diff --git a/src/calibre/ebooks/epub/from_html.py b/src/calibre/ebooks/epub/from_html.py index 822346453a..318cf5cc02 100644 --- a/src/calibre/ebooks/epub/from_html.py +++ b/src/calibre/ebooks/epub/from_html.py @@ -8,24 +8,24 @@ Conversion of HTML/OPF files follows several stages: * All links in the HTML files or in the OPF manifest are followed to build up a list of HTML files to be converted. - This stage is implemented by + This stage is implemented by :function:`calibre.ebooks.html.traverse` and :class:`calibre.ebooks.html.HTMLFile`. - * The HTML is pre-processed to make it more semantic. + * The HTML is pre-processed to make it more semantic. All links in the HTML files to other resources like images, - stylesheets, etc. are relativized. The resources are copied + stylesheets, etc. are relativized. The resources are copied into the `resources` sub directory. This is accomplished by - :class:`calibre.ebooks.html.PreProcessor` and + :class:`calibre.ebooks.html.PreProcessor` and :class:`calibre.ebooks.html.Parser`. * The HTML is processed. Various operations are performed. - All style declarations are extracted and consolidated into + All style declarations are extracted and consolidated into a single style sheet. Chapters are auto-detected and marked. Various font related manipulations are performed. See :class:`HTMLProcessor`. - * The processed HTML is saved and the + * The processed HTML is saved and the :module:`calibre.ebooks.epub.split` module is used to split up large HTML files into smaller chunks. @@ -64,7 +64,7 @@ def remove_bad_link(element, attribute, link, pos): def check_links(opf_path, pretty_print): ''' - Find and remove all invalid links in the HTML files + Find and remove all invalid links in the HTML files ''' logger = logging.getLogger('html2epub') logger.info('\tChecking files for bad links...') @@ -78,7 +78,7 @@ def check_links(opf_path, pretty_print): if isinstance(f, str): f = f.decode('utf-8') html_files.append(os.path.abspath(content(f))) - + for path in html_files: if not os.access(path, os.R_OK): continue @@ -113,27 +113,27 @@ def find_html_index(files): return html_files[-1], os.path.splitext(html_files[-1])[1].lower()[1:] class HTMLProcessor(Processor, Rationalizer): - + def __init__(self, htmlfile, opts, tdir, resource_map, htmlfiles, stylesheets): - Processor.__init__(self, htmlfile, opts, tdir, resource_map, htmlfiles, + Processor.__init__(self, htmlfile, opts, tdir, resource_map, htmlfiles, name='html2epub') if opts.verbose > 2: self.debug_tree('parsed') self.detect_chapters() - + self.extract_css(stylesheets) if self.opts.base_font_size2 > 0: - self.font_css = self.rationalize(self.external_stylesheets+[self.stylesheet], + self.font_css = self.rationalize(self.external_stylesheets+[self.stylesheet], self.root, self.opts) if opts.verbose > 2: self.debug_tree('nocss') - + if hasattr(self.body, 'xpath'): for script in list(self.body.xpath('descendant::script')): script.getparent().remove(script) - + self.fix_markup() - + def convert_image(self, img): rpath = img.get('src', '') path = os.path.join(os.path.dirname(self.save_path()), *rpath.split('/')) @@ -150,10 +150,10 @@ class HTMLProcessor(Processor, Rationalizer): if val == rpath: self.resource_map[key] = rpath+'_calibre_converted.jpg' img.set('src', rpath+'_calibre_converted.jpg') - + def fix_markup(self): ''' - Perform various markup transforms to get the output to render correctly + Perform various markup transforms to get the output to render correctly in the quirky ADE. ''' # Replace
