pyx/type1font.py

   1 #!/usr/bin/env python
   2 # -*- coding: ISO-8859-1 -*-
   3 #
   4 #
   5 # Copyright (C) 2005 Jörg Lehmann <joergl@users.sourceforge.net>
   6 # Copyright (C) 2005 André Wobst <wobsta@users.sourceforge.net>
   7 #
   8 # This file is part of PyX (http://pyx.sourceforge.net/).
   9 #
  10 # PyX is free software; you can redistribute it and/or modify
  11 # it under the terms of the GNU General Public License as published by
  12 # the Free Software Foundation; either version 2 of the License, or
  13 # (at your option) any later version.
  14 #
  15 # PyX is distributed in the hope that it will be useful,
  16 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  17 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18 # GNU General Public License for more details.
  19 #
  20 # You should have received a copy of the GNU General Public License
  21 # along with PyX; if not, write to the Free Software
  22 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  23
  24 import re, tempfile, os, binascii
  25 try:
  26     import zlib
  27 except:
  28     pass
  29 import bbox, canvas, pswriter, pdfwriter, t1strip
  30
  31 # _PFB_ASCII = "\200\1"
  32 # _PFB_BIN = "\200\2"
  33 # _PFB_DONE = "\200\3"
  34 # _PFA = "%!"
  35
  36 _StandardEncodingMatch = re.compile(r"\b/Encoding\s+StandardEncoding\s+def\b")
  37 #_FontBBoxMatch = re.compile(r"\bFontBBox\s*\{\s*(-?\d+)\s+(-?\d+)\s+(-?\d+)\s+(-?\d+)\s*\}\s*readonly\s+def\b")
  38 #_ItalicAngle = re.compile(r"\bItalicAngle\s+(-?\d+)\b")
  39
  40
  41 def _pfblength(s):
  42     if len(s) != 4:
  43         raise ValueError("invalid string length")
  44     return (ord(s[0]) +
  45             ord(s[1])*256 +
  46             ord(s[2])*256*256 +
  47             ord(s[3])*256*256*256)
  48
  49 class _tokenfile:
  50     """ ascii file containing tokens separated by spaces.
  51
  52     Comments beginning with % are ignored. Strings containing spaces
  53     are not handled correctly
  54     """
  55
  56     def __init__(self, filename):
  57         self.file = open(filename, "r")
  58         self.line = None
  59
  60     def gettoken(self):
  61         """ return next token or None if EOF """
  62         while not self.line:
  63             line = self.file.readline()
  64             if line == "":
  65                 return None
  66             self.line = line.split("%")[0].split()
  67         token = self.line[0]
  68         self.line = self.line[1:]
  69         return token
  70
  71     def close(self):
  72         self.file.close()
  73
  74
  75 class encoding:
  76
  77     def __init__(self, name, filename):
  78         """ font encoding contained in filename """
  79         self.name = name
  80         self.filename = filename
  81
  82
  83 class encodingfile:
  84
  85     def __init__(self, name, filename):
  86         self.name = name
  87         encfile = _tokenfile(filename)
  88
  89         # name of encoding
  90         self.encname = encfile.gettoken()
  91         token = encfile.gettoken()
  92         if token != "[":
  93             raise RuntimeError("cannot parse encoding file '%s', expecting '[' got '%s'" % (filename, token))
  94         self.encvector = []
  95         for i in range(256):
  96             token = encfile.gettoken()
  97             if token is None or token=="]":
  98                 raise RuntimeError("not enough charcodes in encoding file '%s'" % filename)
  99             self.encvector.append(token)
 100         if encfile.gettoken() != "]":
 101             raise RuntimeError("too many charcodes in encoding file '%s'" % filename)
 102         token = encfile.gettoken()
 103         if token != "def":
 104             raise RuntimeError("cannot parse encoding file '%s', expecting 'def' got '%s'" % (filename, token))
 105         token = encfile.gettoken()
 106         if token != None:
 107             raise RuntimeError("encoding file '%s' too long" % filename)
 108         encfile.close()
 109
 110 #    def decode(self, charcode):
 111 #        return self.encvector[charcode]
 112
 113     def outputPS(self, file, writer, registry):
 114         file.write("%%%%BeginProcSet: %s\n" % self.name)
