioName enum.
+                    if gi.production_name == f"uni{codepoint:04X}":
+                        use_custom = False
+                        fn += "u"
+                    if gi.production_name == f"u{codepoint:04X}":
+                        use_custom = False
+                        fn += "v"
+                if use_custom:
+                    fn += "c"
+                    args += (f"\"{gi.production_name}\"",)
+
+            entry = fn + "(" + ",".join(args) + ")"
 
             codepoint = "None"
             if gi.codepoint is not None:
                 codepoint = f"Some(0x{gi.codepoint})"
 
-            subcategory = "None"
-            if gi.subcategory is not None:
-                subcategory = f"Some({min_subcategories[gi.subcategory]})"
             fragment = f'("{gi.name}", {entry}),'
             if (len(lines[-1]) + len(fragment)) > 100:
                 lines[-1] += "\n"
@@ -196,6 +237,23 @@ def main():
         f.write("\n")
         f.write("];\n")
 
+        f.write(
+            "// Sorted by production name, has unique production names, therefore safe to bsearch\n"
+        )
+        f.write("pub(crate) const PRODUCTION_NAME_TO_INFO_IDX: &[(&str, usize)] = &[\n")
+        lines = [""]
+        for name, i in sorted(production_name_to_info_idx.items()):
+            fragment = f'("{name}", {i}),'
+            if (len(lines[-1]) + len(fragment)) > 100:
+                lines[-1] += "\n"
+                lines.append("")
+            lines[-1] += fragment
+
+        for line in lines:
+            f.write(line)
+        f.write("\n")
+        f.write("];\n")
+
 
 if __name__ == "__main__":
     main()