def get_navbar(self, f, feeds, top=True):
if len(feeds) < 2:
return DIV()
navbar = DIV('| ', CLASS('calibre_navbar', 'calibre_rescale_70',
style='text-align:center'))
if not top:
hr = HR()
navbar.append(hr)
navbar.text = None
hr.tail = '| '
if f+1 < len(feeds):
link = A(_('Next section'), href='../feed_%d/index.html'%(f+1))
link.tail = ' | '
navbar.append(link)
link = A(_('Main menu'), href="../index.html")
link.tail = ' | '
navbar.append(link)
if f > 0:
link = A(_('Previous section'), href='../feed_%d/index.html'%(f-1))
link.tail = ' |'
navbar.append(link)
if top:
navbar.append(HR())
return navbar
def _generate(self,
bottom,
feed,
art,
number_of_articles_in_feed,
two_levels,
url,
__appname__,
prefix='',
center=True,
extra_css=None,
style=None):
head = HEAD(TITLE('navbar'))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
if prefix and not prefix.endswith('/'):
prefix += '/'
align = 'center' if center else 'left'
navbar = DIV(
attrs('calibre_navbar', rescale=70, style='text-align:' + align))
if bottom:
if not url.startswith('file://'):
navbar.append(HR())
text = 'This article was downloaded by '
p = PT(
text,
STRONG(__appname__),
A(url, href=url, rel='calibre-downloaded-from'),
style='text-align:left; max-width: 100%; overflow: hidden;'
)
p[0].tail = ' from '
navbar.append(p)
navbar.append(BR())
navbar.append(BR())
else:
next_art = 'feed_%d'%(feed+1) if art == number_of_articles_in_feed - 1 \
else 'article_%d'%(art+1)
up = '../..' if art == number_of_articles_in_feed - 1 else '..'
href = '%s%s/%s/index.html' % (prefix, up, next_art)
navbar.text = '| '
navbar.append(A(_('Next'), href=href))
href = '%s../index.html#article_%d' % (prefix, art)
next(navbar.iterchildren(reversed=True)).tail = ' | '
navbar.append(A(_('Section menu'), href=href))
href = '%s../../index.html#feed_%d' % (prefix, feed)
next(navbar.iterchildren(reversed=True)).tail = ' | '
navbar.append(A(_('Main menu'), href=href))
if art > 0 and not bottom:
href = '%s../article_%d/index.html' % (prefix, art - 1)
next(navbar.iterchildren(reversed=True)).tail = ' | '
navbar.append(A(_('Previous'), href=href))
next(navbar.iterchildren(reversed=True)).tail = ' | '
if not bottom:
navbar.append(HR())
self.root = HTML(head, BODY(navbar))
def _generate(self,
bottom,
feed,
art,
number_of_articles_in_feed,
two_levels,
url,
__appname__,
prefix='',
center=True,
extra_css=None,
style=None):
head = HEAD(TITLE('navbar'))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
navbar = DIV()
navbar_t = TABLE(CLASS('touchscreen_navbar'))
navbar_tr = TR()
# | Previous
if art > 0:
link = A(CLASS('article_link'),
_('Previous'),
href='%s../article_%d/index.html' % (prefix, art - 1))
navbar_tr.append(TD(CLASS('article_prev'), link))
else:
navbar_tr.append(TD(CLASS('article_prev'), ''))
# | Articles | Sections |
link = A(CLASS('articles_link'),
_('Articles'),
href='%s../index.html#article_%d' % (prefix, art))
navbar_tr.append(TD(CLASS('article_articles_list'), link))
link = A(CLASS('sections_link'),
_('Sections'),
href='%s../../index.html#feed_%d' % (prefix, feed))
navbar_tr.append(TD(CLASS('article_sections_list'), link))
# | Next
next = 'feed_%d'%(feed+1) if art == number_of_articles_in_feed - 1 \
else 'article_%d'%(art+1)
up = '../..' if art == number_of_articles_in_feed - 1 else '..'
link = A(CLASS('article_link'),
_('Next'),
href='%s%s/%s/index.html' % (prefix, up, next))
navbar_tr.append(TD(CLASS('article_next'), link))
navbar_t.append(navbar_tr)
navbar.append(navbar_t)
#print "\n%s\n" % etree.tostring(navbar, pretty_print=True)
self.root = HTML(head, BODY(navbar))
def _generate(self,
title,
masthead,
datefmt,
feeds,
extra_css=None,
style=None):
self.IS_HTML = False
if isinstance(datefmt, unicode):
datefmt = datefmt.encode(preferred_encoding)
date = strftime(datefmt)
head = HEAD(TITLE(title))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
ul = UL(CLASS('calibre_feed_list'))
for i, feed in enumerate(feeds):
if feed:
li = LI(A(
feed.title,
CLASS('feed',
'calibre_rescale_120',
href='feed_%d/index.html' % i)),
id='feed_%d' % i)
ul.append(li)
div = DIV(
PT(IMG(src=masthead, alt="masthead"), style='text-align:center'),
PT(date, style='text-align:right'), ul,
CLASS('calibre_rescale_100'))
self.root = HTML(head, BODY(div))
if self.html_lang:
self.root.set('lang', self.html_lang)
def inject_toc(tree, rules, ty_map_fn=None):
modified = False
for rule in rules:
nodes = tree.xpath(rule["xpath"])
for node in nodes:
pair = rule["fn"](node)
if not pair:
continue
(name, child_ty) = pair
toc_node = A(CLASS('dashAnchor'))
if ty_map_fn:
ty = ty_map_fn(child_ty)
else:
ty = child_ty
toc_node.attrib["name"] = "//apple_ref/cpp/%s/%s" % (ty, name)
place_fn = rule.get('place_fn', None)
if place_fn:
place_node = place_fn(node)
else:
place_node = node
modified = True
place_node.addprevious(toc_node)
return modified
def _generate(self, title, masthead, datefmt, feeds, extra_css=None, style=None):
self.IS_HTML = False
date = '%s, %s %s, %s' % (strftime('%A'), strftime('%B'), strftime('%d').lstrip('0'), strftime('%Y'))
masthead_p = etree.Element("p")
masthead_p.set("style","text-align:center")
masthead_img = etree.Element("img")
masthead_img.set("src",masthead)
masthead_img.set("alt","masthead")
masthead_p.append(masthead_img)
head = HEAD(TITLE(title))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
toc = TABLE(CLASS('toc'),width="100%",border="0",cellpadding="3px")
for i, feed in enumerate(feeds):
if feed:
tr = TR()
tr.append(TD(CLASS('calibre_rescale_120'), A(feed.title, href='feed_%d/index.html'%i)))
tr.append(TD('%s' % len(feed.articles), style="text-align:right"))
toc.append(tr)
div = DIV(
masthead_p,
H3(CLASS('publish_date'),date),
DIV(CLASS('divider')),
toc)
self.root = HTML(head, BODY(div))
if self.html_lang:
self.root.set('lang', self.html_lang)
def _generate(self,
title,
masthead,
datefmt,
feeds,
extra_css=None,
style=None):
self.IS_HTML = False
date = strftime(datefmt)
head = HEAD(TITLE(title))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
ul = UL(attrs('calibre_feed_list'))
for i, feed in enumerate(feeds):
if len(feed):
li = LI(A(
feed.title,
attrs('feed', rescale=120, href='feed_%d/index.html' % i)),
id='feed_%d' % i)
ul.append(li)
div = DIV(
PT(IMG(src=masthead, alt="masthead"), style='text-align:center'),
PT(date, style='text-align:right'), ul, attrs(rescale=100))
self.root = HTML(head, BODY(div))
if self.html_lang:
self.root.set('lang', self.html_lang)
def _generate(self, bottom, feed, art, number_of_articles_in_feed,
two_levels, url, __appname__, prefix='', center=True,
extra_css=None, style=None):
head = HEAD(TITLE('navbar'))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
navbar = DIV()
navbar_t = TABLE(CLASS('touchscreen_navbar'))
navbar_tr = TR()
if bottom and not url.startswith('file://'):
navbar.append(HR())
text = 'This article was downloaded by '
p = PT(text, STRONG(__appname__), A(url, href=url, rel='calibre-downloaded-from'),
style='text-align:left; max-width: 100%; overflow: hidden;')
p[0].tail = ' from '
navbar.append(p)
navbar.append(BR())
# | Previous
if art > 0:
link = A(CLASS('article_link'),_('Previous'),href='%s../article_%d/index.html'%(prefix, art-1))
navbar_tr.append(TD(CLASS('article_prev'),link))
else:
navbar_tr.append(TD(CLASS('article_prev'),''))
# | Articles | Sections |
link = A(CLASS('articles_link'),_('Articles'), href='%s../index.html#article_%d'%(prefix, art))
navbar_tr.append(TD(CLASS('article_articles_list'),link))
link = A(CLASS('sections_link'),_('Sections'), href='%s../../index.html#feed_%d'%(prefix, feed))
navbar_tr.append(TD(CLASS('article_sections_list'),link))
# | Next
next = 'feed_%d'%(feed+1) if art == number_of_articles_in_feed - 1 \
else 'article_%d'%(art+1)
up = '../..' if art == number_of_articles_in_feed - 1 else '..'
link = A(CLASS('article_link'), _('Next'), href='%s%s/%s/index.html'%(prefix, up, next))
navbar_tr.append(TD(CLASS('article_next'),link))
navbar_t.append(navbar_tr)
navbar.append(navbar_t)
# print "\n%s\n" % etree.tostring(navbar, pretty_print=True)
self.root = HTML(head, BODY(navbar))
def build_navigation(start, num, total, url_base): # {{{
end = min((start + num - 1), total)
tagline = SPAN('Books %d to %d of %d' % (start, end, total),
style='display: block; text-align: center;')
left_buttons = TD(CLASS('button', style='text-align:left'))
right_buttons = TD(CLASS('button', style='text-align:right'))
if start > 1:
for t, s in [('First', 1), ('Previous', max(start - num, 1))]:
left_buttons.append(A(t, href='%s;start=%d' % (url_base, s)))
if total > start + num:
for t, s in [('Next', start + num), ('Last', total - num + 1)]:
right_buttons.append(A(t, href='%s;start=%d' % (url_base, s)))
buttons = TABLE(TR(left_buttons, right_buttons), CLASS('buttons'))
return DIV(tagline, buttons, CLASS('navigation'))
def convert_run(self, run):
ans = SPAN()
self.object_map[ans] = run
text = Text(ans, 'text', [])
for child in run:
if is_tag(child, 'w:t'):
if not child.text:
continue
space = child.get(XML('space'), None)
preserve = False
if space == 'preserve':
# Only use a <span> with white-space:pre-wrap if this element
# actually needs it, i.e. if it has more than one
# consecutive space or it has newlines or tabs.
multi_spaces = self.ms_pat.search(child.text) is not None
preserve = multi_spaces or self.ws_pat.search(child.text) is not None
if preserve:
text.add_elem(SPAN(child.text, style="white-space:pre-wrap"))
ans.append(text.elem)
else:
text.buf.append(child.text)
elif is_tag(child, 'w:cr'):
text.add_elem(BR())
ans.append(text.elem)
elif is_tag(child, 'w:br'):
typ = get(child, 'w:type')
if typ in {'column', 'page'}:
br = BR(style='page-break-after:always')
else:
clear = child.get('clear', None)
if clear in {'all', 'left', 'right'}:
br = BR(style='clear:%s'%('both' if clear == 'all' else clear))
else:
br = BR()
text.add_elem(br)
ans.append(text.elem)
elif is_tag(child, 'w:drawing') or is_tag(child, 'w:pict'):
for img in self.images.to_html(child, self.current_page, self.docx, self.dest_dir):
text.add_elem(img)
ans.append(text.elem)
elif is_tag(child, 'w:footnoteReference') or is_tag(child, 'w:endnoteReference'):
anchor, name = self.footnotes.get_ref(child)
if anchor and name:
l = SUP(A(name, href='#' + anchor, title=name), id='back_%s' % anchor)
l.set('class', 'noteref')
text.add_elem(l)
ans.append(text.elem)
if text.buf:
setattr(text.elem, text.attr, ''.join(text.buf))
style = self.styles.resolve_run(run)
if style.vert_align in {'superscript', 'subscript'}:
ans.tag = 'sub' if style.vert_align == 'subscript' else 'sup'
if style.lang is not inherit:
ans.lang = style.lang
return ans
def convert_run(self, run):
ans = SPAN()
self.object_map[ans] = run
text = Text(ans, 'text', [])
for child in run:
if is_tag(child, 'w:t'):
if not child.text:
continue
space = child.get(XML('space'), None)
if space == 'preserve':
text.add_elem(SPAN(child.text, style="white-space:pre-wrap"))
ans.append(text.elem)
else:
text.buf.append(child.text)
elif is_tag(child, 'w:cr'):
text.add_elem(BR())
ans.append(text.elem)
elif is_tag(child, 'w:br'):
typ = child.get('type', None)
if typ in {'column', 'page'}:
br = BR(style='page-break-after:always')
else:
clear = child.get('clear', None)
if clear in {'all', 'left', 'right'}:
br = BR(style='clear:%s'%('both' if clear == 'all' else clear))
else:
br = BR()
text.add_elem(br)
ans.append(text.elem)
elif is_tag(child, 'w:drawing') or is_tag(child, 'w:pict'):
for img in self.images.to_html(child, self.current_page, self.docx, self.dest_dir):
text.add_elem(img)
ans.append(text.elem)
elif is_tag(child, 'w:footnoteReference') or is_tag(child, 'w:endnoteReference'):
anchor, name = self.footnotes.get_ref(child)
if anchor and name:
