Python DOCX.read方法代码示例

# 需要导入模块: from calibre.ebooks.docx.container import DOCX [as 别名]
# 或者: from calibre.ebooks.docx.container.DOCX import read [as 别名]
def set_metadata(stream, mi):
    from calibre.utils.zipfile import safe_replace
    c = DOCX(stream, extract=False)
    dp_name, ap_name = c.get_document_properties_names()
    dp_raw = c.read(dp_name)
    try:
        ap_raw = c.read(ap_name)
    except Exception:
        ap_raw = None
    cp = etree.fromstring(dp_raw)
    update_doc_props(cp, mi)
    replacements = {}
    if ap_raw is not None:
        ap = etree.fromstring(ap_raw)
        comp = ap.makeelement('{%s}Company' % namespaces['ep'])
        for child in tuple(ap):
            if child.tag == comp.tag:
                ap.remove(child)
        comp.text = mi.publisher
        ap.append(comp)
        replacements[ap_name] = BytesIO(xml2str(ap))
    stream.seek(0)
    safe_replace(stream, dp_name, BytesIO(xml2str(cp)), extra_replacements=replacements)

# 需要导入模块: from calibre.ebooks.docx.container import DOCX [as 别名]
# 或者: from calibre.ebooks.docx.container.DOCX import read [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.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.styles = Styles()
        self.images = Images()
        self.tables = Tables()
        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
            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())
                in_table = False
                for wp in note:
                    if wp.tag.endswith("}tbl"):
                        self.tables.register(wp)
                        in_table = True
                        continue
                    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"
#.........这里部分代码省略.........

# 需要导入模块: from calibre.ebooks.docx.container import DOCX [as 别名]
# 或者: from calibre.ebooks.docx.container.DOCX import read [as 别名]
class Convert(object):

    def __init__(self, path_or_stream, dest_dir=None, log=None, detect_cover=True, 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.detect_cover = detect_cover
        self.notes_text = notes_text or _('Notes')
        self.dest_dir = dest_dir or os.getcwdu()
        self.mi = self.docx.metadata
        self.body = BODY()
        self.theme = Theme()
        self.settings = Settings()
        self.tables = Tables()
        self.fields = Fields()
        self.styles = Styles(self.tables)
        self.images = Images(self.log)
        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.fields(doc, self.log)
        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.link_source_map = {}
        paras = []

        self.log.debug('Converting Word markup to HTML')

        self.read_page_properties(doc)
        self.current_rels = relationships_by_id
        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)
                paras.append(wp)

        self.read_block_anchors(doc)
        self.styles.apply_contextual_spacing(paras)
        # Apply page breaks at the start of every section, except the first
        # section (since that will be the start of the file)
        self.styles.apply_section_page_breaks(self.section_starts[1:])

        notes_header = None
        orig_rid_map = self.images.rid_map
        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())
                paras = []
                self.images.rid_map = self.current_rels = note.rels[0]
                for wp in note:
                    if wp.tag.endswith('}tbl'):
                        self.tables.register(wp, self.styles)
                        self.page_map[wp] = self.current_page
                    else:
                        p = self.convert_p(wp)
                        dl[-1].append(p)
                        paras.append(wp)
                self.styles.apply_contextual_spacing(paras)

        for p, wp in self.object_map.iteritems():
            if len(p) > 0 and not p.text and len(p[0]) > 0 and not p[0].text and p[0][0].get('class', None) == 'tab':
                # Paragraph uses tabs for indentation, convert to text-indent
                parent = p[0]
#.........这里部分代码省略.........

# 需要导入模块: from calibre.ebooks.docx.container import DOCX [as 别名]
# 或者: from calibre.ebooks.docx.container.DOCX import read [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.........