vodorod/Dorod-Sky
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

⚡️ Speed up function fix_unescaped_quotes_in_json by 106% (#1891)

Browse Source 
Co-authored-by: codeflash-ai[bot] <148906541+codeflash-ai[bot]@users.noreply.github.com>
Co-authored-by: Shuchang Zheng <shu@skyvern.com>
This commit is contained in:
Saurabh Misra
2025-03-08 02:34:58 -08:00
committed by GitHub
parent fcc87243c1
commit 21cedc43d1
1 changed files with 8 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
skyvern/forge/sdk/api/llm/utils.py
View File

@@ -122,12 +122,10 @@ def fix_unescaped_quotes_in_json(json_string: str) -> str:
str: The JSON-like string with unescaped quotation marks within strings. str: The JSON-like string with unescaped quotation marks within strings.
""" """
escape_char = "\\" escape_char = "\\"
# Indices to add the escape character to. Since we're processing the string from left to right, we need to sort
# the indices in descending order to avoid index shifting.
indices_to_add_escape_char = []
in_string = False in_string = False
escape = False escape = False
json_structure_chars = {",", ":", "}", "]", "{", "["} json_structure_chars = {",", ":", "}", "]", "{", "["}
result = []
i = 0 i = 0
while i < len(json_string): while i < len(json_string):
@@ -146,23 +144,22 @@ def fix_unescaped_quotes_in_json(json_string: str) -> str:
in_string = False in_string = False
else: else:
# If the next character is not a JSON structure character, the quote is part of the string # If the next character is not a JSON structure character, the quote is part of the string
# Update the indices to add the escape character with the current index # Add the escape character before the quote
indices_to_add_escape_char.append(i) result.append(escape_char)
else: else:
# Start of the JSON string # Start of the JSON string
in_string = True in_string = True
else: else:
escape = False escape = False
# Append the current character to the result
result.append(char)
i += 1 i += 1
# Sort the indices in descending order to avoid index shifting then add the escape character to the string if len(result) != len(json_string):
if indices_to_add_escape_char:
LOG.warning("Unescaped quotes found in JSON string. Adding escape character to fix the issue.") LOG.warning("Unescaped quotes found in JSON string. Adding escape character to fix the issue.")
indices_to_add_escape_char.sort(reverse=True)
for index in indices_to_add_escape_char:
json_string = json_string[:index] + escape_char + json_string[index:]
return json_string return "".join(result)
def fix_and_parse_json_string(json_string: str) -> dict[str, Any]: def fix_and_parse_json_string(json_string: str) -> dict[str, Any]: