diff options
| -rw-r--r-- | FL/Fl_PostScript.H | 20 | ||||
| -rw-r--r-- | src/Fl_PostScript.cxx | 266 |
2 files changed, 236 insertions, 50 deletions
diff --git a/FL/Fl_PostScript.H b/FL/Fl_PostScript.H index 447a8bd1c..45eeef0da 100644 --- a/FL/Fl_PostScript.H +++ b/FL/Fl_PostScript.H @@ -40,9 +40,23 @@ /** \brief PostScript graphical backend. * - PostScript text output is presently implemented only for the latin character set. - FLTK's standard fonts are output using PostScript's standard fonts: Helvetica, Courier, - Times (and their bold, oblique, italic variants), Symbol, ZapfDingbats. + PostScript text uses vectorial fonts for the latin alphabet only (exactly: all unicode + characters between U+0020 and U+017F, that is, ASCII, Latin-1 Supplement and Latin Extended-A charts) + plus a few other punctuation and symbols: + U+0192 (ƒ florin), U+02C6 (ˆ circumflex), U+02C7 (ˇ caron), + U+02D8 (breve), U+02D9 (dotaccent), U+02DA (ring), U+02DB (ogonek), U+02DC (tilde), U+02DD (hungarumlaut), + U+2013 (– endash), U+2014 (— emdash), U+2018 (‘ quoteleft), U+2019 (’ quoteright), + U+201A (‚ quotesinglbase), U+201C (“ quotedblleft), U+201D (” quotedblright), U+0201E („ quotedblbase), + U+2020 († dagger), U+2021 (‡ daggerdbl), U+2022 (• bullet), + U+2026 (… ellipsis), U+2030 (‰ perthousand), U+2039 (‹ guilsinglleft), U+203A (› guilsinglright), + U+2044 (⁄ fraction), U+20AC (€ Euro), U+2122 (™ trademark), + U+2202 (∂ partialdiff), U+2206 (Δ Delta), U+2211 (∑ summation), U+0221A (√ radical), + U+221E (∞ infinity), U+2260 (≠ notequal), U+2264 (≤ lessequal), + U+2265 (≥ greaterequal), + U+25CA (◊ lozenge), U+FB01 (fi), U+FB02 (fl), U+F8FF ( apple symbol, Mac OS X only). + <br> All other unicode characters are output as a bitmap. + <br> FLTK standard fonts are output using PostScript standard fonts: Helvetica, Courier, + Times (and their bold, oblique, italic variants). */ class Fl_PostScript_Graphics_Driver : public Fl_Graphics_Driver { public: diff --git a/src/Fl_PostScript.cxx b/src/Fl_PostScript.cxx index cc088555d..089febd77 100644 --- a/src/Fl_PostScript.cxx +++ b/src/Fl_PostScript.cxx @@ -349,23 +349,87 @@ static const char * prolog_2 = // prolog relevant only if lang_level >1 "end\n" "IDD image GR} bind def\n" -// procedure to modify a font to use ISOLatin1 encoding (iso-8859-1) -// and to keep its name unchanged -"/ToLatin1 { dup findfont dup length dict " -"begin {def} forall /Encoding ISOLatin1Encoding def currentdict end definefont pop } def\n" -// modify all fonts to use ISOLatin1 encoding -"/Helvetica ToLatin1 " -"/Helvetica-Bold ToLatin1 " -"/Helvetica-Oblique ToLatin1 " -"/Helvetica-BoldOblique ToLatin1 \n" -"/Courier ToLatin1 " -"/Courier-Bold ToLatin1 " -"/Courier-Oblique ToLatin1 " -"/Courier-BoldOblique ToLatin1 \n" -"/Times ToLatin1 " -"/Times-Bold ToLatin1 " -"/Times-Italic ToLatin1 " -"/Times-BoldItalic ToLatin1 \n" +// Create a custom PostScript font derived from PostScript standard text fonts +// The encoding of this custom font is as follows: +// 0000-00FF coincides with Unicode, that is to ASCII + Latin-1 +// 0100-017F coincides with Unicode, that is to Latin Extended-A +// 0180-01A6 encodes miscellaneous characters present in PostScript standard text fonts + +// use ISOLatin1Encoding for all text fonts +"/ToISO { dup findfont dup length dict copy begin /Encoding ISOLatin1Encoding def currentdict end definefont pop } def\n" +"/Helvetica ToISO /Helvetica-Bold ToISO /Helvetica-Oblique ToISO /Helvetica-BoldOblique ToISO \n" +"/Courier ToISO /Courier-Bold ToISO /Courier-Oblique ToISO /Courier-BoldOblique ToISO \n" +"/Times-Roman ToISO /Times-Bold ToISO /Times-Italic ToISO /Times-BoldItalic ToISO \n" + +// define LatinExtA, the encoding of Latin-extended-A + some additional characters +// see http://www.adobe.com/devnet/opentype/archives/glyphlist.txt for their names +"/LatinExtA \n" +"[ " +" /Amacron /amacron /Abreve /abreve /Aogonek /aogonek\n" // begin of Latin Extended-A code page +" /Cacute /cacute /Ccircumflex /ccircumflex /Cdotaccent /cdotaccent /Ccaron /ccaron \n" +" /Dcaron /dcaron /Dcroat /dcroat\n" +" /Emacron /emacron /Ebreve /ebreve /Edotaccent /edotaccent /Eogonek /eogonek /Ecaron /ecaron\n" +" /Gcircumflex /gcircumflex /Gbreve /gbreve /Gdotaccent /gdotaccent /Gcommaaccent /gcommaaccent \n" +" /Hcircumflex /hcircumflex /Hbar /hbar \n" +" /Itilde /itilde /Imacron /imacron /Ibreve /ibreve /Iogonek /iogonek /Idotaccent /dotlessi \n" +" /IJ /ij /Jcircumflex /jcircumflex\n" +" /Kcommaaccent /kcommaaccent /kgreenlandic \n" +" /Lacute /lacute /Lcommaaccent /lcommaaccent /Lcaron /lcaron /Ldotaccent /ldotaccent /Lslash /lslash \n" +" /Nacute /nacute /Ncommaaccent /ncommaaccent /Ncaron /ncaron /napostrophe /Eng /eng \n" +" /Omacron /omacron /Obreve /obreve /Ohungarumlaut /ohungarumlaut /OE /oe \n" +" /Racute /racute /Rcommaaccent /rcommaaccent /Rcaron /rcaron \n" +" /Sacute /sacute /Scircumflex /scircumflex /Scedilla /scedilla /Scaron /scaron \n" +" /Tcommaaccent /tcommaaccent /Tcaron /tcaron /Tbar /tbar \n" +" /Utilde /utilde /Umacron /umacron /Ubreve /ubreve /Uring /uring /Uhungarumlaut /uhungarumlaut /Uogonek /uogonek \n" +" /Wcircumflex /wcircumflex /Ycircumflex /ycircumflex /Ydieresis \n" +" /Zacute /zacute /Zdotaccent /zdotaccent /Zcaron /zcaron \n" +" /longs \n" // end of Latin Extended-A code page +" /florin /circumflex /caron /breve /dotaccent /ring \n" // remaining characters from PostScript standard text fonts +" /ogonek /tilde /hungarumlaut /endash /emdash \n" +" /quoteleft /quoteright /quotesinglbase /quotedblleft /quotedblright \n" +" /quotedblbase /dagger /daggerdbl /bullet /ellipsis \n" +" /perthousand /guilsinglleft /guilsinglright /fraction /Euro \n" +" /trademark /partialdiff /Delta /summation /radical \n" +" /infinity /notequal /lessequal /greaterequal /lozenge \n" +" /fi /fl /apple \n" +" ] def \n" +// deal with alternative PostScript names of some characters +" /mycharstrings /Helvetica findfont /CharStrings get def\n" +" /PSname2 { dup mycharstrings exch known {LatinExtA 3 -1 roll 