refactor : parse.pyをクラスで整理した。それにともない依存しているdownload, generate, open, m…

…arkdown, problemなど広範囲の修正を行なった

atcdr/download.py

@@ -8,12 +8,7 @@
 from rich.prompt import Prompt
 
 from atcdr.util.filetype import FILE_EXTENSIONS, Lang
-from atcdr.util.parse import (
-    get_title_from_html,
-    make_problem_markdown,
-    repair_html,
-    title_to_filename,
-)
+from atcdr.util.parse import ProblemHTML
 from atcdr.util.problem import Contest, Diff, Problem
 from atcdr.util.session import load_session
 
@@ -22,15 +17,15 @@ class Downloader:
     def __init__(self) -> None:
         self.session = load_session()
 
-    def get(self, problem: Problem) -> str:
+    def get(self, problem: Problem) -> ProblemHTML:
         session = self.session
         retry_attempts = 3
         retry_wait = 1  # 1 second
 
         for _ in range(retry_attempts):
             response = session.get(problem.url)
             if response.status_code == 200:
-                return response.text
+                return ProblemHTML(response.text)
             elif response.status_code == 429:
                 print(
                     f'[bold yellow][Error {response.status_code}][/bold yellow] 再試行します。{problem}'
@@ -55,13 +50,7 @@ def get(self, problem: Problem) -> str:
                     f'[bold red][Error {response.status_code}][/bold red] {problem}に対応するHTMLファイルを取得できませんでした。'
                 )
                 break
-        return ''
-
-
-def save_file(file_path: str, html: str) -> None:
-    with open(file_path, 'w', encoding='utf-8') as file:
-        file.write(html)
-    print(f'[bold green][+][/bold green] ファイルを保存しました :{file_path}')
+        return ProblemHTML('')
 
 
 def mkdir(path: str) -> None:
@@ -88,32 +77,38 @@ def gene_path_on_num(base: str, problem: Problem) -> str:
         )
 
 
+def title_to_filename(title: str) -> str:
+    title = re.sub(r'[\\/*?:"<>| !@#$%^&()+=\[\]{};,\']', '', title)
+    title = re.sub(r'.*?-', '', title)
+    return title
+
+
 def generate_problem_directory(
     base_path: str, problems: List[Problem], gene_path: Callable[[str, Problem], str]
 ) -> None:
     downloader = Downloader()
     for problem in problems:
         dir_path = gene_path(base_path, problem)
+        mkdir(dir_path)
+        problem_content = downloader.get(problem)
+        if not problem_content:
+            print(f'[bold red][Error][/] {problem}の保存に失敗しました')
+            return
+        problem_content.repair_me()
 
-        html = downloader.get(problem)
-        if not html:
-            continue
-
-        title = get_title_from_html(html)
-        if not title:
-            print('[bold red][Error][/bold red] タイトルが取得できませんでした')
-            title = f'{problem}'
-
+        title = problem_content.title or problem.label
         title = title_to_filename(title)
 
-        mkdir(dir_path)
-        repaired_html = repair_html(html)
-
         html_path = os.path.join(dir_path, title + FILE_EXTENSIONS[Lang.HTML])
-        save_file(html_path, repaired_html)
-        md = make_problem_markdown(html, 'ja')
+        with open(html_path, 'w', encoding='utf-8') as file:
+            file.write(problem_content.html)
+        print(f'[bold green][+][/bold green] ファイルを保存しました :{html_path}')
+
+        md = problem_content.make_problem_markdown('ja')
         md_path = os.path.join(dir_path, title + FILE_EXTENSIONS[Lang.MARKDOWN])
-        save_file(md_path, md)
+        with open(md_path, 'w', encoding='utf-8') as file:
+            file.write(md)
+        print(f'[bold green][+][/bold green] ファイルを保存しました :{md_path}')
 
 
 def parse_range(match: re.Match) -> List[int]:

atcdr/generate.py

@@ -10,7 +10,6 @@
     LabeledTestCaseResult,
     ResultStatus,
     TestInformation,
-    create_testcases_from_html,
     judge_code_from,
 )
 from atcdr.util.execute import execute_files
@@ -22,7 +21,7 @@
     str2lang,
 )
 from atcdr.util.gpt import ChatGPT, set_api_key
-from atcdr.util.parse import make_problem_markdown
+from atcdr.util.parse import ProblemHTML
 
 
 def get_code_from_gpt_output(output: str) -> str:
@@ -65,7 +64,7 @@ def generate_code(file: Filename, lang: Lang) -> None:
     console = Console()
     with open(file, 'r') as f:
         html_content = f.read()
-    md = make_problem_markdown(html_content, 'en')
+    md = ProblemHTML(html_content).make_problem_markdown('en')
 
     if set_api_key() is None:
         return
@@ -96,7 +95,7 @@ def generate_template(file: Filename, lang: Lang) -> None:
     console = Console()
     with open(file, 'r') as f:
         html_content = f.read()
-    md = make_problem_markdown(html_content, 'en')
+    md = ProblemHTML(html_content).make_problem_markdown('en')
 
     if set_api_key() is None:
         return
@@ -131,9 +130,10 @@ def generate_template(file: Filename, lang: Lang) -> None:
 def solve_problem(file: Filename, lang: Lang) -> None:
     console = Console()
     with open(file, 'r') as f:
-        html_content = f.read()
-    md = make_problem_markdown(html_content, 'en')
-    labeled_cases = create_testcases_from_html(html_content)
+        html = ProblemHTML(f.read())
+
+    md = html.make_problem_markdown('en')
+    labeled_cases = html.load_labeled_testcase()
 
     if set_api_key() is None:
         return

atcdr/markdown.py

@@ -5,14 +5,14 @@
 
 from atcdr.util.execute import execute_files
 from atcdr.util.filetype import FILE_EXTENSIONS, Lang
-from atcdr.util.parse import make_problem_markdown
+from atcdr.util.parse import ProblemHTML
 
 
 def save_markdown(html_path: str, lang: str) -> None:
     console = Console()
     with open(html_path, 'r', encoding='utf-8') as f:
-        html = f.read()
-    md = make_problem_markdown(html, lang)
+        html = ProblemHTML(f.read())
+    md = html.make_problem_markdown(lang)
     file_without_ext = os.path.splitext(html_path)[0]
     md_path = file_without_ext + FILE_EXTENSIONS[Lang.MARKDOWN]
 

atcdr/open.py

@@ -4,7 +4,7 @@
 
 from atcdr.util.filetype import Lang
 from atcdr.util.execute import execute_files
-from atcdr.util.parse import find_link_from_html
+from atcdr.util.parse import ProblemHTML
 
 
 def open_html(file: str) -> None:
@@ -21,7 +21,7 @@ def open_html(file: str) -> None:
         )
         return
 
-    url = find_link_from_html(html_content)
+    url = ProblemHTML(html_content).link
     if url:
         webbrowser.open_new_tab(url)
         console.print(

atcdr/test.py

@@ -6,7 +6,6 @@
 from enum import Enum
 from typing import Dict, Generator, List, Optional, Tuple, Union
 
-from bs4 import BeautifulSoup as bs
 from rich.console import Group, RenderableType
 from rich.live import Live
 from rich.markup import escape
@@ -26,6 +25,7 @@
     detect_language,
     lang2str,
 )
+from atcdr.util.parse import ProblemHTML
 
 
 @dataclass
@@ -76,37 +76,6 @@ class TestInformation:
     compile_time: Optional[int] = None
 
 
-def create_testcases_from_html(html: str) -> List[LabeledTestCase]:
-    soup = bs(html, 'html.parser')
-    test_cases = []
-
-    for i in range(1, 20):
-        sample_input_section = soup.find('h3', text=f'Sample Input {i}')
-        sample_output_section = soup.find('h3', text=f'Sample Output {i}')
-        if not sample_input_section or not sample_output_section:
-            break
-
-        sample_input_pre = sample_input_section.find_next('pre')
-        sample_output_pre = sample_output_section.find_next('pre')
-
-        sample_input = (
-            sample_input_pre.get_text(strip=True)
-            if sample_input_pre is not None
-            else ''
-        )
-        sample_output = (
-            sample_output_pre.get_text(strip=True)
-            if sample_output_pre is not None
-            else ''
-        )
-
-        test_case = TestCase(input=sample_input, output=sample_output)
-        labeled_test_case = LabeledTestCase(label=f'Sample {i}', case=test_case)
-        test_cases.append(labeled_test_case)
-
-    return test_cases
-
-
 def run_code(cmd: list, case: TestCase) -> TestCaseResult:
     start_time = time.time()
     try:
@@ -454,9 +423,10 @@ def run_test(path_of_code: str) -> None:
     with open(html_paths[0], 'r') as file:
         html = file.read()
 
-    test_cases = create_testcases_from_html(html)
-    test_results = judge_code_from(test_cases, path_of_code)
-    render_results(test_results)
+    ltest_cases = ProblemHTML(html).load_labeled_testcase()
+
+    ltest_results = judge_code_from(ltest_cases, path_of_code)
+    render_results(ltest_results)
 
 
 def test(*args: str) -> None: