From: mittis <gerhardmitterlechner@gmail.com>
Date: Sun, 4 Jan 2026 06:55:35 +0000 (+0100)
Subject: Regex search: allow for literal search of special regex chars
X-Git-Url: https://git.sthu.org/?a=commitdiff_plain;h=f98fb6435a00b51e4321efde39dc0d5e0185a60d;p=oe1archive.git

Regex search: allow for literal search of special regex chars

Parentheses "(",")" and curly braces "{","}"are escaped now
and not interpreted as regex characters.
This allows for searching broadcasts with -s "Das.*Leben (2)"
which matches the title "Das gute Leben (2)".
Previously "Das gute Leben (2)" led to empty search result
as the paraenthes are used in regex to capture groups.
---

diff --git a/README.markdown b/README.markdown
index 2bd52f3..9d488f3 100644
--- a/README.markdown
+++ b/README.markdown
@@ -62,16 +62,63 @@ extended search:
 # Fast search by title/subtitle
 ./oe1archive -s "Show Name"
 
+# Search for show with special characters (automatically escaped)
+./oe1archive -s "Das gute Leben (2)"
+
+# Search using regex patterns (wildcards still work)
+./oe1archive -s "Das.*Leben"
+
+# Search with alternation (OR) patterns
+./oe1archive -s "Music|Talk"
+
 # Deep search including descriptions
 ./oe1archive -s "Some Keyword" -e -r 30
 
 # Batch download all matches (fast search)
 ./oe1archive -d "Show Name" -p "prefix" -r 30
 
+# Batch download with special characters in title
+./oe1archive -d "Das gute Leben (2)" -p "prefix"
+
 # Batch download all matches (deep search)
 ./oe1archive -d "Some Keyword" -p "prefix" -e -r 30
 ```
 
+## Search Behavior and Regex Escaping
+
+The search feature uses **smart escaping** to make searching for titles with special characters intuitive, while preserving regex pattern support.
+
+### Automatically Escaped Characters
+The following characters are **automatically escaped** (treated as literals):
+- `(` `)` - parentheses
+- `{` `}` - curly braces
+
+This means you can search for titles like "Das gute Leben (2)" without needing to manually escape the parentheses.
+
+### Supported Regex Patterns
+The following regex patterns **still work** and are NOT escaped:
+- `.*` - match any characters (wildcard)
+- `.+` - match one or more of any character
+- `|` - alternation (OR operator)
+- `^` - start of string
+- `$` - end of string
+- `[...]` - character classes
+
+### Examples
+```bash
+# Literal search - parentheses are automatically escaped
+./oe1archive -s "Das gute Leben (2)"
+
+# Regex wildcard - matches "Das gute Leben", "Das schöne Leben", etc.
+./oe1archive -s "Das.*Leben"
+
+# Combine both - wildcard with escaped parentheses
+./oe1archive -s "Das.*Leben (2)"
+
+# Alternation - matches either "Music" or "Talk"
+./oe1archive -s "Music|Talk"
+```
+
 ## How It Works
 
 The OE1 API provides a rolling weekly window of current broadcasts. Streams
diff --git a/oe1archive b/oe1archive
index 4c2d33b..1f6ed08 100755
--- a/oe1archive
+++ b/oe1archive
@@ -265,17 +265,35 @@ class Archive:
             akm = ""
         return description + "<br>" + akm
 
+    def _smart_escape_for_regex(self, pattern):
+        """Escape literal chars () {} while preserving regex patterns like .*
+        
+        Args:
+            pattern: Search pattern string
+
+        Returns:
+            Pattern with literal characters escaped for regex
+        """
+        chars_to_escape = ['(', ')', '{', '}']
+        result = pattern
+        for char in chars_to_escape:
+            result = result.replace(char, '\\' + char)
+        return result
+
     def get_broadcasts_by_regex(self, key, deep_search=False):
-        """Find broadcasts matching a regex pattern.
+        """Find broadcasts matching a search pattern (with smart escaping).
 
         Args:
-            key: Search pattern (regex)
+            key: Search pattern (literal by default, but supports regex patterns)
+                 Literal characters like () and {} are automatically escaped,
+                 while regex patterns like .* and | continue to work.
             deep_search: If True, search in title, subtitle, and description.
                         If False, search only in title and subtitle (faster).
 
         Skips placeholder entries.
         """
-        rex = re.compile(key, re.IGNORECASE)
+        escaped_key = self._smart_escape_for_regex(key)
+        rex = re.compile(escaped_key, re.IGNORECASE)
 
         res = []
         total = sum(len(djson["broadcasts"]) for djson in self.json)