blacklight/madblog
blacklight
/
madblog
1
2
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Removed `alt` attribute from LaTeX rendered images. 

The `alt` contains the LaTeX formula rendered as Unicode characters, and
that may include non-standard characters that break the RSS feed encoding.
This commit is contained in:
Fabio Manganiello 2023-04-24 19:58:36 +02:00
parent cfd929f36f
commit 261e5bfc0e
1 changed files with 46 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

73
madblog/latex.py
View File

@ -31,11 +31,11 @@ def call(*args, **kwargs):
# Defines our basic inline image # Defines our basic inline image
img_expr = '<img class="latex inline math-%s" alt="%s" id="%s" src="data:image/png;base64,%s">' img_expr = '<img class="latex inline math-%s" id="%s" src="data:image/png;base64,%s">'
# Defines multiline expression image # Defines multiline expression image
multiline_img_expr = '''<div class="multiline-wrapper"> multiline_img_expr = """<div class="multiline-wrapper">
<img class="latex multiline math-%s" alt="%s" id="%s" src="data:image/png;base64,%s"></div>''' <img class="latex multiline math-%s" id="%s" src="data:image/png;base64,%s"></div>"""
# Base CSS template # Base CSS template
img_css = """<style scoped> img_css = """<style scoped>
@ -55,8 +55,8 @@ img.latex.inline {
</style>""" </style>"""
# Cache and temp file paths # Cache and temp file paths
tmpdir = tempfile.gettempdir() + '/markdown-latex' tmpdir = tempfile.gettempdir() + "/markdown-latex"
cache_file = tmpdir + '/latex.cache' cache_file = tmpdir + "/latex.cache"
class LaTeXPreprocessor(markdown.preprocessors.Preprocessor): class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
@ -75,16 +75,20 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
""" """
# Math TeX extraction regex # Math TeX extraction regex
math_extract_regex = re.compile(r'(.+?)((\\\(.+?\\\))|(\$\$\n.+?\n\$\$\n))(.*)', re.MULTILINE | re.DOTALL) math_extract_regex = re.compile(
r"(.+?)((\\\(.+?\\\))|(\$\$\n.+?\n\$\$\n))(.*)", re.MULTILINE | re.DOTALL
)
# Math TeX matching regex # Math TeX matching regex
math_match_regex = re.compile(r'\s*(\\\(.+?\\\))|(\$\$\n.+?\n\$\$\n)\s*', re.MULTILINE | re.DOTALL) math_match_regex = re.compile(
r"\s*(\\\(.+?\\\))|(\$\$\n.+?\n\$\$\n)\s*", re.MULTILINE | re.DOTALL
)
def __init__(self, *_, **__): def __init__(self, *_, **__):
if not os.path.isdir(tmpdir): if not os.path.isdir(tmpdir):
os.makedirs(tmpdir) os.makedirs(tmpdir)
try: try:
with open(cache_file, 'r') as f: with open(cache_file, "r") as f:
self.cached = json.load(f) self.cached = json.load(f)
except (IOError, json.JSONDecodeError): except (IOError, json.JSONDecodeError):
self.cached = {} self.cached = {}
@ -94,7 +98,8 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
("dvipng", "args"): "-q -T tight -bg Transparent -z 9 -D 200", ("dvipng", "args"): "-q -T tight -bg Transparent -z 9 -D 200",
("delimiters", "text"): "%", ("delimiters", "text"): "%",
("delimiters", "math"): "$", ("delimiters", "math"): "$",
("delimiters", "preamble"): "%%"} ("delimiters", "preamble"): "%%",
}
def _latex_to_base64(self, tex): def _latex_to_base64(self, tex):
"""Generates a base64 representation of TeX string""" """Generates a base64 representation of TeX string"""
@ -104,18 +109,24 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
with os.fdopen(tmp_file_fd, "w") as tmp_file: with os.fdopen(tmp_file_fd, "w") as tmp_file:
tmp_file.write(self.tex_preamble) tmp_file.write(self.tex_preamble)
tmp_file.write(tex) tmp_file.write(tex)
tmp_file.write('\n\\end{document}') tmp_file.write("\n\\end{document}")
# compile LaTeX document. A DVI file is created # compile LaTeX document. A DVI file is created
status = call(('latex -halt-on-error -output-directory={:s} {:s}' status = call(
.format(tmpdir, path)).split(), (
stdout=PIPE, timeout=10) "latex -halt-on-error -output-directory={:s} {:s}".format(tmpdir, path)
).split(),
stdout=PIPE,
timeout=10,
)
# clean up if the above failed # clean up if the above failed
if status: if status:
self._cleanup(path, err=True) self._cleanup(path, err=True)
raise Exception("Couldn't compile LaTeX document." + raise Exception(
"Please read '%s.log' for more detail." % path) "Couldn't compile LaTeX document."
+ "Please read '%s.log' for more detail." % path
)
# Run dvipng on the generated DVI file. Use tight bounding box. # Run dvipng on the generated DVI file. Use tight bounding box.
# Magnification is set to 1200 # Magnification is set to 1200
@ -129,8 +140,10 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
# clean up if we couldn't make the above work # clean up if we couldn't make the above work
if status: if status:
self._cleanup(path, err=True) self._cleanup(path, err=True)
raise Exception("Couldn't convert LaTeX to image." + raise Exception(
"Please read '%s.log' for more detail." % path) "Couldn't convert LaTeX to image."
+ "Please read '%s.log' for more detail." % path
)
# Read the png and encode the data # Read the png and encode the data
try: try:
@ -157,7 +170,7 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
def run(self, lines): def run(self, lines):
"""Parses the actual page""" """Parses the actual page"""
# Checks for the LaTeX header # Checks for the LaTeX header
use_latex = any(line == '[//]: # (latex: 1)' for line in lines) use_latex = any(line == "[//]: # (latex: 1)" for line in lines)
if not use_latex: if not use_latex:
return lines return lines
@ -165,7 +178,9 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
page = "\n".join(lines) page = "\n".join(lines)
# Adds a preamble mode # Adds a preamble mode
self.tex_preamble += self.config[("general", "preamble")] + "\n\\begin{document}\n" self.tex_preamble += (
self.config[("general", "preamble")] + "\n\\begin{document}\n"
)
# Figure out our text strings and math-mode strings # Figure out our text strings and math-mode strings
tex_expr = self.math_extract_regex.findall(page) tex_expr = self.math_extract_regex.findall(page)
@ -176,7 +191,7 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
# Parse the expressions # Parse the expressions
new_cache = {} new_cache = {}
new_page = '' new_page = ""
n_multiline_expressions = 0 n_multiline_expressions = 0
while page: while page:
@ -200,21 +215,25 @@ class LaTeXPreprocessor(markdown.preprocessors.Preprocessor):
new_cache[tex_hash] = data new_cache[tex_hash] = data
if is_multiline and n_multiline_expressions > 0: if is_multiline and n_multiline_expressions > 0:
new_page += '</p>' new_page += "</p>"
new_page += (multiline_img_expr if is_multiline else img_expr) % ('true', expr, tex_hash, data) new_page += (multiline_img_expr if is_multiline else img_expr) % (
"true",
tex_hash,
data,
)
if is_multiline: if is_multiline:
new_page += '<p>' new_page += "<p>"
n_multiline_expressions += 1 n_multiline_expressions += 1
page = m.group(5) page = m.group(5)
if n_multiline_expressions > 0: if n_multiline_expressions > 0:
new_page += '</p>' new_page += "</p>"
# Cache our data # Cache our data
self.cached.update(new_cache) self.cached.update(new_cache)
with open(cache_file, 'w') as f: with open(cache_file, "w") as f:
json.dump(self.cached, f) json.dump(self.cached, f)
# Make sure to re-split the lines # Make sure to re-split the lines
@ -230,7 +249,7 @@ class MarkdownLatex(markdown.Extension):
def extendMarkdown(self, md): def extendMarkdown(self, md):
md.preprocessors.register( md.preprocessors.register(
LaTeXPreprocessor(self), LaTeXPreprocessor(self),
'latex', "latex",
1, 1,
) )