From 4c586a113ca9fa8332a63e4c23bd58faa4592a34 Mon Sep 17 00:00:00 2001
From: benoit74 <benoit74@users.noreply.github.com>
Date: Wed, 7 Aug 2024 12:28:07 +0000
Subject: [PATCH] Cache HTML rewriting function signature

---
 CHANGELOG.md                           |  1 +
 src/warc2zim/content_rewriting/html.py | 24 +++++++++++++++++-------
 2 files changed, 18 insertions(+), 7 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0bf619e..49de2ce 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -33,6 +33,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Remove subsequent slashes in URLs, both in Python and JS (#365)
 - Ignore non HTTP(S) WARC records (#351)
 - Fix `vimeo_cdn_fix` fuzzy rule for proper operation in Javascript (#348)
+- Performance issue linked to new "extensible" HTML rewriting rules (#370)
 
 ## [2.0.3] - 2024-07-24
 
diff --git a/src/warc2zim/content_rewriting/html.py b/src/warc2zim/content_rewriting/html.py
index abb9dd4..c7232b8 100644
--- a/src/warc2zim/content_rewriting/html.py
+++ b/src/warc2zim/content_rewriting/html.py
@@ -3,9 +3,10 @@
 from collections import namedtuple
 from collections.abc import Callable
 from dataclasses import dataclass
+from functools import cache
 from html import escape
 from html.parser import HTMLParser
-from inspect import signature
+from inspect import Signature, signature
 
 from bs4 import BeautifulSoup
 
@@ -85,6 +86,15 @@ def extract_base_href(content: str) -> str | None:
     return None
 
 
+@cache
+def _cached_signature(func: Callable) -> Signature:
+    """Returns the signature of a given callable
+
+    Result is cached to save performance when reused multiple times
+    """
+    return signature(func)
+
+
 class HtmlRewriter(HTMLParser):
     def __init__(
         self,
@@ -259,8 +269,8 @@ def _check_decorated_func_signature(expected_func: Callable, decorated_func: Cal
 
     It checks that decorated function parameters have known names and proper types
     """
-    expected_params = signature(expected_func).parameters
-    func_params = signature(decorated_func).parameters
+    expected_params = _cached_signature(expected_func).parameters
+    func_params = _cached_signature(decorated_func).parameters
     for name, param in func_params.items():
         if name not in expected_params:
             raise TypeError(
@@ -356,7 +366,7 @@ def _do_drop_attribute(
                         "attr_value": attr_value,
                         "attrs": attrs,
                     }.items()
-                    if arg_name in signature(rule.func).parameters
+                    if arg_name in _cached_signature(rule.func).parameters
                 }
             )
             is True
@@ -399,7 +409,7 @@ def _do_attribute_rewrite(
                             "base_href": base_href,
                             "notify_js_module": notify_js_module,
                         }.items()
-                        if arg_name in signature(rule.func).parameters
+                        if arg_name in _cached_signature(rule.func).parameters
                     }
                 )
             ) is not None:
@@ -429,7 +439,7 @@ def _do_tag_rewrite(
                             "attrs": attrs,
                             "auto_close": auto_close,
                         }.items()
-                        if arg_name in signature(rule.func).parameters
+                        if arg_name in _cached_signature(rule.func).parameters
                     }
                 )
             ) is not None:
@@ -460,7 +470,7 @@ def _do_data_rewrite(
                             "js_rewriter": js_rewriter,
                             "url_rewriter": url_rewriter,
                         }.items()
-                        if arg_name in signature(rule.func).parameters
+                        if arg_name in _cached_signature(rule.func).parameters
                     }
                 )
             ) is not None: