#2687 - Include mimeo python module

modules/adblockplus/files/mimeo.py

Index: modules/adblockplus/files/mimeo.py
===================================================================
new file mode 100644
--- /dev/null
+++ b/modules/adblockplus/files/mimeo.py
@@ -0,0 +1,84 @@
+#!/usr/bin/env python3
+
+import argparse
+import re
+import sys
+import threading
+
+from http.server import BaseHTTPRequestHandler, HTTPServer
+from string import Template
+
+# The token used for the headers passed by nginx is: http_
+REGEX = '\\$http_[a-z_0-9]+\\b'

[Vasily Kuznetsov, 2017/08/14 10:59:56]

It's a bit more elegant to use a raw string here: r'\$http_[a-z_0-9]+\b'. Also
you can use '\w' special symbol that is equivalent to '[a-zA-Z0-9_]' unless you
specifically want to exclude capital letters.

+DEFAULT_LOG = '$remote_addr - - [$time_local] "$request" $status $bytes_sent'
+_lock = threading.Lock()
+
+
+class Handler(BaseHTTPRequestHandler):
+    def get_header_values(self):
+        values = {}
+        headers = re.findall(REGEX, self.format)
+        for name in headers:
+            new_var = name[6:]
+            values[name[1:]] = self.headers.get(new_var, '-')
+        return values
+
+    def write_info(self, args):
+        _lock.acquire()
+        try:
+            message = Template(self.format)
+            self.output.write(message.safe_substitute(args))
+            self.output.flush()
+        except Exception as e:
+            sys.stderr.write(e)

[Vasily Kuznetsov, 2017/08/14 10:59:56]

I've just tested this and it seems to raise a TypeError "write() argument must
be str, not Exception". You should probably look into `traceback` module if you
want to do this: https://docs.python.org/3/library/traceback.html

Speaking about wanting to do this, I hope you realize that you are changing
behavior here. Before, if some part of `write_info` would fail, it would cause
and error and the user would see an HTTP500 response. Now the exception is
swallowed and the user sees a normal response. If that's what you want, all is
cool though.

+        finally:
+            _lock.release()
+
+    def do_POST(self):
+        status = 200
+        content = bytes(self.response, 'UTF-8')
+        values = {
+            'remote_addr': self.address_string(),
+            'time_local': self.log_date_time_string(),
+            'request': self.requestline,
+            'status': status,
+            'bytes_sent': len(content),
+        }
+        values.update(self.get_header_values())
+        self.write_info(values)
+        self.send_response(status)
+        self.end_headers()
+        self.wfile.write(content)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--port', action='store',
+                        default=8000, type=int,
+                        nargs='?',
+                        help='Port to use [default: 8000]')
+    parser.add_argument('--response', action='store',
+                        type=str, nargs='?', default='OK',
+                        help='The response send to the client')
+    parser.add_argument('--format', action='store',
+                        type=str, nargs='?',
+                        default=DEFAULT_LOG,
+                        help='Format of the log ouput')
+    parser.add_argument('output', action='store',
+                        type=str, nargs='?', default='-',
+                        help='The file where the logs will be written')
+    args = parser.parse_args()
+    if args.output and args.output != '-':
+        fh = open(args.output, 'a')
+    else:
+        fh = sys.stdout
+    setattr(Handler, 'output', fh)
+    setattr(Handler, 'format', args.format)
+    setattr(Handler, 'response', args.response)
+    server_address = ('', args.port)
+    httpd = HTTPServer(server_address, Handler)
+    try:
+        httpd.serve_forever()
+    except:
+        if args.output != '-':
+            fh.close()