server development

Signed-off-by: Edvard Rejthar <edvard.rejthar@nic.cz>

INSTALL

@@ -10,7 +10,7 @@ apt install software-properties-common
 add-apt-repository "deb http://archive.ubuntu.com/ubuntu $(lsb_release -sc) main universe restricted multiverse"
 apt update
 apt install firefox python3 mariadb-server xvfb
-pip3 install xvfbwrapper pymysql peewee flask wtforms pyyaml bs4 pygments pillow requests
+pip3 install xvfbwrapper pymysql peewee flask wtforms pyyaml bs4 pygments pillow requests humanize filelock
 
 # current dir
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"

firefox-mdmaug-extension/main.js

@@ -33,7 +33,7 @@ browser.runtime.onMessage.addListener(function (message) {
 /**
  * traffic listener
  */
-let trafficBlacklist = ["http://www.google.com/adsense/", "https://fbstatic-a.akamaihd.net/rsrc.php", "http://clients1.google.com/ocsp", "https://safebrowsing-cache.google.com/safebrowsing/", "https://safebrowsing.google.com/safebrowsing/", "https://tiles.services.mozilla.com/", "http://localhost/redirect/"];
+// let trafficBlacklist = ["http://www.google.com/adsense/", "https://fbstatic-a.akamaihd.net/rsrc.php", "http://clients1.google.com/ocsp", "https://safebrowsing-cache.google.com/safebrowsing/", "https://safebrowsing.google.com/safebrowsing/", "https://tiles.services.mozilla.com/", "http://localhost/redirect/"];
 browser.webRequest.onBeforeRequest.addListener(
         function (details) {
             // log resources that we want

firefox-mdmaug-extension/spy.js

@@ -104,7 +104,7 @@ function screenshot() {
     } else {
         screenshotted = true;
     }
-    console.warn("Starting screenshot countdown...");
+    console.warn("MDMAUG: Starting screenshot countdown...");
     setTimeout(function () {
         // total scroll height of tab
         let height = Math.max(document.body.scrollHeight, document.body.offsetHeight, document.documentElement.clientHeight, document.documentElement.scrollHeight, document.documentElement.offsetHeight);

mdmaug/__main__.py

@@ -2,32 +2,34 @@
 import logging
 import os
 
-from flask import Flask, request
+#logging.basicConfig(level=logging.DEBUG, format="%(message)s")  # init before flask so that logging is shown in the Terminal
+
+from flask import Flask
 from xvfbwrapper import Xvfb
 
 from .lib.config import Config
 from .lib.controller.api import Api
 
-# import ipdb; ipdb.set_trace()
-
-
-logging.basicConfig(level=logging.DEBUG, format="%(message)s")
-
 # assure the logging dir
 if not os.path.exists(Config.LOG_DIR):
     os.makedirs(Config.LOG_DIR)
 
+logger = logging.getLogger("mdmaug")
+logger.setLevel(logging.DEBUG)
+logger.addHandler(logging.StreamHandler())
+
 # server setup
 Api.reset()
 # address = '0.0.0.0'
 
 app = Flask(__name__, static_url_path="/static")
 app.secret_key = b'as8!r"afERaa5'
-app.config["preferences"] = {
-    "safebrowsing": True,
-    "pdns": True,
-    "geoip": False
-}
+
+# app.config["preferences"] = {
+#     "safebrowsing": True,
+#     "pdns": True,
+#     "geoip": False
+# }
 
