Created
June 25, 2023 13:59
-
-
Save magiskboy/3f734d28b9d244f318c1dbac8d5624ad to your computer and use it in GitHub Desktop.
Simple HTTP parser, 1M req/s
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# STATE | |
STATE_METHOD = 0 | |
STATE_PATH = 1 | |
STATE_HTTP_VERSION = 2 | |
STATE_HEADER = 3 | |
STATE_BODY = 4 | |
class ParserBase: | |
def __init__(self, protocol): | |
self.protocol = protocol | |
self.current_token = b"" | |
self.current_state = STATE_METHOD | |
def parse_header(self, header: bytes) -> (bytes, bytes): | |
idx = header.find(b":") | |
if idx < 0: | |
raise ValueError(f"Header {header} is invalid") | |
name, value = header[:idx], header[idx+1:] | |
return name.strip(), value.strip() | |
def feed_data(self, data: bytes): | |
for c in data.decode("utf-8"): | |
self.current_token += c | |
if self.current_state == STATE_METHOD: | |
if self.current_token.endswith(" "): | |
self.protocol.on_method(self.current_token[:-1]) | |
self.current_token = "" | |
self.current_state = STATE_PATH | |
continue | |
if self.current_state == STATE_PATH: | |
if self.current_token.endswith(" "): | |
self.protocol.on_path(self.current_token[:-1]) | |
self.current_token = "" | |
self.current_state = STATE_HTTP_VERSION | |
continue | |
if self.current_state == STATE_HTTP_VERSION: | |
if self.current_token.endswith("\r\n"): | |
self.protocol.on_http_version(self.current_token[:-2]) | |
self.current_token = "" | |
self.current_state = STATE_HEADER | |
continue | |
if self.current_state == STATE_HEADER: | |
if self.current_token.endswith("\r\n"): | |
header = self.current_token[:-2] | |
if header: | |
self.protocol.on_header(*self.parse_header(header)) | |
self.current_token = "" | |
self.current_state = STATE_HEADER | |
else: # blank line | |
self.protocol.on_headers_completed() | |
self.current_token = "" | |
self.current_state = STATE_BODY | |
continue | |
if self.current_state == STATE_BODY: | |
self.protocol.on_body(self.current_token) | |
self.current_token = "" | |
class Parser(ParserBase): | |
def __init__(self, protocol): | |
super().__init__(protocol) | |
self.current_token = b"" | |
def feed_data(self, data: bytes): | |
idx = 0 | |
len_data = len(data) | |
while idx < len_data: | |
if self.current_state == STATE_METHOD: | |
i = data.find(b" ", idx) | |
if i == -1: | |
self.current_token += data[idx:] | |
break | |
else: | |
self.current_token += data[idx:i] | |
self.protocol.on_method(self.current_token) | |
self.current_token = b"" | |
idx = i + 1 | |
self.current_state = STATE_PATH | |
continue | |
if self.current_state == STATE_PATH: | |
i = data.find(b" ", idx) | |
if i == -1: | |
self.current_token += data[idx:] | |
break | |
else: | |
self.current_token += data[idx:i] | |
self.protocol.on_path(self.current_token) | |
self.current_token = b"" | |
idx = i + 1 | |
self.current_state = STATE_HTTP_VERSION | |
continue | |
if self.current_state == STATE_HTTP_VERSION: | |
i = data.find(b"\n", idx) | |
if i == -1: | |
self.current_token += data[idx:] | |
break | |
else: | |
self.current_token += data[idx:i] | |
self.protocol.on_http_version(self.current_token[:-1]) | |
self.current_token = b"" | |
idx = i + 1 | |
self.current_state = STATE_HEADER | |
continue | |
if self.current_state == STATE_HEADER: | |
i = data.find(b"\n", idx) | |
if i == -1: | |
self.current_token += data[idx:] | |
break | |
else: | |
self.current_token += data[idx:i] | |
if self.current_token == b"\r\n": | |
self.current_state = STATE_BODY | |
self.protocol.on_headers_completed() | |
else: | |
header = self.current_token[:-1] | |
if header: | |
self.protocol.on_header(*self.parse_header(header)) | |
self.current_token = b"" | |
idx = i + 1 | |
if self.current_state == STATE_BODY: | |
self.protocol.on_body(data[idx:]) | |
break |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment