pyx/pdfwriter.py

   1 # -*- encoding: utf-8 -*-
   2 #
   3 #
   4 # Copyright (C) 2005-2011 Jörg Lehmann <joergl@users.sourceforge.net>
   5 # Copyright (C) 2007 Michael Schindler <m-schindler@users.sourceforge.net>
   6 # Copyright (C) 2005-2011 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., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA
  23
  24 import io, copy, logging, time
  25 logger = logging.getLogger("pyx")
  26 try:
  27     import zlib
  28     haszlib = True
  29 except:
  30     haszlib = False
  31
  32 from . import bbox, config, style, unit, version, trafo, writer
  33
  34
  35
  36 class PDFregistry:
  37
  38     def __init__(self):
  39         self.types = {}
  40         # we want to keep the order of the resources
  41         self.objects = []
  42         self.resources = {}
  43         self.procsets = {"PDF": 1}
  44         self.merged = None
  45
  46     def add(self, object):
  47         """ register object, merging it with an already registered object of the same type and id """
  48         sameobjects = self.types.setdefault(object.type, {})
  49         if object.id in sameobjects:
  50             sameobjects[object.id].merge(object)
  51         else:
  52             self.objects.append(object)
  53             sameobjects[object.id] = object
  54
  55     def getrefno(self, object):
  56         if self.merged:
  57             return self.merged.getrefno(object)
  58         else:
  59             return self.types[object.type][object.id].refno
  60
  61     def mergeregistry(self, registry):
  62         for object in registry.objects:
  63             self.add(object)
  64         registry.merged = self
  65
  66     def write(self, file, writer, catalog):
  67         # first we set all refnos
  68         refno = 1
  69         for object in self.objects:
  70             object.refno = refno
  71             refno += 1
  72
  73         # second, all objects are written, keeping the positions in the output file
  74         fileposes = []
  75         for object in self.objects:
  76             fileposes.append(file.tell())
  77             file.write("%i 0 obj\n" % object.refno)
  78             object.write(file, writer, self)
  79             file.write("endobj\n")
  80
  81         # xref
  82         xrefpos = file.tell()
  83         file.write("xref\n"
  84                    "0 %d\n"
  85                    "0000000000 65535 f \n" % refno)
  86
  87         for filepos in fileposes:
  88             file.write("%010i 00000 n \n" % filepos)
  89
  90         # trailer
  91         file.write("trailer\n"
  92                    "<<\n"
  93                    "/Size %i\n" % refno)
  94         file.write("/Root %i 0 R\n" % self.getrefno(catalog))
  95         file.write("/Info %i 0 R\n" % self.getrefno(catalog.PDFinfo))
  96         file.write(">>\n"
  97                    "startxref\n"
  98                    "%i\n" % xrefpos)
  99         file.write("%%EOF\n")
 100
 101     def addresource(self, resourcetype, resourcename, object, procset=None):
 102         self.resources.setdefault(resourcetype, {})[resourcename] = object
 103         if procset:
 104             self.procsets[procset] = 1
 105
 106     def writeresources(self, file):
 107         file.write("<<\n")
 108         file.write("/ProcSet [ %s ]\n" % " ".join(["/%s" % p for p in list(self.procsets.keys())]))
 109         if self.resources:
 110             for resourcetype, resources in list(self.resources.items()):
 111                 file.write("/%s <<\n%s\n>>\n" % (resourcetype, "\n".join(["/%s %i 0 R" % (name, self.getrefno(object))
 112                                                                           for name, object in list(resources.items())])))
 113         file.write(">>\n")
 114
 115
 116 class PDFobject:
 117
 118     def __init__(self, type, _id=None):
 119         """create a PDFobject
 120           - type has to be a string describing the type of the object
 121           - _id is a unique identification used for the object if it is not None.