 """
 httpd = HTTPServer((address, Config.APP_PORT), Server)
@@ -45,8 +47,9 @@ try:
     from .lib.controller.server import app as server
 
     app.register_blueprint(server)
-    app.run(host='0.0.0.0', ssl_context=(Config.DIR + 'cert-mdmaug.pem', Config.DIR + 'key-mdmaug.pem'), threaded=True)
-    # for _ in range(Config.profileCount):
+    if __name__ == "__main__":
+        app.run(host=Config.APP_IP, port=Config.APP_PORT, ssl_context=(Config.DIR + 'cert-mdmaug.pem', Config.DIR + 'key-mdmaug.pem'), threaded=True)
+    # for _ in range(Config.profile_count):
     #    threading.Thread(target=httpd.serve_forever).start()
 except (KeyboardInterrupt, SystemExit):
     display.stop()

mdmaug/bin/firefox_mdmaug_writer.py

 #!/usr/bin/env python3
+# Browser extension can write to a file thanks to this.
 
 import json
 import os
 import struct
 import sys
 
-# Read a message from stdin and decode it.
 profile = os.environ["PROFILE"] if "PROFILE" in os.environ else "unknown-profile"
-#log_dir = "/tmp/mdmaug/.cache/mdmaug-scans/_tmp/"  # /tmp/ is small (200 MB) and takes precious RAM
 cache_dir = os.environ["CACHE_DIR"] if "CACHE_DIR" in os.environ else "/tmp/"
 files_encountered = set()
-#cache_dir = log_dir # // default dir to store the analysis is log_dir if we fail to identify a better storage point
 
-#with open(log_dir+"cache.dir", "r") as f:
-#    cache_dir = f.read().strip()
 
 def get_message():
     raw_length = sys.stdin.buffer.read(4)
     if len(raw_length) == 0:
         sys.exit(0)
     message_length = struct.unpack('@I', raw_length)[0]
-    message = sys.stdin.buffer.read(message_length).decode('utf-8')
-    return json.loads(message)
+    return json.loads(sys.stdin.buffer.read(message_length).decode('utf-8'))
+
 
 while True:
     message = get_message()
-    file = cache_dir+message["filename"]
+    file = cache_dir + message["filename"]
     if file in files_encountered:
-        method = "a"        
+        method = "a"
     else:
         method = "w"
         files_encountered.add(file)
     with open(file, method) as f:
         f.write(message["text"])
-
-    with open("/tmp/ram/zde.txt", "a") as f:
-        f.write(f"{file} {method}\n")

mdmaug/lib/config.py

@@ -2,15 +2,27 @@ import logging
 import os
 import threading
 
-from flask import g
 from peewee import MySQLDatabase
 
+logger = logging.getLogger("mdmaug")
+
+
+class Pref:
+    safebrowsing = True
+    pdns = True
+    geoip = False
+
+    @classmethod
+    def val2html(cls, param):
+        return "1" if getattr(cls, param) in [True, "true", "1"] else "0"
+
 
 class Config:
-    profileCount = 21  # number of Firefox profiles. Its name is just a number – 0,1...
+    profile_count = 2  # number of Firefox profiles. Its name is just a number – 0,1...
     browser = 'firefox'  # iceweasel, firefox. What browser gets launched.
-    configFile = '/opt/mdmaug/.cache/mdmaug-scans/_tmp/queue.cache'  # RAM disk was too small: '/tmp/mdm/queue.cache'
-    APP_PORT = 8000
+    config_file = '/opt/mdmaug/.cache/mdmaug-scans/_tmp/queue.cache'  # RAM disk was too small: '/tmp/mdm/queue.cache'
+    APP_PORT = 5000
+    APP_IP = "127.0.0.1"
     APP_DOMAIN = 'https://217.31.202.41:' + str(APP_PORT)  # csirt.csirt.office.nic.cz
     LOG_DIR = "/opt/mdmaug/.cache/mdmaug-scans/_tmp/"
     CACHE_DIR = "/opt/mdmaug/.cache/mdmaug-scans/"
@@ -22,10 +34,11 @@ class Config:
     MAX_BROWSER_RUN_TIME = 25  # maximum time for a browser to run
     MAX_BROWSER_EXPIRATION = 15  # seconds that we wait before killing the browser (waiting for the files to be written)
 
