/* Copyright (C) 2006-2010 Artifex Software, Inc. All Rights Reserved. This software is provided AS-IS with no warranty, either express or implied. This software is distributed under license and may not be copied, modified or distributed except as expressly authorized under the terms of that license. Refer to licensing information at http://www.artifex.com/ or contact Artifex Software, Inc., 7 Mt. Lassen Drive - Suite A-134, San Rafael, CA 94903, U.S.A., +1(415)492-9861, for further information. */ /* XPS interpreter - truetype font support */ #include "ghostxps.h" #include /* * Some extra TTF parsing magic that isn't covered by the graphics library. */ static inline int u16(const byte *p) { return (p[0] << 8) | p[1]; } static inline int u32(const byte *p) { return (p[0] << 24) | (p[1] << 16) | (p[2] << 8) | p[3]; } static const char *pl_mac_names[258] = { ".notdef", ".null", "nonmarkingreturn", "space", "exclam", "quotedbl", "numbersign", "dollar", "percent", "ampersand", "quotesingle", "parenleft", "parenright", "asterisk", "plus", "comma", "hyphen", "period", "slash", "zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "colon", "semicolon", "less", "equal", "greater", "question", "at", "A", "B", "C", "D", "E", "F", "G", "H", "I", "J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T", "U", "V", "W", "X", "Y", "Z", "bracketleft", "backslash", "bracketright", "asciicircum", "underscore", "grave", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "braceleft", "bar", "braceright", "asciitilde", "Adieresis", "Aring", "Ccedilla", "Eacute", "Ntilde", "Odieresis", "Udieresis", "aacute", "agrave", "acircumflex", "adieresis", "atilde", "aring", "ccedilla", "eacute", "egrave", "ecircumflex", "edieresis", "iacute", "igrave", "icircumflex", "idieresis", "ntilde", "oacute", "ograve", "ocircumflex", "odieresis", "otilde", "uacute", "ugrave", "ucircumflex", "udieresis", "dagger", "degree", "cent", "sterling", "section", "bullet", "paragraph", "germandbls", "registered", "copyright", "trademark", "acute", "dieresis", "notequal", "AE", "Oslash", "infinity", "plusminus", "lessequal", "greaterequal", "yen", "mu", "partialdiff", "summation", "product", "pi", "integral", "ordfeminine", "ordmasculine", "Omega", "ae", "oslash", "questiondown", "exclamdown", "logicalnot", "radical", "florin", "approxequal", "Delta", "guillemotleft", "guillemotright", "ellipsis", "nonbreakingspace", "Agrave", "Atilde", "Otilde", "OE", "oe", "endash", "emdash", "quotedblleft", "quotedblright", "quoteleft", "quoteright", "divide", "lozenge", "ydieresis", "Ydieresis", "fraction", "currency", "guilsinglleft", "guilsinglright", "fi", "fl", "daggerdbl", "periodcentered", "quotesinglbase", "quotedblbase", "perthousand", "Acircumflex", "Ecircumflex", "Aacute", "Edieresis", "Egrave", "Iacute", "Icircumflex", "Idieresis", "Igrave", "Oacute", "Ocircumflex", "apple", "Ograve", "Uacute", "Ucircumflex", "Ugrave", "dotlessi", "circumflex", "tilde", "macron", "breve", "dotaccent", "ring", "cedilla", "hungarumlaut", "ogonek", "caron", "Lslash", "lslash", "Scaron", "scaron", "Zcaron", "zcaron", "brokenbar", "Eth", "eth", "Yacute", "yacute", "Thorn", "thorn", "minus", "multiply", "onesuperior", "twosuperior", "threesuperior", "onehalf", "onequarter", "threequarters", "franc", "Gbreve", "gbreve", "Idotaccent", "Scedilla", "scedilla", "Cacute", "cacute", "Ccaron", "ccaron", "dcroat" }; /* * A bunch of callback functions that the ghostscript * font machinery will call. The most important one * is the build_char function. These are specific to * truetype (loca/glyf) flavored opentypes. */ static unsigned int xps_true_get_glyph_index(gs_font_type42 *pfont42, gs_glyph glyph) { /* identity */ return glyph; } static int xps_true_callback_string_proc(gs_font_type42 *p42, ulong offset, uint length, const byte **pdata) { xps_font_t *font = p42->client_data; if (offset < 0 || offset + length > font->length) { *pdata = NULL; return gs_throw2(-1, "font data access out of bounds (offset=%lu size=%u)", offset, length); } *pdata = font->data + offset; return 0; } static gs_char xps_last_char = GS_NO_CHAR; /* same hack as in PCL */ static gs_glyph xps_true_callback_encode_char(gs_font *pfont, gs_char chr, gs_glyph_space_t spc) { xps_font_t *font = pfont->client_data; int value; xps_last_char = chr; /* save the char we're encoding for the decode_glyph hack */ value = xps_encode_font_char(font, chr); if (value == 0) return gs_no_glyph; return value; } static gs_char xps_true_callback_decode_glyph(gs_font *pfont, gs_glyph glyph, int ch) { /* We should do a reverse cmap lookup here to match PS/PDF. * However, a complete rearchitecture of our text and font processing * would be necessary to match XPS unicode mapping with the * cluster maps. Alas, we cheat similarly to PCL. */ return xps_last_char; } static int xps_true_callback_glyph_name(gs_font *pfont, gs_glyph glyph, gs_const_string *pstr) { /* This function is copied verbatim from plfont.c */ int table_length; int table_offset; ulong format; uint numGlyphs; uint glyph_name_index; const byte *postp; /* post table pointer */ /* guess if the font type is not truetype */ if ( pfont->FontType != ft_TrueType ) { glyph -= 29; if ( glyph >= 0 && glyph < 258 ) { pstr->data = (byte*) pl_mac_names[glyph]; pstr->size = strlen((char*)pstr->data); return 0; } else { return gs_throw1(-1, "glyph index %lu out of range", (ulong)glyph); } } table_offset = xps_find_sfnt_table((xps_font_t*)pfont->client_data, "post", &table_length); /* no post table */ if (table_offset < 0) return gs_throw(-1, "no post table"); /* this shoudn't happen but... */ if ( table_length == 0 ) return gs_throw(-1, "zero-size post table"); ((gs_font_type42 *)pfont)->data.string_proc((gs_font_type42 *)pfont, table_offset, table_length, &postp); format = u32(postp); /* Format 1.0 (mac encoding) is a simple table see the TT spec. * We don't implement this because we don't see it in practice. * Format 2.5 is deprecated. * Format 3.0 means that there is no post data in the font file. * We see this a lot but can't do much about it. * The only format we support is 2.0. */ if ( format != 0x20000 ) { /* Invent a name if we don't know the table format. */ char buf[32]; sprintf(buf, "glyph%d", (int)glyph); pstr->data = (byte*)buf; pstr->size = strlen((char*)pstr->data); return 0; } /* skip over the post header */ numGlyphs = u16(postp + 32); if ( glyph < 0 || glyph > numGlyphs - 1) { return gs_throw1(-1, "glyph index %lu out of range", (ulong)glyph); } /* glyph name index starts at post + 34 each entry is 2 bytes */ glyph_name_index = u16(postp + 34 + (glyph * 2)); /* this shouldn't happen */ if ( glyph_name_index < 0 && glyph_name_index > 0x7fff ) return gs_throw(-1, "post table format error"); /* mac easy */ if ( glyph_name_index < 258 ) { // dprintf2("glyph name (mac) %d = %s\n", glyph, pl_mac_names[glyph_name_index]); pstr->data = (byte*) pl_mac_names[glyph_name_index]; pstr->size = strlen((char*)pstr->data); return 0; } /* not mac */ else { byte *mydata; /* and here's the tricky part */ const byte *pascal_stringp = postp + 34 + (numGlyphs * 2); /* 0 - 257 lives in the mac table above */ glyph_name_index -= 258; /* The string we want is the index'th pascal string, * so we "hop" to each length byte "index" times. */ while (glyph_name_index > 0) { pascal_stringp += ((int)(*pascal_stringp)+1); glyph_name_index--; } /* length byte */ pstr->size = (int)(*pascal_stringp); /* + 1 is for the length byte */ pstr->data = pascal_stringp + 1; /* sanity check */ if ( pstr->data + pstr->size > postp + table_length || pstr->data - 1 < postp) return gs_throw(-1, "data out of range"); /* sigh - we have to allocate a copy of the data - by the * time a high level device makes use of it the font data * may be freed. This is a necessary leak. */ mydata = gs_alloc_bytes(pfont->memory, pstr->size + 1, "glyph to name"); if ( mydata == 0 ) return -1; memcpy(mydata, pascal_stringp + 1, pstr->size); pstr->data = mydata; mydata[pstr->size] = 0; return 0; } } static int xps_true_callback_build_char(gs_show_enum *penum, gs_state *pgs, gs_font *pfont, gs_char chr, gs_glyph glyph) { gs_font_type42 *p42 = (gs_font_type42*)pfont; const gs_rect *pbbox; float sbw[4], w2[6]; gs_fixed_point saved_adjust; int code; // dprintf1("build char ttf %d\n", glyph); code = gs_type42_get_metrics(p42, glyph, sbw); if (code < 0) return code; w2[0] = sbw[2]; w2[1] = sbw[3]; pbbox = &p42->FontBBox; w2[2] = pbbox->p.x; w2[3] = pbbox->p.y; w2[4] = pbbox->q.x; w2[5] = pbbox->q.y; /* Expand the bbox when stroking */ if ( pfont->PaintType ) { float expand = max(1.415, gs_currentmiterlimit(pgs)) * gs_currentlinewidth(pgs) / 2; w2[2] -= expand, w2[3] -= expand; w2[4] += expand, w2[5] += expand; } if ( (code = gs_moveto(pgs, 0.0, 0.0)) < 0 ) return code; if ( (code = gs_setcachedevice(penum, pgs, w2)) < 0 ) return code; code = gs_type42_append(glyph, pgs, gx_current_path(pgs), (gs_text_enum_t*)penum, (gs_font*)p42, gs_show_in_charpath(penum) != cpm_show); if (code < 0) return code; /* Indicate that dropout prevention should be enabled by setting fill_adjust to the special value -1. */ saved_adjust = pgs->fill_adjust; pgs->fill_adjust.x = -1; pgs->fill_adjust.y = -1; code = (pfont->PaintType ? gs_stroke(pgs) : gs_fill(pgs)); if (code < 0) return code; pgs->fill_adjust = saved_adjust; return 0; } /* * Initialize the ghostscript font machinery for a truetype * (type42 in postscript terminology) font. */ int xps_init_truetype_font(xps_context_t *ctx, xps_font_t *font) { font->font = (void*) gs_alloc_struct(ctx->memory, gs_font_type42, &st_gs_font_type42, "xps_font type42"); if (!font->font) return gs_throw(gs_error_VMerror, "out of memory"); /* no shortage of things to initialize */ { gs_font_type42 *p42 = (gs_font_type42*) font->font; /* Common to all fonts: */ p42->next = 0; p42->prev = 0; p42->memory = ctx->memory; p42->dir = ctx->fontdir; /* NB also set by gs_definefont later */ p42->base = font->font; /* NB also set by gs_definefont later */ p42->is_resource = false; gs_notify_init(&p42->notify_list, gs_memory_stable(ctx->memory)); p42->id = gs_next_ids(ctx->memory, 1); p42->client_data = font; /* that's us */ /* this is overwritten in grid_fit() */ gs_make_identity(&p42->FontMatrix); gs_make_identity(&p42->orig_FontMatrix); /* NB ... original or zeroes? */ p42->FontType = ft_TrueType; p42->BitmapWidths = false; p42->ExactSize = fbit_use_outlines; p42->InBetweenSize = fbit_use_outlines; p42->TransformedChar = fbit_use_outlines; p42->WMode = 0; p42->PaintType = 0; p42->StrokeWidth = 0; p42->is_cached = 0; p42->procs.define_font = gs_no_define_font; p42->procs.make_font = gs_no_make_font; p42->procs.font_info = gs_type42_font_info; p42->procs.same_font = gs_default_same_font; p42->procs.encode_char = xps_true_callback_encode_char; p42->procs.decode_glyph = xps_true_callback_decode_glyph; p42->procs.enumerate_glyph = gs_type42_enumerate_glyph; p42->procs.glyph_info = gs_type42_glyph_info; p42->procs.glyph_outline = gs_type42_glyph_outline; p42->procs.glyph_name = xps_true_callback_glyph_name; p42->procs.init_fstack = gs_default_init_fstack; p42->procs.next_char_glyph = gs_default_next_char_glyph; p42->procs.build_char = xps_true_callback_build_char; memset(p42->font_name.chars, 0, sizeof(p42->font_name.chars)); xps_load_sfnt_name(font, (char*)p42->font_name.chars); p42->font_name.size = strlen((char*)p42->font_name.chars); memset(p42->key_name.chars, 0, sizeof(p42->key_name.chars)); strcpy((char*)p42->key_name.chars, (char*)p42->font_name.chars); p42->key_name.size = strlen((char*)p42->key_name.chars); /* Base font specific: */ p42->FontBBox.p.x = 0; p42->FontBBox.p.y = 0; p42->FontBBox.q.x = 0; p42->FontBBox.q.y = 0; uid_set_UniqueID(&p42->UID, p42->id); p42->encoding_index = ENCODING_INDEX_UNKNOWN; p42->nearest_encoding_index = ENCODING_INDEX_ISOLATIN1; p42->FAPI = 0; p42->FAPI_font_data = 0; /* Type 42 specific: */ p42->data.string_proc = xps_true_callback_string_proc; p42->data.proc_data = font; gs_type42_font_init(p42, font->subfontid); p42->data.get_glyph_index = xps_true_get_glyph_index; } gs_definefont(ctx->fontdir, font->font); return 0; }