mirror/rendercv
1
0
Fork 0
mirror of https://github.com/rendercv/rendercv.git synced 2026-04-23 08:24:21 -04:00
Code Issues Packages Projects Releases Wiki Activity

fix characters are not escaped in LaTeX, leading to render errors #129

Browse Source
This commit is contained in:
Sina Atalay
2024-11-01 19:14:38 -04:00
parent b0c7d77e2c
commit b6200d8621
3 changed files with 22 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

34
rendercv/renderer/templater.py
View File

@@ -317,7 +317,7 @@ def revert_nested_latex_style_commands(latex_string: str) -> str:
return latex_string
def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
def escape_latex_characters(latex_string: str) -> str:
"""Escape $\\LaTeX$ characters in a string by adding a backslash before them.
Example:
@@ -329,8 +329,6 @@ def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
Args:
latex_string: The string to escape.
strict: Whether to escape all the special $\\LaTeX$ characters or not. If you
want to allow math input, set it to False.
Returns:
The escaped string.
@@ -342,24 +340,13 @@ def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
"%": "\\%",
"&": "\\&",
"~": "\\textasciitilde{}",
}
strict_escape_characters = {
"$": "\\$",
"_": "\\_",
"^": "\\textasciicircum{}",
}
if strict:
# To allow math input, users can use this function with strict = False
escape_characters.update(strict_escape_characters)
translation_map = str.maketrans(escape_characters)
strict_translation_map = str.maketrans(strict_escape_characters)
# Don't escape urls as hyperref package will do it automatically:
# Also always escape link placeholders strictly (as we don't expect any math in
# them):
# Find all the links in the sentence:
links = re.findall(r"\[(.*?)\]\((.*?)\)", latex_string)
@@ -367,8 +354,7 @@ def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
new_links = []
for i, link in enumerate(links):
placeholder = link[0]
escaped_placeholder = placeholder.translate(strict_translation_map)
escaped_placeholder = escaped_placeholder.translate(translation_map)
escaped_placeholder = placeholder.translate(translation_map)
url = link[1]
original_link = f"[{placeholder}]({url})"
@@ -377,6 +363,12 @@ def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
new_link = f"[{escaped_placeholder}]({url})"
new_links.append(new_link)
# If there are equations in the sentence, don't escape the special characters:
# Find all the equations in the sentence:
equations = re.findall(r"(\$\$.*?\$\$)", latex_string)
for i, equation in enumerate(equations):
latex_string = latex_string.replace(equation, f"!!-equation{i}-!!")
# Loop through the letters of the sentence and if you find an escape character,
# replace it with its LaTeX equivalent:
latex_string = latex_string.translate(translation_map)
@@ -385,6 +377,10 @@ def escape_latex_characters(latex_string: str, strict: bool = True) -> str:
for i, new_link in enumerate(new_links):
latex_string = latex_string.replace(f"!!-link{i}-!!", new_link)
# Replace !!-equation{i}-!!" with the original equations:
for i, equation in enumerate(equations):
latex_string = latex_string.replace(f"!!-equation{i}-!!", equation)
return latex_string
@@ -475,7 +471,7 @@ def transform_markdown_sections_to_latex_sections(
for entry in value:
if isinstance(entry, str):
# Then it means it's a TextEntry.
result = markdown_to_latex(escape_latex_characters(entry, strict=False))
result = markdown_to_latex(escape_latex_characters(entry))
transformed_list.append(result)
else:
# Then it means it's one of the other entries.
@@ -483,14 +479,14 @@ def transform_markdown_sections_to_latex_sections(
for entry_key, value in entry_as_dict.items():
if isinstance(value, str):
result = markdown_to_latex(
escape_latex_characters(value, strict=False)
escape_latex_characters(value)
)
setattr(entry, entry_key, result)
elif isinstance(value, list):
for j, item in enumerate(value):
if isinstance(item, str):
value[j] = markdown_to_latex(
escape_latex_characters(item, strict=False)
escape_latex_characters(item)
)
setattr(entry, entry_key, value)
transformed_list.append(entry)

2
tests/conftest.py
View File

@@ -23,7 +23,7 @@ from rendercv.renderer import templater
# the output. Setting update_testdata to True will update the reference files with
# the latest RenderCV. This should be done with caution, as it will overwrite the
# reference files with the latest output.
update_testdata = True
update_testdata = False
# copy sample entries from docs/update_rendercv_files.py:
education_entry_dictionary = {

15
tests/test_renderer.py
View File

@@ -120,17 +120,14 @@ def test_markdown_file_class(tmp_path, rendercv_data_model, jinja2_environment):
" [second link](https://myurl.com)"
),
),
("$a=5_4^3$", "$a=5_4^3$"),
(
"$$a=5_4^3 % & #$$ # $$aaaa ___ &&$$",
"$$a=5_4^3 % & #$$ \\# $$aaaa ___ &&$$",
),
],
)
def test_escape_latex_characters_not_strict(string, expected_string):
assert templater.escape_latex_characters(string, strict=False) == expected_string
def test_escape_latex_characters_strict():
string = "$a=5_4^3$"
expected_string = "\\$a=5\\_4\\textasciicircum{}3\\$"
assert templater.escape_latex_characters(string, strict=True) == expected_string
def test_escape_latex_characters(string, expected_string):
assert templater.escape_latex_characters(string) == expected_string
@pytest.mark.parametrize(