init datamate

runtime/ops/mapper/url_cleaner/__init__.py

@@ -0,0 +1,6 @@
+# -*- coding: utf-8 -*-
+
+from datamate.core.base_op import OPERATORS
+
+OPERATORS.register_module(module_name='AnonymizedUrlCleaner',
+                          module_path="ops.mapper.url_cleaner.process")

runtime/ops/mapper/url_cleaner/metadata.yml

@@ -0,0 +1,16 @@
+name: 'URL网址匿名化'
+name_en: 'URL Anonymization'
+description: '将文档中的url网址匿名化。'
+description_en: 'Anonymizes URLs in documents.'
+language: 'python'
+vendor: 'huawei'
+raw_id: 'AnonymizedUrlCleaner'
+version: '1.0.0'
+types:
+  - 'cleanse'
+modal: 'text'
+effect:
+  before: '需要被屏蔽的url：https://www.huawei.com'
+  after: '需要被屏蔽的url：<url>'
+inputs: 'text'
+outputs: 'text'

runtime/ops/mapper/url_cleaner/process.py

@@ -0,0 +1,36 @@
+#!/user/bin/python
+# -*- coding: utf-8 -*-
+
+"""
+Description: URL网址匿名化
+Create: 2024/12/26 15:43
+"""
+import re
+import time
+from typing import Dict, Any
+
+from loguru import logger
+
+from datamate.core.base_op import Mapper
+
+
+class AnonymizedUrlCleaner(Mapper):
+    """将文档中的网址匿名化"""
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.url_pattern = r'((?:(?:https?|ftp|file)://|(?<![a-zA-Z\-\.])www\.)' \
+                           r'[\-A-Za-z0-9\+&@\(\)#/%\?=\^~_|!:\,\.\;]+[\-A-Za-z0-9\+&@#/%=\~_\|])' \
+                           r'(?![\-A-Za-z0-9\+&@#/%=\~_\|])'
+        self.url_re_compile = re.compile(self.url_pattern, re.MULTILINE)
+
+    def execute(self, sample: Dict[str, Any]) -> Dict[str, Any]:
+        start = time.time()
+        sample[self.text_key] = self._url_filter(sample[self.text_key])
+        logger.info(f"fileName: {sample[self.filename_key]}, method: UrlCleaner costs {time.time() - start:6f} s")
+        return sample
+
+    def _url_filter(self, input_data: str):
+        input_data = ''.join(['【', input_data, '】'])
+        text = self.url_re_compile.sub("<url>", input_data)
+        return text[1:-1]