github/FlareSolverr
1
1
Fork 0
mirror of https://github.com/FlareSolverr/FlareSolverr.git synced 2025-07-24 19:10:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

Refactor POST request handling to streamline logic and add contentType validation. Remove obsolete request.postJSON logic and integrate selenium_fetch. Update requirements.txt with selenium_fetch dependency and enhance DTOs to support contentType and response fields.

Browse Source
This commit is contained in:
Wyatt Henke 2025-06-23 20:05:11 -04:00
parent 48383a459d
commit 668e0aeca1
3 changed files with 51 additions and 177 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
requirements.txt
View File

@ -11,3 +11,5 @@ websockets==11.0.3
xvfbwrapper==0.2.9; platform_system != "Windows"
# only required for windows
pefile==2023.2.7; platform_system == "Windows"
selenium_fetch

2
src/dtos.py
View File

@ -19,6 +19,7 @@ class ChallengeResolutionT:
status: str = None
message: str = None
result: ChallengeResolutionResultT = None
response: str = None
def __init__(self, _dict):
self.__dict__.update(_dict)
@ -39,6 +40,7 @@ class V1RequestBase(object):
# V1Request
url: str = None
contentType: str = None
postData: str = None
returnOnlyCookies: bool = None
download: bool = None # deprecated v2.0.0, not used

224
src/flaresolverr_service.py
View File

@ -1,8 +1,8 @@
import json
import logging
import platform
import sys
import time
import json
from datetime import timedelta
from html import escape
from urllib.parse import unquote, quote
@ -10,13 +10,13 @@ from urllib.parse import unquote, quote
from func_timeout import FunctionTimedOut, func_timeout
from selenium.common import TimeoutException
from selenium.webdriver.chrome.webdriver import WebDriver
from selenium.webdriver.common.action_chains import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support.expected_conditions import (
presence_of_element_located, staleness_of, title_is)
from selenium.webdriver.common.action_chains import ActionChains
from selenium.webdriver.support.wait import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium_fetch import fetch, Options
import utils
from dtos import (STATUS_ERROR, STATUS_OK, ChallengeResolutionResultT,
@ -138,8 +138,6 @@ def _controller_v1_handler(req: V1RequestBase) -> V1ResponseBase:
res = _cmd_request_get(req)
elif req.cmd == 'request.post':
res = _cmd_request_post(req)
elif req.cmd == 'request.postJSON':
res = _cmd_request_postJSON(req)
else:
raise Exception(f"Request parameter 'cmd' = '{req.cmd}' is invalid.")
@ -152,6 +150,8 @@ def _cmd_request_get(req: V1RequestBase) -> V1ResponseBase:
raise Exception("Request parameter 'url' is mandatory in 'request.get' command.")
if req.postData is not None:
raise Exception("Cannot use 'postBody' when sending a GET request.")
if req.contentType is not None:
raise Exception("Cannot use 'contentType' when sending a GET request.")
if req.returnRawHtml is not None:
logging.warning("Request parameter 'returnRawHtml' was removed in FlareSolverr v2.")
if req.download is not None:
@ -169,6 +169,8 @@ def _cmd_request_post(req: V1RequestBase) -> V1ResponseBase:
# do some validations
if req.postData is None:
raise Exception("Request parameter 'postData' is mandatory in 'request.post' command.")
if req.contentType is None:
raise Exception("Request parameter 'contentType' is mandatory in 'request.post' command.")
if req.returnRawHtml is not None:
logging.warning("Request parameter 'returnRawHtml' was removed in FlareSolverr v2.")
if req.download is not None:
@ -312,9 +314,7 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
# navigate to the page
logging.debug(f'Navigating to... {req.url}')
if method == 'POST':
_post_request(req, driver)
elif method == 'POSTJSON':
_post_request_json(req, driver)
res.response = _post_request(req, driver)
else:
driver.get(req.url)
@ -326,7 +326,7 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
driver.add_cookie(cookie)
# reload the page
if method == 'POST':
_post_request(req, driver)
res.response = _post_request(req, driver)
else:
driver.get(req.url)
@ -420,172 +420,42 @@ def _evil_logic(req: V1RequestBase, driver: WebDriver, method: str) -> Challenge
def _post_request(req: V1RequestBase, driver: WebDriver):
post_form = f'<form id="hackForm" action="{req.url}" method="POST">'
query_string = req.postData if req.postData[0] != '?' else req.postData[1:]
pairs = query_string.split('&')
for pair in pairs:
parts = pair.split('=')
# noinspection PyBroadException
try:
name = unquote(parts[0])
except Exception:
name = parts[0]
if name == 'submit':
continue
# noinspection PyBroadException
try:
value = unquote(parts[1])
except Exception:
value = parts[1]
post_form += f'<input type="text" name="{escape(quote(name))}" value="{escape(quote(value))}"><br>'
post_form += '</form>'
html_content = f"""
<!DOCTYPE html>
<html>
<body>
{post_form}
<script>document.getElementById('hackForm').submit();</script>
</body>
</html>"""
driver.get("data:text/html;charset=utf-8,{html_content}".format(html_content=html_content))
def _post_request_json(req: V1RequestBase, driver: WebDriver):
try:
# Step 2a: JSON Payload Preparation
# Convert Python dictionary to JSON string.
json_payload_str_raw = json.dumps(req.postData)
# Step 2b: HTML Document Generation
# Construct a minimal HTML5 document with embedded JavaScript.
html_content = f"""
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>POST Request Executor</title>
</head>
<body>
<h1>Executing POST Request...</h1>
<div id="status-message">Request initiated...</div>
<script>
// JavaScript implementation
(async function() {{
const url = '{req.url}';
// The JSON payload string is directly embedded here.
// Note: It's wrapped in single quotes in the JS.
const jsonPayloadStr = {json_payload_str_raw};
const statusDiv = document.getElementById('status-message');
console.log('Request URL:', url);
console.log('JSON Payload (raw string for JS):', jsonPayloadStr);
try {{
// Parse the JSON string to ensure it's valid before sending
// This also correctly unescapes characters for the actual HTTP body
const jsonData = JSON.parse(jsonPayloadStr);
console.log('JSON Payload (parsed JS object):', jsonData);
statusDiv.textContent = 'Sending request to ' + url + '...';
// fetch API Call
const response = await fetch(url, {{
method: 'POST',
headers: {{
'Content-Type': 'application/json',
// Add any other headers if necessary, e.g., 'Accept': 'application/json'
}},
body: JSON.stringify(jsonData) // Send the parsed and re-stringified JSON object
}});
statusDiv.textContent = 'Waiting for response... Status: ' + response.status;
const responseText = await response.text(); // Get response as text
if (!response.ok) {{
// If response is not ok, throw an error to be caught by the catch block
throw new Error(`HTTP error! Status: ${{response.status}} - ${{response.statusText}}. Response: ${{responseText.substring(0, 500)}}`);
}}
// Success Display
// Displaying a snippet of the response.
const responseSnippet = responseText.substring(0, 200);
statusDiv.textContent = `Success! Status: ${{response.status}}. Response Snippet: ${{responseSnippet}}`;
console.log('Full Response:', responseText);
}} catch (error) {{
// Error Handling
console.error('Fetch Error:', error);
statusDiv.textContent = 'Error: ' + error.message;
}}
}})();
</script>
</body>
</html>
"""
# Step 2c: WebDriver Execution
# URL-encode the HTML content for the data URI.
# `safe=''` ensures that characters like '/', '?', '&', '=', ':' are also encoded.
# However, for data URIs, common characters in HTML are generally fine.
# The primary concern is characters that would break the URI structure itself.
# Using `quote` without `safe=''` is usually sufficient for `data:text/html`.
# Let's be more specific with `safe` if issues arise, but default should be fine.
encoded_html_content = quote(html_content)
data_uri = f"data:text/html,{encoded_html_content}"
# Load the data URI into the WebDriver.
driver.get(data_uri)
# Wait for the JavaScript to update the status message.
# This indicates that the fetch operation has likely completed (or failed).
# We wait for "Success!", "Error:", or "HTTP error!" to appear in the div.
# Increased timeout for potentially slow network requests.
wait = WebDriverWait(driver, 30) # 30 seconds timeout
try:
status_element = wait.until(
EC.presence_of_element_located((By.ID, "status-message"))
)
# Wait until text contains one of the terminal keywords
wait.until(
lambda d: "Success!" in status_element.text or \
"Error:" in status_element.text or \
"HTTP error!" in status_element.text or \
"Request initiated..." not in status_element.text # fallback if it never changes from initial
)
# A small explicit sleep to allow final JS updates to the DOM if any race condition.
time.sleep(1)
final_status_text = status_element.text
except Exception as e:
print(f"Timeout or error waiting for status message: {e}")
# Try to get current status message anyway or logs
if req.contentType == 'application/x-www-form-urlencoded':
post_form = f'<form id="hackForm" action="{req.url}" method="POST">'
query_string = req.postData if req.postData[0] != '?' else req.postData[1:]
pairs = query_string.split('&')
for pair in pairs:
parts = pair.split('=')
# noinspection PyBroadException
try:
final_status_text = driver.find_element(By.ID, "status-message").text
except:
final_status_text = "Error: Could not retrieve final status message."
# Retrieve browser console logs for debugging (optional but good for complex cases)
try:
browser_logs = driver.get_log('browser')
if browser_logs:
print("Browser Console Logs:")
for entry in browser_logs:
print(f" [{entry['level']}] {entry['message']}")
except Exception as e:
print(f"Could not retrieve browser logs: {e} (This might be normal if not supported by driver/config)")
return final_status_text
except json.JSONDecodeError as e:
error_msg = f"JSON Encoding Error: {e}"
print(error_msg)
return error_msg
except Exception as e:
# Catch any other exceptions during the process.
error_msg = f"An unexpected error occurred in solve_post_json: {e}"
print(error_msg)
# It might be useful to also get the current page source for debugging
# print("Current Page Source on Error:\n", driver.page_source)
return error_msg
name = unquote(parts[0])
except Exception:
name = parts[0]
if name == 'submit':
continue
# noinspection PyBroadException
try:
value = unquote(parts[1])
except Exception:
value = parts[1]
post_form += f'<input type="text" name="{escape(quote(name))}" value="{escape(quote(value))}"><br>'
post_form += '</form>'
html_content = f"""
<!DOCTYPE html>
<html>
<body>
{post_form}
<script>document.getElementById('hackForm').submit();</script>
</body>
</html>"""
driver.get("data:text/html;charset=utf-8,{html_content}".format(html_content=html_content))
return "Success"
elif req.contentType == 'application/json':
post_data = json.loads(unquote(req.postData))
options = Options(method="POST", body=post_data)
logging.debug(f"Request => POST /v1 options: {utils.object_to_dict(options)}")
response = fetch(driver, req.url, options)
logging.debug(f"Response => POST /v1 response: {utils.object_to_dict(response)}")
return response.text
else:
raise Exception(f"Request parameter 'contentType' = '{req.contentType}' is invalid.")