3 -1 roll put}{pop pop} ifelse } def \n" +" 16#20 /Gdot PSname2 16#21 /gdot PSname2 16#30 /Idot PSname2 16#3F /Ldot PSname2 16#40 /ldot PSname2 16#7F /slong PSname2 \n" + +// proc that gives LatinExtA encoding to a font +"/ToLatinExtA { findfont dup length dict copy begin /Encoding LatinExtA def currentdict end definefont pop } def\n" +// create Ext-versions of standard fonts that use LatinExtA encoding \n" +"/HelveticaExt /Helvetica ToLatinExtA \n" +"/Helvetica-BoldExt /Helvetica-Bold ToLatinExtA /Helvetica-ObliqueExt /Helvetica-Oblique ToLatinExtA \n" +"/Helvetica-BoldObliqueExt /Helvetica-BoldOblique ToLatinExtA \n" +"/CourierExt /Courier ToLatinExtA /Courier-BoldExt /Courier-Bold ToLatinExtA \n" +"/Courier-ObliqueExt /Courier-Oblique ToLatinExtA /Courier-BoldObliqueExt /Courier-BoldOblique ToLatinExtA \n" +"/Times-RomanExt /Times-Roman ToLatinExtA /Times-BoldExt /Times-Bold ToLatinExtA \n" +"/Times-ItalicExt /Times-Italic ToLatinExtA /Times-BoldItalicExt /Times-BoldItalic ToLatinExtA \n" + +// proc to create a Type 0 font with 2-byte encoding +// that merges a text font with ISO encoding + same font with LatinExtA encoding +"/To2byte { 6 dict begin /FontType 0 def \n" +"/FDepVector 3 1 roll findfont exch findfont 2 array astore def \n" +"/FontMatrix [1 0 0 1 0 0] def /FMapType 6 def /Encoding [ 0 1 0 ] def\n" +// 100: Hexa count of ISO array; A7: hexa count of LatinExtA array +"/SubsVector < 01 0100 00A7 > def\n" +"currentdict end definefont pop } def\n" +// create Type 0 versions of standard fonts +"/Helvetica2B /HelveticaExt /Helvetica To2byte \n" +"/Helvetica-Bold2B /Helvetica-BoldExt /Helvetica-Bold To2byte \n" +"/Helvetica-Oblique2B /Helvetica-ObliqueExt /Helvetica-Oblique To2byte \n" +"/Helvetica-BoldOblique2B /Helvetica-BoldObliqueExt /Helvetica-BoldOblique To2byte \n" +"/Courier2B /CourierExt /Courier To2byte \n" +"/Courier-Bold2B /Courier-BoldExt /Courier-Bold To2byte \n" +"/Courier-Oblique2B /Courier-ObliqueExt /Courier-Oblique To2byte \n" +"/Courier-BoldOblique2B /Courier-BoldObliqueExt /Courier-BoldOblique To2byte \n" +"/Times-Roman2B /Times-RomanExt /Times-Roman To2byte \n" +"/Times-Bold2B /Times-BoldExt /Times-Bold To2byte \n" +"/Times-Italic2B /Times-ItalicExt /Times-Italic To2byte \n" +"/Times-BoldItalic2B /Times-BoldItalicExt /Times-BoldItalic To2byte \n" ; static const char * prolog_2_pixmap = // prolog relevant only if lang_level == 2 for pixmaps/masked color images @@ -875,21 +939,21 @@ void Fl_PostScript_Graphics_Driver::line_style(int style, int width, char* dashe }; static const char *_fontNames[] = { -"Helvetica", -"Helvetica-Bold", -"Helvetica-Oblique", -"Helvetica-BoldOblique", -"Courier", -"Courier-Bold", -"Courier-Oblique", -"Courier-BoldOblique", -"Times", -"Times-Bold", -"Times-Italic", -"Times-BoldItalic", +"Helvetica2B", +"Helvetica-Bold2B", +"Helvetica-Oblique2B", +"Helvetica-BoldOblique2B", +"Courier2B", +"Courier-Bold2B", +"Courier-Oblique2B", +"Courier-BoldOblique2B", +"Times-Roman2B", +"Times-Bold2B", +"Times-Italic2B", +"Times-BoldItalic2B", "Symbol", -"Courier", -"Courier-Bold", +"Courier2B", +"Courier-Bold2B", "ZapfDingbats" }; @@ -929,28 +993,136 @@ void Fl_PostScript_Graphics_Driver::draw(int angle, const char *str, int n, int fprintf(output, "GR\n"); } -// outputs in PostScript a UTF8 string replacing non-Latin1 characters by ? -// and using the same width in points as on display + +// computes the mask for the RGB image img of all pixels with color != bg +static uchar *calc_mask(uchar *img, int w, int h, Fl_Color bg) +{ + uchar red, green, blue, r, g, b; + uchar bit, byte, *q; + Fl::get_color(bg, red, green, blue); + int W = (w+7)/8; // width of mask + uchar* mask = new uchar[W * h]; + q = mask; + while (h-- > 0) { // for each row + bit = 0x80; // byte with last bit set + byte = 0; // next mask byte to compute + for (int j = 0; j < w; j++) { // for each column + r = *img++; // the pixel color components + g = *img++; + b = *img++; + // if pixel doesn't have bg color, put it in mask + if (r != red || g != green || b != blue) byte |= bit; + bit = bit>>1; // shift bit one step to the right + if (bit == 0) { // single set bit has fallen out + *q++ = byte; // enter byte in mask + byte = 0; // reset next mask byte to zero + bit = 0x80; // and this byte + } + } + if (bit != 0x80) *q++ = byte; // enter last columns' byte in mask + } + return mask; +} + +// write to PostScript a bitmap image of a UTF8 string +static void transformed_draw_extra(const char* str, int n, double x, double y, int w, FILE *output) { + const float scale = 3; // scale for bitmask computation + Fl_Fontsize old_size = fl_size(); + fl_font(fl_font(), (Fl_Fontsize)(scale * old_size) ); + w = (int)(w *scale + 0.5); + int h = fl_height(); + // create an offscreen image of the string + Fl_Color text_color = fl_color(); + Fl_Color bg_color = fl_contrast(FL_WHITE, text_color); + Fl_Offscreen off = fl_create_offscreen(w+2, (int)(h+3*scale) ); + fl_begin_offscreen(off); + fl_color(bg_color); + // color offscreen background with a shade contrasting with the text color + fl_rectf(0, 0, w+2, (int)(h+3*scale) ); + fl_color(text_color); + fl_draw(str, n, 1, (int)(h * 0.8) ); // draw string in offscreen + // read (most of) the offscreen image + uchar *img = fl_read_image(NULL, 1, 1, w, h, 0); + fl_end_offscreen(); + fl_font(fl_font(), old_size); + fl_delete_offscreen(off); + // compute the mask of what is not the background + uchar *mask = calc_mask(img, w, h, bg_color); + delete img; + // write the string image to PostScript as a scaled bitmask + fprintf(output, "%g %g %g %g %d %d MI\n", x, y - h*0.77/scale, w/scale, h/scale, w, h); + uchar *di; + int wmask = (w+7)/8; + for (int j = h - 1; j >= 0; j--){ + di = mask + j * wmask; + for (int i = 0; i < wmask; i++){ + //if (!