l = SUP(A(name, href='#' + anchor, title=name), id='back_%s' % anchor)
l.set('class', 'noteref')
text.add_elem(l)
ans.append(text.elem)
elif is_tag(child, 'w:fldChar') and get(child, 'w:fldCharType') == 'separate':
text.buf.append('\xa0')
if text.buf:
setattr(text.elem, text.attr, ''.join(text.buf))
style = self.styles.resolve_run(run)
if style.vert_align in {'superscript', 'subscript'}:
ans.tag = 'sub' if style.vert_align == 'subscript' else 'sup'
if style.lang is not inherit:
ans.lang = style.lang
return ans
def _generate(self, f, feeds, cutoff, extra_css=None, style=None):
from calibre.utils.cleantext import clean_xml_chars
feed = feeds[f]
head = HEAD(TITLE(feed.title))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
body = BODY()
body.append(self.get_navbar(f, feeds))
div = DIV(
H2(feed.title,
CLASS('calibre_feed_title', 'calibre_rescale_160')),
CLASS('calibre_rescale_100')
)
body.append(div)
if getattr(feed, 'image', None):
div.append(DIV(IMG(
alt=feed.image_alt if feed.image_alt else '',
src=feed.image_url
),
CLASS('calibre_feed_image')))
if getattr(feed, 'description', None):
d = DIV(clean_xml_chars(feed.description), CLASS('calibre_feed_description',
'calibre_rescale_80'))
d.append(BR())
div.append(d)
ul = UL(CLASS('calibre_article_list'))
for i, article in enumerate(feed.articles):
if not getattr(article, 'downloaded', False):
continue
li = LI(
A(article.title, CLASS('article calibre_rescale_120',
href=article.url)),
SPAN(article.formatted_date, CLASS('article_date')),
CLASS('calibre_rescale_100', id='article_%d'%i,
style='padding-bottom:0.5em')
)
if article.summary:
li.append(DIV(clean_xml_chars(cutoff(article.text_summary)),
CLASS('article_description', 'calibre_rescale_70')))
ul.append(li)
div.append(ul)
div.append(self.get_navbar(f, feeds, top=False))
self.root = HTML(head, body)
if self.html_lang:
self.root.set('lang', self.html_lang)
def donode(item, parent, base, subpath):
for child in item:
title = child.title
if not title:
continue
raw = unquote_path(child.href or '')
rsrcname = os.path.basename(raw)
rsrcpath = os.path.join(subpath, rsrcname)
if (not os.path.exists(os.path.join(base, rsrcpath)) and os.path.exists(os.path.join(base, raw))):
rsrcpath = raw
if '%' not in rsrcpath:
rsrcpath = urlquote(rsrcpath)
if not raw:
rsrcpath = ''
c = DIV(A(title, href=rsrcpath))
donode(child, c, base, subpath)
parent.append(c)
def clean_stars(div: HtmlElement) -> None:
for e in elements(div, "p[strong[em]]"):
e.drop_tree()
for e in elements(div, ".//hr"):
e.drop_tree()
div.insert(0, H1("Unspeakable Desolation Pouring Down From the Stars"))
e = element(div, "./p[1]")
h2 = H2(e.text_content().title())
replace(e, h2)
e = element(div, "./p[strong[a]]")
a = element(div, "./p/strong/a")
p = P(
CLASS("breakabove"),
A(e.text_content(), CLASS("internal"), href=a.attrib["href"]),
)
replace(e, p)
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.frame_map = {}
self.framed_map = {}
self.anchor_map = {}
self.link_map = defaultdict(list)
self.link_source_map = {}
self.toc_anchor = None
self.block_runs = []
paras = []
self.log.debug('Converting Word markup to HTML')
self.read_page_properties(doc)
self.resolve_alternate_content(doc)
self.current_rels = relationships_by_id
for wp, page_properties in iteritems(self.page_map):
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)
self.mark_block_runs(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:
self.body.append(H1(self.notes_text))
notes_header = self.body[-1]
notes_header.set('class', 'notes-header')
for anchor, text, note in self.footnotes:
dl = DL(id=anchor)
dl.set('class', 'footnote')
self.body.append(dl)
dl.append(DT('[', A('←' + text, href='#back_%s' % anchor, title=text)))
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)
self.mark_block_runs(paras)
for p, wp in iteritems(self.object_map):
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]
tabs = []
for child in parent:
