lpeg_http_parser.lua

local lpeg = require "lpeg"
local re = require "re"
local json = require "json"

local P = lpeg.P
local S = lpeg.S
local R = lpeg.R
local C = lpeg.C
local Ct = lpeg.Ct
local Cmt = lpeg.Cmt

local function snake_upper(s)
  s = string.upper(s)
  return (s:gsub("%-", "_"))
end

-- HTTP PROTOCOL GRAMMAR
--line endings
CRLF = P"\r\n"

-- character types
any = P(1)
cntrl = R(string.char(0) .. string.char(31))
CTL = cntrl + string.char(127)
safe = S"$-_."
extra = S"!*'(),"
reserved = S";/?:@&=+"
sorta_safe = S"\"<>"
unsafe = CTL + S" #%" + sorta_safe
alpha = R("az", "AZ")
digit = R"09"
national = any - (alpha + digit + reserved + extra + safe + unsafe)
unreserved = alpha + digit + safe + extra + national
xdigit = R("09", "af", "AF")
escape = P"%" * xdigit * xdigit
uchar = unreserved + escape + sorta_safe
pchar = uchar + S":@&=+"
tspecials = S"()<>@,;:\\\"/[]?={} \t"
lws = S" \t"

-- elements
ascii = R(string.char(0) .. string.char(127))
token = ascii - (CTL + tspecials)

-- URI schemes and absolute paths
scheme = S"hH" * S"tT" * S"tT" * S"pP" * S"sS"^-1 / function (s) return string.lower(s) end
alnum = R("09", "az", "AZ")
hostname = (alnum + S"-._")^1
host_with_port = C(hostname) * (":" * C(digit^1))^-1 / function (h, p)
							 return { host = h, port = tonumber(p) or 80 } 
						       end
userinfo = ((unreserved + escape + S";:&=+")^1 * "@")^0

path = pchar^1 * ( P"/" * pchar^0 )^0
query = (uchar + reserved)^0
param = C((pchar + "/")^0)
params = param * (";" * param)^0
rel_path = (C(path^-1) * Ct((";" * params)^-1)) * ("?" * C(query))^-1 / function (path, params, query)
								      return { path = path, params = params,
									       query = query }
								    end
absolute_path =  P("/")^1 * rel_path / function (rel_path) 
					 rel_path.path = "/" .. rel_path.path
					 return rel_path
				       end
path_uri = absolute_path
Absolute_URI = scheme * "://" * userinfo * host_with_port * path_uri / function (scheme, host, path)
									 return {
									   scheme = scheme,
									   host = host.host,
									   port = host.port,
									   path = path.path,
									   params = path.params,
									   query = path.query
									 }
								       end

Request_URI = absolute_path + (P"*" / function () return {} end) + Absolute_URI
Fragment = (uchar + reserved)^0;
Method = token^1 / function (s) return string.upper(s) end
GetOnly = C"GET"

http_number = digit^1 * "." * digit^1
HTTP_Version = C("HTTP/" * http_number)
Request_Line = Method * " " * Request_URI * ("#" * C(Fragment))^-1 *
                   " " * HTTP_Version * CRLF / function (method, uri, fragment, version)
						 if fragment and not version then
						   version, fragment = fragment, version
						 end
						 return { method = method, uri = uri, version = version,
							  fragment = fragment }
					       end

field_name = token^1 / function (s)
			 return snake_upper(s)
		       end

field_value = C((any - CRLF)^0)

value_cont = lws^1 * C((any - CRLF)^0) * CRLF

message_header = field_name * ":" * P(" ")^0 * field_value * CRLF *
                    Ct(value_cont^0) / function (name, value, conts)
					 value = value .. table.concat(conts)
					 return { name = name, value = value }
				       end

FullRequest = Request_Line * Ct(message_header^0) * CRLF / function (reqline, headers)
							     return { line = reqline, headers = headers }
							   end
SimpleRequest = GetOnly * " " * Request_URI * (P"#" * C(Fragment))^-1 * CRLF * CRLF / function (uri, fragment)
											return {
											  line  = {
											    method = "GET",
											    uri = uri,
											    fragment =
											      fragment
											  },
											  headers = {}
											}
										      end

