From d63fc56c05d05e7fa848a7082af91f63ca674624 Mon Sep 17 00:00:00 2001 From: Adham 3llam <102502789+Ad7amstein@users.noreply.github.com> Date: Mon, 7 Jul 2025 12:33:37 +0300 Subject: [PATCH 1/4] fix: correct opencv-python syntax in pyproject.toml to prevent TOMLDecodeError --- pyproject.toml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/pyproject.toml b/pyproject.toml index e99d062..2a2b43b 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -18,7 +18,7 @@ dependencies = [ "pyyaml", "frontend", "pymupdf", - opencv-python = "^4.6.0" + "opencv-python>=4.6.0" # Add other common dependencies ] From a7c3f655ee9c539cc67f1bc491f137dc89302ded Mon Sep 17 00:00:00 2001 From: Adham 3llam <102502789+Ad7amstein@users.noreply.github.com> Date: Mon, 7 Jul 2025 13:03:43 +0300 Subject: [PATCH 2/4] fix: resolve editable install errors in pyproject.toml --- pyproject.toml | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/pyproject.toml b/pyproject.toml index 2a2b43b..bf3a7a4 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -10,7 +10,7 @@ authors = [ ] description = "A Comprehensive Toolkit for High-Quality PDF Content Extraction." readme = "README.md" -license = { file="LICENSE" } +license = { file="LICENSE.md" } requires-python = ">=3.10" dependencies = [ "PyPDF2", @@ -32,3 +32,5 @@ formula_detection = [ # Add other dependencies for formula detection ] # Add additional dependencies for other models +[tool.setuptools.packages.find] +include = ["pdf_extract_kit"] From 1b277e735cdc97fdcf3ee54b7acca2948d86c7ae Mon Sep 17 00:00:00 2001 From: Ad7amstein Date: Thu, 10 Jul 2025 12:19:54 +0300 Subject: [PATCH 3/4] fix: update doclayout-yolo version to 0.0.4 in requirements.txt --- requirements.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/requirements.txt b/requirements.txt index ba8ab9f..e1eff8b 100644 --- a/requirements.txt +++ b/requirements.txt @@ -2,7 +2,7 @@ omegaconf matplotlib PyMuPDF ultralytics>=8.2.85 -doclayout-yolo==0.0.2 +doclayout-yolo==0.0.4 unimernet==0.2.1 paddlepaddle-gpu paddleocr==2.7.3 From 0f27237a7cef0b8eb9fbfb6b5b6a01568828e1e9 Mon Sep 17 00:00:00 2001 From: Ad7amstein Date: Thu, 10 Jul 2025 12:36:37 +0300 Subject: [PATCH 4/4] fix: add .venv and .env to .gitignore to prevent environment files from being tracked --- .gitignore | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/.gitignore b/.gitignore index ea20b51..917531f 100644 --- a/.gitignore +++ b/.gitignore @@ -20,3 +20,7 @@ models/* # Sphinx documentation docs/*/_build/ + +# environment +.venv +.env