(i%80)) fprintf(output, "\n"); // don't have lines longer than 255 chars + fprintf(output, "%2.2x", *di ); + di++; + } + fprintf(output,"\n"); + } + fprintf(output,">\n"); + delete mask; +} + +static int is_in_table(unsigned utf) { + unsigned i; + static unsigned extra_table_roman[] = { // unicodes/*names*/ of other characters from PostScript standard fonts + 0x192/*florin*/, 0x2C6/*circumflex*/, 0x2C7/*caron*/, + 0x2D8/*breve*/, 0x2D9/*dotaccent*/, 0x2DA/*ring*/, 0x2DB/*ogonek*/, 0x2DC/*tilde*/, 0x2DD/*hungarumlaut*/, + 0x2013/*endash*/, 0x2014/*emdash*/, 0x2018/*quoteleft*/, 0x2019/*quoteright*/, + 0x201A/*quotesinglbase*/, 0x201C/*quotedblleft*/, 0x201D/*quotedblright*/, 0x201E/*quotedblbase*/, + 0x2020/*dagger*/, 0x2021/*daggerdbl*/, 0x2022/*bullet*/, + 0x2026/*ellipsis*/, 0x2030/*perthousand*/, 0x2039/*guilsinglleft*/, 0x203A/*guilsinglright*/, + 0x2044/*fraction*/, 0x20AC/*Euro*/, 0x2122/*trademark*/, + 0x2202/*partialdiff*/, 0x2206/*Delta*/, 0x2211/*summation*/, 0x221A/*radical*/, + 0x221E/*infinity*/, 0x2260/*notequal*/, 0x2264/*lessequal*/, + 0x2265/*greaterequal*/, + 0x25CA/*lozenge*/, 0xFB01/*fi*/, 0xFB02/*fl*/, + 0xF8FF/*apple*/ + }; + for ( i = 0; i < sizeof(extra_table_roman)/sizeof(int); i++) { + if (extra_table_roman[i] == utf) return i + 0x180; + } + return 0; +} + +// outputs in PostScript a UTF8 string using the same width in points as on display void Fl_PostScript_Graphics_Driver::transformed_draw(const char* str, int n, double x, double y) { - int len; + int len, code; if (!n || !str || !*str) return; - const char *last = str + n; // compute display width of string - fprintf(output,"%g (", fl_width(str, n)); - while (str < last) { + int width = (int)fl_width(str, n); + if (width == 0) return; + fprintf(output, "%d <", width); + // transforms UTF8 encoding to our custom PostScript encoding as follows: + // extract each unicode character + // if unicode <= 0x17F, unicode and PostScript codes are identical + // if unicode is one of the values listed in extra_table_roman above + // its PostScript code is 0x180 + the character's rank in extra_table_roman + // if unicode is something else, draw all string as bitmap image + + const char *last = str + n; + const char *str2 = str; + while (str2 < last) { // Extract each unicode character of string. - // Until 0xFF, UTF codes coincide with iso-Latin1 (iso-8859-1) - unsigned utf = fl_utf8decode(str, last, &len); - str += len; - if (utf > 0xFF) { - utf = '?'; // replace non Latin-1 unicodes by ? + unsigned utf = fl_utf8decode(str2, last, &len); + str2 += len; + if (utf <= 0x17F) { // until Latin Extended-A + ; } - else if (utf == '(' || utf == ')' || utf == '\\') { - putc('\\' , output); // these chars need be escaped - } - putc(utf, output); // output the latin character + else if ( (code = is_in_table(utf)) != 0) { // other handled characters + utf = code; + } + else { // unhandled character: draw all string as bitmap image + fprintf(output, "> pop pop\n"); // close and ignore the opened hex string + transformed_draw_extra(str, n, x, y, width, output); + return; + } + fprintf(output, "%4.4X", utf); } - fprintf(output, ") %g %g show_pos_width\n", x, y); + fprintf(output, "> %g %g show_pos_width\n", x, y); } struct matrix {double a, b, c, d, x, y;}; |