[PDFBOX-5328] Failing to get multiple encodings from cmap table

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.8.16, 2.0.24
• Fix Version/s: 2.0.25, 3.0.0 PDFBox
• Component/s: FontBox
• Assignee: Tilman Hausherr
• Reporter: Tilman Hausherr
• Created: 20/Nov/21 12:06
• Updated: 17/Dec/21 06:50
• Resolved: 20/Nov/21 14:56

Description

As reported by Ty Lewis in the users mailing list, see here

Unicode encodings for GID 8712: List(U+f967)
Unicode encodings for GID 8712 from table (platformId = 0 encodingId = 3):
List(U+4e0d, U+f967)
Unicode encodings for GID 8712 from table (platformId = 0 encodingId = 4):
List(U+f967)

I made some java code to reproduce this:

File fontFile = new File("NotoSansSC-Regular.otf");
OTFParser otfParser = new OTFParser(false);
OpenTypeFont otf = otfParser.parse(fontFile);

CmapLookup unicodeCmapLookup = otf.getUnicodeCmapLookup();
List<Integer> charCodes = unicodeCmapLookup.getCharCodes(8712);
System.out.println(charCodes);

CmapTable cmapTable = otf.getCmap();
CmapSubtable unicodeFullCmapTable = cmapTable.getSubtable(CmapTable.PLATFORM_UNICODE, CmapTable.ENCODING_UNICODE_2_0_FULL);

CmapSubtable unicodeBmpCmapTable = cmapTable.getSubtable(CmapTable.PLATFORM_UNICODE, CmapTable.ENCODING_UNICODE_2_0_BMP);

List<Integer> unicodeBmpCharCodes = unicodeBmpCmapTable.getCharCodes(8712);
List<Integer> unicodeFullCharCodes = unicodeFullCmapTable.getCharCodes(8712);

System.out.println(unicodeBmpCharCodes);
System.out.println(unicodeFullCharCodes);

A look in the tables with DTL OTMaster 3.7 light shows there are indeed two entries. A search for them (in hex) shows the characters 不 and 不.