xlin 0.1.1__py3-none-any.whl → 0.1.3__py3-none-any.whl

xlin/jsonl.py

@@ -47,6 +47,28 @@ def jsonlist_to_dataframe(json_list: List[Dict[str, str]]):
     return pd.DataFrame(json_list)
 
 
+def is_jsonl(filepath: str):
+    with open(filepath) as f:
+        try:
+            l = next(f)  # 读取一行，用来判断文件是json还是jsonl格式
+            f.seek(0)
+        except:
+            return False
+
+        try:
+            _ = json.loads(l)
+        except ValueError:
+            return False  # 第一行不是json，所以是json格式
+        else:
+            return True  # 第一行是json，所以是jsonl格式
+
+
+def load_json_or_jsonl(filepath: str):
+    if is_jsonl(filepath):
+        return load_json_list(filepath)
+    return load_json(filepath)
+
+
 def load_json(filename: str):
     with open(filename, "r", encoding="utf-8") as f:
         return json.load(f)
@@ -66,6 +88,8 @@ def load_json_list(filename: str):
             try:
                 obj = json.loads(i.strip())
             except:
+                print("格式损坏数据，无法加载")
+                print(i)
                 continue
             json_list.append(obj)
         return json_list

xlin/multiprocess_mapping.py

@@ -12,18 +12,11 @@ from loguru import logger
 from xlin.jsonl import dataframe_to_json_list, load_json_list, save_json_list, load_json, save_json
 
 
-cpu_count = multiprocessing.cpu_count()
-# pool = ThreadPool(cpu_count)  # 大模型接口辣鸡，太快会截断答案
-thread_pool_size = int(os.getenv("THREAD_POOL_SIZE", 5))
-pool = ThreadPool(thread_pool_size)
-logger.debug(f"pool size: {thread_pool_size}, cpu count: {cpu_count}")
-
-
 def multiprocessing_mapping_jsonlist(
     jsonlist: List[Any],
     output_path: Optional[Union[str, Path]],
     partial_func,
-    batch_size=cpu_count * 2,
+    batch_size=multiprocessing.cpu_count(),
     cache_batch_num=1,
     thread_pool_size=int(os.getenv("THREAD_POOL_SIZE", 5)),
 ):
@@ -47,7 +40,7 @@ def multiprocessing_mapping_jsonlist(
         else:
             output_path.parent.mkdir(parents=True, exist_ok=True)
     pool = ThreadPool(thread_pool_size)
-    logger.debug(f"pool size: {thread_pool_size}, cpu count: {cpu_count}")
+    logger.debug(f"pool size: {thread_pool_size}, cpu count: {multiprocessing.cpu_count()}")
     start_time = time.time()
     last_save_time = start_time
     for i, line in tqdm(list(enumerate(jsonlist))):
@@ -78,7 +71,7 @@ def multiprocessing_mapping(
     df: pd.DataFrame,
     output_path: Optional[Union[str, Path]],
     partial_func,
-    batch_size=cpu_count * 2,
+    batch_size=multiprocessing.cpu_count(),
     cache_batch_num=1,
     thread_pool_size=int(os.getenv("THREAD_POOL_SIZE", 5)),
 ):
@@ -104,7 +97,7 @@ def multiprocessing_mapping(
         else:
             output_path.parent.mkdir(parents=True, exist_ok=True)
     pool = ThreadPool(thread_pool_size)
-    logger.debug(f"pool size: {thread_pool_size}, cpu count: {cpu_count}")
+    logger.debug(f"pool size: {thread_pool_size}, cpu count: {multiprocessing.cpu_count()}")
     start_time = time.time()
     last_save_time = start_time
     for i, line in tqdm(list(df.iterrows())):
@@ -135,29 +128,51 @@ def multiprocessing_mapping(
     return output_df, output_list
 
 
-def dataframe_by_row_mapping(
+def dataframe_with_row_mapping(
     df: pd.DataFrame,
-    mapping_func: Callable[[dict], Tuple[bool, dict]],
+    mapping_func: Callable[[int, dict], Tuple[bool, dict]],
     use_multiprocessing=True,
     thread_pool_size=int(os.getenv("THREAD_POOL_SIZE", 5)),
 ):
     rows = []
     if use_multiprocessing:
         pool = ThreadPool(thread_pool_size)
-        logger.debug(f"pool size: {thread_pool_size}, cpu count: {cpu_count}")
-        results = pool.map(mapping_func, dataframe_to_json_list(df))
+        logger.debug(f"pool size: {thread_pool_size}, cpu count: {multiprocessing.cpu_count()}")
+        results = pool.map(mapping_func, enumerate(dataframe_to_json_list(df)))
         for ok, row in results:
             if ok:
                 rows.append(row)
     else:
-        for i, row in df.iterrows():
-            ok, row = mapping_func(row)
+        for i, row in tqdm(df.iterrows()):
+            ok, row = mapping_func(i, row)
             if ok:
                 rows.append(row)
     df = pd.DataFrame(rows)
     return df
 
