def convert_to_body_resources(notebook_filename, exporter_class=PDFExporter):
"""Convert notebook to body and resources... replaces markdown local images on the way."""
## Read the actual notebook
notebook = nbformat.read(notebook_filename, as_version=4)
notebook, resources = preprocess_markdown_local_images(
notebook, notebook_filename)
# Overwrite article style
dl = DictLoader({
'article.tplx':
"""
% Default to the notebook output style
((* if not cell_style is defined *))
((* set cell_style = 'style_ipython.tplx' *))
((* endif *))
% Inherit from the specified cell style.
((* extends cell_style *))
%===============================================================================
% Latex Article
%===============================================================================
((* block docclass *))
% In case you want to change it
\documentclass{article}
((* endblock docclass *))
((* block header *))
((( super() )))
% Indentation, no indetation for paragraphs, but blank lines
\setlength{\parskip}{\medskipamount}
\setlength{\parindent}{0pt}
((* endblock header *))
"""
})
if exporter_class is None:
exporter = LatexExporter(extra_loaders=[dl])
else:
exporter = exporter_class(extra_loaders=[dl])
(body, resources) = exporter.from_notebook_node(notebook,
resources=resources)
return body, resources
def convert_to_body_resources(notebook_filename, exporter_class=PDFExporter):
"""Convert notebook to body and resources... replaces markdown local images on the way."""
## Read the actual notebook
notebook = nbformat.read(notebook_filename, as_version=4)
notebook, resources = preprocess_markdown_local_images(notebook, notebook_filename)
# Overwrite article style
dl = DictLoader({'article.tplx':
"""
% Default to the notebook output style
((* if not cell_style is defined *))
((* set cell_style = 'style_ipython.tplx' *))
((* endif *))
% Inherit from the specified cell style.
((* extends cell_style *))
%===============================================================================
% Latex Article
%===============================================================================
((* block docclass *))
% In case you want to change it
\documentclass{article}
((* endblock docclass *))
((* block header *))
((( super() )))
% Indentation, no indetation for paragraphs, but blank lines
\setlength{\parskip}{\medskipamount}
\setlength{\parindent}{0pt}
((* endblock header *))
"""})
if exporter_class is None:
exporter = LatexExporter(extra_loaders=[dl])
else:
exporter = exporter_class(extra_loaders=[dl])
(body, resources) = exporter.from_notebook_node(notebook,resources=resources)
return body, resources
def convert_to_body_resources(notebook_filename, bibtex_filename):
## Initializing resources to have correct output directory
notebook_name = notebook_filename.split('/')[-1].replace('.ipynb', '')
#see https://github.com/jupyter/nbconvert/blob/fcc3a831295b373a7a9ee5e8e0dea175475f8f26/nbconvert/nbconvertapp.py#L288
resources = {}
#resources['config_dir'] = self.config_dir
resources['unique_key'] = notebook_name
resources['output_files_dir'] = '%s_files' % notebook_name
own_notebook = nbformat.read(notebook_filename, as_version=4)
# replace cite keys by bibtex citekeys:
if bibtex_filename is not None and 'cite2c' in own_notebook['metadata']:
with open(bibtex_filename, 'r') as bibtex_file:
bibtex = bibtexparser.load(bibtex_file)
cite2_key_to_bibtex_key = dict()
for key, cite2c_entry in own_notebook['metadata']['cite2c']['citations'].iteritems():
equal_bibtex_entries = [b for b in bibtex.entries if cite2c_bibtex_equal(cite2c_entry,b)]
if (len(equal_bibtex_entries) == 1):
cite2_key_to_bibtex_key[key] = equal_bibtex_entries[0]['ID']
assert len(equal_bibtex_entries) < 2, ("expected at most "
"one equal bibtex entry, got {:s}".format(str(equal_bibtex_entries)))
