tablemaster 2.1.5__tar.gz → 2.1.7__tar.gz

{tablemaster-2.1.5 → tablemaster-2.1.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tablemaster
-Version: 2.1.5
+Version: 2.1.7
 Summary: tablemaster is a Python toolkit for moving and managing tabular data across databases, Feishu/Lark, Google Sheets, and local files with one consistent API.
 Author-email: Livid <livid.su@gmail.com>
 Project-URL: Homepage, https://github.com/ilivid/tablemaster

{tablemaster-2.1.5 → tablemaster-2.1.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tablemaster"
-version = "2.1.5"
+version = "2.1.7"
 description = "tablemaster is a Python toolkit for moving and managing tabular data across databases, Feishu/Lark, Google Sheets, and local files with one consistent API."
 readme = "README.md"
 requires-python = ">=3.9"

{tablemaster-2.1.5 → tablemaster-2.1.7}/tablemaster/feishu.py

@@ -10,12 +10,53 @@ logger = logging.getLogger(__name__)
 _TOKEN_CACHE = {}
 
 
+def _normalize_url(url):
+    normalized = url.strip()
+    if '`' in normalized:
+        logger.warning('url contains backticks, auto sanitize')
+        normalized = normalized.replace('`', '').strip()
+    return normalized
+
+
+def _parse_json_response(response, context):
+    if response.status_code >= 400:
+        body_preview = response.text[:500].replace('\n', ' ')
+        raise requests.HTTPError(
+            f'{context} failed with status={response.status_code}, body={body_preview}',
+            response=response
+        )
+
+    content_type = response.headers.get('Content-Type', '')
+    if 'application/json' not in content_type.lower():
+        body_preview = response.text[:500].replace('\n', ' ')
+        raise ValueError(
+            f'{context} expected JSON but got Content-Type={content_type}, body={body_preview}'
+        )
+
+    try:
+        return response.json()
+    except ValueError as exc:
+        body_preview = response.text[:500].replace('\n', ' ')
+        raise ValueError(f'{context} invalid JSON response, body={body_preview}') from exc
+
+
+def _ensure_feishu_success(body, context):
+    if not isinstance(body, dict):
+        raise ValueError(f'{context} expected dict JSON body')
+    code = body.get('code')
+    if code != 0:
+        msg = body.get('msg', 'Unknown error')
+        raise RuntimeError(f'{context} failed with code={code}, msg={msg}')
+    return body
+
+
 def _request_with_retry(method, url, headers=None, params=None, json_data=None, data=None, timeout=30):
     max_retries = 3
+    normalized_url = _normalize_url(url)
     for attempt in range(max_retries):
         response = requests.request(
             method=method,
-            url=url,
+            url=normalized_url,
             headers=headers,
             params=params,
             json=json_data,
@@ -45,8 +86,11 @@ def _get_tenant_access_token(feishu_cfg):
         "app_secret": feishu_cfg.feishu_app_secret
     }
     r = _request_with_retry("post", feishu_url, json_data=post_data)
-    body = r.json()
-    token = body["tenant_access_token"]
+    body = _parse_json_response(r, 'get tenant access token')
+    _ensure_feishu_success(body, 'get tenant access token')
+    token = body.get("tenant_access_token")
+    if not token:
+        raise KeyError('tenant_access_token is missing in auth response')
     expire_seconds = int(body.get('expire', 7200))
     _TOKEN_CACHE[cache_key] = {
         'token': token,
@@ -91,7 +135,17 @@ def fs_read_df(sheet_address, feishu_cfg):
         + "?valueRenderOption=ToString&dateTimeRenderOption=FormattedString"
     )
     r = _request_with_retry("get", url, headers=header)
-    pull_data = r.json()['data']['valueRange']['values']
+    body = _parse_json_response(r, 'read sheets values')
+    _ensure_feishu_success(body, 'read sheets values')
+
+    pull_data = body.get('data', {}).get('valueRange', {}).get('values')
+    if not pull_data:
+        logger.info('sheet is empty: %s', sheet_address)
+        return pd.DataFrame()
+    if not isinstance(pull_data, list):
+        raise ValueError('unexpected sheets values structure: values is not a list')
+    if not pull_data[0]:
+        return pd.DataFrame(pull_data[1:] if len(pull_data) > 1 else [])
     return pd.DataFrame(pull_data[1:], columns=pull_data[0])
 