if child.get('class', None) == 'tab':
tabs.append(child)
if child.tail:
break
else:
break
indent = len(tabs) * self.settings.default_tab_stop
style = self.styles.resolve(wp)
if style.text_indent is inherit or (hasattr(style.text_indent, 'endswith') and style.text_indent.endswith('pt')):
if style.text_indent is not inherit:
indent = float(style.text_indent[:-2]) + indent
style.text_indent = '%.3gpt' % indent
parent.text = tabs[-1].tail or ''
list(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 iteritems(self.object_map):
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 iteritems(self.object_map):
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 iteritems(self.framed_map):
cls = self.styles.class_name(css)
if cls:
html_obj.set('class', cls)
if notes_header is not None:
for h in self.namespace.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, self.namespace.XPath)
return self.write(doc)
def build_index(books, num, search, sort, order, start, total, url_base, CKEYS,
prefix):
logo = DIV(IMG(src=prefix + '/static/calibre.png', alt=__appname__),
id='logo')
search_box = build_search_box(num, search, sort, order, prefix)
navigation = build_navigation(start, num, total, prefix + url_base)
navigation2 = build_navigation(start, num, total, prefix + url_base)
bookt = TABLE(id='listing')
body = BODY(logo, search_box, navigation, HR(CLASS('spacer')), bookt,
HR(CLASS('spacer')), navigation2)
# Book list {{{
for book in books:
thumbnail = TD(
IMG(type='image/jpeg',
border='0',
src=prefix + '/get/thumb/%s' % book['id']), CLASS('thumbnail'))
data = TD()
for fmt in book['formats'].split(','):
if not fmt or fmt.lower().startswith('original_'):
continue
a = quote(ascii_filename(book['authors']))
t = quote(ascii_filename(book['title']))
s = SPAN(
A(fmt.lower(),
href=prefix + '/get/%s/%s-%s_%d.%s' %
(fmt, a, t, book['id'], fmt.lower())), CLASS('button'))
s.tail = u''
data.append(s)
div = DIV(CLASS('data-container'))
data.append(div)
series = u'[%s - %s]'%(book['series'], book['series_index']) \
if book['series'] else ''
tags = u'Tags=[%s]' % book['tags'] if book['tags'] else ''
ctext = ''
for key in CKEYS:
val = book.get(key, None)
if val:
ctext += '%s=[%s] ' % tuple(val.split(':#:'))
first = SPAN(
u'\u202f%s %s by %s' % (book['title'], series, book['authors']),
CLASS('first-line'))
div.append(first)
second = SPAN(
u'%s - %s %s %s' % (book['size'], book['timestamp'], tags, ctext),
CLASS('second-line'))
div.append(second)
bookt.append(TR(thumbnail, data))
# }}}
body.append(
DIV(A(_('Switch to the full interface (non-mobile interface)'),
href=prefix + "/browse",
style="text-decoration: none; color: blue",
title=_('The full interface gives you many more features, '
'but it may not work well on a small screen')),
style="text-align:center"))
return HTML(
HEAD(
TITLE(__appname__ + ' Library'),
LINK(rel='icon',
href='http://calibre-ebook.com/favicon.ico',
type='image/x-icon'),
LINK(rel='stylesheet',
type='text/css',
href=prefix + '/mobile/style.css'),
LINK(rel='apple-touch-icon',
href="/static/calibre.png")), # End head
body) # End html
def resolve_links(self):
self.resolved_link_map = {}
for hyperlink, spans in iteritems(self.link_map):
relationships_by_id = self.link_source_map[hyperlink]
span = spans[0]
if len(spans) > 1:
span = self.wrap_elems(spans, SPAN())
span.tag = 'a'
self.resolved_link_map[hyperlink] = span
tgt = self.namespace.get(hyperlink, 'w:tgtFrame')
if tgt:
span.set('target', tgt)
tt = self.namespace.get(hyperlink, 'w:tooltip')
if tt:
span.set('title', tt)
rid = self.namespace.get(hyperlink, 'r:id')
if rid and rid in relationships_by_id:
span.set('href', relationships_by_id[rid])
continue
anchor = self.namespace.get(hyperlink, 'w:anchor')
if anchor and anchor in self.anchor_map:
span.set('href', '#' + self.anchor_map[anchor])
continue
self.log.warn('Hyperlink with unknown target (rid=%s, anchor=%s), ignoring' %
(rid, anchor))
# hrefs that point nowhere give epubcheck a hernia. The element
# should be styled explicitly by Word anyway.
# span.set('href', '#')
rmap = {v:k for k, v in iteritems(self.object_map)}
for hyperlink, runs in self.fields.hyperlink_fields:
spans = [rmap[r] for r in runs if r in rmap]
if not spans:
continue
span = spans[0]
if len(spans) > 1:
span = self.wrap_elems(spans, SPAN())
span.tag = 'a'
tgt = hyperlink.get('target', None)
if tgt:
span.set('target', tgt)
tt = hyperlink.get('title', None)
if tt:
span.set('title', tt)
url = hyperlink.get('url', None)
if url is None:
anchor = hyperlink.get('anchor', None)
if anchor in self.anchor_map:
span.set('href', '#' + self.anchor_map[anchor])
continue
self.log.warn('Hyperlink field with unknown anchor: %s' % anchor)
else:
if url in self.anchor_map:
span.set('href', '#' + self.anchor_map[url])
continue
span.set('href', url)
for img, link, relationships_by_id in self.images.links:
parent = img.getparent()
idx = parent.index(img)
a = A(img)
a.tail, img.tail = img.tail, None
parent.insert(idx, a)
tgt = link.get('target', None)
if tgt:
a.set('target', tgt)
tt = link.get('title', None)
if tt:
a.set('title', tt)
rid = link['id']
if rid in relationships_by_id:
dest = relationships_by_id[rid]
if dest.startswith('#'):
if dest[1:] in self.anchor_map:
a.set('href', '#' + self.anchor_map[dest[1:]])
else:
a.set('href', dest)
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)
paras = []
self.log.debug('Converting Word markup to HTML')
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)
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
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 = []
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)
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.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.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 _generate(self, f, feeds, cutoff, extra_css=None, style=None):
from calibre.utils.cleantext import clean_xml_chars
def trim_title(title,clip=18):
if len(title)>clip:
tokens = title.split(' ')
new_title_tokens = []
new_title_len = 0
if len(tokens[0]) > clip:
return tokens[0][:clip] + '...'
for token in tokens:
if len(token) + new_title_len < clip:
new_title_tokens.append(token)
new_title_len += len(token)
else:
new_title_tokens.append('...')
title = ' '.join(new_title_tokens)
break
return title
self.IS_HTML = False
feed = feeds[f]
# Construct the navbar
navbar_t = TABLE(CLASS('touchscreen_navbar'))
navbar_tr = TR()
# Previous Section
link = ''
if f > 0:
link = A(CLASS('feed_link'),
trim_title(feeds[f-1].title),
href='../feed_%d/index.html' % int(f-1))
navbar_tr.append(TD(CLASS('feed_prev'),link))
# Up to Sections
link = A(_('Sections'), href="../index.html")
navbar_tr.append(TD(CLASS('feed_up'),link))
# Next Section
link = ''
if f < len(feeds)-1:
link = A(CLASS('feed_link'),
trim_title(feeds[f+1].title),
href='../feed_%d/index.html' % int(f+1))
navbar_tr.append(TD(CLASS('feed_next'),link))
navbar_t.append(navbar_tr)
top_navbar = navbar_t
bottom_navbar = copy.copy(navbar_t)
# print "\n%s\n" % etree.tostring(navbar_t, pretty_print=True)
# Build the page
head = HEAD(TITLE(feed.title))
if style:
head.append(STYLE(style, type='text/css'))
if extra_css:
head.append(STYLE(extra_css, type='text/css'))
body = BODY()
div = DIV(
top_navbar,
H2(feed.title, CLASS('feed_title'))
)
body.append(div)
if getattr(feed, 'image', None):
div.append(DIV(IMG(
alt=feed.image_alt if feed.image_alt else '',
src=feed.image_url
),
CLASS('calibre_feed_image')))
if getattr(feed, 'description', None):
d = DIV(clean_xml_chars(feed.description), CLASS('calibre_feed_description',
'calibre_rescale_80'))
d.append(BR())
div.append(d)
for i, article in enumerate(feed.articles):
if not getattr(article, 'downloaded', False):
continue
div_td = DIV(CLASS('article_summary'),
A(article.title, CLASS('summary_headline','calibre_rescale_120',
href=article.url)))
if article.author:
div_td.append(DIV(article.author,
CLASS('summary_byline', 'calibre_rescale_100')))
if article.summary:
div_td.append(DIV(cutoff(article.text_summary),
CLASS('summary_text', 'calibre_rescale_100')))
div.append(div_td)
div.append(bottom_navbar)
self.root = HTML(head, body)
if self.html_lang:
self.root.set('lang', self.html_lang)
def convert_run(self, run):
ans = SPAN()
self.object_map[ans] = run
text = Text(ans, 'text', [])
for child in run:
if self.namespace.is_tag(child, 'w:t'):
if not child.text:
continue
space = child.get(XML('space'), None)
preserve = False
ctext = child.text
if space != 'preserve':
# Remove leading and trailing whitespace. Word ignores
# leading and trailing whitespace without preserve
ctext = ctext.strip(' \n\r\t')
# Only use a <span> with white-space:pre-wrap if this element
# actually needs it, i.e. if it has more than one
# consecutive space or it has newlines or tabs.
multi_spaces = self.ms_pat.search(ctext) is not None
preserve = multi_spaces or self.ws_pat.search(ctext) is not None
if preserve:
text.add_elem(SPAN(ctext, style="white-space:pre-wrap"))
ans.append(text.elem)
else:
text.buf.append(ctext)
elif self.namespace.is_tag(child, 'w:cr'):
text.add_elem(BR())
ans.append(text.elem)
elif self.namespace.is_tag(child, 'w:br'):
typ = self.namespace.get(child, 'w:type')
if typ in {'column', 'page'}:
br = BR(style='page-break-after:always')
else:
clear = child.get('clear', None)
if clear in {'all', 'left', 'right'}:
br = BR(style='clear:%s'%('both' if clear == 'all' else clear))
else:
br = BR()
text.add_elem(br)
ans.append(text.elem)
elif self.namespace.is_tag(child, 'w:drawing') or self.namespace.is_tag(child, 'w:pict'):
for img in self.images.to_html(child, self.current_page, self.docx, self.dest_dir):
text.add_elem(img)
ans.append(text.elem)
elif self.namespace.is_tag(child, 'w:footnoteReference') or self.namespace.is_tag(child, 'w:endnoteReference'):
anchor, name = self.footnotes.get_ref(child)
if anchor and name:
l = A(SUP(name, id='back_%s' % anchor), href='#' + anchor, title=name)
l.set('class', 'noteref')
text.add_elem(l)
ans.append(text.elem)
elif self.namespace.is_tag(child, 'w:tab'):
spaces = int(math.ceil((self.settings.default_tab_stop / 36) * 6))
text.add_elem(SPAN(NBSP * spaces))
ans.append(text.elem)
ans[-1].set('class', 'tab')
elif self.namespace.is_tag(child, 'w:noBreakHyphen'):
text.buf.append('\u2011')
elif self.namespace.is_tag(child, 'w:softHyphen'):
text.buf.append('\u00ad')
if text.buf:
setattr(text.elem, text.attr, ''.join(text.buf))
style = self.styles.resolve_run(run)
if style.vert_align in {'superscript', 'subscript'}:
ans.tag = 'sub' if style.vert_align == 'subscript' else 'sup'
if style.lang is not inherit:
lang = html_lang(style.lang)
if lang is not None and lang != self.doc_lang:
ans.set('lang', lang)
if style.rtl is True:
ans.set('dir', 'rtl')
if is_symbol_font(style.font_family):
for elem in text:
if elem.text:
elem.text = map_symbol_text(elem.text, style.font_family)
if elem.tail:
elem.tail = map_symbol_text(elem.tail, style.font_family)
style.font_family = 'sans-serif'
return ans
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'
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()
