def resolve_paragraph(self, p):
ans = self.para_cache.get(p, None)
if ans is None:
ans = self.para_cache[p] = ParagraphStyle()
ans.style_name = None
direct_formatting = None
for pPr in XPath('./w:pPr')(p):
ps = ParagraphStyle(pPr)
if direct_formatting is None:
direct_formatting = ps
else:
direct_formatting.update(ps)
if direct_formatting is None:
direct_formatting = ParagraphStyle()
parent_styles = []
if self.default_paragraph_style is not None:
parent_styles.append(self.default_paragraph_style)
ts = self.tables.para_style(p)
if ts is not None:
parent_styles.append(ts)
default_para = self.default_styles.get('paragraph', None)
if direct_formatting.linked_style is not None:
ls = self.get(direct_formatting.linked_style)
if ls is not None:
ans.style_name = ls.name
ps = ls.paragraph_style
if ps is not None:
parent_styles.append(ps)
if ls.character_style is not None:
self.para_char_cache[p] = ls.character_style
elif default_para is not None:
if default_para.paragraph_style is not None:
parent_styles.append(default_para.paragraph_style)
if default_para.character_style is not None:
self.para_char_cache[p] = default_para.character_style
is_numbering = direct_formatting.numbering is not inherit
if is_numbering:
num_id, lvl = direct_formatting.numbering
if num_id is not None:
p.set('calibre_num_id', '%s:%s' % (lvl, num_id))
if num_id is not None and lvl is not None:
ps = self.numbering.get_para_style(num_id, lvl)
if ps is not None:
parent_styles.append(ps)
for attr in ans.all_properties:
if not (is_numbering and attr
== 'text_indent'): # skip text-indent for lists
setattr(
ans, attr,
self.para_val(parent_styles, direct_formatting, attr))
ans.linked_style = direct_formatting.linked_style
return ans
def resolve_based_on(self, parent):
if parent.table_style is not None:
if self.table_style is None:
self.table_style = TableStyle(self.namespace)
self.table_style.resolve_based_on(parent.table_style)
if parent.paragraph_style is not None:
if self.paragraph_style is None:
self.paragraph_style = ParagraphStyle(self.namespace)
self.paragraph_style.resolve_based_on(parent.paragraph_style)
if parent.character_style is not None:
if self.character_style is None:
self.character_style = RunStyle(self.namespace)
self.character_style.resolve_based_on(parent.character_style)
def __init__(self, namespace, tblPr=None):
self.namespace = namespace
if tblPr is None:
for p in self.all_properties:
setattr(self, p, inherit)
else:
self.overrides = inherit
self.bidi = binary_property(tblPr, 'bidiVisual', namespace.XPath,
namespace.get)
for x in ('width', 'float', 'padding', 'shd', 'justification',
'spacing', 'indent', 'borders', 'band_size', 'look'):
f = globals()['read_%s' % x]
f(tblPr, self, self.namespace.XPath, self.namespace.get)
parent = tblPr.getparent()
if self.namespace.is_tag(parent, 'w:style'):
self.overrides = {}
for tblStylePr in self.namespace.XPath(
'./w:tblStylePr[@w:type]')(parent):
otype = self.namespace.get(tblStylePr, 'w:type')
orides = self.overrides[otype] = {}
for tblPr in self.namespace.XPath('./w:tblPr')(tblStylePr):
orides['table'] = TableStyle(self.namespace, tblPr)
for trPr in self.namespace.XPath('./w:trPr')(tblStylePr):
orides['row'] = RowStyle(self.namespace, trPr)
for tcPr in self.namespace.XPath('./w:tcPr')(tblStylePr):
orides['cell'] = CellStyle(self.namespace, tcPr)
for pPr in self.namespace.XPath('./w:pPr')(tblStylePr):
orides['para'] = ParagraphStyle(self.namespace, pPr)
for rPr in self.namespace.XPath('./w:rPr')(tblStylePr):
orides['run'] = RunStyle(self.namespace, rPr)
self._css = None
def __init__(self, tblPr=None):
if tblPr is None:
for p in self.all_properties:
setattr(self, p, inherit)
else:
self.overrides = inherit
for x in ('width', 'padding', 'shd', 'justification', 'spacing',
'indent', 'borders', 'band_size', 'look'):
f = globals()['read_%s' % x]
f(tblPr, self)
parent = tblPr.getparent()
if is_tag(parent, 'w:style'):
self.overrides = {}
for tblStylePr in XPath('./w:tblStylePr[@w:type]')(parent):
otype = get(tblStylePr, 'w:type')
orides = self.overrides[otype] = {}
for tblPr in XPath('./w:tblPr')(tblStylePr):
orides['table'] = TableStyle(tblPr)
for trPr in XPath('./w:trPr')(tblStylePr):
orides['row'] = RowStyle(trPr)
for tcPr in XPath('./w:tcPr')(tblStylePr):
orides['cell'] = tcPr
for pPr in XPath('./w:pPr')(tblStylePr):
orides['para'] = ParagraphStyle(pPr)
for rPr in XPath('./w:rPr')(tblStylePr):
orides['run'] = RunStyle(rPr)
def read_from_xml(self, lvl, override=False):
XPath, get = self.namespace.XPath, self.namespace.get
for lr in XPath('./w:lvlRestart[@w:val]')(lvl):
try:
self.restart = int(get(lr, 'w:val'))
except (TypeError, ValueError):
pass
for lr in XPath('./w:start[@w:val]')(lvl):
try:
self.start = int(get(lr, 'w:val'))
except (TypeError, ValueError):
pass
for rPr in XPath('./w:rPr')(lvl):
ps = RunStyle(self.namespace, rPr)
if self.character_style is None:
self.character_style = ps
else:
self.character_style.update(ps)
lt = None
for lr in XPath('./w:lvlText[@w:val]')(lvl):
lt = get(lr, 'w:val')
for lr in XPath('./w:numFmt[@w:val]')(lvl):
val = get(lr, 'w:val')
if val == 'bullet':
self.is_numbered = False
cs = self.character_style
if lt in {
'\uf0a7', 'o'
} or (cs is not None and cs.font_family is not inherit
and cs.font_family.lower() in {'wingdings', 'symbol'}):
self.fmt = {
'\uf0a7': 'square',
'o': 'circle'
}.get(lt, 'disc')
else:
self.bullet_template = lt
for lpid in XPath('./w:lvlPicBulletId[@w:val]')(lvl):
self.pic_id = get(lpid, 'w:val')
else:
self.is_numbered = True
self.fmt = STYLE_MAP.get(val, 'decimal')
if lt and re.match(r'%\d+\.$', lt) is None:
self.num_template = lt
for lr in XPath('./w:pStyle[@w:val]')(lvl):
self.para_link = get(lr, 'w:val')
for pPr in XPath('./w:pPr')(lvl):
ps = ParagraphStyle(self.namespace, pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
def resolve_based_on(self, parent):
if parent.table_style is not None:
if self.table_style is None:
self.table_style = TableStyle(self.namespace)
self.table_style.resolve_based_on(parent.table_style)
if parent.paragraph_style is not None:
if self.paragraph_style is None:
self.paragraph_style = ParagraphStyle(self.namespace)
self.paragraph_style.resolve_based_on(parent.paragraph_style)
if parent.character_style is not None:
if self.character_style is None:
self.character_style = RunStyle(self.namespace)
self.character_style.resolve_based_on(parent.character_style)
def __init__(self, namespace, elem):
self.namespace = namespace
self.name_path = namespace.XPath('./w:name[@w:val]')
self.based_on_path = namespace.XPath('./w:basedOn[@w:val]')
self.resolved = False
self.style_id = namespace.get(elem, 'w:styleId')
self.style_type = namespace.get(elem, 'w:type')
names = self.name_path(elem)
self.name = namespace.get(names[-1], 'w:val') if names else None
based_on = self.based_on_path(elem)
self.based_on = namespace.get(based_on[0],
'w:val') if based_on else None
if self.style_type == 'numbering':
self.based_on = None
self.is_default = namespace.get(elem,
'w:default') in {'1', 'on', 'true'}
self.paragraph_style = self.character_style = self.table_style = None
if self.style_type in {'paragraph', 'character', 'table'}:
if self.style_type == 'table':
for tblPr in namespace.XPath('./w:tblPr')(elem):
ts = TableStyle(namespace, tblPr)
if self.table_style is None:
self.table_style = ts
else:
self.table_style.update(ts)
if self.style_type in {'paragraph', 'table'}:
for pPr in namespace.XPath('./w:pPr')(elem):
ps = ParagraphStyle(namespace, pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
for rPr in namespace.XPath('./w:rPr')(elem):
rs = RunStyle(namespace, rPr)
if self.character_style is None:
self.character_style = rs
else:
self.character_style.update(rs)
if self.style_type in {'numbering', 'paragraph'}:
self.numbering_style_link = None
for x in namespace.XPath('./w:pPr/w:numPr/w:numId[@w:val]')(elem):
self.numbering_style_link = namespace.get(x, 'w:val')
def read_from_xml(self, lvl, override=False):
for lr in XPath('./w:lvlRestart[@w:val]')(lvl):
try:
self.restart = int(get(lr, 'w:val'))
except (TypeError, ValueError):
pass
for lr in XPath('./w:start[@w:val]')(lvl):
try:
self.start = int(get(lr, 'w:val'))
except (TypeError, ValueError):
pass
lt = None
for lr in XPath('./w:lvlText[@w:val]')(lvl):
lt = get(lr, 'w:val')
for lr in XPath('./w:numFmt[@w:val]')(lvl):
val = get(lr, 'w:val')
if val == 'bullet':
self.is_numbered = False
self.fmt = {'\uf0a7': 'square', 'o': 'circle'}.get(lt, 'disc')
for lpid in XPath('./w:lvlPicBulletId[@w:val]')(lvl):
self.pic_id = get(lpid, 'w:val')
else:
self.is_numbered = True
self.fmt = STYLE_MAP.get(val, 'decimal')
if lt and re.match(r'%\d+\.$', lt) is None:
self.num_template = lt
for lr in XPath('./w:pStyle[@w:val]')(lvl):
self.para_link = get(lr, 'w:val')
for pPr in XPath('./w:pPr')(lvl):
ps = ParagraphStyle(pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
for rPr in XPath('./w:rPr')(lvl):
ps = RunStyle(rPr)
if self.character_style is None:
self.character_style = ps
else:
self.character_style.update(ps)
def __call__(self, root, fonts, theme):
self.fonts, self.theme = fonts, theme
self.default_paragraph_style = self.default_character_style = None
if root is not None:
for s in self.namespace.XPath('//w:style')(root):
s = Style(self.namespace, s)
if s.style_id:
self.id_map[s.style_id] = s
if s.is_default:
self.default_styles[s.style_type] = s
if getattr(s, 'numbering_style_link', None) is not None:
self.numbering_style_links[
s.style_id] = s.numbering_style_link
for dd in self.namespace.XPath('./w:docDefaults')(root):
for pd in self.namespace.XPath('./w:pPrDefault')(dd):
for pPr in self.namespace.XPath('./w:pPr')(pd):
ps = ParagraphStyle(self.namespace, pPr)
if self.default_paragraph_style is None:
self.default_paragraph_style = ps
else:
self.default_paragraph_style.update(ps)
for pd in self.namespace.XPath('./w:rPrDefault')(dd):
for pPr in self.namespace.XPath('./w:rPr')(pd):
ps = RunStyle(self.namespace, pPr)
if self.default_character_style is None:
self.default_character_style = ps
else:
self.default_character_style.update(ps)
def resolve(s, p):
if p is not None:
if not p.resolved:
resolve(p, self.get(p.based_on))
s.resolve_based_on(p)
s.resolved = True
for s in self:
if not s.resolved:
resolve(s, self.get(s.based_on))
class Style(object):
'''
Class representing a <w:style> element. Can contain block, character, etc. styles.
'''
def __init__(self, namespace, elem):
self.namespace = namespace
self.name_path = namespace.XPath('./w:name[@w:val]')
self.based_on_path = namespace.XPath('./w:basedOn[@w:val]')
self.resolved = False
self.style_id = namespace.get(elem, 'w:styleId')
self.style_type = namespace.get(elem, 'w:type')
names = self.name_path(elem)
self.name = namespace.get(names[-1], 'w:val') if names else None
based_on = self.based_on_path(elem)
self.based_on = namespace.get(based_on[0], 'w:val') if based_on else None
if self.style_type == 'numbering':
self.based_on = None
self.is_default = namespace.get(elem, 'w:default') in {'1', 'on', 'true'}
self.paragraph_style = self.character_style = self.table_style = None
if self.style_type in {'paragraph', 'character', 'table'}:
if self.style_type == 'table':
for tblPr in namespace.XPath('./w:tblPr')(elem):
ts = TableStyle(namespace, tblPr)
if self.table_style is None:
self.table_style = ts
else:
self.table_style.update(ts)
if self.style_type in {'paragraph', 'table'}:
for pPr in namespace.XPath('./w:pPr')(elem):
ps = ParagraphStyle(namespace, pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
for rPr in namespace.XPath('./w:rPr')(elem):
rs = RunStyle(namespace, rPr)
if self.character_style is None:
self.character_style = rs
else:
self.character_style.update(rs)
if self.style_type in {'numbering', 'paragraph'}:
self.numbering_style_link = None
for x in namespace.XPath('./w:pPr/w:numPr/w:numId[@w:val]')(elem):
self.numbering_style_link = namespace.get(x, 'w:val')
def resolve_based_on(self, parent):
if parent.table_style is not None:
if self.table_style is None:
self.table_style = TableStyle(self.namespace)
self.table_style.resolve_based_on(parent.table_style)
if parent.paragraph_style is not None:
if self.paragraph_style is None:
self.paragraph_style = ParagraphStyle(self.namespace)
self.paragraph_style.resolve_based_on(parent.paragraph_style)
if parent.character_style is not None:
if self.character_style is None:
self.character_style = RunStyle(self.namespace)
self.character_style.resolve_based_on(parent.character_style)
class Style(object):
'''
Class representing a <w:style> element. Can contain block, character, etc. styles.
'''
def __init__(self, namespace, elem):
self.namespace = namespace
self.name_path = namespace.XPath('./w:name[@w:val]')
self.based_on_path = namespace.XPath('./w:basedOn[@w:val]')
self.resolved = False
self.style_id = namespace.get(elem, 'w:styleId')
self.style_type = namespace.get(elem, 'w:type')
names = self.name_path(elem)
self.name = namespace.get(names[-1], 'w:val') if names else None
based_on = self.based_on_path(elem)
self.based_on = namespace.get(based_on[0],
'w:val') if based_on else None
if self.style_type == 'numbering':
self.based_on = None
self.is_default = namespace.get(elem,
'w:default') in {'1', 'on', 'true'}
self.paragraph_style = self.character_style = self.table_style = None
if self.style_type in {'paragraph', 'character', 'table'}:
if self.style_type == 'table':
for tblPr in namespace.XPath('./w:tblPr')(elem):
ts = TableStyle(namespace, tblPr)
if self.table_style is None:
self.table_style = ts
else:
self.table_style.update(ts)
if self.style_type in {'paragraph', 'table'}:
for pPr in namespace.XPath('./w:pPr')(elem):
ps = ParagraphStyle(namespace, pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
for rPr in namespace.XPath('./w:rPr')(elem):
rs = RunStyle(namespace, rPr)
if self.character_style is None:
self.character_style = rs
else:
self.character_style.update(rs)
if self.style_type in {'numbering', 'paragraph'}:
self.numbering_style_link = None
for x in namespace.XPath('./w:pPr/w:numPr/w:numId[@w:val]')(elem):
self.numbering_style_link = namespace.get(x, 'w:val')
def resolve_based_on(self, parent):
if parent.table_style is not None:
if self.table_style is None:
self.table_style = TableStyle(self.namespace)
self.table_style.resolve_based_on(parent.table_style)
if parent.paragraph_style is not None:
if self.paragraph_style is None:
self.paragraph_style = ParagraphStyle(self.namespace)
self.paragraph_style.resolve_based_on(parent.paragraph_style)
if parent.character_style is not None:
if self.character_style is None:
self.character_style = RunStyle(self.namespace)
self.character_style.resolve_based_on(parent.character_style)
def resolve_paragraph(self, p):
ans = self.para_cache.get(p, None)
if ans is None:
linked_style = None
ans = self.para_cache[p] = ParagraphStyle(self.namespace)
