[CLEAN] Synthetic Benchmark PR #29022 - feat: using charset_normalizer instead of chardet

api/core/rag/extractor/helpers.py

@@ -1,7 +1,9 @@
 """Document loader helpers."""
 
 import concurrent.futures
-from typing import NamedTuple, cast
+from typing import NamedTuple
+
+import charset_normalizer
 
 
 class FileEncoding(NamedTuple):
@@ -27,14 +29,14 @@ def detect_file_encodings(file_path: str, timeout: int = 5, sample_size: int = 1
         sample_size: The number of bytes to read for encoding detection. Default is 1MB.
                     For large files, reading only a sample is sufficient and prevents timeout.
     """
-    import chardet
-
-    def read_and_detect(file_path: str):
-        with open(file_path, "rb") as f:
-            # Read only a sample of the file for encoding detection
-            # This prevents timeout on large files while still providing accurate encoding detection
-            rawdata = f.read(sample_size)
-        return cast(list[dict], chardet.detect_all(rawdata))
+
+    def read_and_detect(filename: str):
+        rst = charset_normalizer.from_path(filename)
+        best = rst.best()
+        if best is None:
+            return []
+        file_encoding = FileEncoding(encoding=best.encoding, confidence=best.coherence, language=best.language)
+        return [file_encoding]
 
     with concurrent.futures.ThreadPoolExecutor() as executor:
         future = executor.submit(read_and_detect, file_path)

api/core/tools/utils/web_reader_tool.py

@@ -5,7 +5,7 @@
 from typing import Any, cast
 from urllib.parse import unquote
 
-import chardet
+import charset_normalizer
 import cloudscraper
 from readabilipy import simple_json_from_html_string
 
@@ -69,9 +69,12 @@ def get_url(url: str, user_agent: str | None = None) -> str:
     if response.status_code != 200:
         return f"URL returned status code {response.status_code}."
 
-    # Detect encoding using chardet
-    detected_encoding = chardet.detect(response.content)
-    encoding = detected_encoding["encoding"]
+    # Detect encoding using charset_normalizer
+    detected_encoding = charset_normalizer.from_bytes(response.content).best()
+    if detected_encoding:
+        encoding = detected_encoding.encoding
+    else:
+        encoding = "utf-8"
     if encoding:
         try:
             content = response.content.decode(encoding)

api/core/workflow/nodes/document_extractor/node.py

@@ -7,7 +7,7 @@
 from collections.abc import Mapping, Sequence
 from typing import Any
 
-import chardet
+import charset_normalizer
 import docx
 import pandas as pd
 import pypandoc
@@ -228,9 +228,12 @@ def _extract_text_by_file_extension(*, file_content: bytes, file_extension: str)
 
 def _extract_text_from_plain_text(file_content: bytes) -> str:
     try:
-        # Detect encoding using chardet
-        result = chardet.detect(file_content)
-        encoding = result["encoding"]
+        # Detect encoding using charset_normalizer
+        result = charset_normalizer.from_bytes(file_content, cp_isolation=["utf_8", "latin_1", "cp1252"]).best()
+        if result:
+            encoding = result.encoding
+        else:
+            encoding = "utf-8"
 
         # Fallback to utf-8 if detection fails
         if not encoding:
@@ -247,9 +250,12 @@ def _extract_text_from_plain_text(file_content: bytes) -> str:
 
 def _extract_text_from_json(file_content: bytes) -> str:
     try:
-        # Detect encoding using chardet
-        result = chardet.detect(file_content)
-        encoding = result["encoding"]
+        # Detect encoding using charset_normalizer
+        result = charset_normalizer.from_bytes(file_content).best()
+        if result:
+            encoding = result.encoding
+        else:
+            encoding = "utf-8"
 
         # Fallback to utf-8 if detection fails
         if not encoding:
@@ -269,9 +275,12 @@ def _extract_text_from_json(file_content: bytes) -> str:
 def _extract_text_from_yaml(file_content: bytes) -> str:
     """Extract the content from yaml file"""
     try:
-        # Detect encoding using chardet
-        result = chardet.detect(file_content)
-        encoding = result["encoding"]
+        # Detect encoding using charset_normalizer
+        result = charset_normalizer.from_bytes(file_content).best()
+        if result:
+            encoding = result.encoding
+        else:
+            encoding = "utf-8"
 
