Merge pull request #1009 from googlefonts/fontra-u-plus

[fontra format] Prepend U+ to code points in .csv, but still allow 'bare' hex values when reading

src/fontra/backends/fontra.py

@@ -110,7 +110,7 @@ def _readGlyphInfo(self):
             for row in reader:
                 glyphName, *rest = row
                 if rest:
-                    codePoints = [int(cp, 16) for cp in rest[0].split(",") if cp]
+                    codePoints = _parseCodePoints(rest[0])
                 else:
                     codePoints = []
                 self.glyphMap[glyphName] = codePoints
@@ -120,7 +120,7 @@ def _writeGlyphInfo(self):
             writer = csv.writer(file, delimiter=";")
             writer.writerow(["glyph name", "code points"])
             for glyphName, codePoints in sorted(self.glyphMap.items()):
-                codePoints = ",".join(f"{cp:04X}" for cp in codePoints)
+                codePoints = ",".join(f"U+{cp:04X}" for cp in codePoints)
                 writer.writerow([glyphName, codePoints])
 
     def _readFontData(self):
@@ -144,6 +144,19 @@ def getGlyphFilePath(self, glyphName):
         return self.glyphsDir / (stringToFileName(glyphName) + ".json")
 
 
+def _parseCodePoints(cell):
+    codePoints = []
+    cell = cell.strip()
+    if cell:
+        for s in cell.split(","):
+            s = s.strip()
+            # U+ should become mandatory, but for now let's be lenient
+            if s.startswith("U+"):
+                s = s[2:]
+            codePoints.append(int(s, 16))
+    return codePoints
+
+
 def serializeGlyph(glyph, glyphName=None):
     glyph = glyph.convertToPaths()
     jsonGlyph = unstructure(glyph)

test-common/fonts/MutatorSans.fontra/glyph-info.csv

@@ -1,55 +1,55 @@
 glyph name;code points
-A;0041,0061
-Aacute;00C1,00E1
-Adieresis;00C4,00E4
-B;0042,0062
-C;0043,0063
-D;0044,0064
-E;0045,0065
-F;0046,0066
-G;0047,0067
-H;0048,0068
-I;0049,0069
+A;U+0041,U+0061
+Aacute;U+00C1,U+00E1
+Adieresis;U+00C4,U+00E4
+B;U+0042,U+0062
+C;U+0043,U+0063
+D;U+0044,U+0064
+E;U+0045,U+0065
+F;U+0046,U+0066
+G;U+0047,U+0067
+H;U+0048,U+0068
+I;U+0049,U+0069
 I.narrow;
 IJ;
-J;004A,006A
+J;U+004A,U+006A
 J.narrow;
-K;004B,006B
-L;004C,006C
-M;004D,006D
-N;004E,006E
-O;004F,006F
-P;0050,0070
-Q;0051,0071
-R;0052,0072
+K;U+004B,U+006B
+L;U+004C,U+006C
+M;U+004D,U+006D
+N;U+004E,U+006E
+O;U+004F,U+006F
+P;U+0050,U+0070
+Q;U+0051,U+0071
+R;U+0052,U+0072
 R.alt;
-S;0053,0073
+S;U+0053,U+0073
 S.closed;
-T;0054,0074
-U;0055,0075
-V;0056,0076
-W;0057,0077
-X;0058,0078
-Y;0059,0079
-Z;005A,007A
-acute;00B4
-arrowdown;2193
-arrowleft;2190
-arrowright;2192
-arrowup;2191
-colon;003A
-comma;002C
-dieresis;00A8
-dot;27D1
+T;U+0054,U+0074
+U;U+0055,U+0075
+V;U+0056,U+0076
+W;U+0057,U+0077
+X;U+0058,U+0078
+Y;U+0059,U+0079
+Z;U+005A,U+007A
+acute;U+00B4
+arrowdown;U+2193
+arrowleft;U+2190
+arrowright;U+2192
+arrowup;U+2191
+colon;U+003A
+comma;U+002C
+dieresis;U+00A8
+dot;U+27D1
 em;
 nestedcomponents;
 nlitest;
-period;002E
-quotedblbase;201E
-quotedblleft;201C
-quotedblright;201D
-quotesinglbase;201A
-semicolon;003B
-space;0020
-varcotest1;E000
-varcotest2;E001
+period;U+002E
+quotedblbase;U+201E
+quotedblleft;U+201C
+quotedblright;U+201D
+quotesinglbase;U+201A
+semicolon;U+003B
+space;U+0020
+varcotest1;U+E000
+varcotest2;U+E001