for cell in own_notebook['cells']:
for cite2_key, bibtex_key in cite2_key_to_bibtex_key.iteritems():
if cell['cell_type'] == 'markdown':
cell['source'] = cell['source'].replace(cite2_key, bibtex_key)
# find imgs and convert to resources
resources['outputs'] = dict()
for cell in own_notebook['cells']:
if cell['cell_type'] == 'markdown':
img_urls_filenames = re.findall(r"<img.*src=\"([^>]*/([^\.]*\.[a-z]*)[^\"]*)\"[^>]*>[^<]*</img>", cell['source'])
for url, img_filename in img_urls_filenames:
response = requests.get(url, stream=True)
all_blocks = []
if not response.ok:
continue
for block in response.iter_content(1024):
all_blocks.append(block)
data = ''.join(all_blocks)
if img_filename.endswith('svg'):
svg_2_pdf = SVG2PDFPreprocessor()
pdfdata = svg_2_pdf.convert_figure(None, data)
pdfdata = base64.decodestring(pdfdata) # it is encoded by svg2pdfpreproc.. not sure if decoding is necessary
img_filename = img_filename.replace('.svg', '.pdf')
data = pdfdata
if img_filename.endswith('gif'):
jpgdata = gif_to_jpg(data)
img_filename = img_filename.replace('.gif', '.jpg')
data = jpgdata
resource_key = os.path.join(resources['output_files_dir'], img_filename)
resources['outputs'][resource_key] = data
# Replace the whole image tag by latex code with the correct filename
cell['source'] = re.sub(r"<img.*src=\"[^>]*/([^\.]*\.[a-z]*)[^>]*>[^<]*</img>",
"\\\\begin{center}\n" +
"\\\\adjustimage{max size={0.9\\linewidth}{0.9\\paperheight}}{" +
resources['output_files_dir'] + "/" +
r"\1" + # here is the filename
"}\n"+
"\\end{center}\n",
cell['source'])
cell['source'] = cell['source'].replace('.svg', '.pdf')
cell['source'] = cell['source'].replace('.gif', '.jpg')
# remove javascript/html outputs
for cell in own_notebook['cells']:
if cell['cell_type'] == 'code' and 'outputs' in cell:
cell['outputs'] = remove_javascript_html_outputs(cell['outputs'])
# do some custom replacements of html
for cell in own_notebook['cells']:
if cell['cell_type'] == 'markdown':
cell['source'] = cell['source'].replace('<span class="todecide">', '\\begin{comment}\nTODECIDE\n')
cell['source'] = cell['source'].replace('<span class="todo">', '\\begin{comment}\nTODO\n')
cell['source'] = cell['source'].replace('</span>', '\n\\end{comment}\n')
cell['source'] = cell['source'].replace("<div class=\"summary\">", "\\begin{keypointbox}")
cell['source'] = cell['source'].replace("</div>", "\\end{keypointbox}")
cell['source'] = cell['source'].replace("<li>", "\\item ")
cell['source'] = cell['source'].replace("</li>", "").replace("<ul>", "").replace("</ul>", "")
dl = DictLoader({'article.tplx':
"""
((*- extends 'base.tplx' -*))
((* block header *))
((* endblock header *))
% only part-document, not complete document, so call not base constructor, but the one above
((* block body *))
((( super.super() )))
((* endblock body *))
% is this removing code? unclear.. probaby removing stdout/stdin
((* block stream *))
((* endblock stream *))
% Remove execute result stuff
((* block execute_result scoped *))
((* endblock execute_result *))
((* macro draw_figure(filename) -*))
((* set filename = filename | posix_path *))
((*- block figure scoped -*))
%\\begin{figure}[ht]
\\begin{center}
\\adjustimage{max size={0.9\\linewidth}{0.9\\paperheight}}{((( filename )))}
\\end{center}
%\\end{figure}
%{ \\hspace*{\\fill} \\\\}
((*- endblock figure -*))
((*- endmacro *))
((* block markdowncell scoped *))
((( cell.source | citation2latex | strip_files_prefix | markdown2latex(extra_args=["--chapters"]) )))
((* endblock markdowncell *))
"""})
exportLatex = LatexExporter(extra_loaders=[dl])
(body, resources) = exportLatex.from_notebook_node(own_notebook,resources=resources)
# postprocess url links with footnotes
body = re.sub(r"(\\href{([^}]*)}{[^}]*})", r"\1\\footnote{\\url{\2}}", body)
return body, resources
