handle json parsing errors (#5371)

adds tests cases, consolidates a lot of PRs

langchain/output_parsers/json.py

@@ -1,20 +1,28 @@
 from __future__ import annotations
 
 import json
+import re
 from typing import List
 
 from langchain.schema import OutputParserException
 
 
 def parse_json_markdown(json_string: str) -> dict:
-    # Remove the triple backticks if present
+    # Try to find JSON string within triple backticks
-    json_string = json_string.replace("```json", "").replace("```", "")
+    match = re.search(r"```(json)?(.*?)```", json_string, re.DOTALL)
+
+    # If no match found, assume the entire string is a JSON string
+    if match is None:
+        json_str = json_string
+    else:
+        # If match found, use the content within the backticks
+        json_str = match.group(2)
 
     # Strip whitespace and newlines from the start and end
-    json_string = json_string.strip()
+    json_str = json_str.strip()
 
     # Parse the JSON string into a Python dictionary
-    parsed = json.loads(json_string)
+    parsed = json.loads(json_str)
 
     return parsed
 

tests/unit_tests/output_parsers/test_json.py

@@ -64,6 +64,31 @@ NO_TICKS_WHITE_SPACE = """
 }
 """
 
+TEXT_BEFORE = """Thought: I need to use the search tool
+
+Action:
+```
+{
+  "foo": "bar"
+}
+```"""
+
+TEXT_AFTER = """```
+{
+  "foo": "bar"
+}
+```
+This should do the trick"""
+
+TEXT_BEFORE_AND_AFTER = """Action: Testing
+
+```
+{
+  "foo": "bar"
+}
+```
+This should do the trick"""
+
 TEST_CASES = [
     GOOD_JSON,
     JSON_WITH_NEW_LINES,
@@ -72,6 +97,8 @@ TEST_CASES = [
     TICKS_WITH_NEW_LINES_EVERYWHERE,
     NO_TICKS,
     NO_TICKS_WHITE_SPACE,
+    TEXT_BEFORE,
+    TEXT_AFTER,
 ]