 
+def list_with_element_mapping(
+    iterator: List[Any],
+    mapping_func: Callable[[int, Any], Tuple[bool, Any]],
+    use_multiprocessing=True,
+    thread_pool_size=int(os.getenv("THREAD_POOL_SIZE", 5)),
+):
+    rows = []
+    if use_multiprocessing:
+        pool = ThreadPool(thread_pool_size)
+        logger.debug(f"pool size: {thread_pool_size}, cpu count: {multiprocessing.cpu_count()}")
+        results = pool.map(mapping_func, enumerate(iterator))
+        for ok, row in results:
+            if ok:
+                rows.append(row)
+    else:
+        for i, row in tqdm(enumerate(iterator)):
+            ok, row = mapping_func(i, row)
+            if ok:
+                rows.append(row)
+    return rows
+
+
 def continue_run(
     jsonfiles: List[str],
     save_dir: str,

{xlin-0.1.1.dist-info → xlin-0.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: xlin
-Version: 0.1.1
+Version: 0.1.3
 Summary: toolbox for LinXueyuan
 License: MIT
 Author: XiChen

{xlin-0.1.1.dist-info → xlin-0.1.3.dist-info}/RECORD

@@ -1,7 +1,7 @@
 xlin/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 xlin/ischinese.py,sha256=Ia9IMQ6q-UHkdLwqS70L1fTnfSPbluFrv_I1UqsKquo,293
-xlin/jsonl.py,sha256=Fg9JqyzV9gmBW7R4P8aqdURyM3pD_pS2c2wVeFK2MVY,5938
-xlin/multiprocess_mapping.py,sha256=DE-liX94UlWJUuDAgIb_FpM6oaiNZ5Hq-HapdbU-DAg,6971
+xlin/jsonl.py,sha256=oE8w8IFVEnBQdWUCMGYF9BlE3wtEhFsmjaLpZPKwSXg,6605
+xlin/multiprocess_mapping.py,sha256=mvkxra4uPvHfwHfI38FMEZU4dVYf6mu3gMVMAC1-P4o,7502
 xlin/read_as_dataframe.py,sha256=ir3HUT6dt3crqa3xnlcNn8j3wqjSIGJgiIVLP3KkBaQ,8678
 xlin/statistic.py,sha256=BLj8hszlbBT5xDIfd70_YtOb8QgZEvYXiFJDGXBwCfw,881
 xlin/terminal_color.py,sha256=nfE-CY2BzjY2eZbm9yk8r-AuyJ-hchmLXhASCb4HAIA,191
@@ -9,7 +9,7 @@ xlin/util.py,sha256=hme7Zl4Sa_-FTA9TEVzr1qTdaKW1eq5dTWZgd4owcDc,11303
 xlin/uuid.py,sha256=gouvm7_DL22sIhXl-g4e6S2qzIZtmE3SEp00xy1upyg,271
 xlin/xls2xlsx.py,sha256=5zfcM0gmunFQOcOj9nYd9Dj0HMhU7-cPKnPIy6Ot9iU,930
 xlin/yaml.py,sha256=kICi7G3Td5q2MaSXXt85qNTWoHMgjzt7pvn7r3C4dME,183
-xlin-0.1.1.dist-info/LICENSE,sha256=KX0dDCYlO4DskqMZY8qeY94EZMrDRNnNqlGLkXVlKyM,1063
-xlin-0.1.1.dist-info/METADATA,sha256=_Kg8_eS9Atxo_2PF_h80KEeu-WG1Y6r0_uvYdlczy48,772
-xlin-0.1.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-xlin-0.1.1.dist-info/RECORD,,
+xlin-0.1.3.dist-info/LICENSE,sha256=KX0dDCYlO4DskqMZY8qeY94EZMrDRNnNqlGLkXVlKyM,1063
+xlin-0.1.3.dist-info/METADATA,sha256=Z0r6K1HueSI2T-U8I3xxV5Vl-92NZuwnQVuztiXClIk,772
+xlin-0.1.3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+xlin-0.1.3.dist-info/RECORD,,