 115         file.write("/%s\n"
 116                    "[" % self.name)
 117         for i, glyphname in enumerate(self.encvector):
 118             if i and not (i % 8):
 119                 file.write("\n")
 120             else:
 121                 file.write(" ")
 122             file.write(glyphname)
 123         file.write(" ] def\n"
 124                    "%%EndProcSet\n")
 125
 126     def outputPDF(self, file, writer, registry):
 127         file.write("<<\n"
 128                    "/Type /Encoding\n"
 129                    "/Differences\n"
 130                    "[ 0")
 131         for i, glyphname in enumerate(self.encvector):
 132             if i and not (i % 8):
 133                 file.write("\n")
 134             else:
 135                 file.write(" ")
 136             file.write(glyphname)
 137         file.write(" ]\n"
 138                    ">>\n")
 139
 140
 141 class font:
 142
 143     def __init__(self, basefontname, filename, encoding, metric):
 144         self.basefontname = basefontname
 145         self.filename = filename
 146         self.encoding = encoding
 147         self.metric = metric
 148
 149         if encoding is None:
 150             self.name = basefontname
 151         else:
 152             self.name = "%s-%s" % (basefontname, encoding.name)
 153
 154
 155 class fontfile:
 156
 157     # TODO: own stripping using glyph names instead of char codes; completely remove encoding
 158
 159     def __init__(self, fontname, fontfilename, usedchars, encodingfilename):
 160         self.fontname = fontname
 161         self.fontfilename = fontfilename
 162         self.usedchars = [0]*256
 163         for charcode in usedchars.keys():
 164             self.usedchars[charcode] = 1
 165         self.encodingfilename = encodingfilename
 166
 167     def getflags(self):
 168         # As a simple heuristics we assume non-symbolic fonts if and only
 169         # if the Adobe standard encoding is used. All other font flags are
 170         # not specified here.
 171         fontfile = open(self.fontfilename, "rb")
 172         fontdata = fontfile.read()
 173         fontfile.close()
 174         if _StandardEncodingMatch.search(fontdata):
 175             return 32
 176         return 4
 177
 178     def outputPS(self, file, writer, registry):
 179         file.write("%%%%BeginFont: %s\n" % self.fontname)
 180         file.write("%Included char codes:")
 181         for i in range(len(self.usedchars)):
 182             if self.usedchars[i]:
 183                 file.write(" %d" % i)
 184         file.write("\n")
 185         if self.encodingfilename is not None:
 186             t1strip.t1strip(file, self.fontfilename, self.usedchars, self.encodingfilename)
 187         else:
 188             t1strip.t1strip(file, self.fontfilename, self.usedchars)
 189         file.write("%%EndFont\n")
 190
 191     def outputPDF(self, file, writer, registry):
 192         strippedfontfilename = tempfile.mktemp()
 193         strippedfontfile = open(strippedfontfilename, "w")
 194         if self.usedchars:
 195             if self.encodingfilename is not None:
 196                 t1strip.t1strip(strippedfontfile, self.fontfilename, self.usedchars, self.encodingfilename)
 197             else:
 198                 t1strip.t1strip(strippedfontfile, self.fontfilename, self.usedchars)
 199         strippedfontfile.close()
 200         strippedfontfile = open(strippedfontfilename, "r")
 201         fontdata = strippedfontfile.read()
 202         strippedfontfile.close()
 203         os.unlink(strippedfontfilename)
 204
 205         # split the font into its three parts
 206         length1 = fontdata.index("currentfile eexec") + 18
 207         length2 = fontdata.index("0"*20)
 208         fontdata1 = fontdata[:length1]
 209         fontdata2 = fontdata[length1:length2]
 210         fontdata3 = fontdata[length2:]
 211         length3 = len(fontdata3)
 212
 213         # convert to pfa
 214         fontdata2 = binascii.a2b_hex(fontdata2.replace(" ", "").replace("\r", "").replace("\n", "").replace("\t", ""))
 215         length2 = len(fontdata2)
 216
 217         # we might be allowed to skip the third part ...
 218         if (fontdata3.replace("\n", "")
 219                      .replace("\r", "")
 220                      .replace("\t", "")
 221                      .replace(" ", "")) == "0"*512 + "cleartomark":
 222             length3 = 0
 223             fontdata3 = ""
 224
 225         data = fontdata1 + fontdata2 + fontdata3
 226         if writer.compress:
 227             data = zlib.compress(fontdata1 + fontdata2 + fontdata3)
 228
 229         file.write("<<\n"
 230                    "/Length %d\n"
 231                    "/Length1 %d\n"
 232                    "/Length2 %d\n"
 233                    "/Length3 %d\n" % (len(data), length1, length2, length3))
 234         if writer.compress:
 235             file.write("/Filter /FlateDecode\n")
 236         file.write(">>\n"
 237                    "stream\n")
 238         file.write(data)
 239         file.write("\nendstream\n")
 240
 241
 242 class text_pt(canvas.canvasitem):
 243
 244     def __init__(self, x_pt, y_pt, font):
 245         self.font = font
 246         self.x_pt = x_pt
 247         self.y_pt = y_pt
 248         self.width_pt = 0
 249         self.height_pt = 0
 250         self.depth_pt = 0
 251         self.chars = []
 252
 253     def addchar(self, char):
 254         metric = self.font.metric
 255         self.width_pt += metric.getwidth_pt(char)
 256         cheight_pt = metric.getwidth_pt(char)
 257         if cheight_pt > self.height_pt:
 258             self.height_pt = cheight_pt
 259         cdepth_pt = metric.getdepth_pt(char)
 260         if cdepth_pt > self.depth_pt:
 261             self.depth_pt = cdepth_pt
 262         self.chars.append(char)
 263
 264     def bbox(self):
 265         return bbox.bbox_pt(self.x_pt, self.y_pt-self.depth_pt, self.x_pt+self.width_pt, self.y_pt+self.height_pt)
 266
 267     def registerPS(self, registry):
 268         # note that we don't register PSfont as it is just a helper resource
 269         # which registers the needed components
 270         pswriter.PSfont(self.font, self.chars, registry)
 271
 272     def registerPDF(self, registry):
 273         registry.add(pdfwriter.PDFfont(self.font, self.chars, registry))
 274
 275     def outputPS(self, file, writer, context):
 276         if ( context.font is None or
 277              context.font.name != self.font.name or
 278              context.font.metric.getsize_pt() !=  self.font.metric.getsize_pt() ):
 279             file.write("/%s %f selectfont\n" % (self.font.name, self.font.metric.getsize_pt()))
 280             context.font = self.font
 281         outstring = ""
 282         for char in self.chars:
 283             if char > 32 and char < 127 and chr(char) not in "()[]<>\\":
 284                 ascii = "%s" % chr(char)
 285             else:
 286                 ascii = "\\%03o" % char
 287             outstring += ascii
 288         file.write("%g %g moveto (%s) show\n" % (self.x_pt, self.y_pt, outstring))
 289
 290
 291     def outputPDF(self, file, writer, context):
 292         if ( context.font is None or
 293              context.font.name != self.font.name or
 294              context.font.metric.getsize_pt() !=  self.font.metric.getsize_pt() ):
 295             file.write("/%s %f Tf\n" % (self.font.name, self.font.metric.getsize_pt()))
 296             context.font = self.font
 297         outstring = ""
 298         for char in self.chars:
 299             if char > 32 and char < 127 and chr(char) not in "()[]<>\\":
 300                 ascii = "%s" % chr(char)
 301             else:
 302                 ascii = "\\%03o" % char
 303             outstring += ascii
 304         file.write("1 0 0 1 %f %f Tm (%s) Tj\n" % (self.x_pt, self.y_pt, outstring))
 305