+    @staticmethod
     def connect():
         # XX resim problem peewee.OperationalError: (2006, "MySQL server has gone away (BrokenPipeError(32, 'Broken pipe'))") po 7 hodinach timeoutu
         # XX kupodivu pripojeni nemuze byt v dbp DBModel.connect. Prestoze type je pak spravne (MySQLDatabase), nic udelat nejde a pokusy o select konci NoneType.
-        logging.debug("Connecting to DB.")
+        logger.debug("Connecting to DB.")
         Config.myDB = MySQLDatabase("mdmaug", host='localhost', port=3306, user="mdmaug",
                                     passwd="fidFDSs676")  # XX dal jsem pryc: , threadlocals=False
 

mdmaug/lib/controller/api.py

@@ -5,13 +5,14 @@ from urllib.parse import urlparse
 
 from peewee import IntegrityError
 
-from ..domains import Domains
+from ..domains import domain2dir
 from .scan_controller import ScanController
 from ..config import Config
 from ..model.dbp import Turris, Whitelist
 from ..parser.traffic_log_parser import TrafficLogParser
 from ...templates.crawl_view import CrawlView
 
+logger = logging.getLogger("mdmaug")
 
 class Api:
     website = ""  # http://site.cz
@@ -19,7 +20,7 @@ class Api:
 
     def run(self, request):
         """ Accept command
-        :type path: dict from URL request. /api/analyze=cache/http://example.com → {"api": True, "analyze": cache, "page": "http://example.com"}
+        :type request: dict from URL request. /api/analyze=cache/http://example.com → {"api": True, "analyze": cache, "page": "http://example.com"}
         """
 
         crawl = None
@@ -34,13 +35,13 @@ class Api:
         elif "decide" in request:  # XX deprecated?
             return self.get_undecided()
         elif "nicify" in request:
-            return TrafficLogParser.getStylesheet() + TrafficLogParser.nicifyFile(request["page"])
+            return TrafficLogParser.getStylesheet() + TrafficLogParser.nicify_file(request["page"])
         elif "vote" in request:  # /api/vote/block/example.org/10.0.0.1
-            logging.debug("vote cmd")
+            logger.debug("vote cmd")
             return Turris.vote(request["vote"], request["page"])
         elif "scans" in request:
-            if "url" in request: # /api/scans/url/http://example.com
-                domain = Domains.domain2dir(request["page"])
+            if "url" in request:  # /api/scans/url/http://example.com
+                domain = domain2dir(request["page"])
                 if not domain:
                     return "Wrong domain"
                 return ScanController().get_domain_snapdirs(domain, full_dirs=False)
@@ -50,21 +51,23 @@ class Api:
             """url = path.split("/", 3)
             if len(url) > 3:
                 self._setWebsite(url[2])  # osetrit self.website, ze je URL, a nikoli shell
-                logging.debug("XXX nejsem si jist, zda url je spravne na url[2]")  # XXX
-                logging.debug(url)  # XXX
+                logger.debug("XXX nejsem si jist, zda url je spravne na url[2]")  # XXX
+                logger.debug(url)  # XXX
                 quit()  # XXX
-                logging.debug(self.website)
-                logging.debug(self.websiteDomain)
+                logger.debug(self.website)
+                logger.debug(self.websiteDomain)
             return self.whitelist()"""
             return "Implement first if needed."
         elif "reset" in request:
             self.reset()
             return "reset"
         else:
-            return "Unknown method."
+            return "Unknown API method."
 
         if crawl:
-            if request["api"] == "json":
+            if type(crawl) is str:  # probably an error
+                return crawl
+            elif request["api"] == "json":
                 return CrawlView.output_json(crawl)
             elif request["api"] == "mdmaug":
                 return CrawlView.output_mdmaug(crawl)
@@ -73,16 +76,16 @@ class Api:
 