 122             Otherwise id(self) is used
 123         """
 124         self.type = type
 125         if _id is None:
 126             self.id = id(self)
 127         else:
 128             self.id = _id
 129
 130     def merge(self, other):
 131         pass
 132
 133     def write(self, file, writer, registry):
 134         raise NotImplementedError("write method has to be provided by PDFobject subclass")
 135
 136
 137 class PDFcatalog(PDFobject):
 138
 139     def __init__(self, document, writer, registry):
 140         PDFobject.__init__(self, "catalog")
 141         self.PDFform = PDFform(writer, registry)
 142         registry.add(self.PDFform)
 143         self.PDFpages = PDFpages(document, writer, registry)
 144         registry.add(self.PDFpages)
 145         self.PDFinfo = PDFinfo()
 146         registry.add(self.PDFinfo)
 147
 148     def write(self, file, writer, registry):
 149         file.write("<<\n"
 150                    "/Type /Catalog\n"
 151                    "/Pages %i 0 R\n" % registry.getrefno(self.PDFpages))
 152         if not self.PDFform.empty():
 153             file.write("/AcroForm %i 0 R\n" % registry.getrefno(self.PDFform))
 154         if writer.fullscreen:
 155             file.write("/PageMode /FullScreen\n")
 156         file.write(">>\n")
 157
 158
 159 class PDFinfo(PDFobject):
 160
 161     def __init__(self):
 162         PDFobject.__init__(self, "info")
 163
 164     def write(self, file, writer, registry):
 165         if time.timezone < 0:
 166             # divmod on positive numbers, otherwise the minutes have a different sign from the hours
 167             timezone = "-%02i'%02i'" % divmod(-time.timezone/60, 60)
 168         elif time.timezone > 0:
 169             timezone = "+%02i'%02i'" % divmod(time.timezone/60, 60)
 170         else:
 171             timezone = "Z00'00'"
 172
 173         def pdfstring(s):
 174             r = ""
 175             for c in s:
 176                 if 32 <= ord(c) <= 127 and c not in "()[]<>\\":
 177                     r += c
 178                 else:
 179                     r += "\\%03o" % ord(c)
 180             return r
 181
 182         file.write("<<\n")
 183         if writer.title:
 184             file.write("/Title (%s)\n" % pdfstring(writer.title))
 185         if writer.author:
 186             file.write("/Author (%s)\n" % pdfstring(writer.author))
 187         if writer.subject:
 188             file.write("/Subject (%s)\n" % pdfstring(writer.subject))
 189         if writer.keywords:
 190             file.write("/Keywords (%s)\n" % pdfstring(writer.keywords))
 191         file.write("/Creator (PyX %s)\n" % version.version)
 192         file.write("/CreationDate (D:%s%s)\n" % (time.strftime("%Y%m%d%H%M"), timezone))
 193         file.write(">>\n")
 194
 195
 196 class PDFpages(PDFobject):
 197
 198     def __init__(self, document, writer, registry):
 199         PDFobject.__init__(self, "pages")
 200         self.PDFpagelist = []
 201         for pageno, page in enumerate(document.pages):
 202             page = PDFpage(page, pageno, self, writer, registry)
 203             registry.add(page)
 204             self.PDFpagelist.append(page)
 205
 206     def write(self, file, writer, registry):
 207         file.write("<<\n"
 208                    "/Type /Pages\n"
 209                    "/Kids [%s]\n"
 210                    "/Count %i\n"
 211                    ">>\n" % (" ".join(["%i 0 R" % registry.getrefno(page)
 212                                        for page in self.PDFpagelist]),
 213                              len(self.PDFpagelist)))
 214
 215
 216 class PDFpage(PDFobject):
 217
 218     def __init__(self, page, pageno, PDFpages, writer, registry):
 219         PDFobject.__init__(self, "page")
 220         self.PDFpages = PDFpages
 221         self.page = page
 222
 223         # every page uses its own registry in order to find out which
 224         # resources are used within the page. However, the
 225         # pageregistry is also merged in the global registry
 226         self.pageregistry = PDFregistry()
 227         self.pageregistry.add(self)
 228
 229         self.PDFannotations = PDFannotations()
 230         self.pageregistry.add(self.PDFannotations)
 231         # we eventually need the form dictionary to append formfields
 232         for object in registry.objects:
 233             if object.type == "form":
 234                 self.pageregistry.add(object)
 235
 236         self.PDFcontent = PDFcontent(page, writer, self.pageregistry)
 237         self.pageregistry.add(self.PDFcontent)
 238         registry.mergeregistry(self.pageregistry)
 239
 240     def write(self, file, writer, registry):
 241         file.write("<<\n"
 242                    "/Type /Page\n"
 243                    "/Parent %i 0 R\n" % registry.getrefno(self.PDFpages))
 244         paperformat = self.page.paperformat
 245         if paperformat:
 246             file.write("/MediaBox [0 0 %f %f]\n" % (unit.topt(paperformat.width), unit.topt(paperformat.height)))
 247         else:
 248             file.write("/MediaBox [%f %f %f %f]\n" % self.PDFcontent.bbox.highrestuple_pt())
 249         if self.PDFcontent.bbox and writer.writebbox:
 250             file.write("/CropBox [%f %f %f %f]\n" % self.PDFcontent.bbox.highrestuple_pt())
 251         if self.page.rotated:
 252             file.write("/Rotate 90\n")
 253         if not self.PDFannotations.empty():
 254             file.write("/Annots %i 0 R\n" % registry.getrefno(self.PDFannotations))
 255         file.write("/Contents %i 0 R\n" % registry.getrefno(self.PDFcontent))
 256         file.write("/Resources ")
 257         self.pageregistry.writeresources(file)
 258         file.write(">>\n")
 259
 260
 261 class PDFcontent(PDFobject):
 262
 263     def __init__(self, page, awriter, registry):
 264         PDFobject.__init__(self, registry, "content")
 265         contentfile = writer.writer(io.BytesIO())
 266         self.bbox = bbox.empty()
 267         acontext = context()
 268         page.processPDF(contentfile, awriter, acontext, registry, self.bbox)
 269         self.content = contentfile.file.getvalue()
 270
 271     def write(self, file, awriter, registry):
 272         if awriter.compress:
 273             content = zlib.compress(self.content)
 274         else:
 275             content = self.content
 276         file.write("<<\n"
 277                    "/Length %i\n" % len(content))
 278         if awriter.compress:
 279             file.write("/Filter /FlateDecode\n")
 280         file.write(">>\n"
 281                    "stream\n")
 282         file.write_bytes(content)
 283         file.write("endstream\n")
 284
 285
 286 class PDFwriter:
 287
 288     def __init__(self, document, file,
 289                        title=None, author=None, subject=None, keywords=None,
 290                        fullscreen=False, writebbox=False, compress=True, compresslevel=6,
 291                        strip_fonts=True, text_as_path=False, mesh_as_bitmap=False, mesh_as_bitmap_resolution=300):
 292         self._fontmap = None
 293
 294         self.title = title
 295         self.author = author
 296         self.subject = subject
 297         self.keywords = keywords
 298         self.fullscreen = fullscreen
 299         self.writebbox = writebbox
 300         if compress and not haszlib:
 301             compress = 0
 302             logger.warning("PDFwriter: compression disabled due to missing zlib module")
 303         self.compress = compress
 304         self.compresslevel = compresslevel
 305         self.strip_fonts = strip_fonts
 306         self.text_as_path = text_as_path
 307         self.mesh_as_bitmap = mesh_as_bitmap
 308         self.mesh_as_bitmap_resolution = mesh_as_bitmap_resolution
 309
 310         # dictionary mapping font names to dictionaries mapping encoding names to encodings
 311         # encodings themselves are mappings from glyphnames to codepoints
 312         self.encodings = {}
 313
 314         # the PDFcatalog class automatically builds up the pdfobjects from a document
 315         registry = PDFregistry()
 316         catalog = PDFcatalog(document, self, registry)
 317         registry.add(catalog)
 318
 319         file = writer.writer(file)
 320         file.write_bytes(b"%PDF-1.4\n%\xc3\xb6\xc3\xa9\n")
 321         registry.write(file, self, catalog)
 322
 323     def getfontmap(self):
 324         if self._fontmap is None:
 325             # late import due to cyclic dependency
 326             from pyx.dvi import mapfile
 327             fontmapfiles = config.getlist("text", "pdffontmaps", ["pdftex.map"])
 328             self._fontmap = mapfile.readfontmap(fontmapfiles)
 329         return self._fontmap
 330
 331
 332 class PDFannotations(PDFobject):
 333
 334     def __init__(self):
 335         PDFobject.__init__(self, "annotations")
 336         self.annots = []
 337
 338     def append(self, item):
 339         if item not in self.annots:
 340             self.annots.append(item)
 341
 342     def empty(self):
 343         return len(self.annots) == 0
 344
 345     def write(self, file, writer, registry):
 346         # XXX problem: This object will be written to the file even if it is useless (empty)
 347         file.write("[ %s ]\n" % " ".join(["%d 0 R" % registry.getrefno(annot) for annot in self.annots]))
 348
 349
 350 class PDFform(PDFobject):
 351
 352     def __init__(self, writer, registry):
 353         PDFobject.__init__(self, "form")
 354         self.fields = []
 355
 356     def merge(self, other):
 357         for field in other.fields:
 358             self.append(field)
 359
 360     def append(self, field):
 361         if field not in self.fields:
 362             self.fields.append(field)
 363
 364     def empty(self):
 365         return len(self.fields) == 0
 366
 367     def write(self, file, writer, registry):
 368         # XXX problem: This object will be written to the file even if it is useless (empty)
 369         file.write("<<")
 370         file.write("/Fields [")
 371         for field in self.fields:
 372             file.write(" %d 0 R" % registry.getrefno(field))
 373         file.write(" ]\n")
 374         file.write(">>\n")
 375
 376
 377 class context:
 378
 379     def __init__(self):
 380         self.linewidth_pt = None
 381         # XXX there are both stroke and fill color spaces
 382         self.colorspace = None
 383         self.strokeattr = 1
 384         self.fillattr = 1
 385         self.selectedfont = None
 386         self.trafo = trafo.trafo()
 387         self.fillstyles = []
 388         self.fillrule = 0
 389
 390     def __call__(self, **kwargs):
 391         newcontext = copy.copy(self)
 392         for key, value in list(kwargs.items()):
 393             setattr(newcontext, key, value)
 394         return newcontext
 395