ans.style_name = None
direct_formatting = None
is_section_break = False
for pPr in self.namespace.XPath('./w:pPr')(p):
ps = ParagraphStyle(self.namespace, pPr)
if direct_formatting is None:
direct_formatting = ps
else:
direct_formatting.update(ps)
if self.namespace.XPath('./w:sectPr')(pPr):
is_section_break = True
if direct_formatting is None:
direct_formatting = ParagraphStyle(self.namespace)
parent_styles = []
if self.default_paragraph_style is not None:
parent_styles.append(self.default_paragraph_style)
ts = self.tables.para_style(p)
if ts is not None:
parent_styles.append(ts)
default_para = self.default_styles.get('paragraph', None)
if direct_formatting.linked_style is not None:
ls = linked_style = self.get(direct_formatting.linked_style)
if ls is not None:
ans.style_name = ls.name
ps = ls.paragraph_style
if ps is not None:
parent_styles.append(ps)
if ls.character_style is not None:
self.para_char_cache[p] = ls.character_style
elif default_para is not None:
if default_para.paragraph_style is not None:
parent_styles.append(default_para.paragraph_style)
if default_para.character_style is not None:
self.para_char_cache[p] = default_para.character_style
def has_numbering(block_style):
num_id, lvl = getattr(block_style, 'numbering_id',
inherit), getattr(
block_style, 'numbering_level',
inherit)
return num_id is not None and num_id is not inherit and lvl is not None and lvl is not inherit
is_numbering = has_numbering(direct_formatting)
is_section_break = is_section_break and not self.namespace.XPath(
'./w:r')(p)
if is_numbering and not is_section_break:
num_id, lvl = direct_formatting.numbering_id, direct_formatting.numbering_level
p.set('calibre_num_id', '%s:%s' % (lvl, num_id))
ps = self.numbering.get_para_style(num_id, lvl)
if ps is not None:
parent_styles.append(ps)
if (not is_numbering and not is_section_break
and linked_style is not None
and has_numbering(linked_style.paragraph_style)):
num_id, lvl = linked_style.paragraph_style.numbering_id, linked_style.paragraph_style.numbering_level
p.set('calibre_num_id', '%s:%s' % (lvl, num_id))
is_numbering = True
ps = self.numbering.get_para_style(num_id, lvl)
if ps is not None:
parent_styles.append(ps)
for attr in ans.all_properties:
if not (is_numbering and attr
== 'text_indent'): # skip text-indent for lists
setattr(
ans, attr,
self.para_val(parent_styles, direct_formatting, attr))
ans.linked_style = direct_formatting.linked_style
return ans
class Style(object):
"""
Class representing a <w:style> element. Can contain block, character, etc. styles.
"""
name_path = XPath("./w:name[@w:val]")
based_on_path = XPath("./w:basedOn[@w:val]")
def __init__(self, elem):
self.resolved = False
self.style_id = get(elem, "w:styleId")
self.style_type = get(elem, "w:type")
names = self.name_path(elem)
self.name = get(names[-1], "w:val") if names else None
based_on = self.based_on_path(elem)
self.based_on = get(based_on[0], "w:val") if based_on else None
if self.style_type == "numbering":
self.based_on = None
self.is_default = get(elem, "w:default") in {"1", "on", "true"}
self.paragraph_style = self.character_style = self.table_style = None
if self.style_type in {"paragraph", "character", "table"}:
if self.style_type == "table":
for tblPr in XPath("./w:tblPr")(elem):
ts = TableStyle(tblPr)
if self.table_style is None:
self.table_style = ts
else:
self.table_style.update(ts)
if self.style_type in {"paragraph", "table"}:
for pPr in XPath("./w:pPr")(elem):
ps = ParagraphStyle(pPr)
if self.paragraph_style is None:
self.paragraph_style = ps
else:
self.paragraph_style.update(ps)
for rPr in XPath("./w:rPr")(elem):
rs = RunStyle(rPr)
if self.character_style is None:
self.character_style = rs
else:
self.character_style.update(rs)
if self.style_type == "numbering":
self.numbering_style_link = None
for x in XPath("./w:pPr/w:numPr/w:numId[@w:val]")(elem):
self.numbering_style_link = get(x, "w:val")
def resolve_based_on(self, parent):
if parent.table_style is not None:
if self.table_style is None:
self.table_style = TableStyle()
self.table_style.resolve_based_on(parent.table_style)
if parent.paragraph_style is not None:
if self.paragraph_style is None:
self.paragraph_style = ParagraphStyle()
self.paragraph_style.resolve_based_on(parent.paragraph_style)
if parent.character_style is not None:
if self.character_style is None:
self.character_style = RunStyle()
self.character_style.resolve_based_on(parent.character_style)