that are children of as ADE doesn't handle them @@ -179,8 +179,8 @@ class HTMLProcessor(Processor, Rationalizer): if not br.tail: br.tail = '' br.tail += sibling.tail - - + + if self.opts.profile.remove_object_tags: for tag in self.root.xpath('//embed'): tag.getparent().remove(tag) @@ -188,42 +188,46 @@ class HTMLProcessor(Processor, Rationalizer): if tag.get('type', '').lower().strip() in ('image/svg+xml',): continue tag.getparent().remove(tag) - - + + for tag in self.root.xpath('//title|//style'): if not tag.text: tag.getparent().remove(tag) for tag in self.root.xpath('//script'): if not tag.text and not tag.get('src', False): tag.getparent().remove(tag) - + for tag in self.root.xpath('//form'): tag.getparent().remove(tag) - + for tag in self.root.xpath('//center'): tag.tag = 'div' tag.set('style', 'text-align:center') - + if self.opts.linearize_tables: for tag in self.root.xpath('//table | //tr | //th | //td'): tag.tag = 'div' - - + + # ADE can't handle & in an img url + for tag in self.root.xpath('//img[@src]'): + tag.set('src', tag.get('src', '').replace('&', '')) + + def save(self): for meta in list(self.root.xpath('//meta')): meta.getparent().remove(meta) # Strip all comments since Adobe DE is petrified of them Processor.save(self, strip_comments=True) - + def remove_first_image(self): images = self.root.xpath('//img') if images: images[0].getparent().remove(images[0]) return True return False - - - + + + def config(defaults=None): return common_config(defaults=defaults) @@ -235,7 +239,7 @@ def option_parser(): Convert a HTML file to an EPUB ebook. Recursively follows links in the HTML file. If you specify an OPF file instead of an HTML file, the list of links is takes from -the element of the OPF file. +the element of the OPF file. ''')) def parse_content(filelist, opts, tdir): @@ -246,7 +250,7 @@ def parse_content(filelist, opts, tdir): first_image_removed = False for htmlfile in filelist: logging.getLogger('html2epub').debug('Processing %s...'%htmlfile) - hp = HTMLProcessor(htmlfile, opts, os.path.join(tdir, 'content'), + hp = HTMLProcessor(htmlfile, opts, os.path.join(tdir, 'content'), resource_map, filelist, stylesheets) if not first_image_removed and opts.remove_first_image: first_image_removed = hp.remove_first_image() @@ -254,7 +258,7 @@ def parse_content(filelist, opts, tdir): hp.save() stylesheet_map[os.path.basename(hp.save_path())] = \ [s for s in hp.external_stylesheets + [hp.stylesheet, hp.font_css, hp.override_css] if s is not None] - + logging.getLogger('html2epub').debug('Saving stylesheets...') if opts.base_font_size2 > 0: Rationalizer.remove_font_size_information(stylesheets.values()) @@ -268,7 +272,7 @@ def parse_content(filelist, opts, tdir): if toc.count('chapter') + toc.count('file') > opts.toc_threshold: toc.purge(['link', 'unknown']) toc.purge(['link'], max=opts.max_toc_links) - + return resource_map, hp.htmlfile_map, toc, stylesheet_map TITLEPAGE = '''\ @@ -325,26 +329,26 @@ def process_title_page(mi, filelist, htmlfilemap, opts, tdir): metadata_cover = mi.cover if metadata_cover and not os.path.exists(metadata_cover): metadata_cover = None - + cpath = '/'.join(('resources', '_cover_.jpg')) cover_dest = os.path.join(tdir, 'content', *cpath.split('/')) if metadata_cover is not None: - if not create_cover_image(metadata_cover, cover_dest, + if not create_cover_image(metadata_cover, cover_dest, opts.profile.screen_size): metadata_cover = None specified_cover = opts.cover if specified_cover and not os.path.exists(specified_cover): specified_cover = None if specified_cover is not None: - if not create_cover_image(specified_cover, cover_dest, + if not create_cover_image(specified_cover, cover_dest, opts.profile.screen_size): specified_cover = None - + cover = metadata_cover if specified_cover is None or (opts.prefer_metadata_cover and metadata_cover is not None) else specified_cover if cover is not None: titlepage = TITLEPAGE%cpath - tp = 'calibre_title_page.html' if old_title_page is None else old_title_page + tp = 'calibre_title_page.html' if old_title_page is None else old_title_page tppath = os.path.join(tdir, 'content', tp) with open(tppath, 'wb') as f: f.write(titlepage) @@ -370,7 +374,7 @@ def condense_ncx(ncx_path): compressed = etree.tostring(tree.getroot(), encoding='utf-8') open(ncx_path, 'wb').write(compressed) -def convert(htmlfile, opts, notification=None, create_epub=True, +def convert(htmlfile, opts, notification=None, create_epub=True, oeb_cover=False, extract_to=None): htmlfile = os.path.abspath(htmlfile) if opts.output is None: @@ -399,16 +403,16 @@ def convert(htmlfile, opts, notification=None, create_epub=True, else: opf, filelist = get_filelist(htmlfile, opts) mi = merge_metadata(htmlfile, opf, opts) - opts.chapter = XPath(opts.chapter, + opts.chapter = XPath(opts.chapter, namespaces={'re':'http://exslt.org/regular-expressions'}) for x in (1, 2, 3): attr = 'level%d_toc'%x if getattr(opts, attr): - setattr(opts, attr, XPath(getattr(opts, attr), + setattr(opts, attr, XPath(getattr(opts, attr), namespaces={'re':'http://exslt.org/regular-expressions'})) else: - setattr(opts, attr, None) - + setattr(opts, attr, None) + with TemporaryDirectory(suffix='_html2epub', keep=opts.keep_intermediate) as tdir: if opts.keep_intermediate: print 'Intermediate files in', tdir @@ -416,16 +420,16 @@ def convert(htmlfile, opts, notification=None, create_epub=True, parse_content(filelist, opts, tdir) logger = logging.getLogger('html2epub') resources = [os.path.join(tdir, 'content', f) for f in resource_map.values()] - - + + title_page, has_title_page = process_title_page(mi, filelist, htmlfile_map, opts, tdir) spine = [htmlfile_map[f.path] for f in filelist] if not oeb_cover and title_page is not None: spine = [title_page] + spine mi.cover = None mi.cover_data = (None, None) - - + + mi = create_metadata(tdir, mi, spine, resources) buf = cStringIO.StringIO() if mi.toc: @@ -453,7 +457,7 @@ def convert(htmlfile, opts, notification=None, create_epub=True, logger.info('\tBuilding page map...') add_page_map(opf_path, opts) check_links(opf_path, opts.pretty_print) - + opf = OPF(opf_path, tdir) opf.remove_guide() oeb_cover_file = None @@ -465,7 +469,7 @@ def convert(htmlfile, opts, notification=None, create_epub=True, opf.add_guide_item('cover', 'Cover', 'content/'+spine[0]) if oeb_cover and oeb_cover_file: opf.add_guide_item('cover', 'Cover', 'content/'+oeb_cover_file) - + cpath = os.path.join(tdir, 'content', 'resources', '_cover_.jpg') if os.path.exists(cpath): opf.add_path_to_manifest(cpath, 'image/jpeg') @@ -477,29 +481,29 @@ def convert(htmlfile, opts, notification=None, create_epub=True, condense_ncx(ncx_path) if os.stat(ncx_path).st_size > opts.profile.flow_size: logger.warn('NCX still larger than allowed size at %d bytes. Menu based Table of Contents may not work on device.'%os.stat(ncx_path).st_size) - + if create_epub: epub = initialize_container(opts.output) epub.add_dir(tdir) epub.close() run_plugins_on_postprocess(opts.output, 'epub') logger.info(_('Output written to ')+opts.output) - + if opts.show_opf: print open(opf_path, 'rb').read() - + if opts.extract_to is not None: if os.path.exists(opts.extract_to): shutil.rmtree(opts.extract_to) shutil.copytree(tdir, opts.extract_to) - + if extract_to is not None: if os.path.exists(extract_to): shutil.rmtree(extract_to) shutil.copytree(tdir, extract_to) - - - + + + def main(args=sys.argv): parser = option_parser() opts, args = parser.parse_args(args) @@ -509,6 +513,6 @@ def main(args=sys.argv): return 1 convert(args[1], opts) return 0 - + if __name__ == '__main__': sys.exit(main()) diff --git a/src/calibre/ebooks/lrf/objects.py b/src/calibre/ebooks/lrf/objects.py index ff9edc4e47..0693b9a508 100644 --- a/src/calibre/ebooks/lrf/objects.py +++ b/src/calibre/ebooks/lrf/objects.py @@ -15,13 +15,13 @@ ruby_tags = { } class LRFObject(object): - + tag_map = { 0xF500: ['', ''], 0xF502: ['infoLink', 'D'], 0xF501: ['', ''], } - + @classmethod def descramble_buffer(cls, buf, l, xorKey): i = 0 @@ -35,7 +35,7 @@ class LRFObject(object): @classmethod def parse_empdots(self, tag, f): self.refEmpDotsFont, self.empDotsFontName, self.empDotsCode = tag.contents - + @staticmethod def tag_to_val(h, obj, tag, stream): val = None @@ -45,7 +45,7 @@ class LRFObject(object): val = tag.word elif h[1] == 'w': val = tag.word - if val > 0x8000: + if val > 0x8000: val -= 0x10000 elif h[1] == 'B': val = tag.byte @@ -56,44 +56,44 @@ class LRFObject(object): if len(h) > 2: val = h[2](val) if callable(h[2]) else h[2][val] return val - + def __init__(self, document, stream, id, scramble_key, boundary): self._scramble_key = scramble_key self._document = document self.id = id - + while stream.tell() < boundary: tag = Tag(stream) self.handle_tag(tag, stream) - + def parse_bg_image(self, tag, f): self.bg_image_mode, self.bg_image_id = struct.unpack(" 0x400 and (isinstance(self, ImageStream) or isinstance(self, Font) or isinstance(self, SoundStream)): - l = 0x400; + l = 0x400 self.stream = self.descramble_buffer(self.stream, l, key) if self.stream_flags & 0x100 !=0: decomp_size = struct.unpack("\n' return s - + def as_dict(self): d = {} for h in self.tag_map.values(): @@ -232,7 +236,7 @@ class PageAttr(StyleObject, LRFObject): 0xF529: ['', "parse_bg_image"], } tag_map.update(LRFObject.tag_map) - + @classmethod def to_css(cls, obj, inline=False): return '' @@ -241,79 +245,79 @@ class PageAttr(StyleObject, LRFObject): class Color(object): def __init__(self, val): self.a, self.r, self.g, self.b = val & 0xFF, (val>>8)&0xFF, (val>>16)&0xFF, (val>>24)&0xFF - + def __unicode__(self): return u'0x%02x%02x%02x%02x'%(self.a, self.r, self.g, self.b) - + def __str__(self): return unicode(self) - - def __len__(self): + + def __len__(self): return 4 - + def __getitem__(self, i): # Qt compatible ordering and values return (self.r, self.g, self.b, 0xff-self.a)[i] # In Qt 0xff is opaque while in LRS 0x00 is opaque - + def to_html(self): return 'rgb(%d, %d, %d)'%(self.r, self.g, self.b) - + class EmptyPageElement(object): def __iter__(self): for i in range(0): yield i - + def __str__(self): return unicode(self) class PageDiv(EmptyPageElement): - + def __init__(self, pain, spacesize, linewidth, linecolor): self.pain, self.spacesize, self.linewidth = pain, spacesize, linewidth self.linecolor = Color(linecolor) - + def __unicode__(self): return u'\n\n'%\ (self.pain, self.spacesize, self.linewidth, self.color) - - + + class RuledLine(EmptyPageElement): - + linetype_map = {0x00: 'none', 0x10: 'solid', 0x20: 'dashed', 0x30: 'double', 0x40: 'dotted', 0x13: 'unknown13'} - + def __init__(self, linelength, linetype, linewidth, linecolor): self.linelength, self.linewidth = linelength, linewidth self.linetype = self.linetype_map[linetype] self.linecolor = Color(linecolor) self.id = -1 - + def __unicode__(self): return u'\n\n'%\ (self.linelength, self.linetype, self.linewidth, self.linecolor) - + class Wait(EmptyPageElement): - + def __init__(self, time): self.time = time - + def __unicode__(self): return u'\n\n'%(self.time) - + class Locate(EmptyPageElement): - + pos_map = {1:'bottomleft', 2:'bottomright',3:'topright',4:'topleft', 5:'base'} - + def __init__(self, pos): self.pos = self.pos_map[pos] - + def __unicode__(self): return u'\n\n'%(self.pos) - + class BlockSpace(EmptyPageElement): - + def __init__(self, xspace, yspace): self.xspace, self.yspace = xspace, yspace - + def __unicode__(self): return u'\n\n'%\ (self.xspace, self.yspace) @@ -332,7 +336,7 @@ class Page(LRFStream): evenfooter = property(fget=lambda self : self._document.objects[self.style.evenfooterid]) oddheader = property(fget=lambda self : self._document.objects[self.style.oddheaderid]) oddfooter = property(fget=lambda self : self._document.objects[self.style.oddfooterid]) - + class Content(LRFContentObject): tag_map = { 0xF503: 'link', @@ -344,44 +348,44 @@ class Page(LRFStream): 0xF5D4: 'wait', 0xF5D6: 'sound_stop', } - + def __init__(self, bytes, objects): self.in_blockspace = False LRFContentObject.__init__(self, bytes, objects) - + def link(self, tag): self.close_blockspace() self._contents.append(self.objects[tag.dword]) - + def page_div(self, tag): self.close_blockspace() pars = struct.unpack("\n'%(self.style_id, self.id) for i in self: s += unicode(i) s += '\n\n' return s - + def __str__(self): return unicode(self) - + def to_html(self): s = u'' for i in self: s += i.to_html() return s - - + + class BlockAttr(StyleObject, LRFObject): tag_map = { @@ -443,7 +447,7 @@ class BlockAttr(StyleObject, LRFObject): 0xF529: ['', 'parse_bg_image'], } tag_map.update(LRFObject.tag_map) - + @classmethod def to_css(cls, obj, inline=False): ans = '' @@ -451,7 +455,7 @@ class BlockAttr(StyleObject, LRFObject): ans += '' if inline else '\t' ans += line ans += ' ' if inline else '\n' - + if hasattr(obj, 'sidemargin'): margin = str(obj.sidemargin) + 'px' item('margin-left: %(m)s; margin-right: %(m)s;'%dict(m=margin)) @@ -465,13 +469,13 @@ class BlockAttr(StyleObject, LRFObject): item('border-color: %s;'%obj.framecolor.to_html()) if hasattr(obj, 'bgcolor') and obj.bgcolor.a < 255: item('background-color: %s;'%obj.bgcolor.to_html()) - + return ans - - + + class TextCSS(object): - + @classmethod def to_css(cls, obj, inline=False): ans = '' @@ -479,7 +483,7 @@ class TextCSS(object): ans += '' if inline else '\t' ans += line ans += ' ' if inline else '\n' - + fs = getattr(obj, 'fontsize', None) if fs is not None: item('font-size: %fpt;'%(int(fs)/10.)) @@ -508,18 +512,18 @@ class TextCSS(object): pi = getattr(obj, 'parindent', None) if pi is not None: item('text-indent: %fpt;'%(int(pi)/10.)) - + return ans - - + + class TextAttr(StyleObject, LRFObject, TextCSS): - + FONT_MAP = collections.defaultdict(lambda : 'serif') for key, value in PRS500_PROFILE.default_fonts.items(): FONT_MAP[value] = key - + tag_map = { 0xF511: ['fontsize', 'w'], 0xF512: ['fontwidth', 'w'], @@ -544,8 +548,8 @@ class TextAttr(StyleObject, LRFObject, TextCSS): } tag_map.update(ruby_tags) tag_map.update(LRFObject.tag_map) - - + + class Block(LRFStream, TextCSS): tag_map = { @@ -556,10 +560,10 @@ class Block(LRFStream, TextCSS): tag_map.update(LRFStream.tag_map) extra_attrs = [i[0] for i in BlockAttr.tag_map.values()] extra_attrs.extend([i[0] for i in TextAttr.tag_map.values()]) - + style = property(fget=lambda self : self._document.objects[self.style_id]) textstyle = property(fget=lambda self : self._document.objects[self.textstyle_id]) - + def initialize(self): self.attrs = {} stream = cStringIO.StringIO(self.stream) @@ -572,7 +576,7 @@ class Block(LRFStream, TextCSS): self.textstyle_id = obj.style_id elif isinstance(obj, Text): self.name = 'TextBlock' - self.textstyle_id = obj.style_id + self.textstyle_id = obj.style_id elif isinstance(obj, Image): self.name = 'ImageBlock' for attr in ('x0', 'x1', 'y0', 'y1', 'xsize', 'ysize', 'refstream'): @@ -582,14 +586,14 @@ class Block(LRFStream, TextCSS): self.name = 'ButtonBlock' else: raise LRFParseError("Unexpected block type: "+obj.__class__.__name__) - + self.content = obj - - + + for attr in self.extra_attrs: if hasattr(self, attr): self.attrs[attr] = getattr(self, attr) - + def __unicode__(self): s = u'\n<%s objid="%d" blockstyle="%d" '%(self.name, self.id, self.style_id) if hasattr(self, 'textstyle_id'): @@ -602,12 +606,12 @@ class Block(LRFStream, TextCSS): s += '\n'%(self.name,) return s return s.rstrip() + ' />\n' - + def to_html(self): if self.name == 'TextBlock': return u'
%s
'%(self.style_id, self.textstyle_id, self.content.to_html()) return u'' - + class MiniPage(LRFStream): tag_map = { @@ -625,12 +629,12 @@ class Text(LRFStream): } tag_map.update(TextAttr.tag_map) tag_map.update(LRFStream.tag_map) - + style = property(fget=lambda self : self._document.objects[self.style_id]) - + text_map = { 0x22: u'"', 0x26: u'&', 0x27: u'\'', 0x3c: u'<', 0x3e: u'>' } entity_pattern = re.compile(r'&(\S+?);') - + text_tags = { 0xF581: ['simple_container', 'Italic'], 0xF582: 'end_container', @@ -668,47 +672,47 @@ class Text(LRFStream): 0xF5D1: 'plot', 0xF5D2: 'cr', } - + class TextTag(object): - + def __init__(self, name, attrs={}, self_closing=False): self.name = name self.attrs = attrs self.self_closing = self_closing - + def __unicode__(self): s = u'<%s '%(self.name,) for name, val in self.attrs.items(): s += '%s="%s" '%(name, val) return s.rstrip() + (u' />' if self.self_closing else u'>') - + def to_html(self): s = u'' return s - + def close_html(self): return u'' - + class Span(TextTag): pass - - + + linetype_map = {0: 'none', 0x10: 'solid', 0x20: 'dashed', 0x30: 'double', 0x40: 'dotted'} adjustment_map = {1: 'top', 2: 'center', 3: 'baseline', 4: 'bottom'} lineposition_map = {1:'before', 2:'after'} - + def add_text(self, text): s = unicode(text, "utf-16-le") if s: s = s.translate(self.text_map) self.content.append(self.entity_pattern.sub(entity_to_unicode, s)) - + def end_container(self, tag, stream): self.content.append(None) - + def start_para(self, tag, stream): self.content.append(self.__class__.TextTag('P')) - + def close_containers(self, start=0): if len(self.content) == 0: return @@ -723,8 +727,8 @@ class Text(LRFStream): open_containers += 1 start += 1 self.content.extend(None for i in range(open_containers)) - - + + def end_para(self, tag, stream): i = len(self.content)-1 while i > -1: @@ -732,22 +736,22 @@ class Text(LRFStream): break i -= 1 self.close_containers(start=i) - + def cr(self, tag, stream): self.content.append(self.__class__.TextTag('CR', self_closing=True)) - + def char_button(self, tag, stream): - self.content.append(self.__class__.TextTag( + self.content.append(self.__class__.TextTag( 'CharButton', attrs={'refobj':tag.dword})) - + def simple_container(self, tag, name): self.content.append(self.__class__.TextTag(name)) - + def empline(self, tag, stream): def invalid(op): stream.seek(op) #self.simple_container(None, 'EmpLine') - + oldpos = stream.tell() try: t = Tag(stream) @@ -759,7 +763,7 @@ class Text(LRFStream): h = TextAttr.tag_map[t.id] attrs = {} attrs[h[0]] = TextAttr.tag_to_val(h, None, t, None) - oldpos = stream.tell() + oldpos = stream.tell() try: t = Tag(stream) if t.id not in [0xF579, 0xF57A]: @@ -768,31 +772,31 @@ class Text(LRFStream): attrs[h[0]] = TextAttr.tag_to_val(h, None, t, None) except LRFParseError: stream.seek(oldpos) - + if attrs: - self.content.append(self.__class__.TextTag( + self.content.append(self.__class__.TextTag( 'EmpLine', attrs=attrs)) - + def space(self, tag, stream): - self.content.append(self.__class__.TextTag('Space', - attrs={'xsize':tag.sword}, - self_closing=True)) - + self.content.append(self.__class__.TextTag('Space', + attrs={'xsize':tag.sword}, + self_closing=True)) + def plot(self, tag, stream): xsize, ysize, refobj, adjustment = struct.unpack("= start_pos: if tag_pos > start_pos: @@ -833,9 +837,9 @@ class Text(LRFStream): self.add_text(self.stream) stream.seek(0, 2) break - + tag = Tag(stream) - + if tag.id == 0xF5CC: self.add_text(stream.read(tag.word)) elif tag.id in self.__class__.text_tags: # A Text tag @@ -857,9 +861,9 @@ class Text(LRFStream): self.content.append(self.__class__.Span('Span', {name:val})) current_style[name] = val if len(self.content) > 0: - self.close_containers() + self.close_containers() self.stream = None - + def __unicode__(self): s = u'' open_containers = collections.deque() @@ -868,16 +872,16 @@ class Text(LRFStream): s += c elif c is None: p = open_containers.pop() - s += u''%(p.name,) + s += u''%(p.name,) else: s += unicode(c) - if not c.self_closing: + if not c.self_closing: open_containers.append(c) - + if len(open_containers) > 0: - raise LRFParseError('Malformed text stream %s'%([i.name for i in open_containers if isinstance(i, Text.TextTag)],)) + raise LRFParseError('Malformed text stream %s'%([i.name for i in open_containers if isinstance(i, Text.TextTag)],)) return s - + def to_html(self): s = u'' open_containers = collections.deque() @@ -889,19 +893,19 @@ class Text(LRFStream): if c.name == 'P': in_p = False p = open_containers.pop() - s += p.close_html() + s += p.close_html() else: if c.name == 'P': in_p = True elif c.name == 'CR': s += '
' if in_p else '

' - else: + else: s += c.to_html() - if not c.self_closing: + if not c.self_closing: open_containers.append(c) - + if len(open_containers) > 0: - raise LRFParseError('Malformed text stream %s'%([i.name for i in open_containers if isinstance(i, Text.TextTag)],)) + raise LRFParseError('Malformed text stream %s'%([i.name for i in open_containers if isinstance(i, Text.TextTag)],)) return s @@ -909,30 +913,30 @@ class Image(LRFObject): tag_map = { 0xF54A: ['', 'parse_image_rect'], 0xF54B: ['', 'parse_image_size'], - 0xF54C: ['refstream', 'D'], + 0xF54C: ['refstream', 'D'], 0xF555: ['comment', 'P'], } - + def parse_image_rect(self, tag, f): self.x0, self.y0, self.x1, self.y1 = struct.unpack("\n'%\ (self.id, self.x0, self.y0, self.x1, self.y1, self.xsize, self.ysize, self.refstream) class PutObj(EmptyPageElement): - + def __init__(self, objects, x1, y1, refobj): self.x1, self.y1, self.refobj = x1, y1, refobj self.object = objects[refobj] - + def __unicode__(self): return u''%(self.x1, self.y1, self.refobj) @@ -949,14 +953,14 @@ class Canvas(LRFStream): 0xF52E: ['framemode', 'W', {0: 'none', 2: 'curve', 1:'square'}], } tag_map.update(LRFStream.tag_map) - extra_attrs = ['canvaswidth', 'canvasheight', 'blockrule', 'layout', + extra_attrs = ['canvaswidth', 'canvasheight', 'blockrule', 'layout', 'framewidth', 'framecolor', 'framemode'] - + def parse_waits(self, tag, f): val = tag.word self.setwaitprop = val&0xF self.setwaitsync = val&0xF0 - + def initialize(self): self.attrs = {} for attr in self.extra_attrs: @@ -967,7 +971,7 @@ class Canvas(LRFStream): while stream.tell() < len(self.stream): tag = Tag(stream) self._contents.append(PutObj(self._document.objects, *struct.unpack("\n'%\ (self.id, self.encoding, self.file) @@ -1036,55 +1040,55 @@ class Button(LRFObject): 0xF5F9: ['','parse_run'], #Run } tag_map.update(LRFObject.tag_map) - + def __init__(self, document, stream, id, scramble_key, boundary): self.xml = u'' self.refimage = {} self.actions = {} self.to_dump = True LRFObject.__init__(self, document, stream, id, scramble_key, boundary) - + def do_ref_image(self, tag, f): self.refimage[self.button_type] = tag.dword - + def do_base_button(self, tag, f): self.button_type = 0 self.actions[self.button_type] = [] - + def do_focus_in_button(self, tag, f): self.button_type = 1 - + def do_push_button(self, tag, f): self.button_type = 2 - + def do_up_button(self, tag, f): self.button_type = 3 - + def do_start_actions(self, tag, f): self.actions[self.button_type] = [] - + def parse_jump_to(self, tag, f): self.actions[self.button_type].append((1, struct.unpack("\n'%(self.id,) if self.button_flags & 0x10 != 0: @@ -1098,10 +1102,10 @@ class Button(LRFObject): raise LRFParseError('Unsupported button type') s += '\n' return s - + refpage = property(fget=lambda self : self.jump_action(2)[0]) refobj = property(fget=lambda self : self.jump_action(2)[1]) - + class Window(LRFObject): pass @@ -1122,12 +1126,12 @@ class Font(LRFStream): } tag_map.update(LRFStream.tag_map) data = property(fget=lambda self: self.stream) - + def end_stream(self, *args): LRFStream.end_stream(self, *args) self._document.font_map[self.fontfacename] = self self.file = self.fontfacename + '.ttf' - + def __unicode__(self): s = '\n'%\ (self.id, self.fontfilename, self.fontfacename, self.file) @@ -1146,20 +1150,20 @@ class BookAttr(StyleObject, LRFObject): tag_map.update(ruby_tags) tag_map.update(LRFObject.tag_map) binding_map = {1: 'Lr', 16 : 'Rl'} - + def __init__(self, document, stream, id, scramble_key, boundary): self.font_link_list = [] LRFObject.__init__(self, document, stream, id, scramble_key, boundary) - + def add_font(self, tag, f): self.font_link_list.append(tag.dword) - + def __unicode__(self): s = u'\n'%(self.id, self.id) s += u'\n'%(self._tags_to_xml(),) doc = self._document s += u'\n'%\ - (self.binding_map[doc.binding], doc.dpi, doc.width, doc.height, doc.color_depth) + (self.binding_map[doc.binding], doc.dpi, doc.width, doc.height, doc.color_depth) for font in self._document.font_map.values(): s += unicode(font) s += '\n' @@ -1169,15 +1173,15 @@ class SimpleText(Text): pass class TocLabel(object): - + def __init__(self, refpage, refobject, label): self.refpage, self.refobject, self.label = refpage, refobject, label - + def __unicode__(self): return u'%s\n'%(self.refpage, self.refobject, self.label) class TOCObject(LRFStream): - + def initialize(self): stream = cStringIO.StringIO(self.stream) c = struct.unpack("\n' for i in self: s += unicode(i) return s + '\n' - + object_map = [ None, #00 @@ -1246,7 +1250,7 @@ def get_object(document, stream, id, offset, size, scramble_key): obj_id, obj_type = struct.unpack("