apt-panopticon/apt-panopticon.lua

#!/usr/bin/env luajit

local now = os.time()

local APT = require 'apt-panopticommon'
local D = APT.D
local I = APT.I
local T = APT.T
local W = APT.W
local E = APT.E
local C = APT.C
local arg, sendArgs = APT.parseArgs({...})
APT.html = true


local defaultURL = {scheme = "http"}
local releases = {"jessie", "ascii", "beowulf", "ceres"}
local releaseFiles =
{
    -- Release file.
    "Release",				-- 3.7 MB
    "Release.gpg",			--
--    "InRelease",			-- 3.7 MB
--    "main/binary-all/Packages.xz",	-- 2.6 GB for all that changed recently.
    -- Contents files.			-- 3.3 GB
--    "main/Contents-all.xz",
--    "main/Contents-amd64.xz",
--    "main/Contents-arm64.xz",
--    "-security/main/Contents-all.xz",
--    "-security/main/Contents-amd64.xz",
--    "-security/main/Contents-arm64.xz",
}
local notExist =
{
    "ceres-security"	-- This will never exist, it's our code name for the testing suite.
}
local referenceDebs =
{
    -- Debian package.
    "merged/pool/DEBIAN/main/d/debian-keyring/debian-keyring_2019.02.25_all.deb",
    -- Debian security package.  NOTE this one should always be redirected?
    "merged/pool/DEBIAN-SECURITY/updates/main/a/apt/apt-transport-https_1.4.9_amd64.deb",
}
local referenceDevs =
{
    -- Devuan package.  NOTE this one should not get redirected, but that's more a warning than an error.
    "merged/pool/DEVUAN/main/d/devuan-keyring/devuan-keyring_2017.10.03_all.deb",
}

local curlStatus = 
{
    [1 ] = "Unsupported protocol. This build of curl has no support for this protocol.",
    [2 ] = "Failed to initialize.",
    [3 ] = "URL malformed. The syntax was not correct.",
    [4 ] = "A feature or option that was needed to perform the desired request was not enabled or was explicitly disabled at build-time. To make curl able to do this, you probably need another build of libcurl!",
    [5 ] = "Couldn't resolve proxy. The given proxy host could not be resolved.",
    [6 ] = "Couldn't resolve host. The given remote host was not resolved.",
    [7 ] = "Failed to connect to host.",
    [8 ] = "Weird server reply. The server sent data curl couldn't parse.",
    [9 ] = "FTP access denied. The server denied login or denied access to the particular resource or directory you wanted to reach. Most often you tried to change to a directory that doesn't exist on the server.",
    [10] = "While waiting for the server to connect back when an active FTP session is used, an error code was sent over the control connection or similar.",
    [11] = "FTP weird PASS reply. Curl couldn't parse the reply sent to the PASS request.",
    [12] = "During an active FTP session while waiting for the server to connect, the CURLOPT_ACCEPTTIMEOUT_MS (or the internal default) timeout expired.",
    [13] = "FTP weird PASV reply, Curl couldn't parse the reply sent to the PASV request.",
    [14] = "FTP weird 227 format. Curl couldn't parse the 227-line the server sent.",
    [15] = "FTP can't get host. Couldn't resolve the host IP we got in the 227-line.",
    [16] = "A problem was detected in the HTTP2 framing layer. This is somewhat generic and can be one out of several problems, see the error buffer for details.",
    [17] = "FTP couldn't set binary. Couldn't change transfer method to binary.",
    [18] = "Partial file. Only a part of the file was transferred.",
    [19] = "FTP couldn't download/access the given file, the RETR (or similar) command failed.",

    [21] = "FTP quote error. A quote command returned error from the server.",
    [22] = "HTTP page not retrieved. The requested url was not found or returned another error with the HTTP error code being 400 or above. This return code only appears if -f, --fail is used.",
    [23] = "Write error. Curl couldn't write data to a local filesystem or similar.",

    [25] = "FTP couldn't STOR file. The server denied the STOR operation, used for FTP uploading.",
    [26] = "Read error. Various reading problems.",
    [27] = "Out of memory. A memory allocation request failed.",
    [28] = "Operation timeout. The specified time-out period was reached according to the conditions.",

    [30] = "FTP PORT failed. The PORT command failed. Not all FTP servers support the PORT command, try doing a transfer using PASV instead!",
    [31] = "FTP couldn't use REST. The REST command failed. This command is used for resumed FTP transfers.",

    [33] = "HTTP range error. The range \"command\" didn't work.",
    [34] = "HTTP post error. Internal post-request generation error.",
    [35] = "SSL connect error. The SSL handshaking failed.",
    [36] = "FTP bad download resume. Couldn't continue an earlier aborted download.",
    [37] = "FILE couldn't read file. Failed to open the file. Permissions?",
    [38] = "LDAP cannot bind. LDAP bind operation failed.",
    [39] = "LDAP search failed.",

    [41] = "Function not found. A required LDAP function was not found.",
    [42] = "Aborted by callback. An application told curl to abort the operation.",
    [43] = "Internal error. A function was called with a bad parameter.",

    [45] = "Interface error. A specified outgoing interface could not be used.",

    [47] = "Too many redirects. When following redirects, curl hit the maximum amount.",
    [48] = "Unknown option specified to libcurl. This indicates that you passed a weird option to curl that was passed on to libcurl and rejected. Read up in the manual!",
    [49] = "Malformed telnet option.",

    [51] = "The peer's SSL certificate or SSH MD5 fingerprint was not OK.",
    [52] = "The server didn't reply anything, which here is considered an error.",
    [53] = "SSL crypto engine not found.",
    [54] = "Cannot set SSL crypto engine as default.",
    [55] = "Failed sending network data.",
    [56] = "Failure in receiving network data.",

    [58] = "Problem with the local certificate.",
    [59] = "Couldn't use specified SSL cipher.",
    [60] = "Peer certificate cannot be authenticated with known CA certificates.",
    [61] = "Unrecognized transfer encoding.",
    [62] = "Invalid LDAP URL.",
    [63] = "Maximum file size exceeded.",
    [64] = "Requested FTP SSL level failed.",
    [65] = "Sending the data requires a rewind that failed.",
    [66] = "Failed to initialise SSL Engine.",
    [67] = "The user name, password, or similar was not accepted and curl failed to log in.",
    [68] = "File not found on TFTP server.",
    [69] = "Permission problem on TFTP server.",
    [70] = "Out of disk space on TFTP server.",
    [71] = "Illegal TFTP operation.",
    [72] = "Unknown TFTP transfer ID.",
    [73] = "File already exists (TFTP).",
    [74] = "No such user (TFTP).",
    [75] = "Character conversion failed.",
    [76] = "Character conversion functions required.",
    [77] = "Problem with reading the SSL CA cert (path? access rights?).",
    [78] = "The resource referenced in the URL does not exist.",
    [79] = "An unspecified error occurred during the SSH session.",
    [80] = "Failed to shut down the SSL connection.",
    [81] = "Socket is not ready for send/recv wait till it's ready and try again. This return code is only returned from curl_easy_recv and curl_easy_send.",
    [82] = "Could not load CRL file, missing or wrong format (added in 7.19.0).",
    [83] = "Issuer check failed (added in 7.19.0).",
    [84] = "The FTP PRET command failed",
    [85] = "RTSP: mismatch of CSeq numbers",
    [86] = "RTSP: mismatch of Session Identifiers",
    [87] = "unable to parse FTP file list",
    [88] = "FTP chunk callback reported error",
    [89] = "No connection available, the session will be queued",
    [90] = "SSL public key does not matched pinned public key",
    [91] = "Status returned failure when asked with CURLOPT_SSL_VERIFYSTATUS.",
    [92] = "Stream error in the HTTP/2 framing layer.",
    [93] = "An API function was called from inside a callback.",
    [94] = "An authentication function returned an error.",
    [95] = "A problem was detected in the HTTP/3 layer. This is somewhat generic and can be one out of several problems, see the error buffer for details.",
}


local socket = require 'socket'
local ftp = require 'socket.ftp'
local http = require 'socket.http'
local url = require 'socket.url'


local ip = ""
local cor = nil
local Updating = false

local downloadLock = "flock -n results/curl-"
local arw = ' &nbsp; <font color="magenta"><b>-></b></font> &nbsp; '

local repoExists = function (r)
    r = r:match("([%a-]*)")
    if nil == r then return false end
    for k, v in pairs(notExist) do
	if v == r then return false end
    end
    return true
end

local IP = {}
gatherIPs = function (host)
    if nil == IP[host] then
        local IPs
        -- Takes about 30 seconds to look up the lot.
        -- I tested using dig's -f option, it didn't seem much faster.
        -- The sort -r assumes that deb.devuan.org is the first alphabetically.
	local dig = io.popen('dig +keepopen +noall +nottlid +answer ' .. host .. ' A ' .. host .. ' AAAA ' .. host .. ' CNAME ' .. host .. ' SRV | sort -r | uniq')
	repeat
	    IPs = dig:read("*l")
	    if nil ~= IPs then
		for k, t, v in IPs:gmatch("([%w_%-%.]*)%.%s*IN%s*(%a*)%s*(.*)") do
		    if "." == v:sub(-1, -1) then v = v:sub(1, -2) end
		    if nil == IP[k] then IP[k] = {} end
		    IP[k][v] = t
		    D("  DNS record " .. host .. " == " .. k .. " type " .. t .. " -> " .. v)
		    if t == "CNAME" then
			gatherIPs(v)
			IP[k][v] = IP[v]
		    elseif v == "SRV" then
			print("SVR record found, now what do we do?")
		    end
		end
	    end
	until nil == IPs
    end

    -- If this is the DNS-RR domain name, gather the IPs for the mirrors that mirror_list.txt says should be in it.
    if host == APT.options.roundRobin.value then
	for k, m in pairs(APT.mirrors) do
	    if "yes" == m.DNSRR then
		gatherIPs(m.FQDN)
		IP[host][m.FQDN] = IP[m.FQDN]
	    end
	end
    end
    return IP[host]
end

-- Returns FTP directory listing
local nlst = function (u)
    local t = {}
    local p = url.parse(u)
    p.command = "nlst"
    p.sink = ltn12.sink.table(t)
    local r, e = ftp.get(p)
    return r and table.concat(t), e
end

local timeouts = 0;
local totalTimeouts = 0
local spcd = ' &nbsp; '
checkHEAD = function (host, URL, r, retry, sanity)
    if nil == r then r = 0 end
    if nil == retry then retry = 0 end
    if true == sanity then sanity = 'URLSanity' else sanity = '' end
    local check = "HEAD testing file"
    local PU = url.parse(URL, defaultURL)
    local pu = url.parse(PU.scheme .. "://" .. host, defaultURL)
    local fname = host .. "_" .. PU.host .. "_" .. PU.path:gsub("/", "_") .. ".log.txt"
    local hdr = ""
    local IP = ""
    if pu.host ~= PU.host then
	if "http" == PU.scheme then
	    hdr = '-H "Host: ' .. host .. '"'
	end
	IP = '--connect-to "' .. pu.host .. '::' .. PU.host .. ':"'
	fname = host .. "_" .. pu.host .. '_' .. PU.host .. "_" .. PU.path:gsub("/", "_") .. ".txt"
    end
    os.execute('rm -f results/HEADERS_' .. fname .. ' 2>/dev/null; rm -f results/STATUS_' .. fname .. ' 2>/dev/null; touch results/STATUS_' .. fname)
    if not APT.testing(PU.scheme, host) and APT.redir then I(spcd .. string.upper(PU.scheme) .. " not supported, not tested. &nbsp; " .. URL, host); return end
    if 0 < r then
	check = "Redirecting to"
    end
    if 0 < retry then
	os.execute("sleep " .. math.random(1, 3))
	check = "Retry " .. retry .. " " .. check
    end
    if 2 <= timeouts then
	E(spcd .. spcd .. "too many timeouts!  " .. check .. " " .. host .. arw .. URL, PU.scheme, "", host)
	return
    end
    if APT.options.timeouts.value <= (totalTimeouts) then
	E(spcd .. spcd .. "Way too many timeouts!", PU.scheme, "", host)
	return
    end
    if 20 <= r then
	E(spcd .. spcd .. "too many redirects!  " .. check .. " " .. host .. arw .. URL, PU.scheme, "", host)
	return
    end
    if APT.options.retries.value <= retry then
	E(spcd .. spcd .. "too many retries!  " .. check .. " " .. host .. arw .. URL, PU.scheme, "", host)
	return
    end
    if "https" == PU.scheme and APT.options.roundRobin.value == host then
	I(spcd .. "Not testing " .. APT.lnk(URL) .. " mirrors wont have the correct HTTPS certificate for the round robin.", host)
	return
    else
	I(spcd .. check .. " " .. APT.lnk(URL), host)
    end

    --[[ Using curl command line -
	-I - HEAD
	--connect-to domain:port:IP:port - connect to IP, but use SNI from URL.
	-header "" - add extra headers.
	-L - DO follow redirects.
	--max-redirs n - set maximum redirects, default is 50, -1 = unlimited.
	-4 or -6 - Only use IPv4 or IPv6
	--retry n - maximum retries, default is 0, no retries.
	-o file - write to file instead of stdout.
	--path-as-is - https://curl.haxx.se/libcurl/c/CURLOPT_PATH_AS_IS.html might be useful for URLSanity.
	-s silent - don't output progress or error messages.
	--connect-timeout n - timeout in seconds.
	    Should return with error code 28 on a timeout?
	-D file - write the received headers to a file.  This includes the status code and string.
    ]]
    local status = APT.exe(
	'curl -I --retry 0 -s --path-as-is --connect-timeout ' .. APT.options.timeout.value .. ' --max-redirs 0 ' .. APT.IPv46 .. ' ' ..
	IP .. ' ' .. '-o /dev/null -D results/"HEADERS_' .. fname .. '" ' ..
	hdr .. ' -w "#%{http_code} %{ssl_verify_result} %{url_effective}\\n" ' .. PU.scheme .. '://' .. host .. PU.path .. ' >>results/"STATUS_' .. fname .. '"'
    ):Nice():log():Do().status
    if 0 < r then
	APT.tested(PU.scheme, 'Redirects', host)
    else
	APT.tested(PU.scheme, '', host)
    end
    local code = "???"
    local cstr = ""
    local location = nil
    local tmot = 1
    while not APT.checkFile('results/STATUS_' .. fname) do
	D(spcd .. spcd .. 'Waiting for results/STATUS_' .. fname .. ' file.')
	os.execute('sleep ' .. tmot)
	tmot = tmot * 2
	if 8 < tmot then
	    T(spcd .. spcd .. "TIMEOUT " .. timeouts + 1 .. ", retry " .. retry + 1 .. ' ' .. APT.lnk(URL), PU.scheme, sanity, host)
	    timeouts = timeouts + 1
	    checkHEAD(host, URL, r, retry + 1, '' ~= sanity)
	    os.execute('cat results/"HEADERS_' .. fname .. '" >>results/"STATUS_' .. fname .. '" 2>/dev/null; rm -f results/"HEADERS_' .. fname .. '" 2>/dev/null')
	    return
	end
    end
    os.execute('cat results/"HEADERS_' .. fname .. '" >>results/"STATUS_' .. fname .. '" 2>/dev/null; rm -f results/"HEADERS_' .. fname .. '" 2>/dev/null')
    if 0 ~= status then
	local msg = curlStatus[status]
	if nil == msg then msg = "UNKNOWN CURL STATUS CODE!" end
	if (28 == status) or (7 == status) then
	    T(spcd .. spcd .. "TIMEOUT " .. timeouts + 1 .. ", retry " .. retry + 1 .. ' ' .. APT.lnk(URL), PU.scheme, sanity, host)
	    timeouts = timeouts + 1
	else
	    E(spcd .. spcd .. "The curl command return an error code of " .. status .. " - " .. msg .. ' for '.. APT.lnk(URL), PU.scheme, sanity, host)
	end
	if 60 == status then return end	-- Certificate is invalid, don't bother retrying.
	checkHEAD(host, URL, r, retry + 1, '' ~= sanity)
	return
    end
    local rfile, e = io.open("results/STATUS_" .. fname, "r")
    if nil == rfile then W("opening results/STATUS_" .. fname .. " file - " .. e) else
	for line in rfile:lines("*l") do
	    if "#" == line:sub(1, 1) then
		code = line:sub(2, 4)
		if ("https" == PU.scheme) and ("0" ~= line:sub(6, 6)) then
		    os.execute('cp results/STATUS_' .. fname .. ' results/STATUS_' .. fname .. '_SAVED')
		    if '' ~= sanity then
			E(spcd .. spcd .. "The certificate is invalid.", PU.scheme, sanity, host)
		    else
			E(spcd .. spcd .. "The certificate is invalid.", PU.scheme, "https", host)
		    end
		end
	    elseif "http" == line:sub(1, 4):lower() then
		-- -2 coz the headers file gets a \r at the end.
		cstr = line:sub(14, -2)
	    elseif "location" == line:sub(1, 8):lower() then
		location = line:sub(11, -2)
	    end
	end
	if '???' == code then
	    W(spcd .. spcd .. 'Could not find response code. ' .. APT.lnk(URL), PU.scheme, sanity, host)
	end
    end
    os.execute('cat results/STATUS_' .. fname .. ' >> results/curl_HEAD_' .. fname .. '; rm -f results/STATUS_' .. fname .. ' 2>/dev/null')

