#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 +1,85 @@
+#!/usr/bin/env python3
+
+import argparse
+import re
+import sys
+import threading
+
+from http.server import BaseHTTPRequestHandler, HTTPServer
+from string import Template
+
+PROXY_TOKEN = 'http_'
+REGEX = '\\$' + PROXY_TOKEN + '[a-z_0-9]+\\b'

[Vasily Kuznetsov, 2017/08/10 15:05:57]

How about making the regex r'\$http_(\w+)\b'? Then you can just grab the first
group from the match and you don't need the PROXY_TOKEN constant.

[Vasily Kuznetsov, 2017/08/11 17:31:47]

You don't like my idea? Wouldn't the code become more compact and readable?

+DEFAULT_LOG = '$remote_addr - - [$time_local] "$request" $status $bytes_sent'
+LOCK = threading.Lock()

[mathias, 2017/08/09 21:09:07]

Not sure if an upper-case name is appropriate for this one, but either way it
should have a prefixing underscore to designate it as an internal thingy.

[Vasily Kuznetsov, 2017/08/10 15:05:57]

Yeah, this is not a constant in my book either.

[f.lopez, 2017/08/10 19:06:34]

Done.

+
+
+class Handler(BaseHTTPRequestHandler):
+    def get_header_values(self):
+        values = {}
+        headers = re.findall(REGEX, self.format)
+        for name in headers:
+            new_var = name[len(PROXY_TOKEN):].title()

[mathias, 2017/08/09 21:09:06]

Shouldn't this take the `$` into account, because otherwise it is searching for
a header that starts with an underscore? And why use method .title() here? I am
pretty sure this is not properly recognizing Foo-Bar headers, which should get
mapped as $http_foo_bar.

[f.lopez, 2017/08/10 19:06:33]

Done.

+            values[name] = self.headers.get(new_var, '-')
+        return values
+
+    def write_info(self, args):
+        LOCK.acquire()
+        message = Template(self.format)
+        if self.log_file and self.log_file is not sys.stdout:

[mathias, 2017/08/09 21:09:06]

This should probably check for `not in ('/dev/stdout', '-')` or something, not
compare a string for identity with a file handle.

[Vasily Kuznetsov, 2017/08/10 15:05:57]

If it's '/dev/stdout', then you just open it, no? Otherwise we'd have to support
all funny ways that OSes have to designate stdout.

I agree with the rest of the comment though, it makes sense to support '-' as a
file name and just do `fh = sys.stdout` instead of open in this case. Also, the
default would be '-' instead of sys.stdout, which is also more appropriate for a
command line argument with the type 'str'.

[f.lopez, 2017/08/10 19:06:34]

Done.

+            fh = open(self.log_file, 'a')

[mathias, 2017/08/09 21:09:06]

Why open and close the file handle with every invocation / request?

[Vasily Kuznetsov, 2017/08/10 15:05:58]

Just checked, and apparently this class is instantiated for every request. This
means we want to create the lock (as we already do) and open/close the file
outside of the class.

[f.lopez, 2017/08/10 19:06:34]

Done.

+        else:
+            fh = sys.stdout
+        fh.write(message.safe_substitute(args))
+
+        if self.log_file is not sys.stdout:
+            fh.close()
+        else:
+            fh.flush()
+
+        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),
+            'content_type': 'text/plain',
+            'content_length': len(content),

[mathias, 2017/08/09 21:09:07]

Since we're emulating Nginx behavior this should actually be prefixed:
http://nginx.org/en/docs/http/ngx_http_core_module.html#var_sent_http_

[f.lopez, 2017/08/10 19:06:33]

Done.

+        }
+        values.update(self.get_header_values())
+        self.write_info(values)
+        self.send_response(status)
+        self.send_header('Content-Type', values['content_type'])
+        self.send_header('Content-Length', values['content_length'])
+        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('log_file', action='store',
+                        type=str, nargs='?', default=sys.stdout,
+                        help='The file where the logs will be written')
+    args = parser.parse_args()
+    setattr(Handler, 'format', args.format)
+    setattr(Handler, 'response', args.response)
+    setattr(Handler, 'log_file', args.log_file)

[mathias, 2017/08/09 21:09:06]

Again this is not a log, but regular 'output' or a 'target' or, analogous with
to method .write_info(), 'info_file', or similar. 

[f.lopez, 2017/08/10 19:06:34]

Done.

+    server_address = ('', args.port)
+    httpd = HTTPServer(server_address, Handler)
+    httpd.serve_forever()