[deps] Pin transformers to 4.47

* [chore] Fix ruff typo

* [robot] Fix welcome robot

.github/workflows/deploy-doc.yml

@@ -11,18 +11,13 @@ jobs:
     - uses: actions/checkout@v4
       with:
         persist-credentials: false
-    - name: Set up Python
-      uses: actions/setup-python@v4
-      with:
-        python-version: '3.10'
-    - name: Install uv
-      run: pip install uv
-    - name: Install docs dependencies
-      run: uv pip install --system -e ".[docs]"
     - name: Build HTML
-      run: |
-        cd docs
-        make html
+      uses: ammaraskar/sphinx-action@7.0.0
+      with:
+        pre-build-command: |
+          apt-get update && apt-get install -y git
+          pip install uv
+          uv pip install --system . .[docs]
     - name: Upload artifacts
       uses: actions/upload-artifact@v4
       with:
@@ -33,4 +28,4 @@ jobs:
       if: github.ref == 'refs/heads/main'
       with:
         github_token: ${{ secrets.GITHUB_TOKEN }}
-        publish_dir: docs/build/html/
+        publish_dir: docs/build/html

.github/workflows/pr-welcome.yml

@@ -4,6 +4,10 @@ on:
   pull_request:
     types: [opened]
 
+permissions:
+  pull-requests: write
+  issues: write
+
 jobs:
   welcome:
     runs-on: ubuntu-latest

.github/workflows/publish.yml

@@ -2,8 +2,6 @@ name: Publish to PyPI
 
 on:
   push:
-    branches:
-      - 'main'
     tags:
       - 'v*'
 

.github/workflows/test.yaml

@@ -28,8 +28,8 @@ jobs:
 
     - name: Install dependencies
       run: |
-        uv sync --group test
+        uv sync --extra test
 
     - name: Run tests with pytest
       run: |
-        uv run pytest tests/
+        uv run pytest -v tests/

README.md

@@ -56,37 +56,43 @@ TexTeller was trained with **80M image-formula pairs** (previous dataset can be
 </tr>
 </table>
 
-## 🔄 Change Log
+## 📮 Change Log
 
-- 📮[2024-06-06] **TexTeller3.0 released!** The training data has been increased to **80M** (**10x more than** TexTeller2.0 and also improved in data diversity). TexTeller3.0's new features:
+- [2024-06-06] **TexTeller3.0 released!** The training data has been increased to **80M** (**10x more than** TexTeller2.0 and also improved in data diversity). TexTeller3.0's new features:
 
   - Support scanned image, handwritten formulas, English(Chinese) mixed formulas.
 
   - OCR abilities in both Chinese and English for printed images.
 
-- 📮[2024-05-02] Support **paragraph recognition**.
+- [2024-05-02] Support **paragraph recognition**.
 
-- 📮[2024-04-12] **Formula detection model** released!
+- [2024-04-12] **Formula detection model** released!
 
-- 📮[2024-03-25] TexTeller2.0 released! The training data for TexTeller2.0 has been increased to 7.5M (15x more than TexTeller1.0 and also improved in data quality). The trained TexTeller2.0 demonstrated **superior performance** in the test set, especially in recognizing rare symbols, complex multi-line formulas, and matrices.
+- [2024-03-25] TexTeller2.0 released! The training data for TexTeller2.0 has been increased to 7.5M (15x more than TexTeller1.0 and also improved in data quality). The trained TexTeller2.0 demonstrated **superior performance** in the test set, especially in recognizing rare symbols, complex multi-line formulas, and matrices.
 
   > [Here](./assets/test.pdf) are more test images and a horizontal comparison of various recognition models.
 
 ## 🚀 Getting Started
 
-1. Install the project's dependencies:
+1. Install uv:
 
    ```bash
-   pip install texteller
+   pip install uv
    ```
 
-2. If your are using CUDA backend, you may need to install `onnxruntime-gpu`:
+2. Install the project's dependencies:
 
    ```bash
-   pip install texteller[onnxruntime-gpu]
+   uv pip install texteller
    ```
 
-3. Run the following command to start inference:
+3. If your are using CUDA backend, you may need to install `onnxruntime-gpu`:
+
+   ```bash
+   uv pip install texteller[onnxruntime-gpu]
+   ```
+
+4. Run the following command to start inference:
 
    ```bash
    texteller inference "/path/to/image.{jpg,png}"
@@ -164,7 +170,7 @@ Please setup your environment before training:
 1. Install the dependencies for training:
 
    ```bash
-   pip install texteller[train]
+   uv pip install texteller[train]
    ```
 
 2. Clone the repository:

assets/README_zh.md

@@ -74,19 +74,25 @@ TexTeller 使用 **8千万图像-公式对** 进行训练（前代数据集可
 
 ## 🚀 快速开始
 
-1. 安装项目依赖：
+1. 安装uv：
 
    ```bash
-   pip install texteller
+   pip install uv
    ```
 
-2. 若使用 CUDA 后端，可能需要安装 `onnxruntime-gpu`：
+2. 安装项目依赖：
 
    ```bash
-   pip install texteller[onnxruntime-gpu]
+   uv pip install texteller
    ```
 
-3. 运行以下命令开始推理：
+3. 若使用 CUDA 后端，可能需要安装 `onnxruntime-gpu`：
+
+   ```bash
+   uv pip install texteller[onnxruntime-gpu]
+   ```
+
+4. 运行以下命令开始推理：
 
    ```bash
    texteller inference "/path/to/image.{jpg,png}"
@@ -96,7 +102,7 @@ TexTeller 使用 **8千万图像-公式对** 进行训练（前代数据集可
 
 ## 🌐 网页演示
 
-运行命令：
+命令行运行：
 
 ```bash
 texteller web
@@ -152,7 +158,7 @@ print(response.text)
 TexTeller的公式检测模型在3415张中文资料图像和8272张[IBEM数据集](https://zenodo.org/records/4757865)图像上训练。
 
 <div align="center">
-    <img src="./assets/det_rec.png" width=250>
+    <img src="./det_rec.png" width=250>
 </div>
 
 我们在Python接口中提供了公式检测接口，详见[接口文档](https://oleehyo.github.io/TexTeller/)。
@@ -164,7 +170,7 @@ TexTeller的公式检测模型在3415张中文资料图像和8272张[IBEM数据
 1. 安装训练依赖：
 
    ```bash
-   pip install texteller[train]
+   uv pip install texteller[train]
    ```
 
 2. 克隆仓库：

assets/fire.svg

@@ -457,4 +457,4 @@
   <animate attributeName="cy" values="114.80243604193255;7.19374553530416" keyTimes="0;1" dur="1s" repeatCount="indefinite" begin="-0.6866227460985781s"></animate>
   <animate attributeName="r" values="9;0;0" keyTimes="0;0.6690048284116141;1" dur="1s" repeatCount="indefinite" begin="-0.6866227460985781s"></animate>
 </circle></g>
-</svg>
+</svg>

assets/logo.svg

@@ -1,9 +1,10 @@
-<svg xmlns="http://www.w3.org/2000/svg" width="354" height="100" viewBox="0 0 354 100">
+
+<svg xmlns="http://www.w3.org/2000/svg" width="430" height="80" viewBox="0 0 430 80">
 
   <text
     x="50%"
     y="50%"
-    font-family="Arial, sans-serif"
+    font-family="monaco"
     font-size="55"
     text-anchor="middle"
     dominant-baseline="middle">

docs/source/conf.py

@@ -12,64 +12,64 @@
 import os
 import sys
 
-sys.path.insert(0, os.path.abspath('../..'))
+sys.path.insert(0, os.path.abspath("../.."))
 
 # -- Project information -----------------------------------------------------
 
-project = 'TexTeller'
-copyright = '2025, TexTeller Team'
-author = 'TexTeller Team'
+project = "TexTeller"
+copyright = "2025, TexTeller Team"
+author = "TexTeller Team"
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration
 
 extensions = [
-    'myst_parser',
-    'sphinx.ext.duration',
-    'sphinx.ext.intersphinx',
-    'sphinx.ext.autosectionlabel',
-    'sphinx.ext.autodoc',
-    'sphinx.ext.viewcode',
-    'sphinx.ext.napoleon',
-    'sphinx.ext.autosummary',
-    'sphinx_copybutton',
+    "myst_parser",
+    "sphinx.ext.duration",
+    "sphinx.ext.intersphinx",
+    "sphinx.ext.autosectionlabel",
+    "sphinx.ext.autodoc",
+    "sphinx.ext.viewcode",
+    "sphinx.ext.napoleon",
+    "sphinx.ext.autosummary",
+    "sphinx_copybutton",
     # 'sphinx.ext.linkcode',
     # 'sphinxarg.ext',
-    'sphinx_design',
-    'nbsphinx',
+    "sphinx_design",
+    "nbsphinx",
 ]
 
-templates_path = ['_templates']
+templates_path = ["_templates"]
 exclude_patterns = []
 
 # Autodoc settings
-autodoc_member_order = 'bysource'
+autodoc_member_order = "bysource"
 add_module_names = False
-autoclass_content = 'both'
+autoclass_content = "both"
 autodoc_default_options = {
-    'members': True,
-    'member-order': 'bysource',
-    'undoc-members': True,
-    'show-inheritance': True,
-    'imported-members': True,
+    "members": True,
+    "member-order": "bysource",
+    "undoc-members": True,
+    "show-inheritance": True,
+    "imported-members": True,
 }
 
 # Intersphinx settings
 intersphinx_mapping = {
-    'python': ('https://docs.python.org/3', None),
-    'numpy': ('https://numpy.org/doc/stable', None),
-    'torch': ('https://pytorch.org/docs/stable', None),
-    'transformers': ('https://huggingface.co/docs/transformers/main/en', None),
+    "python": ("https://docs.python.org/3", None),
+    "numpy": ("https://numpy.org/doc/stable", None),
+    "torch": ("https://pytorch.org/docs/stable", None),
+    "transformers": ("https://huggingface.co/docs/transformers/main/en", None),
 }
 
-html_theme = 'sphinx_book_theme'
+html_theme = "sphinx_book_theme"
 
 html_theme_options = {
-    'repository_url': 'https://github.com/OleehyO/TexTeller',
-    'use_repository_button': True,
-    'use_issues_button': True,
-    'use_edit_page_button': True,
-    'use_download_button': True,
+    "repository_url": "https://github.com/OleehyO/TexTeller",
+    "use_repository_button": True,
+    "use_issues_button": True,
+    "use_edit_page_button": True,
+    "use_download_button": True,
 }
 
 html_logo = "../../assets/logo.svg"

docs/source/index.rst

@@ -40,7 +40,7 @@ Converting an image to LaTeX:
 
 Processing a mixed text/formula image:
 
-.. code-block::python
+.. code-block:: python
 
    from texteller import (
        load_model, load_tokenizer, load_latexdet_model,

examples/client_demo.py

@@ -3,8 +3,8 @@ import requests
 server_url = "http://127.0.0.1:8000/predict"
 
 img_path = "/path/to/your/image"
-with open(img_path, 'rb') as img:
-    files = {'img': img}
+with open(img_path, "rb") as img:
+    files = {"img": img}
     response = requests.post(server_url, files=files)
 
 print(response.text)

pyproject.toml

@@ -22,7 +22,7 @@ dependencies = [
     "streamlit-paste-button>=0.1.2",
     "torch>=2.6.0",
     "torchvision>=0.21.0",
-    "transformers==4.45.2",
+    "transformers==4.47",
     "wget>=3.2",
     "optimum[onnxruntime]>=1.24.0",
     "python-multipart>=0.0.20",

texteller/api/format.py

@@ -19,8 +19,8 @@ TEXT_LINE_START = ""
 COMMENT_LINE_START = "% "
 
 # Opening and closing delimiters
-OPENS = ['{', '(', '[']
-CLOSES = ['}', ')', ']']
+OPENS = ["{", "(", "["]
+CLOSES = ["}", ")", "]"]
 
 # Names of LaTeX verbatim environments
 VERBATIMS = ["verbatim", "Verbatim", "lstlisting", "minted", "comment"]
@@ -138,7 +138,7 @@ class Pattern:
                 contains_env_end=ENV_END in s,
                 contains_item=ITEM in s,
                 contains_splitting=True,
-                contains_comment='%' in s,
+                contains_comment="%" in s,
             )
         else:
             return cls(
@@ -146,7 +146,7 @@ class Pattern:
                 contains_env_end=False,
                 contains_item=False,
                 contains_splitting=False,
-                contains_comment='%' in s,
+                contains_comment="%" in s,
             )
 
 
@@ -169,11 +169,11 @@ def find_comment_index(line: str, pattern: Pattern) -> Optional[int]:
 
     in_command = False
     for i, c in enumerate(line):
-        if c == '\\':
+        if c == "\\":
             in_command = True
         elif in_command and not c.isalpha():
             in_command = False
-        elif c == '%' and not in_command:
+        elif c == "%" and not in_command:
             return i
 
     return None
@@ -390,10 +390,10 @@ def find_wrap_point(line: str, indent_length: int, args: Args) -> Optional[int]:
         line_width += 1
         if line_width > wrap_boundary and wrap_point is not None:
             break
-        if c == ' ' and prev_char != '\\':
+        if c == " " and prev_char != "\\":
             if after_char:
                 wrap_point = i
-        elif c != '%':
+        elif c != "%":
             after_char = True
         prev_char = c
 
@@ -483,8 +483,8 @@ def split_line(line: str, state: State, file: str, args: Args, logs: List[Log])
     if not match:
         return line, ""
 
-    prev = match.group('prev')
-    rest = match.group('env')
+    prev = match.group("prev")
+    rest = match.group("env")
 
     if args.verbosity >= 3:  # Trace level
         logs.append(
@@ -517,8 +517,8 @@ def clean_text(text: str, args: Args) -> str:
     text = RE_NEWLINES.sub(f"{LINE_END}{LINE_END}", text)
 
     # Remove tabs if they shouldn't be used
-    if args.tabchar != '\t':
-        text = text.replace('\t', ' ' * args.tabsize)
+    if args.tabchar != "\t":
+        text = text.replace("\t", " " * args.tabsize)
 
     # Remove trailing spaces
     text = RE_TRAIL.sub(LINE_END, text)
@@ -577,7 +577,7 @@ def _format_latex(old_text: str, file: str, args: Args) -> Tuple[str, List[Log]]
     new_text = ""
 
     # Select the character used for indentation
-    indent_char = '\t' if args.tabchar == '\t' else ' '
+    indent_char = "\t" if args.tabchar == "\t" else " "
 
     # Get any extra environments to be indented as lists
     lists_begin = [f"\\begin{{{l}}}" for l in args.lists]

texteller/api/katex.py

@@ -5,13 +5,13 @@ from .format import format_latex
 
 
 def _rm_dollar_surr(content):
-    pattern = re.compile(r'\\[a-zA-Z]+\$.*?\$|\$.*?\$')
+    pattern = re.compile(r"\\[a-zA-Z]+\$.*?\$|\$.*?\$")
     matches = pattern.findall(content)
 
     for match in matches:
-        if not re.match(r'\\[a-zA-Z]+', match):
-            new_match = match.strip('$')
-            content = content.replace(match, ' ' + new_match + ' ')
+        if not re.match(r"\\[a-zA-Z]+", match):
+            new_match = match.strip("$")
+            content = content.replace(match, " " + new_match + " ")
 
     return content
 
@@ -33,97 +33,97 @@ def to_katex(formula: str) -> str:
     """
     res = formula
     # remove mbox surrounding
-    res = change_all(res, r'\mbox ', r' ', r'{', r'}', r'', r'')
-    res = change_all(res, r'\mbox', r' ', r'{', r'}', r'', r'')
+    res = change_all(res, r"\mbox ", r" ", r"{", r"}", r"", r"")
+    res = change_all(res, r"\mbox", r" ", r"{", r"}", r"", r"")
     # remove hbox surrounding
-    res = re.sub(r'\\hbox to ?-? ?\d+\.\d+(pt)?\{', r'\\hbox{', res)
-    res = change_all(res, r'\hbox', r' ', r'{', r'}', r'', r' ')
+    res = re.sub(r"\\hbox to ?-? ?\d+\.\d+(pt)?\{", r"\\hbox{", res)
+    res = change_all(res, r"\hbox", r" ", r"{", r"}", r"", r" ")
     # remove raise surrounding
-    res = re.sub(r'\\raise ?-? ?\d+\.\d+(pt)?', r' ', res)
+    res = re.sub(r"\\raise ?-? ?\d+\.\d+(pt)?", r" ", res)
     # remove makebox
-    res = re.sub(r'\\makebox ?\[\d+\.\d+(pt)?\]\{', r'\\makebox{', res)
-    res = change_all(res, r'\makebox', r' ', r'{', r'}', r'', r' ')
+    res = re.sub(r"\\makebox ?\[\d+\.\d+(pt)?\]\{", r"\\makebox{", res)
+    res = change_all(res, r"\makebox", r" ", r"{", r"}", r"", r" ")
     # remove vbox surrounding, scalebox surrounding
-    res = re.sub(r'\\raisebox\{-? ?\d+\.\d+(pt)?\}\{', r'\\raisebox{', res)
-    res = re.sub(r'\\scalebox\{-? ?\d+\.\d+(pt)?\}\{', r'\\scalebox{', res)
-    res = change_all(res, r'\scalebox', r' ', r'{', r'}', r'', r' ')
-    res = change_all(res, r'\raisebox', r' ', r'{', r'}', r'', r' ')
-    res = change_all(res, r'\vbox', r' ', r'{', r'}', r'', r' ')
+    res = re.sub(r"\\raisebox\{-? ?\d+\.\d+(pt)?\}\{", r"\\raisebox{", res)
+    res = re.sub(r"\\scalebox\{-? ?\d+\.\d+(pt)?\}\{", r"\\scalebox{", res)
+    res = change_all(res, r"\scalebox", r" ", r"{", r"}", r"", r" ")
+    res = change_all(res, r"\raisebox", r" ", r"{", r"}", r"", r" ")
+    res = change_all(res, r"\vbox", r" ", r"{", r"}", r"", r" ")
 
     origin_instructions = [
-        r'\Huge',
-        r'\huge',
-        r'\LARGE',
-        r'\Large',
-        r'\large',
-        r'\normalsize',
-        r'\small',
-        r'\footnotesize',
-        r'\tiny',
+        r"\Huge",
+        r"\huge",
+        r"\LARGE",
+        r"\Large",
+        r"\large",
+        r"\normalsize",
+        r"\small",
+        r"\footnotesize",
+        r"\tiny",
     ]
     for old_ins, new_ins in zip(origin_instructions, origin_instructions):
-        res = change_all(res, old_ins, new_ins, r'$', r'$', '{', '}')
-    res = change_all(res, r'\mathbf', r'\bm', r'{', r'}', r'{', r'}')
-    res = change_all(res, r'\boldmath ', r'\bm', r'{', r'}', r'{', r'}')
-    res = change_all(res, r'\boldmath', r'\bm', r'{', r'}', r'{', r'}')
-    res = change_all(res, r'\boldmath ', r'\bm', r'$', r'$', r'{', r'}')
-    res = change_all(res, r'\boldmath', r'\bm', r'$', r'$', r'{', r'}')
-    res = change_all(res, r'\scriptsize', r'\scriptsize', r'$', r'$', r'{', r'}')
-    res = change_all(res, r'\emph', r'\textit', r'{', r'}', r'{', r'}')
-    res = change_all(res, r'\emph ', r'\textit', r'{', r'}', r'{', r'}')
+        res = change_all(res, old_ins, new_ins, r"$", r"$", "{", "}")
+    res = change_all(res, r"\mathbf", r"\bm", r"{", r"}", r"{", r"}")
+    res = change_all(res, r"\boldmath ", r"\bm", r"{", r"}", r"{", r"}")
+    res = change_all(res, r"\boldmath", r"\bm", r"{", r"}", r"{", r"}")
+    res = change_all(res, r"\boldmath ", r"\bm", r"$", r"$", r"{", r"}")
+    res = change_all(res, r"\boldmath", r"\bm", r"$", r"$", r"{", r"}")
+    res = change_all(res, r"\scriptsize", r"\scriptsize", r"$", r"$", r"{", r"}")
+    res = change_all(res, r"\emph", r"\textit", r"{", r"}", r"{", r"}")
+    res = change_all(res, r"\emph ", r"\textit", r"{", r"}", r"{", r"}")
 
     # remove bold command
-    res = change_all(res, r'\bm', r' ', r'{', r'}', r'', r'')
+    res = change_all(res, r"\bm", r" ", r"{", r"}", r"", r"")
 
     origin_instructions = [
-        r'\left',
-        r'\middle',
-        r'\right',
-        r'\big',
-        r'\Big',
-        r'\bigg',
-        r'\Bigg',
-        r'\bigl',
-        r'\Bigl',
-        r'\biggl',
-        r'\Biggl',
-        r'\bigm',
-        r'\Bigm',
-        r'\biggm',
-        r'\Biggm',
-        r'\bigr',
-        r'\Bigr',
-        r'\biggr',
-        r'\Biggr',
+        r"\left",
+        r"\middle",
+        r"\right",
+        r"\big",
+        r"\Big",
+        r"\bigg",
+        r"\Bigg",
+        r"\bigl",
+        r"\Bigl",
+        r"\biggl",
+        r"\Biggl",
+        r"\bigm",
+        r"\Bigm",
+        r"\biggm",
+        r"\Biggm",
+        r"\bigr",
+        r"\Bigr",
+        r"\biggr",
+        r"\Biggr",
     ]
     for origin_ins in origin_instructions:
-        res = change_all(res, origin_ins, origin_ins, r'{', r'}', r'', r'')
+        res = change_all(res, origin_ins, origin_ins, r"{", r"}", r"", r"")
 
-    res = re.sub(r'\\\[(.*?)\\\]', r'\1\\newline', res)
+    res = re.sub(r"\\\[(.*?)\\\]", r"\1\\newline", res)
 
-    if res.endswith(r'\newline'):
+    if res.endswith(r"\newline"):
         res = res[:-8]
 
     # remove multiple spaces
-    res = re.sub(r'(\\,){1,}', ' ', res)
-    res = re.sub(r'(\\!){1,}', ' ', res)
-    res = re.sub(r'(\\;){1,}', ' ', res)
-    res = re.sub(r'(\\:){1,}', ' ', res)
-    res = re.sub(r'\\vspace\{.*?}', '', res)
+    res = re.sub(r"(\\,){1,}", " ", res)
+    res = re.sub(r"(\\!){1,}", " ", res)
+    res = re.sub(r"(\\;){1,}", " ", res)
+    res = re.sub(r"(\\:){1,}", " ", res)
+    res = re.sub(r"\\vspace\{.*?}", "", res)
 
     # merge consecutive text
     def merge_texts(match):
         texts = match.group(0)
-        merged_content = ''.join(re.findall(r'\\text\{([^}]*)\}', texts))
-        return f'\\text{{{merged_content}}}'
+        merged_content = "".join(re.findall(r"\\text\{([^}]*)\}", texts))
+        return f"\\text{{{merged_content}}}"
 
-    res = re.sub(r'(\\text\{[^}]*\}\s*){2,}', merge_texts, res)
+    res = re.sub(r"(\\text\{[^}]*\}\s*){2,}", merge_texts, res)
 
-    res = res.replace(r'\bf ', '')
+    res = res.replace(r"\bf ", "")
     res = _rm_dollar_surr(res)
 
     # remove extra spaces (keeping only one)
-    res = re.sub(r' +', ' ', res)
+    res = re.sub(r" +", " ", res)
 
     # format latex
     res = res.strip()

texteller/models/__init__.py

@@ -1,3 +1,3 @@
 from .texteller import TexTeller
 
-__all__ = ['TexTeller']
+__all__ = ["TexTeller"]

texteller/utils/image.py

@@ -41,7 +41,7 @@ def readimgs(image_paths: list[str]) -> list[np.ndarray]:
         if image is None:
             raise ValueError(f"Image at {path} could not be read.")
         if image.dtype == np.uint16:
-            _logger.warning(f'Converting {path} to 8-bit, image may be lossy.')
+            _logger.warning(f"Converting {path} to 8-bit, image may be lossy.")
             image = cv2.convertScaleAbs(image, alpha=(255.0 / 65535.0))
 
         channels = 1 if len(image.shape) == 2 else image.shape[2]
@@ -112,7 +112,7 @@ def transform(images: List[Union[np.ndarray, Image.Image]]) -> List[torch.Tensor
 
     assert IMG_CHANNELS == 1, "Only support grayscale images for now"
     images = [
-        np.array(img.convert('RGB')) if isinstance(img, Image.Image) else img for img in images
+        np.array(img.convert("RGB")) if isinstance(img, Image.Image) else img for img in images
     ]
     images = [trim_white_border(image) for image in images]
     images = [general_transform_pipeline(image) for image in images]

texteller/utils/latex.py

@@ -21,7 +21,7 @@ def _change(input_str, old_inst, new_inst, old_surr_l, old_surr_r, new_surr_l, n
             j = start + 1
             escaped = False
             while j < n and count > 0:
-                if input_str[j] == '\\' and not escaped:
+                if input_str[j] == "\\" and not escaped:
                     escaped = True
                     j += 1
                     continue
@@ -71,10 +71,10 @@ def change_all(input_str, old_inst, new_inst, old_surr_l, old_surr_r, new_surr_l
     for p in pos[::-1]:
         res[p:] = list(
             _change(
-                ''.join(res[p:]), old_inst, new_inst, old_surr_l, old_surr_r, new_surr_l, new_surr_r
+                "".join(res[p:]), old_inst, new_inst, old_surr_l, old_surr_r, new_surr_l, new_surr_r
             )
         )
-    res = ''.join(res)
+    res = "".join(res)
     return res
 
 
@@ -121,7 +121,7 @@ def add_newlines(latex_str: str) -> str:
 
     # 4. Cleanup: Collapse multiple consecutive newlines into a single newline.
     # This handles cases where the replacements above might have created \n\n.
-    processed_str = re.sub(r'\n{2,}', '\n', processed_str)
+    processed_str = re.sub(r"\n{2,}", "\n", processed_str)
 
     # Remove leading/trailing whitespace (including potential single newlines
     # at the very start/end resulting from the replacements) from the entire result.