Remove obsoleted output files (#153)

statements_manager/src/convert_task_runner.py

@@ -1,7 +1,9 @@
 from __future__ import annotations
 
+import glob
 import hashlib
 import json
+import os
 import pathlib
 import shutil
 from enum import Enum
@@ -155,9 +157,10 @@ def copy_assets(
                 logger.info("copy assets file")
                 if assets_dst_path.exists():
                     logger.warning(
-                        f"assets directory '{assets_dst_path}' already exists."
+                        f"assets directory '{assets_dst_path}' already exists. overwriting..."
                     )
-                shutil.copytree(assets_src_path, assets_dst_path, dirs_exist_ok=True)
+                    shutil.rmtree(assets_dst_path)
+                shutil.copytree(assets_src_path, assets_dst_path)
 
                 for path in assets_dst_path.glob("**/*"):
                     with open(path, "rb") as f:
@@ -293,6 +296,15 @@ def run(
                 reference_cache = json.load(open(output_dir / "cache.json"))
             reference_cache.setdefault(output_ext, {})
             reference_cache[output_ext].setdefault(problem_id, {})
+            problem_group = self.problemset_config.get_problem_group(problem_id)
+            for ext in reference_cache.keys():
+                obsoleted_ids = list(
+                    filter(
+                        lambda id: id not in problem_group, reference_cache[ext].keys()
+                    )
+                )
+                for id in obsoleted_ids:
+                    reference_cache[ext].pop(id)
 
             problem_cache["assets"] = self.copy_assets(
                 problem_id, output_dir / "assets"
@@ -313,6 +325,21 @@ def run(
                 sort_keys=True,
             )
             dict_merge(problemset_cache, reference_cache[output_ext])
+
+            filenames = list(
+                filter(
+                    lambda filename: pathlib.Path(filename).stem not in problem_group,
+                    sum(
+                        [
+                            list(glob.glob(str(output_dir) + f"/*.{ext}"))
+                            for ext in ["html", "pdf", "md"]
+                        ],
+                        [],
+                    ),
+                )
+            )
+            for filename in filenames:
+                os.remove(filename)
             logger.info("")
 
         # 問題セットに対応するものを出力

statements_manager/src/execute_config.py

@@ -175,6 +175,7 @@ def __init__(self, problemset_filename: pathlib.Path, config: dict) -> None:
         )
         self.pdf_config = PDFRenderingConfig(problemset_filename, config.get("pdf", {}))
         self.known_ids: set[str] = set()
+        self.id_groups: list[list[str]] = list()
         self.problem_configs: dict[str, ProblemConfig] = dict()
 
         dirname = problemset_filename.parent.resolve()
@@ -192,10 +193,18 @@ def get_problem(self, id: str) -> ProblemConfig:
     def get_problem_ids(self) -> list[str]:
         return sorted(list(self.problem_configs.keys()))
 
+    def get_problem_group(self, id: str) -> list[str]:
+        for group in self.id_groups:
+            if id in group:
+                return group
+        logger.error("problem id not found in any group")
+        raise ValueError("problem id not found in any group")
+
     def add_problem_configs(self, filename: pathlib.Path) -> None:
         raw_config = RawProblemConfig(filename, toml.load(filename))
         self._check_id(raw_config)
         numbered_ids = get_numbered_ids(raw_config)
+        self.id_groups.append(numbered_ids)
         for id, statement_config in zip(numbered_ids, raw_config.statements):
             self.problem_configs[id] = ProblemConfig(
                 filename, id, raw_config, statement_config