Python Styles.generate_classes方法代码示例

# 需要导入模块: from calibre.ebooks.docx.styles import Styles [as 别名]
# 或者: from calibre.ebooks.docx.styles.Styles import generate_classes [as 别名]

#.........这里部分代码省略.........
                        indent = float(style.text_indent[:-2]) + indent
                    style.text_indent = '%.3gpt' % indent
                    parent.text = tabs[-1].tail or ''
                    map(parent.remove, tabs)

        self.images.rid_map = orig_rid_map

        self.resolve_links()

        self.styles.cascade(self.layers)

        self.tables.apply_markup(self.object_map, self.page_map)

        numbered = []
        for html_obj, obj in self.object_map.iteritems():
            raw = obj.get('calibre_num_id', None)
            if raw is not None:
                lvl, num_id = raw.partition(':')[0::2]
                try:
                    lvl = int(lvl)
                except (TypeError, ValueError):
                    lvl = 0
                numbered.append((html_obj, num_id, lvl))
        self.numbering.apply_markup(numbered, self.body, self.styles, self.object_map, self.images)
        self.apply_frames()

        if len(self.body) > 0:
            self.body.text = '\n\t'
            for child in self.body:
                child.tail = '\n\t'
            self.body[-1].tail = '\n'

        self.log.debug('Converting styles to CSS')
        self.styles.generate_classes()
        for html_obj, obj in self.object_map.iteritems():
            style = self.styles.resolve(obj)
            if style is not None:
                css = style.css
                if css:
                    cls = self.styles.class_name(css)
                    if cls:
                        html_obj.set('class', cls)
        for html_obj, css in self.framed_map.iteritems():
            cls = self.styles.class_name(css)
            if cls:
                html_obj.set('class', cls)

        if notes_header is not None:
            for h in children(self.body, 'h1', 'h2', 'h3'):
                notes_header.tag = h.tag
                cls = h.get('class', None)
                if cls and cls != 'notes-header':
                    notes_header.set('class', '%s notes-header' % cls)
                break

        self.fields.polish_markup(self.object_map)

        self.log.debug('Cleaning up redundant markup generated by Word')
        self.cover_image = cleanup_markup(self.log, self.html, self.styles, self.dest_dir, self.detect_cover)

        return self.write(doc)

    def read_page_properties(self, doc):
        current = []
        self.page_map = OrderedDict()
        self.section_starts = []

# 需要导入模块: from calibre.ebooks.docx.styles import Styles [as 别名]
# 或者: from calibre.ebooks.docx.styles.Styles import generate_classes [as 别名]

#.........这里部分代码省略.........
                    if in_table:
                        if ancestor(wp, "w:tbl") is not None:
                            self.tables.add(wp)
                        else:
                            in_table = False
                    p = self.convert_p(wp)
                    dl[-1].append(p)

        self.resolve_links(relationships_by_id)

        self.styles.cascade(self.layers)

        self.tables.apply_markup(self.object_map)

        numbered = []
        for html_obj, obj in self.object_map.iteritems():
            raw = obj.get("calibre_num_id", None)
            if raw is not None:
                lvl, num_id = raw.partition(":")[0::2]
                try:
                    lvl = int(lvl)
                except (TypeError, ValueError):
                    lvl = 0
                numbered.append((html_obj, num_id, lvl))
        self.numbering.apply_markup(numbered, self.body, self.styles, self.object_map)
        self.apply_frames()

        if len(self.body) > 0:
            self.body.text = "\n\t"
            for child in self.body:
                child.tail = "\n\t"
            self.body[-1].tail = "\n"

        self.styles.generate_classes()
        for html_obj, obj in self.object_map.iteritems():
            style = self.styles.resolve(obj)
            if style is not None:
                css = style.css
                if css:
                    cls = self.styles.class_name(css)
                    if cls:
                        html_obj.set("class", cls)
        for html_obj, css in self.framed_map.iteritems():
            cls = self.styles.class_name(css)
            if cls:
                html_obj.set("class", cls)

        if notes_header is not None:
            for h in self.body.iterchildren("h1", "h2", "h3"):
                notes_header.tag = h.tag
                cls = h.get("class", None)
                if cls and cls != "notes-header":
                    notes_header.set("class", "%s notes-header" % cls)
                break

        return self.write()

    def read_page_properties(self, doc):
        current = []
        self.page_map = OrderedDict()

        in_table = False

        for p in descendants(doc, "w:p", "w:tbl"):
            if p.tag.endswith("}tbl"):
                in_table = True

# 需要导入模块: from calibre.ebooks.docx.styles import Styles [as 别名]
# 或者: from calibre.ebooks.docx.styles.Styles import generate_classes [as 别名]
class Convert(object):

    def __init__(self, path_or_stream, dest_dir=None, log=None, notes_text=None):
        self.docx = DOCX(path_or_stream, log=log)
        self.ms_pat = re.compile(r'\s{2,}')
        self.ws_pat = re.compile(r'[\n\r\t]')
        self.log = self.docx.log
        self.notes_text = notes_text or _('Notes')
        self.dest_dir = dest_dir or os.getcwdu()
        self.mi = self.docx.metadata
        self.body = BODY()
        self.tables = Tables()
        self.styles = Styles(self.tables)
        self.images = Images()
        self.object_map = OrderedDict()
        self.html = HTML(
            HEAD(
                META(charset='utf-8'),
                TITLE(self.mi.title or _('Unknown')),
                LINK(rel='stylesheet', type='text/css', href='docx.css'),
            ),
            self.body
        )
        self.html.text='\n\t'
        self.html[0].text='\n\t\t'
        self.html[0].tail='\n'
        for child in self.html[0]:
            child.tail = '\n\t\t'
        self.html[0][-1].tail = '\n\t'
        self.html[1].text = self.html[1].tail = '\n'
        lang = canonicalize_lang(self.mi.language)
        if lang and lang != 'und':
            lang = lang_as_iso639_1(lang)
            if lang:
                self.html.set('lang', lang)

    def __call__(self):
        doc = self.docx.document
        relationships_by_id, relationships_by_type = self.docx.document_relationships
        self.read_styles(relationships_by_type)
        self.images(relationships_by_id)
        self.layers = OrderedDict()
        self.framed = [[]]
        self.framed_map = {}
        self.anchor_map = {}
        self.link_map = defaultdict(list)

        self.read_page_properties(doc)
        for wp, page_properties in self.page_map.iteritems():
            self.current_page = page_properties
            if wp.tag.endswith('}p'):
                p = self.convert_p(wp)
                self.body.append(p)

        notes_header = None
        if self.footnotes.has_notes:
            dl = DL()
            dl.set('class', 'notes')
            self.body.append(H1(self.notes_text))
            notes_header = self.body[-1]
            notes_header.set('class', 'notes-header')
            self.body.append(dl)
            for anchor, text, note in self.footnotes:
                dl.append(DT('[', A('←' + text, href='#back_%s' % anchor, title=text), id=anchor))
                dl[-1][0].tail = ']'
                dl.append(DD())
                for wp in note:
                    if wp.tag.endswith('}tbl'):
                        self.tables.register(wp, self.styles)
                        self.page_map[wp] = self.current_page
                    p = self.convert_p(wp)
                    dl[-1].append(p)

        self.resolve_links(relationships_by_id)

        self.styles.cascade(self.layers)

        self.tables.apply_markup(self.object_map, self.page_map)

        numbered = []
        for html_obj, obj in self.object_map.iteritems():
            raw = obj.get('calibre_num_id', None)
            if raw is not None:
                lvl, num_id = raw.partition(':')[0::2]
                try:
                    lvl = int(lvl)
                except (TypeError, ValueError):
                    lvl = 0
                numbered.append((html_obj, num_id, lvl))
        self.numbering.apply_markup(numbered, self.body, self.styles, self.object_map)
        self.apply_frames()

        if len(self.body) > 0:
            self.body.text = '\n\t'
            for child in self.body:
                child.tail = '\n\t'
            self.body[-1].tail = '\n'

        self.styles.generate_classes()
        for html_obj, obj in self.object_map.iteritems():
#.........这里部分代码省略.........