pyxllib 0.3.127__tar.gz → 0.3.128__tar.gz

{pyxllib-0.3.127/pyxllib.egg-info → pyxllib-0.3.128}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyxllib
-Version: 0.3.127
+Version: 0.3.128
 Summary: 厦门理工模式识别团队通用python代码工具库
 Home-page: https://github.com/XLPRUtils/pyxllib
 Author: code4101

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxllib/data/pglib.py

@@ -194,7 +194,7 @@ WHERE {table_name}.{item_id_name} = cte.{item_id_name}"""
     @classmethod
     def cvt_type(cls, val):
         if isinstance(val, (dict, list)):
-            val = json.dumps(val, ensure_ascii=False)
+            val = json.dumps(val, ensure_ascii=False, default=str)
         # 注意list数组类型读、写都会自动适配py
         return val
 

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxllib/ext/yuquelib.py

@@ -197,7 +197,7 @@ class Yuque:
         elif return_mode == 'md':
             return res['data']['body']
         elif return_mode == 'title_and_md':
-            return f'文章标题：{res["data"]["title"]}\n\n{res["data"]["body"]}'
+            return res["data"]["title"], res["data"]["body"]
 
     def export_markdown(self, url, output_dir=None, post_mode=1):
         """ 导出md格式文件

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxllib/file/xlsxlib.py

@@ -3018,12 +3018,51 @@ def extract_workbook_summary3(file_path, summary_limit_len=4000, **kwargs):
     return data
 
 
+def summary2_add_enums(summary2, enum_values):
+    # 1 预备
+    if enum_values is True:
+        enum_values = (20, 10)
+    max_len, max_num = enum_values
+
+    # 2 枚举值
+    for sheet in summary2['sheets']:
+        # 2.1 遍历计数
+        cols = defaultdict(Counter)
+        for addr, val in sheet['cells'].items():
+            n = len(str(val))
+            if not n or n > max_len:
+                continue
+            col = re.match(r'[A-Z]+', addr).group()
+            cols[col][val] += 1
+
+        # 2.2 添加枚举值列
+        enums = {}
+        keys = sorted(cols.keys(), key=column_index_from_string)
+        for k in keys:
+            ct = cols[k]
+            if len(ct) > max_num:
+                continue
+            vals = ct.most_common()
+            if vals[0][1] == 1:  # 都只出现了一次，也不认为是枚举值，跳过。或者是小数据表，一般也能全量展示。
+                continue
+            enums[k] = [v for v, _ in vals]
+
+        # 2.3 保存
+        if enums:
+            sheet['enums'] = enums
+            # enums2 = json.dumps(enums, ensure_ascii=False, default=str)
+            # sheet['enums'] = json.loads(enums2)
+
+    return summary2
+
+
 def extract_workbook_summary3b(file_path,
                                summary_limit_len=4000,
                                timeout_seconds=10,
                                return_mode=0,
                                debug=False,
                                len_mode=0,
+                               enum_values=False,
                                **kwargs):
     """
 
@@ -3036,6 +3075,9 @@ def extract_workbook_summary3b(file_path,
     :param len_mode:
         0, 使用len作为token长度评估
         1, 使用模型评估实际token长度
+    :param enum_values: 是否展示每列枚举值
+        False, 默认不展示
+        True, 展示，并且默认参数 (20, 10) 表示长度超过20的丢弃，只保留枚举类型不超过10种值的列
     :param kwargs: 其他是summary2读取文件的时候的参数，其实都不太关键，一般不用特地设置
     """
     res = {}
@@ -3056,7 +3098,10 @@ def extract_workbook_summary3b(file_path,
             res, load_time = extract_workbook_summary2(file_path, mode=1, return_mode=1, **kwargs)
             # res = convert_to_json_compatible(res)
             summary2_res = copy.deepcopy(res)
+            if enum_values:
+                res = summary2_add_enums(res, enum_values)
             summary2_time = time.time() - start_time - load_time
+
             start_time = time.time()
             if len_mode == 1:
                 res = WorkbookSummary3plus.summary2_to_summary3b(res, summary_limit_len)

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxllib/prog/pupil.py

@@ -1223,7 +1223,6 @@ def support_multi_processes(default_processes=1):
                         cmds.append(f'--{k}')  # 添加关键字参数的键
                         cmds.append(str(v))  # 添加关键字参数的值
 
-                    mpl.add_process_python_module()
                     mpl.add_process_python(sys.argv[0], cmds, shell=shell)
                 mpl.run_endless()
 

{pyxllib-0.3.127 → pyxllib-0.3.128/pyxllib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyxllib
-Version: 0.3.127
+Version: 0.3.128
 Summary: 厦门理工模式识别团队通用python代码工具库
 Home-page: https://github.com/XLPRUtils/pyxllib
 Author: code4101

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxlpr/openai2/_core/chat.py

@@ -145,7 +145,7 @@ class Chat:
                  model: Literal["gpt-4-1106-preview", "gpt-4-vision-preview",
                  "gpt-4", "gpt-4-0314", "gpt-4-0613",
                  "gpt-4-32k", "gpt-4-32k-0314", "gpt-4-32k-0613", "gpt-3.5-turbo",
-                 "gpt-4o", "gpt-4-turbo"] = "gpt-3.5-turbo",
+                 "gpt-4o", "gpt-4-turbo"] = "gpt-4o-mini",
                  # Chat
                  msg_max_count: int = None,
                  # kwargs
@@ -505,7 +505,7 @@ class Chat2:
                  gizmo_id=None,
                  *,
                  base_url=None,
-                 model='gpt-4o'):
+                 model='gpt-4o-mini'):
         """
         :param gizmo_id: gpts的id，比如 https://chatgpt.com/g/g-ABCD1234-ce-shi。则id就是"g-ABCD1234"
             后面的gpts的名称变了没关系，前面前缀的id不要变就行
@@ -697,7 +697,7 @@ class CompressContent:
               expect_len=None,
               max_round=-1,
               text_spliter=None,
-              model='gpt-3.5-turbo',
+              model='gpt-4o-mini',
               soft_merge_prompt=None,
               calc_len=None,
               logger=None,

{pyxllib-0.3.127 → pyxllib-0.3.128}/setup.py

@@ -48,7 +48,7 @@ _dir = Path(__file__).parent
 
 setup(
     name='pyxllib',  # pip 安装时用的名字
-    version='0.3.127',  # 当前版本，每次更新上传到pypi都需要修改; 第4位版本号一般是修紧急bug
+    version='0.3.128',  # 当前版本，每次更新上传到pypi都需要修改; 第4位版本号一般是修紧急bug
     author='code4101',
     author_email='877362867@qq.com',
     url='https://github.com/XLPRUtils/pyxllib',

{pyxllib-0.3.127 → pyxllib-0.3.128}/pyxllib.egg-info/requires.txt

@@ -24,30 +24,30 @@ pillow
 
 [xlai]
 pillow
-opsdroid-get-image-size
-pynvml
 visdom
+opsdroid-get-image-size
 opencv-python
+pynvml
 
 [xlall]
 pillow
-flask-jwt-extended
-flask
 fastapi
-opsdroid-get-image-size
-html2text
+xlrd
+flask-wtf
+sentencepiece
+flask
+zhconv
 ujson
+flask-jwt-extended
 arts
 uvicorn
-flask-wtf
-zhconv
-sentencepiece
-flask-cors
-xlrd
 flask-restful
+opsdroid-get-image-size
+flask-cors
 opencv-python
+html2text
 
 [xlcv]
+pillow
 opencv-python
 opsdroid-get-image-size
-pillow