feat: add test case (#499)

Co-authored-by: quyuan <quyuan@pjlab.org>

.github/workflows/cli.yml

@@ -6,20 +6,22 @@ on:
   push:
     branches:
       - "master"
+      - "dev"
     paths-ignore:
       - "cmds/**"
       - "**.md"
   pull_request:
     branches:
       - "master"
+      - "dev"
     paths-ignore:
       - "cmds/**"
       - "**.md"
   workflow_dispatch:
 jobs:
   cli-test:
-    runs-on: ubuntu-latest
-    timeout-minutes: 40
+    runs-on: pdf
+    timeout-minutes: 120
     strategy:
       fail-fast: true
 
@@ -28,27 +30,22 @@ jobs:
       uses: actions/checkout@v3
       with:
         fetch-depth: 2
-      
-    - name: check-requirements
-      run: |
-        pip install -r requirements.txt
-        pip install -r requirements-qa.txt
-        pip install magic-pdf
-    - name: test_cli
+
+    - name: install
       run: |
-        cp magic-pdf.template.json ~/magic-pdf.json
-        echo $GITHUB_WORKSPACE
-        cd $GITHUB_WORKSPACE && export PYTHONPATH=. && pytest -s -v tests/test_unit.py
-        cd $GITHUB_WORKSPACE &&  pytest -s -v tests/test_cli/test_cli.py
-                                                                                                                            
-    - name: benchmark
+        echo $GITHUB_WORKSPACE && sh tests/retry_env.sh
+    - name: unit test
+      run: |        
+        cd $GITHUB_WORKSPACE && export PYTHONPATH=. && coverage run -m  pytest  tests/test_unit.py --cov=magic_pdf/ --cov-report term-missing --cov-report html
+        cd $GITHUB_WORKSPACE && python tests/get_coverage.py
+    - name: cli test
       run: |
-        cd $GITHUB_WORKSPACE &&  pytest -s -v tests/test_cli/test_bench.py
+        cd $GITHUB_WORKSPACE &&  pytest -s -v tests/test_cli/test_cli_sdk.py
 
   notify_to_feishu:
     if: ${{ always() && !cancelled() && contains(needs.*.result, 'failure') && (github.ref_name == 'master') }}
-    needs: [cli-test]
-    runs-on: ubuntu-latest
+    needs: cli-test
+    runs-on: pdf
     steps:
     - name: get_actor
       run: |
@@ -67,9 +64,5 @@ jobs:
 
     - name: notify
       run: |
-        curl  ${{ secrets.WEBHOOK_URL }} -H 'Content-Type: application/json'  -d '{
-        "msgtype": "text",
-        "text": {
+        echo ${{ secrets.USER_ID }}
+        curl -X POST -H "Content-Type: application/json" -d '{"msg_type":"post","content":{"post":{"zh_cn":{"title":"'${{ github.repository }}' GitHubAction Failed","content":[[{"tag":"text","text":""},{"tag":"a","text":"Please click here for details ","href":"https://github.com/'${{ github.repository }}'/actions/runs/'${GITHUB_RUN_ID}'"},{"tag":"at","user_id":"'${{ secrets.USER_ID }}'"}]]}}}}'  ${{ secrets.WEBHOOK_URL }}
\ No newline at end of file
-            "mentioned_list": ["${{ env.METIONS }}"] , "content": "'${{ github.repository }}' GitHubAction Failed!\n 细节请查看：https://github.com/'${{ github.repository }}'/actions/runs/'${GITHUB_RUN_ID}'"
-        }
-        }'   
\ No newline at end of file

requirements-qa.txt

@@ -14,4 +14,6 @@ tqdm
 htmltabletomd
 pypandoc
 pyopenssl==24.0.0
-struct-eqtable==0.1.0
+struct-eqtable==0.1.0
\ No newline at end of file
+pytest-cov
+beautifulsoup4
\ No newline at end of file

tests/get_coverage.py

+"""
+get cov
+"""
+from bs4 import BeautifulSoup
+
+def get_covrage():
+    """get covrage"""
+    # 发送请求获取网页内容
+    html_content = open("htmlcov/index.html", "r", encoding="utf-8").read()
+    soup = BeautifulSoup(html_content, 'html.parser')
+
+    # 查找包含"pc_cov"的span标签
+    pc_cov_span = soup.find('span', class_='pc_cov')
+
+    # 提取百分比值
+    percentage_value = pc_cov_span.text.strip()
+    percentage_float = float(percentage_value.rstrip('%'))
+    print ("percentage_float:", percentage_float)
+    assert percentage_float >= 0.2
+
+if __name__ == '__main__':
+    get_covrage()
\ No newline at end of file

tests/pdf_indicator/overall_indicator.py

-import json
-import pandas as pd
-import numpy as np
-from nltk.translate.bleu_score import sentence_bleu
-import argparse
-from sklearn.metrics import classification_report
-from collections import Counter
-from sklearn import metrics
-from pandas import isnull
-
-
-
-def indicator_cal(json_standard,json_test):
-
-    json_standard = pd.DataFrame(json_standard)
-    json_test = pd.DataFrame(json_test)
-
-
-
-    '''数据集总体指标'''
-    
-    a=json_test[['id','mid_json']]
-    b=json_standard[['id','mid_json','pass_label']]
-
-    a=a.drop_duplicates(subset='id',keep='first')
-    a.index=range(len(a))
-    b=b.drop_duplicates(subset='id',keep='first')
-    b.index=range(len(b))
-
-    outer_merge=pd.merge(a,b,on='id',how='outer')
-    outer_merge.columns=['id','standard_mid_json','test_mid_json','pass_label']
-    standard_exist=outer_merge.standard_mid_json.apply(lambda x: not isnull(x))
-    test_exist=outer_merge.test_mid_json.apply(lambda x: not isnull(x))
-
-    overall_report = {}
-    overall_report['accuracy']=metrics.accuracy_score(standard_exist,test_exist)
-    overall_report['precision']=metrics.precision_score(standard_exist,test_exist)
-    overall_report['recall']=metrics.recall_score(standard_exist,test_exist)
-    overall_report['f1_score']=metrics.f1_score(standard_exist,test_exist)
-
-
-    inner_merge=pd.merge(a,b,on='id',how='inner')
-    inner_merge.columns=['id','standard_mid_json','test_mid_json','pass_label']
-    json_standard = inner_merge['standard_mid_json']#check一下是否对齐
-    json_test = inner_merge['test_mid_json']
-
-
-    
-
-    '''批量读取中间生成的json文件'''
-    test_inline_equations=[]
-    test_interline_equations=[]
-    test_inline_euqations_bboxs=[]
-    test_interline_equations_bboxs=[]
-    test_dropped_text_bboxes=[]
-    test_dropped_text_tag=[]
-    test_dropped_image_bboxes=[]
-    test_dropped_table_bboxes=[] 
-    test_preproc_num=[]#阅读顺序
-    test_para_num=[]
-    test_para_text=[]
-
-    for i in json_test:
-        mid_json=pd.DataFrame(i)
-        mid_json=mid_json.iloc[:,:-1]
-        for j1 in mid_json.loc['inline_equations',:]:
-            page_in_text=[]
-            page_in_bbox=[]
-            for k1 in j1:
-                page_in_text.append(k1['latex_text'])
-                page_in_bbox.append(k1['bbox'])
-            test_inline_equations.append(page_in_text)
-            test_inline_euqations_bboxs.append(page_in_bbox)
-        for j2 in mid_json.loc['interline_equations',:]:
-            page_in_text=[]
-            page_in_bbox=[]
-            for k2 in j2:
-                page_in_text.append(k2['latex_text'])
-                page_in_bbox.append(k2['bbox'])
-            test_interline_equations.append(page_in_text)
-            test_interline_equations_bboxs.append(page_in_bbox)
-
-        for j3 in mid_json.loc['droped_text_block',:]:
-            page_in_bbox=[]
-            page_in_tag=[]
-            for k3 in j3:
-                page_in_bbox.append(k3['bbox'])
-                #如果k3中存在tag这个key
-                if 'tag' in k3.keys():
-                    page_in_tag.append(k3['tag'])
-                else:
-                    page_in_tag.append('None')
-            test_dropped_text_tag.append(page_in_tag)
-            test_dropped_text_bboxes.append(page_in_bbox)
-        for j4 in mid_json.loc['droped_image_block',:]:
-                test_dropped_image_bboxes.append(j4)
-        for j5 in mid_json.loc['droped_table_block',:]:
-                test_dropped_table_bboxes.append(j5)
-        for j6 in mid_json.loc['preproc_blocks',:]:
-            page_in=[]
-            for k6 in j6:
-                page_in.append(k6['number'])
-            test_preproc_num.append(page_in)
-
-        test_pdf_text=[]     
-        for j7 in mid_json.loc['para_blocks',:]:
-            test_para_num.append(len(j7))  
-            for k7 in j7:
-                test_pdf_text.append(k7['text'])  
-        test_para_text.append(test_pdf_text)
-
-
-
-    standard_inline_equations=[]
-    standard_interline_equations=[]
-    standard_inline_euqations_bboxs=[]
-    standard_interline_equations_bboxs=[]
-    standard_dropped_text_bboxes=[]
-    standard_dropped_text_tag=[]
-    standard_dropped_image_bboxes=[]
-    standard_dropped_table_bboxes=[] 
-    standard_preproc_num=[]#阅读顺序
-    standard_para_num=[]
-    standard_para_text=[]
-
-    for i in json_standard:
-        mid_json=pd.DataFrame(i)
-        mid_json=mid_json.iloc[:,:-1]
-        for j1 in mid_json.loc['inline_equations',:]:
-            page_in_text=[]
-            page_in_bbox=[]
-            for k1 in j1:
-                page_in_text.append(k1['latex_text'])
-                page_in_bbox.append(k1['bbox'])
-            standard_inline_equations.append(page_in_text)
-            standard_inline_euqations_bboxs.append(page_in_bbox)
-        for j2 in mid_json.loc['interline_equations',:]:
-            page_in_text=[]
-            page_in_bbox=[]
-            for k2 in j2:
-                page_in_text.append(k2['latex_text'])
-                page_in_bbox.append(k2['bbox'])
-            standard_interline_equations.append(page_in_text)
-            standard_interline_equations_bboxs.append(page_in_bbox)
-        for j3 in mid_json.loc['droped_text_block',:]:
-            page_in_bbox=[]
-            page_in_tag=[]
-            for k3 in j3:
-                page_in_bbox.append(k3['bbox'])
-                if 'tag' in k3.keys():
-                    page_in_tag.append(k3['tag'])
-                else:
-                    page_in_tag.append('None')
-            standard_dropped_text_bboxes.append(page_in_bbox)
-            standard_dropped_text_tag.append(page_in_tag)
-        for j4 in mid_json.loc['droped_image_block',:]:
-                standard_dropped_image_bboxes.append(j4)
-        for j5 in mid_json.loc['droped_table_block',:]:
-                standard_dropped_table_bboxes.append(j5)
-        for j6 in mid_json.loc['preproc_blocks',:]:
-            page_in=[]
-            for k6 in j6:
-                page_in.append(k6['number'])
-            standard_preproc_num.append(page_in)     
-
-        standard_pdf_text=[]
-        for j7 in mid_json.loc['para_blocks',:]:
-            standard_para_num.append(len(j7))  
-            for k7 in j7:
-                standard_pdf_text.append(k7['text'])
-        standard_para_text.append(standard_pdf_text)
-
-
-    """
-    在计算指标之前最好先确认基本统计信息是否一致
-    """
-
-
-    '''
-    计算pdf之间的总体编辑距离和bleu
-    这里只计算正例的pdf
-    '''
-    
-    test_para_text=np.asarray(test_para_text, dtype = object)[inner_merge['pass_label']=='yes']
-    standard_para_text=np.asarray(standard_para_text, dtype = object)[inner_merge['pass_label']=='yes']
-
-    pdf_dis=[]
-    pdf_bleu=[]
-    for a,b in zip(test_para_text,standard_para_text):
-        a1=[ ''.join(i) for i in a]
-        b1=[ ''.join(i) for i in b]
-        pdf_dis.append(Levenshtein_Distance(a1,b1))
-        pdf_bleu.append(sentence_bleu([a1],b1))
-    overall_report['pdf间的平均编辑距离']=np.mean(pdf_dis)
-    overall_report['pdf间的平均bleu']=np.mean(pdf_bleu)
-
-
-    '''行内公式和行间公式的编辑距离和bleu'''
-    inline_equations_edit_bleu=equations_indicator(test_inline_euqations_bboxs,standard_inline_euqations_bboxs,test_inline_equations,standard_inline_equations)
-    interline_equations_edit_bleu=equations_indicator(test_interline_equations_bboxs,standard_interline_equations_bboxs,test_interline_equations,standard_interline_equations)
-   
-    '''行内公式bbox匹配相关指标'''
-    inline_equations_bbox_report=bbox_match_indicator(test_inline_euqations_bboxs,standard_inline_euqations_bboxs)
-
-    '''行间公式bbox匹配相关指标'''
-    interline_equations_bbox_report=bbox_match_indicator(test_interline_equations_bboxs,standard_interline_equations_bboxs)
-
-    '''可以先检查page和bbox数量是否一致'''
-    '''dropped_text_block的bbox匹配相关指标'''
-    test_text_bbox=[]
-    standard_text_bbox=[]
-    test_tag=[]
-    standard_tag=[]
-
-    index=0
-    for a,b in zip(test_dropped_text_bboxes,standard_dropped_text_bboxes):
-        test_page_tag=[]
-        standard_page_tag=[]
-        test_page_bbox=[]
-        standard_page_bbox=[]
-        if len(a)==0 and len(b)==0:
-            pass
-        else:
-            for i in range(len(b)):
-                judge=0
-                standard_page_tag.append(standard_dropped_text_tag[index][i])
-                standard_page_bbox.append(1)
-                for j in range(len(a)):
-                    if bbox_offset(b[i],a[j]):
-                        judge=1
-                        test_page_tag.append(test_dropped_text_tag[index][j])
-                        test_page_bbox.append(1)
-                        break
-                if judge==0:
-                    test_page_tag.append('None')
-                    test_page_bbox.append(0)
-
-
-            if len(test_dropped_text_tag[index])+test_page_tag.count('None')>len(standard_dropped_text_tag[index]):#有多删的情况出现
-                test_page_tag1=test_page_tag.copy()
-                if 'None' in test_page_tag:
-                    test_page_tag1=test_page_tag1.remove('None')
-                else:
-                    test_page_tag1=test_page_tag
-
-                diff=list((Counter(test_dropped_text_tag[index]) - Counter(test_page_tag1)).elements())
-              
-                test_page_tag.extend(diff)
-                standard_page_tag.extend(['None']*len(diff))
-                test_page_bbox.extend([1]*len(diff))
-                standard_page_bbox.extend([0]*len(diff))
-
-            test_tag.extend(test_page_tag)
-            standard_tag.extend(standard_page_tag)
-            test_text_bbox.extend(test_page_bbox)
-            standard_text_bbox.extend(standard_page_bbox)
-
-        index+=1
-
-    
-    text_block_report = {}
-    text_block_report['accuracy']=metrics.accuracy_score(standard_text_bbox,test_text_bbox)
-    text_block_report['precision']=metrics.precision_score(standard_text_bbox,test_text_bbox)
-    text_block_report['recall']=metrics.recall_score(standard_text_bbox,test_text_bbox)
-    text_block_report['f1_score']=metrics.f1_score(standard_text_bbox,test_text_bbox)
-
-    '''删除的text_block的tag的准确率,召回率和f1-score'''
-    text_block_tag_report = classification_report(y_true=standard_tag , y_pred=test_tag,output_dict=True)
-    del text_block_tag_report['None']
-    del text_block_tag_report["macro avg"]
-    del text_block_tag_report["weighted avg"]
-
-
-    '''dropped_image_block的bbox匹配相关指标'''
-    '''有数据格式不一致的问题'''
-    image_block_report=bbox_match_indicator(test_dropped_image_bboxes,standard_dropped_image_bboxes)
-    
-    
-    '''dropped_table_block的bbox匹配相关指标'''
-    table_block_report=bbox_match_indicator(test_dropped_table_bboxes,standard_dropped_table_bboxes)
-    
-   
-    '''阅读顺序编辑距离的均值'''
-    preproc_num_dis=[]
-    for a,b in zip(test_preproc_num,standard_preproc_num):
-        preproc_num_dis.append(Levenshtein_Distance(a,b))
-    preproc_num_edit=np.mean(preproc_num_dis)
-
-
-    '''分段准确率'''
-    test_para_num=np.array(test_para_num)
-    standard_para_num=np.array(standard_para_num)
-    acc_para=np.mean(test_para_num==standard_para_num)
-
-    
-    output=pd.DataFrame()
-    output['总体指标']=[overall_report]
-    output['行内公式平均编辑距离']=[inline_equations_edit_bleu[0]]
-    output['行内公式平均bleu']=[inline_equations_edit_bleu[1]]
-    output['行间公式平均编辑距离']=[interline_equations_edit_bleu[0]]
-    output['行间公式平均bleu']=[interline_equations_edit_bleu[1]]
-    output['行内公式识别相关指标']=[inline_equations_bbox_report]
-    output['行间公式识别相关指标']=[interline_equations_bbox_report]
-    output['阅读顺序平均编辑距离']=[preproc_num_edit]
-    output['分段准确率']=[acc_para]
-    output['删除的text block的相关指标']=[text_block_report]
-    output['删除的image block的相关指标']=[image_block_report]
-    output['删除的table block的相关指标']=[table_block_report]
-    output['删除的text block的tag相关指标']=[text_block_tag_report]
-    
-
-    return output
-
-
-
-"""
-计算编辑距离
-"""
-def Levenshtein_Distance(str1, str2):
-    matrix = [[ i + j for j in range(len(str2) + 1)] for i in range(len(str1) + 1)]
-    for i in range(1, len(str1)+1):
-        for j in range(1, len(str2)+1):
-            if(str1[i-1] == str2[j-1]):
-                d = 0
-            else:
-                d = 1
-            matrix[i][j] = min(matrix[i-1][j]+1, matrix[i][j-1]+1, matrix[i-1][j-1]+d)
-    return matrix[len(str1)][len(str2)]
-
-
-'''
-计算bbox偏移量是否符合标准的函数
-'''
-def bbox_offset(b_t,b_s):
-    '''b_t是test_doc里的bbox,b_s是standard_doc里的bbox'''
-    x1_t,y1_t,x2_t,y2_t=b_t
-    x1_s,y1_s,x2_s,y2_s=b_s
-    x1=max(x1_t,x1_s)
-    x2=min(x2_t,x2_s)
-    y1=max(y1_t,y1_s)
-    y2=min(y2_t,y2_s)
-    area_overlap=(x2-x1)*(y2-y1)
-    area_t=(x2_t-x1_t)*(y2_t-y1_t)+(x2_s-x1_s)*(y2_s-y1_s)-area_overlap
-    if area_t-area_overlap==0 or area_overlap/(area_t-area_overlap)>0.95:
-        return True
-    else:
-        return False
-    
-
-'''bbox匹配和对齐函数，输出相关指标'''
-'''输入的是以page为单位的bbox列表'''
-def bbox_match_indicator(test_bbox_list,standard_bbox_list):
-    
-    test_bbox=[]
-    standard_bbox=[]
-    for a,b in zip(test_bbox_list,standard_bbox_list):
-
-        test_page_bbox=[]
-        standard_page_bbox=[]
-        if len(a)==0 and len(b)==0:
-            pass
-        else:
-            for i in b:
-                if len(i)!=4:
-                    continue
-                else:
-                    judge=0
-                    standard_page_bbox.append(1)
-                    for j in a:
-                        if bbox_offset(i,j):
-                            judge=1
-                            test_page_bbox.append(1)
-                            break
-                    if judge==0:
-                        test_page_bbox.append(0)
-                        
-            diff_num=len(a)+test_page_bbox.count(0)-len(b)
-            if diff_num>0:#有多删的情况出现
-                test_page_bbox.extend([1]*diff_num)
-                standard_page_bbox.extend([0]*diff_num)
-
-          
-            test_bbox.extend(test_page_bbox)
-            standard_bbox.extend(standard_page_bbox)
-
-    
-    block_report = {}
-    block_report['accuracy']=metrics.accuracy_score(standard_bbox,test_bbox)
-    block_report['precision']=metrics.precision_score(standard_bbox,test_bbox)
-    block_report['recall']=metrics.recall_score(standard_bbox,test_bbox)
-    block_report['f1_score']=metrics.f1_score(standard_bbox,test_bbox)
-
-    return block_report
-
-'''公式编辑距离和bleu'''
-def equations_indicator(test_euqations_bboxs,standard_euqations_bboxs,test_equations,standard_equations):
-    test_match_equations=[]
-    standard_match_equations=[]
-
-    index=0
-    for a,b in zip(test_euqations_bboxs,standard_euqations_bboxs):
-        if len(a)==0 and len(b)==0:
-            pass
-        else:
-            for i in range(len(b)):
-                for j in range(len(a)):
-                    if bbox_offset(b[i],a[j]):
-                        standard_match_equations.append(standard_equations[index][i])
-                        test_match_equations.append(test_equations[index][j])
-                        break
-        index+=1
-
-    
-    dis=[]
-    bleu=[]
-    for a,b in zip(test_match_equations,standard_match_equations):
-        if len(a)==0 and len(b)==0:
-            continue
-        else:
-            if a==b:
-                dis.append(0)
-                bleu.append(1)
-            else:
-                dis.append(Levenshtein_Distance(a,b))
-                bleu.append(sentence_bleu([a],b))
-
-    equations_edit=np.mean(dis)
-    equations_bleu=np.mean(bleu)
-    return (equations_edit,equations_bleu)
-
-
-
-
-
-   
-parser = argparse.ArgumentParser()
-parser.add_argument('--test', type=str)
-parser.add_argument('--standard', type=str)
-args = parser.parse_args()
-pdf_json_test = args.test
-pdf_json_standard = args.standard
-
-
-
-if __name__ == '__main__':
-    
-   pdf_json_test = [json.loads(line) 
-                        for line in open(pdf_json_test, 'r', encoding='utf-8')]
-   pdf_json_standard = [json.loads(line) 
-                    for line in open(pdf_json_standard, 'r', encoding='utf-8')]
-   
-   overall_indicator=indicator_cal(pdf_json_standard,pdf_json_test)
-
-   '''计算的指标输出到overall_indicator_output.json中'''
-   overall_indicator.to_json('overall_indicator_output.json',orient='records',lines=True,force_ascii=False)
-    
\ No newline at end of file

tests/retry_env.sh

+#!/bin/bash
+
+# 定义最大重试次数
+max_retries=5
+retry_count=0
+
+while true; do
+    # prepare env
+    source activate MinerU
+    pip install -r requirements-qa.txt
+    pip install magic-pdf[full]==0.7.0b1 --extra-index-url https://wheels.myhloli.com -i https://pypi.tuna.tsinghua.edu.cn/simple
+    pip install paddlepaddle-gpu==3.0.0b1 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/
+    exit_code=$?
+    if [ $exit_code -eq 0 ]; then
+        echo "test.sh 成功执行！"
+        break
+    else
+        let retry_count+=1
+        if [ $retry_count -ge $max_retries ]; then
+            echo "达到最大重试次数 ($max_retries)，放弃重试。"
+            exit 1
+        fi
+        echo "test.sh 执行失败 (退出码: $exit_code)。尝试第 $retry_count 次重试..."
+        sleep 5  # 等待 5 秒后重试
+    fi
+done

tests/test_cli/lib/common.py

-import subprocess
+"""common definitions."""
 import os
+import shutil
 
 
 def check_shell(cmd):
-    """
-    shell successful
-    """
+    """shell successful."""
     res = os.system(cmd)
     assert res == 0
 
-def count_folders_and_check_contents(file_path):
-    """"
-    获取文件夹大小
-    """
+
+def cli_count_folders_and_check_contents(file_path):
+    """" count cli files."""
+    if os.path.exists(file_path):
+        for files in os.listdir(file_path):
+            folder_count = os.path.getsize(os.path.join(file_path, files))
+            assert folder_count > 0
+    assert len(os.listdir(file_path)) > 5
+
+def sdk_count_folders_and_check_contents(file_path):
+    """count folders."""
     if os.path.exists(file_path):
-        folder_count = os.path.getsize(file_path)
-        assert folder_count > 0
+        file_count = os.path.getsize(file_path)
+        assert file_count > 0
+    else:
+        exit(1)
 
 
-if __name__ == "__main__":
-    count_folders_and_check_contents("/home/quyuan/code/Magic-PDF/Magic-PDF/Magic-PDF/ci") 
+def delete_file(path):
+    """delete file."""
\ No newline at end of file
+    if not os.path.exists(path):
+        if os.path.isfile(path):
+            try:
+                os.remove(path)
+                print(f"File '{path}' deleted.")
+            except TypeError as e:
+                print(f"Error deleting file '{path}': {e}")
+    elif os.path.isdir(path):
+        try:
+            shutil.rmtree(path)
+            print(f"Directory '{path}' and its contents deleted.")
+        except TypeError as e:
+            print(f"Error deleting directory '{path}': {e}")
\ No newline at end of file

tests/test_cli/pdf_dev/annotations/cleaned/cleaned_ordinary_textbook_1d9a847603a5e37e379738316820850d.md

-# 数学新星问题征解 
-
-第十五期 (2016.06)
-
-主持: 牟晓生
-
-第一题. 设 $z_{1}, z_{2}, z_{3}$ 是单位复数. 证明存在单位复数 $z$ 使得:
-
-$$
-\frac{1}{\left|z-z_{1}\right|^{2}}+\frac{1}{\left|z-z_{2}\right|^{2}}+\frac{1}{\left|z-z_{3}\right|^{2}} \leq \frac{9}{4}
-$$
-
-(湖北武钢三中学生 王逸轩, 上海大学冷岗松 供题)
-
-第二题. 如图, $D$ 是正三角形 $A B C$ 的边 $B C$ 上一点, $B D>C D$. 记 $O_{1}, I_{1}$ 为 $\triangle A B D$ 的外心与内心, $O_{2}, I_{2}$ 为 $\triangle A C D$ 的外心与内心. 圆 $I_{1}$ 与圆 $I_{2}$ 除 $B C$外的另一条外公切线交 $A B, A C$ 于 $P, Q$. 设直线 $P I_{1}$与 $Q I_{2}$ 交于 $R$, 而直线 $O_{1} I_{1}$ 与 $O_{2} I_{2}$ 交于 $T$. 证明: $A T^{2}=A R^{2}+A D \cdot B C$.
-
-(广西钦州 卢圣 供题)
-
-
-
-第三题. 给定正整数 $m, n$, 考虑在 $m \times n$ 白棋盘上先将一些格染成黑色. 在之后的每一时刻, 若存在一个白格至少与两个黑格相邻, 则可将它也染成黑色. 求最初至少要染多少个黑色格才能在某一时刻染黑整个棋盘?
-
-(哈佛大学 牟晓生 供题)
-
-第四题. $A B C$ 是一个三角形, 而 $P, Q, R$ 分别是 $B C, C A, A B$ 上的点。证明 $\triangle P Q R$ 的周长不小于 $\triangle A Q R, \triangle B R P, \triangle C P Q$ 周长的最小值.
-
-(哈佛大学 牟晓生 供题)
-

tests/test_cli/test_bench.py

@@ -37,9 +37,9 @@ class TestBench():
         now_simscore = now_score["average_sim_score"]
         now_editdistance = now_score["average_edit_distance"]
         now_bleu = now_score["average_bleu_score"]
-        #assert last_simscore <= now_simscore
-        #assert last_editdistance <= now_editdistance
-        #assert last_bleu <= now_bleu
+        assert last_simscore <= now_simscore
+        assert last_editdistance <= now_editdistance
+        assert last_bleu <= now_bleu
 
 
 def get_score():