def figure_replacement(self, key, value, format, metadata):
"""Replace figures with appropriate representation.
This works with Figure, which is our special type for images
with attributes. This allows us to set an id in the attributes.
The other way of doing it would be to pull out a '\label{(.*)}'
from the caption of an Image and use that to update the references.
"""
caption, (filename, alt), attr = value
if 'unnumbered' in attr[1]:
fcaption = caption
else:
self.fig_replacement_count += 1
if not attr[0]:
attr[0] = self.auto_fig_id(self.fig_replacement_count)
ref = self.references[attr[0]]
if caption:
fcaption = [
pf.Str('Figure'),
pf.Space(),
pf.Str(str(ref['id']) + ':'),
pf.Space()
] + caption
else:
fcaption = [
pf.Str('Figure'),
pf.Space(),
pf.Str(str(ref['id']))
]
# if 'figure' not in attr[1]:
# # FIXME: Currently adding this in html_figure() and html5_figure()
# attr[1].append('figure')
if format == 'latex' or format == 'beamer':
return latex_figure(attr, filename, caption, alt)
elif format == 'html':
return html_figure(attr, filename, fcaption, alt)
elif format == 'html5':
return html5_figure(attr, filename, fcaption, alt)
elif format == 'markdown':
return markdown_figure(attr, filename, fcaption, alt)
else:
image = pf.Image(attr, fcaption, (filename, alt))
return pf.Para([image])
def process_buffer(buffer, new_value, item=None):
if buffer:
text = ''.join(buffer)
try:
translated_text = translate.translate(text)
except TypeError:
translated_text = text
except json.decoder.JSONDecodeError as e:
print('Failed to translate', str(e), file=sys.stderr)
sys.exit(1)
debug(f'Translate: "{text}" -> "{translated_text}"')
if text and text[0].isupper() and not translated_text[0].isupper():
translated_text = translated_text[0].upper() + translated_text[1:]
if text.startswith(' ') and not translated_text.startswith(' '):
translated_text = ' ' + translated_text
if text.endswith(' ') and not translated_text.endswith(' '):
translated_text = translated_text + ' '
for token in translated_text.split(' '):
new_value.append(pandocfilters.Str(token))
new_value.append(pandocfilters.Space())
if item is None and len(new_value):
new_value.pop(len(new_value) - 1)
else:
new_value[-1] = item
elif item:
new_value.append(item)
def convert_internal_refs(self, key, value, format, metadata):
"""Convert all internal links from '#blah' into format
specified in self.replacements.
"""
if key != 'Cite':
return None
citations, inlines = value
if len(citations) > 1:
'''
Note: Need to check that *all* of the citations in a
multicitation are in the reference list. If not, the citation
is bibliographic, and we want LaTeX to handle it, so just
return unmodified.
'''
for citation in citations:
if citation['citationId'] not in self.references:
return
return self.convert_multiref(key, value, format, metadata)
else:
citation = citations[0]
prefix = citation['citationPrefix']
suffix = citation['citationSuffix']
label = citation['citationId']
if label not in self.references:
return
if prefix:
prefix += [pf.Space()]
rtype = self.references[label]['type']
n = self.references[label]['id']
text = self.replacements[rtype].format(n)
if format in ['latex', 'beamer'] and self.autoref:
link = pf.RawInline('latex',
'\\cref{{{label}}}'.format(label=label))
return prefix + [link] + suffix
elif format in ['latex', 'beamer'] and not self.autoref:
link = pf.RawInline('latex',
'\\ref{{{label}}}'.format(label=label))
return prefix + [link] + suffix
else:
link = pf.Link(["", [], []], [pf.Str(text)], ('#' + label, ''))
return prefix + [link] + suffix
def process_buffer(buffer, new_value, item=None, is_header=False):
if buffer:
text = ''.join(buffer)
try:
translated_text = translate.translate(text)
except TypeError:
translated_text = text
except json.decoder.JSONDecodeError as e:
print('Failed to translate', str(e), file=sys.stderr)
sys.exit(1)
debug(f'Translate: "{text}" -> "{translated_text}"')
if text and text[0].isupper() and not translated_text[0].isupper():
translated_text = translated_text[0].upper() + translated_text[1:]
if text.startswith(' ') and not translated_text.startswith(' '):
translated_text = ' ' + translated_text
if text.endswith(' ') and not translated_text.endswith(' '):
translated_text = translated_text + ' '
if is_header and translated_text.endswith('.'):
translated_text = translated_text.rstrip('.')
title_case = is_header and translate.default_target_language == 'en' and text[
0].isupper()
title_case_whitelist = {
'a', 'an', 'the', 'and', 'or', 'that', 'of', 'on', 'for', 'from',
'with', 'to', 'in'
}
is_first_iteration = True
for token in translated_text.split(' '):
if title_case and token.isascii() and not token.isupper():
if len(token) > 1 and token.lower(
) not in title_case_whitelist:
token = token[0].upper() + token[1:]
elif not is_first_iteration:
token = token.lower()
is_first_iteration = False
new_value.append(pandocfilters.Str(token))
new_value.append(pandocfilters.Space())
if item is None and len(new_value):
new_value.pop(len(new_value) - 1)
else:
new_value[-1] = item
elif item:
new_value.append(item)
def tableattrs_replacement(self, key, value, format, metadata):
"""Replace TableAttrs with appropriate representation.
TableAttrs is our special type for tables with attributes,
allowing us to set an id in the attributes.
"""
caption, alignment, size, headers, rows, (id, classes, kvs) = value
if 'unnumbered' in classes:
fcaption = caption
else:
self.table_replacement_count += 1
if not id:
id = self.auto_table_id(self.table_replacement_count)
ref = self.references[id]
if caption:
fcaption = [
pf.Str('Table'),
pf.Space(),
pf.Str(str(ref['id']) + ':'),
pf.Space()
] + caption
else:
fcaption = [
pf.Str('Table'),
pf.Space(),
pf.Str(str(ref['id']))
]
if format == 'latex' or format == 'beamer':
return latex_table(caption, alignment, size, headers, rows, id,
classes, kvs)
else:
return pf.Div([id, classes, kvs],
[pf.Table(fcaption, alignment, size, headers, rows)])
def process_buffer(buffer, new_value, item=None, is_header=False):
if buffer:
text = ''.join(buffer)
try:
translated_text = translate.translate(text)
except TypeError:
translated_text = text
except json.decoder.JSONDecodeError as e:
print('Failed to translate', str(e), file=sys.stderr)
sys.exit(1)
debug(f'Translate: "{text}" -> "{translated_text}"')
if text and text[0].isupper() and not translated_text[0].isupper():
translated_text = translated_text.capitalize()
if text.startswith(' ') and not translated_text.startswith(' '):
translated_text = ' ' + translated_text
if text.endswith(' ') and not translated_text.endswith(' '):
translated_text = translated_text + ' '
title_case = is_header and translate.default_target_language == 'en' and text[
0].isupper()
title_case_whitelist = {'a', 'an', 'the', 'and', 'or'}
for token in translated_text.split(' '):
if title_case and not token.isupper():
if token not in title_case_whitelist:
token = token.capitalize()
new_value.append(pandocfilters.Str(token))
new_value.append(pandocfilters.Space())
if item is None and len(new_value):
new_value.pop(len(new_value) - 1)
else:
new_value[-1] = item
elif item:
new_value.append(item)
def translate_filter(key, value, _format, _):
if key not in ['Space', 'Str']:
debug(key, value)
try:
cls = getattr(pandocfilters, key)
except AttributeError:
return
if key == 'Para' and value:
marker = value[0].get('c')
if isinstance(marker,
str) and marker.startswith('!!!') and len(value) > 2:
# Admonition case
if marker != '!!!':
# Lost space after !!! case
value.insert(1, pandocfilters.Str(marker[3:]))
value.insert(1, pandocfilters.Space())
value[0]['c'] = '!!!'
admonition_value = []
remaining_para_value = []
in_admonition = True
for item in value:
if in_admonition:
if item.get('t') == 'SoftBreak':
in_admonition = False
else:
admonition_value.append(item)
else:
remaining_para_value.append(item)
break_value = [
pandocfilters.LineBreak(),
pandocfilters.Str(' ' * 4)
]
if admonition_value[-1].get('t') == 'Quoted':
text = process_sentence(admonition_value[-1]['c'][-1])
text[0]['c'] = '"' + text[0]['c']
text[-1]['c'] = text[-1]['c'] + '"'
admonition_value.pop(-1)
admonition_value += text
else:
text = admonition_value[-1].get('c')
if text:
text = translate(text[0].upper() + text[1:])
admonition_value.append(pandocfilters.Space())
admonition_value.append(pandocfilters.Str(f'"{text}"'))
return cls(admonition_value + break_value +
process_sentence(remaining_para_value))
else:
return cls(process_sentence(value))
elif key == 'Plain' or key == 'Strong' or key == 'Emph':
return cls(process_sentence(value))
elif key == 'Link':
try:
# Plain links case
if value[2][0] == value[1][0].get('c'):
return pandocfilters.Str(value[2][0])
except IndexError:
pass
value[1] = process_sentence(value[1])
return cls(*value)
elif key == 'Header':
# TODO: title case header in en
if '_' not in value[1][0]: # Preserve some manually specified anchors
value[1][0] = slugify.slugify(value[1][0],
separator='-',
word_boundary=True,
save_order=True)
value[2] = process_sentence(value[2])
return cls(*value)
elif key == 'SoftBreak':
return pandocfilters.LineBreak()
return
def translate_filter(key, value, _format, _):
if key not in ['Space', 'Str']:
debug(key, value)
try:
cls = getattr(pandocfilters, key)
except AttributeError:
return
if key == 'Para' and value:
marker = value[0].get('c')
if isinstance(marker, str) and marker.startswith('!!!') and len(value) > 2:
# Admonition case
if marker != '!!!':
# Lost space after !!! case
value.insert(1, pandocfilters.Str(marker[3:]))
value.insert(1, pandocfilters.Space())
value[0]['c'] = '!!!'
admonition_value = []
remaining_para_value = []
in_admonition = True
break_value = [pandocfilters.LineBreak(), pandocfilters.Str(' ' * 4)]
for item in value:
if in_admonition:
if item.get('t') == 'SoftBreak':
in_admonition = False
else:
admonition_value.append(item)
else:
if item.get('t') == 'SoftBreak':
remaining_para_value += break_value
else:
remaining_para_value.append(item)
if admonition_value[-1].get('t') == 'Quoted':
text = process_sentence(admonition_value[-1]['c'][-1])
text[0]['c'] = '"' + text[0]['c']
text[-1]['c'] = text[-1]['c'] + '"'
admonition_value.pop(-1)
admonition_value += text
else:
text = admonition_value[-1].get('c')
if text:
text = translate.translate(text[0].upper() + text[1:])
admonition_value.append(pandocfilters.Space())
admonition_value.append(pandocfilters.Str(f'"{text}"'))
return cls(admonition_value + break_value + process_sentence(remaining_para_value))
else:
return cls(process_sentence(value))
elif key == 'Plain' or key == 'Strong' or key == 'Emph':
return cls(process_sentence(value))
elif key == 'Link':
try:
# Plain links case
if value[2][0] == value[1][0].get('c'):
return pandocfilters.Str(value[2][0])
except IndexError:
pass
value[1] = process_sentence(value[1])
href = value[2][0]
if not (href.startswith('http') or href.startswith('#')):
anchor = None
attempts = 10
if '#' in href:
href, anchor = href.split('#', 1)
if href.endswith('.md') and not href.startswith('/'):
parts = [part for part in os.environ['INPUT'].split('/') if len(part) == 2]
lang = parts[-1]
script_path = os.path.dirname(__file__)
base_path = os.path.abspath(f'{script_path}/../../{lang}')
href = os.path.join(
os.path.relpath(base_path, os.path.dirname(os.environ['INPUT'])),
os.path.relpath(href, base_path)
)
if anchor:
href = f'{href}#{anchor}'
value[2][0] = href
return cls(*value)
elif key == 'Header':
if value[1][0].islower() and '_' not in value[1][0]: # Preserve some manually specified anchors
value[1][0] = slugify.slugify(value[1][0], separator='-', word_boundary=True, save_order=True)
# TODO: title case header in en
value[2] = process_sentence(value[2], is_header=True)
return cls(*value)
elif key == 'SoftBreak':
return pandocfilters.LineBreak()
return
def strip_code(key, value, format, meta):
if key in ('CodeBlock', 'Header'):
return pf.Null()
elif key in ('DisplayMath'):
return pf.Space()