 
@@ -127,7 +181,9 @@ def fs_read_base(sheet_address, feishu_cfg):
             query_params += f"&page_token={page_token}"
 
         r = _request_with_retry("get", base_url + query_params, headers=header)
-        data = r.json().get('data', {})
+        body = _parse_json_response(r, 'read bitable records')
+        _ensure_feishu_success(body, 'read bitable records')
+        data = body.get('data', {})
         pull_data.extend(data.get('items', []))
         has_more = data.get('has_more', False)
         page_token = data.get('page_token')
@@ -170,15 +226,38 @@ def fs_write_df(sheet_address, df, feishu_cfg, loc='A1', clear_sheet=True):
     
     # 清空工作表（如果需要）
     if clear_sheet:
-        logger.info('clearing sheet by values_batch_clear api')
+        logger.info('clearing sheet before writing')
         try:
-            clear_url = f"https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{spreadsheet_token}/values_batch_clear"
             clear_data = {"ranges": [f"{sheet_id}!A1:XFD1048576"]}
-            clear_resp = _request_with_retry("post", clear_url, headers=header, json_data=clear_data)
-            if clear_resp.json().get('code') == 0:
-                logger.info('sheet cleared')
-            else:
-                raise RuntimeError(f"failed to clear sheet: {clear_resp.json().get('msg')}")
+            clear_urls = [
+                f"https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{spreadsheet_token}/values_batch_clear",
+                f"https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{spreadsheet_token}/values/batch_clear",
+                f"https://open.feishu.cn/open-apis/sheets/v2/spreadsheets/{spreadsheet_token}/values_batch_clear/",
+            ]
+            clear_ok = False
+            last_error = None
+
+            for clear_url in clear_urls:
+                clear_resp = _request_with_retry("post", clear_url, headers=header, json_data=clear_data)
+                if clear_resp.status_code == 404:
+                    last_error = requests.HTTPError(
+                        f'clear sheet endpoint not found: {clear_url}',
+                        response=clear_resp
+                    )
+                    logger.warning('clear endpoint 404, try next: %s', clear_url)
+                    continue
+                clear_body = _parse_json_response(clear_resp, f'clear sheet ({clear_url})')
+                _ensure_feishu_success(clear_body, f'clear sheet ({clear_url})')
+                clear_ok = True
+                logger.info('sheet cleared by endpoint: %s', clear_url)
+                break
+
+            if not clear_ok:
+                logger.warning(
+                    'skip clear sheet because all clear endpoints returned 404; '
+                    'writing will continue and old trailing cells may remain. last_error=%s',
+                    last_error
+                )
         except Exception as e:
             logger.exception('failed to clear sheet: %s', e)
             raise
@@ -240,14 +319,9 @@ def fs_write_df(sheet_address, df, feishu_cfg, loc='A1', clear_sheet=True):
     
     try:
         r = _request_with_retry("put", url, headers=header, json_data=post_data)
-        response = r.json()
-        
-        if response.get('code') == 0:
-            logger.info('data is written')
-        else:
-            logger.error('failed to write data: %s', response.get('msg', 'Unknown error'))
-            logger.error('error code: %s', response.get('code'))
-        
+        response = _parse_json_response(r, 'write sheets values')
+        _ensure_feishu_success(response, 'write sheets values')
+        logger.info('data is written')
         return response
         
     except Exception as e:
@@ -263,13 +337,10 @@ def _get_bitable_fields(app_token, table_id, header):
     """
     url = f"https://open.feishu.cn/open-apis/bitable/v1/apps/{app_token}/tables/{table_id}/fields"
     r = _request_with_retry("get", url, headers=header)
-    
-    if r.status_code == 200 and r.json().get('code') == 0:
-        items = r.json().get('data', {}).get('items', [])
-        return {item['field_name'] for item in items}
-    else:
-        logger.warning("failed to get fields: %s", r.json().get('msg', 'Unknown error'))
-        return set()
+    body = _parse_json_response(r, 'get bitable fields')
+    _ensure_feishu_success(body, 'get bitable fields')
+    items = body.get('data', {}).get('items', [])
+    return {item['field_name'] for item in items}
 
 def fs_write_base(sheet_address, df, feishu_cfg, clear_table=False):
     """

{tablemaster-2.1.5 → tablemaster-2.1.7}/tablemaster.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tablemaster
-Version: 2.1.5
+Version: 2.1.7
 Summary: tablemaster is a Python toolkit for moving and managing tabular data across databases, Feishu/Lark, Google Sheets, and local files with one consistent API.
 Author-email: Livid <livid.su@gmail.com>
 Project-URL: Homepage, https://github.com/ilivid/tablemaster