main = FullRequest + SimpleRequest

do
  local http = "GET / HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET /foo/bar.html HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST /foo/bar.html HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST /foo/bar.html\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST http://foo.com/foo/bar.html HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST https://foo.com/foo/bar.html HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST https://foo.com:8080/foo/bar.html HTTP/1.1\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\nConnection: close\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\nHost: foo\r\nX-Forwarded-Proto: https\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\naaaaaaaaaaaaa:++++++++++\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\nX-SSL-Bullshit:   -----BEGIN CERTIFICATE-----\r\n\tMIIFbTCCBFWgAwIBAgICH4cwDQYJKoZIhvcNAQEFBQAwcDELMAkGA1UEBhMCVUsx\r\n\tETAPBgNVBAoTCGVTY2llbmNlMRIwEAYDVQQLEwlBdXRob3JpdHkxCzAJBgNVBAMT\r\n\tAkNBMS0wKwYJKoZIhvcNAQkBFh5jYS1vcGVyYXRvckBncmlkLXN1cHBvcnQuYWMu\r\n\tdWswHhcNMDYwNzI3MTQxMzI4WhcNMDcwNzI3MTQxMzI4WjBbMQswCQYDVQQGEwJV\r\n\tSzERMA8GA1UEChMIZVNjaWVuY2UxEzARBgNVBAsTCk1hbmNoZXN0ZXIxCzAJBgNV\r\n\tBAcTmrsogriqMWLAk1DMRcwFQYDVQQDEw5taWNoYWVsIHBhcmQYJKoZIhvcNAQEB\r\n\tBQADggEPADCCAQoCggEBANPEQBgl1IaKdSS1TbhF3hEXSl72G9J+WC/1R64fAcEF\r\n\tW51rEyFYiIeZGx/BVzwXbeBoNUK41OK65sxGuflMo5gLflbwJtHBRIEKAfVVp3YR\r\n\tgW7cMA/s/XKgL1GEC7rQw8lIZT8RApukCGqOVHSi/F1SiFlPDxuDfmdiNzL31+sL\r\n\t0iwHDdNkGjy5pyBSB8Y79dsSJtCW/iaLB0/n8Sj7HgvvZJ7x0fr+RQjYOUUfrePP\r\n\tu2MSpFyf+9BbC/aXgaZuiCvSR+8Snv3xApQY+fULK/xY8h8Ua51iXoQ5jrgu2SqR\r\n\twgA7BUi3G8LFzMBl8FRCDYGUDy7M6QaHXx1ZWIPWNKsCAwEAAaOCAiQwggIgMAwG\r\n\tA1UdEwEB/wQCMAAwEQYJYIZIAYb4QgEBBAQDAgWgMA4GA1UdDwEB/wQEAwID6DAs\r\n\tBglghkgBhvhCAQ0EHxYdVUsgZS1TY2llbmNlIFVzZXIgQ2VydGlmaWNhdGUwHQYD\r\n\tVR0OBBYEFDTt/sf9PeMaZDHkUIldrDYMNTBZMIGaBgNVHSMEgZIwgY+AFAI4qxGj\r\n\tloCLDdMVKwiljjDastqooXSkcjBwMQswCQYDVQQGEwJVSzERMA8GA1UEChMIZVNj\r\n\taWVuY2UxEjAQBgNVBAsTCUF1dGhvcml0eTELMAkGA1UEAxMCQ0ExLTArBgkqhkiG\r\n\t9w0BCQEWHmNhLW9wZXJhdG9yQGdyaWQtc3VwcG9ydC5hYy51a4IBADApBgNVHRIE\r\n\tIjAggR5jYS1vcGVyYXRvckBncmlkLXN1cHBvcnQuYWMudWswGQYDVR0gBBIwEDAO\r\n\tBgwrBgEEAdkvAQEBAQYwPQYJYIZIAYb4QgEEBDAWLmh0dHA6Ly9jYS5ncmlkLXN1\r\n\tcHBvcnQuYWMudmT4sopwqlBWsvcHViL2NybC9jYWNybC5jcmwwPQYJYIZIAYb4QgEDBDAWLmh0\r\n\tdHA6Ly9jYS5ncmlkLXN1cHBvcnQuYWMudWsvcHViL2NybC9jYWNybC5jcmwwPwYD\r\n\tVR0fBDgwNjA0oDKgMIYuaHR0cDovL2NhLmdyaWQt5hYy51ay9wdWIv\r\n\tY3JsL2NhY3JsLmNybDANBgkqhkiG9w0BAQUFAAOCAQEAS/U4iiooBENGW/Hwmmd3\r\n\tXCy6Zrt08YjKCzGNjorT98g8uGsqYjSxv/hmi0qlnlHs+k/3Iobc3LjS5AMYr5L8\r\n\tUO7OSkgFFlLHQyC9JzPfmLCAugvzEbyv4Olnsr8hbxF1MbKZoQxUZtMVu29wjfXk\r\n\thTeApBv7eaKCWpSp7MCbvgzm74izKhu3vlDk9w6qVrxePfGgpKPqfHiOoGhFnbTK\r\n\twTC6o2xq5y0qZ03JonF7OJspEd3I5zKY3E+ov7/ZhW6DqT8UFvsAdjvQbXyhV8Eu\r\n\tYhixw1aKEPzNjNowuIseVogKOLXxWI5vAi5HgXdS0/ES5gDGsABo4fqovUKlgop3\r\n\tRA==\r\n\t-----END CERTIFICATE-----\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\nX-ASDF:      \r\n\t\r\n    \r\n  ASDF\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET / HTTP/1.1\r\nX-ASDF:   hi\r\n    y\r\n\t\r\n       x\r\n  ASDF\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "GET /test HTTP/1.1\r\nUser-Agent: curl/7.18.0 (i486-pc-linux-gnu) libcurl/7.18.0 OpenSSL/0.9.8g zlib/1.2.3.3 libidn/1.1\r\nHost: 0.0.0.0=5000\r\nAccept: */*\r\n\r\n"
  local req = main:match(http)
  print(json.encode(req))
end

do
  local http = "POST /post_identity_body_world?q=search#hey HTTP/1.1\r\nAccept: */*\r\nTransfer-Encoding: identity\r\nContent-Length: 5\r\n\r\n"
  local req
  local n = 100000
  local t1 = os.clock()
  for i = 1, n do
    req = main:match(http)
  end
  local t2 = os.clock()
  print(t2-t1, n/(t2-t1))
  print(json.encode(req))
end