Merge remote-tracking branch 'origin/master'

.github/workflows/benchmark.yml

 # This workflow will install Python dependencies, run tests and lint with a variety of Python versions
 # For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
 
-name: PDF
+name: mineru
 on:
   push:
     branches:
@@ -18,7 +18,7 @@ on:
   workflow_dispatch:
 jobs:
   pdf-test:
-    runs-on: pdf
+    runs-on: mineru
     timeout-minutes: 180
     strategy:
       fail-fast: true
@@ -37,12 +37,8 @@ jobs:
         echo $changed_files
         if [[ $changed_files =~ "requirements.txt" ]]; then
           pip install -r requirements.txt
+          pip install -r requirements-qa.txt
         fi
-
-    - name: config-net-reset
-      run: |
-        export http_proxy=""
-        export https_proxy=""
     - name: get-benchmark-result
       run: |
         echo "start test"

.github/workflows/cli.yml

 # This workflow will install Python dependencies, run tests and lint with a variety of Python versions
 # For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
 
-name: PDF
+name: mineru
 on:
   push:
     branches:
@@ -18,7 +18,7 @@ on:
   workflow_dispatch:
 jobs:
   cli-test:
-    runs-on: pdf
+    runs-on: mineru
     timeout-minutes: 40
     strategy:
       fail-fast: true
@@ -38,6 +38,7 @@ jobs:
         echo $changed_files
         if [[ $changed_files =~ "requirements.txt" ]]; then
           pip install -r requirements.txt
+          pip install -r requirements-qa.txt
         fi
 
     - name: config-net-reset
@@ -53,7 +54,7 @@ jobs:
   notify_to_feishu:
     if: ${{ always() && !cancelled() && contains(needs.*.result, 'failure') && (github.ref_name == 'master') }}
     needs: [cli-test]
-    runs-on: pdf
+    runs-on: mineru
     steps:
     - name: get_actor
       run: |

tests/test_cli/conf/conf.py

@@ -2,7 +2,7 @@ import os
 conf = {
 "code_path": os.environ.get('GITHUB_WORKSPACE'),
 "pdf_dev_path" : os.environ.get('GITHUB_WORKSPACE') + "/tests/test_cli/pdf_dev",
-"pdf_res_path": "/home/quyuan/code/Magic-PDF/Magic-PDF/Magic-PDF/ci"
+"pdf_res_path": "/share/quyuan/mineru/data/mineru"
 
 }
 

tools/benchmark.py

@@ -4,15 +4,13 @@ import shutil
 import json
 import markdown_calculate
 code_path = os.environ.get('GITHUB_WORKSPACE')
-#code_path = "/home/quyuan/actions-runner/_work/Magic-PDF/Magic-PDF.bk"
-#评测集存放路径
-pdf_dev_path = "/home/quyuan/data"
-#magicpdf跑测结果
-pdf_res_path = "/home/quyuan/code/Magic-PDF/Magic-PDF/Magic-PDF/ci/magic-pdf"
+#数据集存放路径
+pdf_dev_path = "/share/quyuan/mineru/data/"
+#magicpdf最终结果
+pdf_res_path = "/share/quyuan/mineru/data/mineru"
 file_types = ["academic_literature", "atlas", "courseware", "colorful_textbook", "historical_documents", "notes", "ordinary_books", "ordinary_exam_paper", "ordinary_textbook", "research_report", "special_exam_paper"]
-#file_types = ["academic_literature"]
-
 def test_cli():
+    #magicpdf模型输出结果
     magicpdf_path = os.path.join(pdf_dev_path, "output")
     rm_cmd = "rm -rf %s" % (pdf_res_path)
     os.system(rm_cmd)