     @staticmethod
     def reset():
-        logging.debug("resetting running browsers")
-        with open(Config.configFile, 'w') as f:  # clear the queue
+        logger.debug("resetting running browsers")
+        with open(Config.config_file, 'w') as f:  # clear the queue
             json.dump({}, f)
         subprocess.call(["pkill", Config.browser])  # kill frozen browsers
 
     # prida 2ld domenu mezi whitelistovane
     def whitelist(self):
-        logging.debug("whitelistuju")
+        logger.debug("whitelistuju")
         # Db.cur = Db.connection.cursor()
-        # self._logging.debug(Db.cur.execute("""REPLACE INTO whitelist set domain = %s""", (self.websiteDomain, )))
+        # self._logger.debug(Db.cur.execute("""REPLACE INTO whitelist set domain = %s""", (self.websiteDomain, )))
         # Db.connection.commit()
         # Db.cur.close()
         try:
@@ -92,5 +95,5 @@ class Api:
 
     @staticmethod
     def get_undecided():
-        logging.debug("XXX jeste jsem neudelal - ma vylezt tabulka vsech nerozhodlych domen od posledniho exportu")
+        logger.debug("XXX jeste jsem neudelal - ma vylezt tabulka vsech nerozhodlych domen od posledniho exportu")
         pass

mdmaug/lib/controller/server.py

 import json
 import logging
-from collections import defaultdict
+import re
 
-from flask import Blueprint, send_from_directory, render_template, request, make_response, current_app
+from flask import Blueprint, send_from_directory, render_template, request, make_response
 from jinja2 import Environment, FileSystemLoader
 from wtforms import Form
 from wtforms.fields import BooleanField
 
-from ..config import Config
+from ..config import Config, Pref
 from ..controller.api import Api
 from ..model.dbp import DbModel
 from ..model.dbp import Export
 
-env = Environment()
-env.loader = FileSystemLoader(Config.DIR + "templates/")
+# env = Environment()
+# env.loader = FileSystemLoader(Config.DIR + "templates/")
 
+logger = logging.getLogger("mdmaug")
 app = Blueprint('app', __name__, template_folder='templates')
 
 
 def update_preferences():
     """ cookies → config """
     for k, v in request.cookies.items():
-        if v is "0":
+        if v in ["0", ""]:
             v = False
         elif v is "1":
             v = True
-        current_app.config["preferences"][k] = v
+        setattr(Pref, k, v)
+        # current_app.config["preferences"][k] = v
 
 
 @app.route('/test')
@@ -41,10 +43,11 @@ def homepage():
     update_preferences()
 
     class OptionsForm(Form):
-        pref = defaultdict(bool, current_app.config["preferences"])
-        safebrowsing = BooleanField('Google Safebrowsing', default=pref["safebrowsing"])
-        pdns = BooleanField('PDNS', default=pref["pdns"])
-        geoip = BooleanField('geoip', default=pref["geoip"])
+        # pref = defaultdict(bool, current_app.config["preferences"])
+        safebrowsing = BooleanField('Google Safebrowsing', default=Pref.safebrowsing,
+                                    render_kw={"value": Pref.val2html("safebrowsing")})
+        pdns = BooleanField('PDNS', default=Pref.pdns, render_kw={"value": Pref.val2html("pdns")})
+        geoip = BooleanField('geoip', default=Pref.geoip, render_kw={"value": Pref.val2html("geoip")})
 
     if request.method == 'POST':
         name = request.form['name']
@@ -77,13 +80,14 @@ def controller(request_url):
     DbModel.assureConnection()
 
     # parse the request url into a friendly dictionary
+    keywords = {"safebrowsing", "geoip", "api", "destination", "analyze", "pdns", "export"}
     request = {"page": ""}
     page = False
     for l in request_url.split("/"):
         if not page:
             c, *d = l.split("=", 1)
-            if c in ["http:", "https:"]:
-                page = True
+            if re.search("[a-zA-Z]+:|\.", c):  # page def begins when we have a transport protocol ':' or a domain dot in param
+                page = True  # Xin ["http:", "https:"]:
             else:
                 request[c] = d[0] if len(d) else True
                 if c == "nicify":
@@ -95,9 +99,14 @@ def controller(request_url):
     if request["page"]:  # strip last slash
         request["page"] = request["page"][:-1]
 
-    logging.debug("Request: {}".format(request))
+    logger.debug("Request: {}".format(request))
 
     if "api" in request:  # /api/analyze/web
+        # set api-preferences from request, non-set are default
+        for s in Pref.__dict__:
+            if not s.startswith("_") and s in request:
+                setattr(Pref, s, request[s] in ["y", "1", True])
+
         output = Api().run(request)
 
         if request["api"] == "json":

mdmaug/lib/domains.py

-import socket
+import logging
 import re
 import socket
-import urllib.request
-
-
-class Domains:
-    """ webove nastroje """
-
-    def get_ips_for_host(host):
-        try:
-            ips = socket.getaddrinfo(host, 80, 0, 0, socket.IPPROTO_TCP)  # XXX co kdyz nepratelsky web reaguje jen na 80, 81, 8080
-        except socket.gaierror:
-            ips = []
-        return ips
-
-    def url2domain(url):
-        """ http://www.seznam.cz/url -> www.seznam.cz """
-        try:
-            return re.search('(([\w\-_]+(?:(?:\.[\w\-_]+)+)))', url).group(0)
-        except TypeError:
-            logging.debug("Domains/url2domain type error")
-            logging.debug(url)
-            raise  # return ""
-
-    def url2path(url):
-        """ http://seznam.cz/url -> /url """
-        url = re.sub('^(http://|https://|ftp://)', '', url)  # odstrihnout protokol
-        url = re.sub('^([^/])*', '', url)  # stojim jen cestu, ne o domenu
-        return url
-
-    def assureUrl(url):  # zajistit, ze se jedna o url a ne treba o shell
-        # XX what about other protocols? smb, sftp? Is there a danger?
-        return re.search(
-            '(((((http|https|ftp)://)?[\w\-_]+(?:(?:\.[\w\-_]+)+)))([\w\-\.,@?^=%&:/~\+#]*[\w\-\@?^=%&/~\+#])?)',
-            url).group(0)
-
-    def domain2dir(url) -> str:
-        """
-        Friendly dir name from a domain contained in an url
-
-        XXmay be replaced by urlparse(url).netloc?
-        """
-        url = url.lower()
-        url = re.sub('^(http://|https://|ftp://)', '', url)  # odstrihnout protokol
-        url = re.sub('(/.*)', '', url)  # nestojim o cestu, jen o domene
-        url = re.sub('[^a-z0-9\.]', '', url)  # nechat jen pratelske znaky
-        return url
-
-    def getPdnsLink(ip):
-        return 'http://pdns.cert.at/p/dns?qry=' + ip
-
-    def ip2pdnsDomains(ip):
-        """
-        try:
-            # XX mohl bych sortovat dle 2nd domeny. Pripadne oriznout 3rd domenu, nechat jen 2nd. Tam ale musim osetrit problemove dvojite tld - co.uk, gov.ua...
-            pdns = urllib.request.urlopen(Domains.getPdnsLink(ip)).read().decode("utf-8")
-            items = re.findall("<div class='x[BA]'>(.*)</div>", pdns)
-            return items
-        except Exception as e:
-            logging.debug("chyba pri kontaktu s PDNS: " + str(e))
+from urllib.parse import urlparse
+
+from .config import Pref
+
+logger = logging.getLogger("mdmaug")
+re_url = re.compile('(((((http|https|ftp)://)?[\w\-_]+(?:(?:\.[\w\-_]+)+)))([\w\-\.,@?^=%&amp;:/~\+#]*[\w\-\@?^=%&amp;/~\+#])?)$')
+
+
+def url2domain(url):
+    """ http://www.seznam.cz/url -> www.seznam.cz """
+    try:
+        return re.search('(([\w\-_]+(?:(?:\.[\w\-_]+)+)))', url).group(0)
+    except TypeError:
+        logger.debug("Domains/url2domain type error")
+        logger.debug(url)
+        raise  # return ""
+
+
+def url2path(url):
+    """ http://seznam.cz/url -> /url """
+    url = re.sub('^(http://|https://|ftp://)', '', url)  # odstrihnout protokol
+    url = re.sub('^([^/])*', '', url)  # stojim jen cestu, ne o domenu
+    return url
+
+
+def assure_url(url):
+    """  Assure this is URL and not a shell.
+
+    # XX what about other protocols? smb, sftp? Is there a danger?
+
+    :param url: any URL
+    :return: url, prepended with "http" if no scheme found, or false
+    """
+    if not urlparse(url).scheme:
+        url = "http://" + url
+    s = re_url.match(url)
+    if s:
+        return s.group(0)
+    return False
+
+
+def domain2dir(url) -> str:
+    """
+    Friendly dir name from a domain contained in an url
+
+    XXmay be replaced by urlparse(url).netloc?
+    """
+    url = url.lower()
+    url = re.sub('^(http://|https://|ftp://)', '', url)  # odstrihnout protokol
+    url = re.sub('(/.*)', '', url)  # nestojim o cestu, jen o domene
+    url = re.sub('[^a-z0-9\.]', '', url)  # nechat jen pratelske znaky
+    return url
+
+
+def ip2pdns_domains(ip):
+    if not Pref.pdns:
         return None
-        """
-        return None  # #24 doesnt work
-
-    def ip2countryAndCity(ip):
-        """
-        try:
-            hostipApi = urllib.request.urlopen('http://api.hostip.info/get_html.php?ip=' + ip + '&position=true').read().decode(
-                "utf-8").split("\n")
-            # ['Country: CZECH REPUBLIC (CZ)', 'City: Prague', '', 'Latitude: 50.0833', 'Longitude: 14.4333', 'IP: 109.123.209.188', '']
-            return hostipApi[0].split(":")[1], hostipApi[1].split(":")[1]
-        except UnicodeDecodeError:  # as e
-            logging.debug("neumim dekodovat")
-        except Exception as e:
-            logging.debug("hostip.info down: " + str(e))
+    """
+    try:
+        # XX mohl bych sortovat dle 2nd domeny. Pripadne oriznout 3rd domenu, nechat jen 2nd. Tam ale musim osetrit problemove dvojite tld - co.uk, gov.ua...
+        pdns = urllib.request.urlopen(Domains.get_pdns_link(ip)).read().decode("utf-8")
+        items = re.findall("<div class='x[BA]'>(.*)</div>", pdns)
+        return items
+    except Exception as e:
+        logger.debug("chyba pri kontaktu s PDNS: " + str(e))
+    return None
+    """
+    return None  # #24 doesnt work
+
+
+def ip2country_and_city(ip):
+    if not Pref.geoip:
         return None, None
-        """
-        return None, None  # #23 service down
-
-    @staticmethod
-    def is_suspicious(domain, output='bool'):
-        """
-        Scrape Safebrowsing service webpages and try to read out if the site is considered dangerous to visit.
-        :param domain:
-        :param output: 'bool' → True/False/None or 'attr' → int "1", "0", "" for an HTML attribute
-        :return:
-        """
-        # contents = urllib.request.urlopen('http://www.google.com/safebrowsing/diagnostic?site=' + domain).read().decode("utf-8")
-        # with open("debugsf.tmp","a") as f:
-        #    f.write(contents  + "\n\n")
-        # if "Site is listed as suspicious" in contents:
-        # elif "This site is not currently listed as suspicious." in contents:
-        import requests
-        r = requests.get("http://www.google.com/safebrowsing/diagnostic?output=jsonp&site=" + domain, timeout=5)
-        if '"listed"' in r.text:
-            return True if output == 'bool' else "1"
-        if '"unlisted"' in r.text:  # vratilo to alespon neco rozumneho
-            return False if output == 'bool' else "0"
-        else:
-            return None if output == 'bool' else ""
+    """
+    try:
+        hostipApi = urllib.request.urlopen('http://api.hostip.info/get_html.php?ip=' + ip + '&position=true').read().decode(
+            "utf-8").split("\n")
+        # ['Country: CZECH REPUBLIC (CZ)', 'City: Prague', '', 'Latitude: 50.0833', 'Longitude: 14.4333', 'IP: 109.123.209.188', '']
+        return hostipApi[0].split(":")[1], hostipApi[1].split(":")[1]
+    except UnicodeDecodeError:  # as e
+        logger.debug("neumim dekodovat")
+    except Exception as e:
+        logger.debug("hostip.info down: " + str(e))
+    return None, None
+    """
+    return None, None  # #23 service down
+
+
+def get_pdns_link(ip):
+    return 'http://pdns.cert.at/p/dns?qry=' + ip
+
+
+def is_suspicious(domain, output='bool'):
+    """
+    Scrape Safebrowsing service webpages and try to read out if the site is considered dangerous to visit.
+    :param domain:
+    :param output: 'bool' → True/False/None or 'attr' → int "1", "0", "" for an HTML attribute
+    :return:
+    """
+    if not Pref.safebrowsing:
+        return None
+    # contents = urllib.request.urlopen('http://www.google.com/safebrowsing/diagnostic?site=' + domain).read().decode("utf-8")
+    # with open("debugsf.tmp","a") as f:
+    #    f.write(contents  + "\n\n")
+    # if "Site is listed as suspicious" in contents:
+    # elif "This site is not currently listed as suspicious." in contents:
+    import requests
+    logger.debug("Safebrowsing %s", domain)
+    r = requests.get("http://www.google.com/safebrowsing/diagnostic?output=jsonp&site=" + domain, timeout=5)
+    if '"listed"' in r.text:
+        return True if output == 'bool' else "1"
+    if '"unlisted"' in r.text:  # vratilo to alespon neco rozumneho
+        return False if output == 'bool' else "0"
+    else:
+        return None if output == 'bool' else ""
+
+
+def get_ips_for_host(host):
+    try:
+        ips = socket.getaddrinfo(host, 80, 0, 0, socket.IPPROTO_TCP)  # XXX co kdyz nepratelsky web reaguje jen na 80, 81, 8080
+    except socket.gaierror:
+        ips = []
+    return ips

mdmaug/lib/model/crawl.py

@@ -28,7 +28,7 @@ class Crawl(defaultdict):
         with open(filename, 'r') as f:
             return Crawl(state=load(f.read(), Loader=Loader))
 
-    def __init__(self, host=None, state=None, log_dir=None, cache_dir=None):
+    def __init__(self, host=None, state=None, log_dir=None, cache_dir=None, profile=None):
         """ State muze obsahovat vystup __getstate__() (serializace YAMLem) """
         self.default_factory = _Domain
         self.screenfile = None  # HTML output XXX
@@ -39,6 +39,7 @@ class Crawl(defaultdict):
         self.logDir = log_dir
         # if cache_dir:
         self.cacheDir = cache_dir
+        self.profile = profile
         if state:
             self.__setstate__(state)
 
@@ -141,6 +142,7 @@ class _Address(set):
 
 """
 Example (and yaml-serialization check):
+logger = logging.getLogger("mdmaug")
 c = Crawl()
 c["seznam.cz"].urls["/aurl"].spyfile = "/soubor-spyfil"
 c["seznam.cz"].urls["/aurl"].sourcefiles.append("/1.source") 
@@ -153,15 +155,15 @@ c["seznam.cz"].addresses["8.8.8.8"].country = "preague"
 e = Crawl()
 e.__setstate__( c.__getstate__() )
 #e = dill.loads(dill.dumps(c))
-logging.debug(str(c) == str(e))
+logger.debug(str(c) == str(e))
 
 
-logging.debug(c)
+logger.debug(c)
 output = dump(c.__getstate__(), Dumper=Dumper)
 e = Crawl()
 e.__setstate__(load(output, Loader=Loader))
-logging.debug(e)
-logging.debug(str(c) == str(e))
+logger.debug(e)
+logger.debug(str(c) == str(e))
 """
 
 """

mdmaug/lib/model/dbp.py

@@ -6,51 +6,50 @@ from peewee import Model, DateTimeField, IntegerField, CharField, JOIN, BigAutoF
 from peewee import RawQuery
 
 from ..config import Config
-from ..domains import Domains
+from ..domains import url2domain
 
 # Do not print all queries to stderr.
-logger = logging.getLogger('peewee')
-logger.setLevel(logging.WARNING)
-
+logging.getLogger('peewee').setLevel(logging.WARNING)
+logger = logging.getLogger("mdmaug")
 
 class DbModel(Model):
 
     @staticmethod
     def assureConnection():
-        logging.debug("Assure connection.")
+        logger.debug("Assure connection.")
         try:  # zkousim spustit nejaky prikaz
-            # logging.debug("1")
+            # logger.debug("1")
             Whitelist.select().count()
-            # logging.debug("2")
+            # logger.debug("2")
         except:  # OperationalError:
-            # logging.debug("3")
+            # logger.debug("3")
             Config.connect()
-            # logging.debug("4")
+            # logger.debug("4")
             try:  # zkousim spustit nejaky prikaz
-                # logging.debug("5")
+                # logger.debug("5")
                 Whitelist.select().count()
-                # logging.debug("6")
+                # logger.debug("6")
             except:
-                logging.debug("7 - failed")
+                logger.debug("7 - failed")
                 raise
-        logging.debug("Connection to DB assured.")
+        logger.debug("Connection to DB assured.")
 
     """A base model that will use our MySQL database"""
 
     @staticmethod
     def connect():
-        logging.debug("connecting db....")
+        logger.debug("connecting db....")
         # DbModel.Meta.myDb = Config.myDB
         # Config.myDB.connect() # XX kupodivu toto neni potreba
-        # logging.debug(Whitelist.select().count())
-        # logging.debug("Connecting to DB.")
-        # logging.debug("Start test.")
-        # logging.debug(Whitelist.matches("www.mozilla.org"))
-        # logging.debug("End test.")
+        # logger.debug(Whitelist.select().count())
+        # logger.debug("Connecting to DB.")
+        # logger.debug("Start test.")
+        # logger.debug(Whitelist.matches("www.mozilla.org"))
+        # logger.debug("End test.")
 
     @staticmethod
     def disconnect():
-        logging.debug("... db ends.")
+        logger.debug("... db ends.")
         Config.myDB.close()
 
     class Meta:
@@ -58,9 +57,9 @@ class DbModel(Model):
         pass
 
 
-# logging.debug("DVA")
+# logger.debug("DVA")
 # DbModel.connect()
-# logging.debug(Config.myDB)
+# logger.debug(Config.myDB)
 
 class Status(DbModel):
     id = BigAutoField()
@@ -111,7 +110,7 @@ class Export(DbModel):
                 q += "(select case when MAX(timestamp IS NULL)=0 THEN max(timestamp) ELSE 0 END from export)"
             q += " GROUP BY concat(`ip`,`port`) "  # group by concat may be a performance issue
             q += " ORDER BY `timestamp` DESC"
-            logging.debug(q)
+            logger.debug(q)
             rq = RawQuery(Turris, q).execute()
             print(rq)
         for r in rq:  # for r in rows: