| This include file will give you a function that lets you get data stored in any TTF file. Now you can read the proper fontname directly from Blitz. :) |
; Example for BlitzPlus ; Include "GetTTFdata.bb" ; filename$=RequestFile$( "Select a TTF file","ttf" ) ; Print filename$ ; For x = 0 To 7 ; fontdata$=GetTTFdata$(filename$,x) ; Print x+"="+fontdata$ ; Next ; WaitKey() ; datatypes ; gfiCopyrightNotice = 0 ; Copyright notice ; gfiFontFamilyName = 1 ; Font Family name ; gfiFontSubfamilyName = 2 ; Font Style (usually) ; gfiUniqueFontIdentifier = 3 ; Some unique ID For this font. ; gfiFullFontName = 4 ; combination of strings 1 And 2. Exception: ; gfiVersionString = 5 ; Version String (sometime includes a date). ; gfiPostscriptName = 6 ; Postscript name. ; gfiTrademark = 7 ; Trademark notice/information. Function GetTTFdata$(filename$, datatype=1) ;Create Banks bnkTableOffsets = CreateBank(12) ; Open the file to read from FontName$="" If Lower(Right(Trim(filename$),4))=".ttf" And FileType(filename$)=1 Then fontfile=ReadFile(filename$) ;Read the OffSetTable ReadBytes bnkTableOffsets,fontfile,0,12 For i = 0 To PeekShort(bnkTableOffsets,5)-1 SeekFile(fontfile,12+i*16) word$="" For x = 1 To 4 word$=word$+Chr$(ReadByte(fontfile)) Next If Lower(word$)="name" Then ChkSum=Int_SwapEndian%(ReadInt(fontfile)) Offset=Int_SwapEndian%(ReadInt(fontfile)) Length=Int_SwapEndian%(ReadInt(fontfile)) ElseIf Lower(word$)="cmap" Then cmapChkSum=Int_SwapEndian%(ReadInt(fontfile)) cmapOffset=Int_SwapEndian%(ReadInt(fontfile)) cmapLength=Int_SwapEndian%(ReadInt(fontfile)) SeekFile(fontfile,cmapOffset) TableVersion = Short_SwapEndian%(ReadShort(fontfile)) TableCount = Short_SwapEndian%(ReadShort(fontfile)) PlatformID = Short_SwapEndian%(ReadShort(fontfile)) EncodingID = Short_SwapEndian%(ReadShort(fontfile)) cmapSubOffset=Int_SwapEndian%(ReadInt(fontfile)) SeekFile(fontfile,cmapSubOffset) cmapFormat=Short_SwapEndian%(ReadShort(fontfile)) words$=Words$+"|"+cmapOffset+">"+cmapFormat EndIf words$=Words$+"|"+word$ Next ;now we find the Correct Name Record SeekFile(fontfile,offset) FormatSelector=Short_SwapEndian%(ReadShort(fontfile)) NumberNameRecords=Short_SwapEndian%(ReadShort(fontfile)) StorageOffset= Short_SwapEndian%(ReadShort(fontfile)) ;loop thru NameRecords While (NameID<>datatype) And (Eof(fontfile) <>1) PlatformID = Short_SwapEndian%(ReadShort(fontfile)) EncodingID = Short_SwapEndian%(ReadShort(fontfile)) LanguageID = Short_SwapEndian%(ReadShort(fontfile)) NameID = Short_SwapEndian%(ReadShort(fontfile)) If NameID=datatype ;This is what we're looking for NameLength = Short_SwapEndian%(ReadShort(fontfile)) NameOffset = Short_SwapEndian%(ReadShort(fontfile)) Else junk = ReadInt(fontfile) ;NameLength + NameOffset EndIf Wend ;Now we can get the data SeekFile(fontfile,offset+StorageOffset+NameOffset) For x = 1 To NameLength newChar=ReadByte(fontfile) If newChar<>0 Then fontdata$=fontdata$+ Chr$(newChar) ;filters unicode chr(0)s Next Return fontdata$ CloseFile(fontfile) EndIf FreeBank bnkTableOffsets End Function ; We use these for "ByteSwapping" since TTF format is other endian Function Int_SwapEndian%(n%) Return ((n And $FF) Shl 24) Or ((n And $FF00) Shl 8) Or ((n And $FF0000) Shr 8) Or ((n And $FF000000) Shr 24) End Function Function Short_SwapEndian%(n%) Return ((n And $FF) Shl 8) Or ((n And $FF00) Shr 8) End Function |