    if ("4" == tostring(code):sub(1, 1)) or ("5" == tostring(code):sub(1, 1)) then
	E(spcd .. spcd .. code .. " " .. cstr .. ".   " .. check .. " " .. APT.lnk(URL), PU.scheme, sanity, host)
    else
	if not APT.testing(PU.scheme, host) then
	    I(spcd .. spcd .. "Not supported, but works " .. PU.scheme .. " " .. APT.lnk(URL), PU.scheme, "", host)
	end
	I(spcd .. spcd .. code .. " " .. cstr .. ".   " .. check .. " " .. APT.lnk(URL), host)
--	timeouts = timeouts - 1	-- Backoff the timeouts count if we managed to get through.
	if nil ~= location then
	    pu = url.parse(location, defaultURL)
	    if (pu.host == APT.options.roundRobin.value) and (nil ~= PU.path:find('merged/pool/DEVUAN/')) then
		E('DEVUAN packages must not be redirected to ' .. APT.options.roundRobin.value .. ' - ' .. APT.lnk(URL) .. arw .. APT.lnk(location), PU.scheme, 'Redirects', host)
	    end
	    if APT.testing("Protocol") then
		if ('http' == location:sub(1, 4)) and (pu.scheme ~= PU.scheme) then	-- Sometimes a location sans scheme is returned, this is not a protocol change.
		    if APT.options.roundRobin.value == host then  -- Coz HTTPS shouldn't happen via the round robin.
			E(spcd .. spcd .. "Protocol changed during redirect!  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), PU.scheme, "Protocol", host)
		    end
			W(spcd .. spcd .. "Protocol changed during redirect!  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), PU.scheme, "Protocol", host)
		    else
		end
		APT.tested(PU.scheme, 'Protocol', host)
	    end

	    if location == URL then
		E(spcd .. spcd .. "Redirect loop!  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), PU.scheme, "", host)
	    elseif nil == pu.host then
		I(spcd .. spcd .. "Relative redirect.  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), host)
		if 1 <= APT.options.bandwidth.value then checkHEAD(host, PU.scheme .. "://" .. PU.host .. location, r + 1, retry, '' ~= sanity) end
	    elseif (PU.host == pu.host) or (host == pu.host) then
		if PU.host ~= host then
		    local t = pu.host
		    pu.host = PU.host
		    location = url.build(pu)
		    pu.host = t
		end
		I(spcd .. spcd .. "Redirect to same host.  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), host)
		if 1 <= APT.options.bandwidth.value then checkHEAD(host, location, r + 1, retry, '' ~= sanity) end
	    else
		I(spcd .. spcd .. "Redirect to different host.  " .. check .. " " .. APT.lnk(URL) .. arw .. APT.lnk(location), host)
		if 1 <= APT.options.bandwidth.value then 
		    --[[ The hard part here is that we end up throwing ALL of the test files at the redirected location.
		        Not good for deb.debian.org, which we should only be throwing .debs at.
		        What we do is loop through the DNS entries, and only test the specific protocol & file being tested here.
		    ]]
		    local u = pu.host .. "/" .. pu.path
		    local file = pu.path:match(".*/([%w%.%+%-_]*)$")	-- Get the filename.
		    local path = pu.path:sub(2, -1 -(#file))
		    local check = u:gsub("/", "_")
		    local extraArgs = sendArgs .. ' -o -r '
		    if 'https' == pu.scheme then extraArgs = extraArgs .. ' --tests=-http' end
		    if 'http'  == pu.scheme then extraArgs = extraArgs .. ' --tests=-https' end
		    local pth = path:match('^(.*/pool/).*$')
		    if nil ~= pth then table.insert(APT.results[PU.scheme].redirects, pu.host .. "/" .. pth) else E(spcd .. spcd .. 'Odd redirect path ' .. path) end
		    I(spcd .. spcd .. "Now checking redirected host " .. u .. ' &nbsp; for &nbsp; ' .. APT.lnk(URL) .. arw .. APT.lnk(location), host)
		    APT.exe(downloadLock .. "REDIR-" .. check .. ".log.txt" .. " ./apt-panopticon.lua " .. extraArgs .. ' ' .. pu.host .. "/" .. path .. " " .. file):Nice():log():fork()
		    D(spcd .. 'logging to ' .. APT.logName(pu.host, nil, file)[2])
		    APT.tested(PU.scheme, 'Redirects', host)
		end
	    end
	elseif nil ~= PU.path:find('merged/pool/DEBIAN-SECURITY/') then
	    W('DEBIAN-SECURITY packages must be redirected to a Debian mirror - ' .. APT.lnk(URL) .. arw .. APT.lnk(location), PU.scheme, 'Redirects', host)
	end
    end
end

local checkTimeouts = function(host, scheme, URL)
    totalTimeouts = totalTimeouts + timeouts; timeouts = 0
    checkHEAD(host, scheme .. "://" .. URL)
    if (1 <= APT.options.bandwidth.value) and APT.testing("URLSanity") then
	URL = URL:gsub("/", "///")
	URL = URL:gsub("///", "/", 1)
	checkHEAD(host, scheme .. "://" .. URL, 0, 0, true)
	APT.tested(scheme, 'URLSanity', host)
    end
    if nil ~= cor then
	D('*&gt;* About to resume coroutine after checkHEAD(' .. host .. ' , ' .. scheme .. ' :// ' .. URL .. ')')
	local ok, message = coroutine.resume(cor)
	if not ok then cor = nil; print(message) end
    end
    if APT.options.timeouts.value <= (totalTimeouts) then
	E("Way too many timeouts!", scheme, "URLSanity", host)
	return true
    end
    return false
end

local checkFiles = function (host, ip, path, file)
    timeouts = 0
    if nil == path then path = "" end
    if nil ~= file then
	if "redir" == ip then ip = host end
	if checkTimeouts(host, "http",  ip .. path .. "/" .. file) then return end
	if checkTimeouts(host, "https", ip .. path .. "/" .. file) then return end
    else
	I("  HEAD testing files for " .. host .. arw .. ip .. " " .. path, host)
	if 1 <= APT.options.bandwidth.value then
	    -- Do these first, coz they are likely to fork off a different server.
	    for i, s in pairs(referenceDebs) do
		if checkTimeouts(host, "http",  ip .. path .. "/" .. s) then return end
		if checkTimeouts(host, "https", ip .. path .. "/" .. s) then return end
	    end
	end
	for i, s in pairs(releases) do
	    for j, k in pairs(releaseFiles) do
		if repoExists(s .. k) then
		    if checkTimeouts(host, "http",  ip .. path .. "/merged/dists/" .. s .. '/' .. k) then return end
		    if 1 <= APT.options.bandwidth.value then
			if checkTimeouts(host, "https", ip .. path .. "/merged/dists/" .. s .. '/' .. k) then return end
		    else
			break
		    end
		end
		if 2 >= APT.options.bandwidth.value then break end
	    end
	    if 2 >= APT.options.bandwidth.value then break end
	end
	if 1 <= APT.options.bandwidth.value then
	    for i, s in pairs(referenceDevs) do
		if checkTimeouts(host, "http",  ip .. path .. "/" .. s) then return end
		if checkTimeouts(host, "https", ip .. path .. "/" .. s) then return end
	    end
	end
    end
end

checkHost = function (orig, host, path, ip, file)
    if nil == host then host = orig end
    if nil == path then path = "" end
    if nil == file then file = "" end
    local ph = url.parse("http://" .. host)
    if (nil ~= ip) and ("redir" ~= ip) then
	local po = url.parse("http://" .. orig)
	if "" ~= file then
	    D("checking redirected file " .. po.host .. " " .. file)
	    checkFiles(po.host, ip, path, file)
	else
	    checkFiles(po.host, ip, path)
	end
    else
	if orig == host then
	    I("Testing mirror " .. orig .. "" .. file)
	    APT.exe("./apt-panopticon.lua " .. sendArgs .. " -o " .. orig .. path .. " " .. file):Nice():log():fork()
	    D('logging to ' .. APT.logName(ph.host, nil, file)[2])
	else D("checkHost " .. orig .. arw .. host) end
    end
end


local addDownload = function(host, URL, f, r, k)
    local file = k:match(".*/([%w%.%+%-_]*)$")	-- Get the filename.
    if APT.checkFile("results/" .. host .. "/merged/dists/" .. r .. '/' .. k) then
	-- Curls "check timestamp and overwrite file" stuff sucks.
	-- -R means the destination file gets the timestamp of the remote file.
	-- Can only do ONE timestamp check per command.
	--  This doesn't work either.  All downloads get all these headers.  Pffft
--	local status, ts = APT.execute('TZ="GMT" ls -l --time-style="+%a, %d %b %Y %T %Z"  results/' .. host .. "/merged/dists/" .. r .. '/' .. k .. ' | cut -d " " -f 6-11')
--	f:write('header "If-Modified-Since: ' .. ts:sub(2, -2) .. '"\n')
	-- Curl will DELETE the existing file if the timestamp fails to download a new one, unless we change directory first, 
	-- which wont work with multiple files in multiple directories.  WTF?
--TODO - change tactic, do a HEAD if-modified test first before adding the file to the list to download.
	os.execute(" mv   results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. 
			" results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. ".old")
    end

    D('Downloading http://' .. host .. URL .. '/merged/dists/' .. r .. '/' .. k)
    f:write('url "' .. 'http://' .. host .. URL .. '/merged/dists/' .. r .. '/' .. k .. '"\n')
    f:write('output "results/' .. host .. '/merged/dists/' .. r .. '/' .. k .. '"\n')
end

local postDownload = function(host, r, k)
    local file = k:match(".*/([%w%.%+%-_]*)$")	-- Get the filename.
    if nil == file then file = k end
    os.execute("if [ -f results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. ".old ]" ..
	    " && [ ! -f results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. " ]; then cp -a" ..
		    "   results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. ".old" ..
		    "   results/" .. host .. "/merged/dists/" .. r .. '/' .. k .. "; fi")
    if APT.checkFile('results/' .. host .. '/merged/dists/' .. r .. '/' .. k) then
	if ".gz"  == k:sub(-3, -1) then APT.exe("gzip -dfk results/" .. host .. "/merged/dists/" .. r .. '/' .. k):Nice():noErr():Do() end
	if ".xz"  == k:sub(-3, -1) then APT.exe("xz   -dfk results/" .. host .. "/merged/dists/" .. r .. '/' .. k):Nice():noErr():Do() end
    end
end

local download = "curl" ..
		 " --connect-timeout " .. APT.options.timeout.value ..
		 " --create-dirs -f -L" ..
		 " --fail-early" ..
		 " --max-time " .. APT.options.maxtime.value ..
		 APT.IPv46 .. ' ' ..
		 " --retry " .. APT.options.retries.value ..
		 " -R -v -z 'results/stamp.old'  --stderr results/"
local downloads = function(host, URL, meta, release, list)
    if nil == URL then URL = "" end
    local files = 'curl-' .. meta .. '-' .. host .. '.files.txt'
    local lock = meta .. "-" .. host .. ".log.txt"
    local log = "curl-" .. meta .. "-" .. host .. ".log.txt"
    local cm = downloadLock .. lock .. " " .. download .. log .. " -K results/" .. files
    if      APT.testing("IPv4")  and (not APT.testing("IPv6")) then cm = cm .. ' -4' end
    if (not APT.testing("IPv4")) and      APT.testing("IPv6")  then cm = cm .. ' -6' end
    f, e = io.open("results/curl-" .. meta .. '-' .. host .. ".files.txt", "a+")
    if nil == f then C("opening curl downloads list file - " .. e); return end

    if nil ~= list then
	if "" ~= list then
	    if nil ~= release then
		for l in list:gmatch("\n*([^\n]+)\n*") do
		    addDownload(host, URL, f, release, "/" .. l)
		end
	    else
		I('Downloading ' .. APT.lnk('http://' .. host .. URL .. '/merged/' .. list))
		f:write('url "' .. 'http://' .. host .. URL .. '/merged/' .. list .. '"\n')
		f:write('output "results/' .. host .. '/merged/' .. list .. '"\n')
	    end
	    f:close()
	    return
	end
    else
	for i, s in pairs(releases) do
	    for j, k in pairs(releaseFiles) do
		if repoExists(s .. k) then
		    addDownload(host, URL, f, s, k)
		end
	    end
	end
    end
    f:close()
    APT.exe(cm):Nice():log():fork()
    D('logging to <a href="' .. log .. '">' .. log .. '</a>, with <a href="' .. files .. '">these files</a>')
end


local validateURL = function(m)
    if " " == m.BaseURL:sub(-1, -1) then
	W("space at end of BaseURL in mirror_list.txt!  " .. m.BaseURL, "", "", m.FQDN)
	m.BaseURL = m.BaseURL:sub(1, -2)
    end
    if "/" == m.BaseURL:sub(-1, -1) then
	W("slash at end of BaseURL in mirror_list.txt!  " .. m.BaseURL, "", "", m.FQDN)
	m.BaseURL = m.BaseURL:sub(1, -2)
    end
    local p = url.parse("http://" .. m.BaseURL)
    if nil == p.path then p.path = '' end
    if nil ~= p.port then p.authority = authority .. ':' .. p.port end
    if nil == m.FDQN then W("Something wrong in FDQN from mirror_list.txt!  nil", "", "", p.authority) else
	if m.FQDN ~= p.authority then W("Something wrong in FDQN from mirror_list.txt!  " .. m.FDQN, "", "", p.authority) end
    end
    if nil == m.BaseURL then W("Something wrong in BaseURL from mirror_list.txt!  nil", "", "", p.authority) else
	if m.BaseURL ~= (p.authority .. p.path) then W("Something wrong in BaseURL from mirror_list.txt!  " .. m.BaseURL, "", "", p.authority) end
    end
    if (nil ~= p.query) or (nil ~= p.fragment) or (nil ~= p.params) then W("Something wrong in BaseURL from mirror_list.txt, should be nothing after the path!  " .. m.BaseURL, "", "", p.authority) end
    if (nil ~= p.user) or (nil ~= p.userinfo) or (nil ~= p.password) then W("Something wrong in BaseURL from mirror_list.txt, should be no credentials!  " .. m.BaseURL, "", "", p.authority) end
    m.FQDN = p.authority
    m.BaseURL = p.authority .. p.path
    return m
end

local getMirrors = function ()
    local mirrors = {}
    local host = ""
    local m = {}
    local active = true
    local URL = 'http://' .. APT.options.referenceSite.value .. '/mirror_list.txt'
    I('Downloading and parsing http://' .. APT.options.referenceSite.value .. '/mirror_list.txt')
    local p, c, h = http.request(URL)
    if nil == p then E(c .. " fetching " .. URL) else

	for l in p:gmatch("\n*([^\n]+)\n*") do
	    local t, d = l:match("(%a*):%s*(.*)")
	    d = string.lower(d)
	    if "FQDN" == t then
		if "" ~= host then
		    mirrors[host] = validateURL(m)
		    m = {}
		    active = true
		end
		host = d
		m[t] = d
	    elseif "Protocols" == t then
		local prot = {}
		for w in d:gmatch("(%w+)") do
		    if APT.search(APT.protocols, w:lower()) then prot[w] = true end
		end
		m[t] = prot
	    elseif "Active" == t and nil == d:sub(1, 3):find("yes", 1, true) then
		W("Mirror " .. host .. " is not active - " .. d, "", "", host)
		active = false
		m[t] = d
-- TODO - Should do some more input validation on everything.
	    elseif "Rate" == t then
		local time, unit = d:match('(%d+) *(%a+)')
		time = tonumber(time)
		unit = unit:sub(1, 1)
		m[t] = time .. ' ' .. unit
		if     'm' == unit then
		    time = time * 60
		elseif 'h' == unit then
		    time = time * 60 * 60
		else
		    C('Unknown Rate for mirror ' .. host)
		end
		m['Updated'] = time
	    else
		m[t] = d
	    end
	end
	if "" ~= host --[[and active]] then
	    mirrors[host] = validateURL(m)
	end
    end
    if APT.testing("DNSRR") then
	mirrors[APT.options.roundRobin.value] = { ["Protocols"] = { ["http"] = true; }; ['Updated'] = 300; ['DNSRR'] = false;
						  ["FQDN"] = APT.options.roundRobin.value; ["Active"] = 'yes'; ["BaseURL"] = APT.options.roundRobin.value; }
    end
    local file, e = io.open("results/mirrors.lua", "w+")
    if nil == file then C("opening mirrors file - " .. e) else
	file:write(APT.dumpTable(mirrors, "mirrors") .. "\nreturn mirrors\n")
        file:close()
    end
    if 42 < #mirrors then print(#mirrors .. ' is too many mirrors!') ; os.exit(false) end
    return mirrors
end


local postParse = function(host, list)
    if APT.options.referenceSite.value == host then
	if nil ~= list then
	    local sem = 'results/NEW_' .. list.out .. '_%s.txt'
	    for i, n in pairs(releases) do
		local f = sem:format(n)
		if APT.checkFile(f .. '.tmp') then
		    os.execute('mv ' .. f .. '.tmp ' .. f)
		else
		    os.execute('touch ' .. f)
		end
	    end
	end
    end
end

local parseDebs = function(host)
    for i, n in pairs(releases) do
	local inFile =  'results/NEW_debs_' .. n .. '.txt'
	local nfile, e = io.open(inFile, "r")
	    if nil == nfile then W("opening " .. inFile .. " file - " .. e) else
	    for l in nfile:lines() do
		local v, p, sz, sha = l:match(' | (.+) | (pool/.+%.deb) | (%d.+) | (%x.+) |')
		if nil ~= p then
		    if APT.checkFile('results/' .. host .. "/merged/" .. p) then
			local fsz = APT.exe('ls -l results/' .. host .. "/merged/" .. p .. ' | cut -d " " -f 5-5'):Do().result
			if APT.testing("Integrity") then
			    if sz ~= fsz:sub(2, -2) then	-- The sub bit is to slice off the EOLs at each end.
				E('Package size mismatch - results/' .. host .. "/merged/" .. p .. ' should be ' .. sz .. ', but is ' .. fsz:sub(2, -2) .. '.', 'http', 'Integrity', host)
			    else
				local fsha = APT.exe('sha256sum results/' .. host .. "/merged/" .. p .. ' | cut -d " " -f 1'):log():Do().result
				if sha ~= fsha:sub(2, -2) then E('Package SHA256 sum mismatch - results/' .. host .. "/merged/" .. p, 'http', 'Integrity', host) end
-- TODO - maybe check the PGP key, though packages are mostly not signed.
			    end
			    APT.tested('http', 'Integrity', host)
			end
			if APT.testing("Updated") then
			    if sz ~= fsz:sub(2, -2) then