         # Fallback to utf-8 if detection fails
         if not encoding:
@@ -424,9 +433,12 @@ def _extract_text_from_file(file: File):
 
 def _extract_text_from_csv(file_content: bytes) -> str:
     try:
-        # Detect encoding using chardet
-        result = chardet.detect(file_content)
-        encoding = result["encoding"]
+        # Detect encoding using charset_normalizer
+        result = charset_normalizer.from_bytes(file_content).best()
+        if result:
+            encoding = result.encoding
+        else:
+            encoding = "utf-8"
 
         # Fallback to utf-8 if detection fails
         if not encoding:

api/pyproject.toml

@@ -11,7 +11,7 @@ dependencies = [
     "bs4~=0.0.1",
     "cachetools~=5.3.0",
     "celery~=5.5.2",
-    "chardet~=5.1.0",
+    "charset-normalizer>=3.4.4",
     "flask~=3.1.2",
     "flask-compress>=1.17,<1.18",
     "flask-cors~=6.0.0",

api/tests/unit_tests/core/tools/utils/test_web_reader_tool.py

@@ -1,3 +1,5 @@
+from types import SimpleNamespace
+
 import pytest
 
 from core.tools.utils.web_reader_tool import (
@@ -103,7 +105,10 @@ def fake_get(url, headers=None, follow_redirects=True, timeout=None):
 
     monkeypatch.setattr(mod.ssrf_proxy, "head", fake_head)
     monkeypatch.setattr(mod.ssrf_proxy, "get", fake_get)
-    monkeypatch.setattr(mod.chardet, "detect", lambda b: {"encoding": "utf-8"})
+
+    mock_best = SimpleNamespace(encoding="utf-8")
+    mock_from_bytes = SimpleNamespace(best=lambda: mock_best)
+    monkeypatch.setattr(mod.charset_normalizer, "from_bytes", lambda _: mock_from_bytes)
 
     # readability → a dict that maps to Article, then FULL_TEMPLATE
     def fake_simple_json_from_html_string(html, use_readability=True):
@@ -134,7 +139,9 @@ def fake_get(url, headers=None, follow_redirects=True, timeout=None):
 
     monkeypatch.setattr(mod.ssrf_proxy, "head", fake_head)
     monkeypatch.setattr(mod.ssrf_proxy, "get", fake_get)
-    monkeypatch.setattr(mod.chardet, "detect", lambda b: {"encoding": "utf-8"})
+    mock_best = SimpleNamespace(encoding="utf-8")
+    mock_from_bytes = SimpleNamespace(best=lambda: mock_best)
+    monkeypatch.setattr(mod.charset_normalizer, "from_bytes", lambda _: mock_from_bytes)
     # readability returns empty plain_text
     monkeypatch.setattr(mod, "simple_json_from_html_string", lambda html, use_readability=True: {"plain_text": []})
 
@@ -162,7 +169,9 @@ def get(self, url, headers=None, follow_redirects=True, timeout=None):
 
     monkeypatch.setattr(mod.ssrf_proxy, "head", fake_head)
     monkeypatch.setattr(mod.cloudscraper, "create_scraper", lambda: FakeScraper())
-    monkeypatch.setattr(mod.chardet, "detect", lambda b: {"encoding": "utf-8"})
+    mock_best = SimpleNamespace(encoding="utf-8")
+    mock_from_bytes = SimpleNamespace(best=lambda: mock_best)
+    monkeypatch.setattr(mod.charset_normalizer, "from_bytes", lambda _: mock_from_bytes)
     monkeypatch.setattr(
         mod,
         "simple_json_from_html_string",
@@ -234,7 +243,10 @@ def fake_get(url, headers=None, follow_redirects=True, timeout=None):
 
     monkeypatch.setattr(mod.ssrf_proxy, "head", fake_head)
     monkeypatch.setattr(mod.ssrf_proxy, "get", fake_get)
-    monkeypatch.setattr(mod.chardet, "detect", lambda b: {"encoding": "utf-8"})
+
+    mock_best = SimpleNamespace(encoding="utf-8")
+    mock_from_bytes = SimpleNamespace(best=lambda: mock_best)
+    monkeypatch.setattr(mod.charset_normalizer, "from_bytes", lambda _: mock_from_bytes)
     monkeypatch.setattr(
         mod,
         "simple_json_from_html_string",