Add support for configuration files.

README

@@ -161,6 +161,43 @@ without parsing more logs::
   apachedex --default foo --state-file save_state.json save_state.2.json
   --period day --out index.html
 
+
+Configuration files
+===================
+
+Providing a filename prefixed by "@" puts the content of that file in place of
+that argument, recursively. Each file is loaded relative to the containing
+directory of referencing file, or current working directory for command line.
+
+- foo/dev.cfg::
+
+  --error-detail
+  @site.cfg
+  --stats
+
+- foo/site.cfg::
+
+  --default Front-office
+  # This is a comment
+  --prefix "+Back office" "/back(/|$|\?)" # This is another comment
+  --skip-prefix "/baz/ignored(/|$|\?)" --prefix +Something "/baz(/|$|\?)"
+
+- command line::
+
+  apachedex --skip-base "/ignored(/|$|\?)" @foo/dev.cfg --out index.html
+  access.log
+
+This is equivalent to::
+
+  apachedex --skip-base "/ignored(/|$|\?)" --error-detail
+  --default Front-office --prefix "+Back office" "/back(/|$|\?)"
+  --skip-prefix "/baz/ignored(/|$|\?)" --prefix +Something "/baz(/|$|\?)"
+  --stats --out index.html access.log
+
+Portability note: the use of paths containing directory elements inside
+configuration files is discouraged, as it's not portable. This may change
+later (ex: deciding that import paths are URLs and applying their rules).
+
 Performance
 ===========
 

apachedex/__init__.py

@@ -44,6 +44,7 @@ import math
 import os
 import platform
 import re
+import shlex
 import sys
 import time
 import traceback
@@ -751,6 +752,49 @@ class AggregateSiteUrl(argparse.Action):
           suffix=lambda x: match_suffix(x).group('suffix'))
       site_list.append((value, match, action))
 
+class ShlexArgumentParser(argparse.ArgumentParser):
+  """
+  Two objectives in this class:
+  - use shlex to split config files
+  - when recursively including files, do it from referer's path instead of
+    current working directory, to facilitate relative inclusion.
+  """
+  # XXX: I whould like to be able to hook inside _read_args_from_files, but
+  # it would be dirtier. Instead, declare a private method doing similar
+  # replacement before handing args to original parse_known_args.
+  def __read_args_from_files(self, args, cwd):
+    new_args = []
+    append = new_args.append
+    extend = new_args.extend
+    for arg in args:
+      if arg[:1] in self.fromfile_prefix_chars:
+        filepath = arg[1:]
+        new_cwd = os.path.normpath(os.path.join(
+          cwd,
+          os.path.dirname(filepath),
+        ))
+        try:
+          with open(os.path.join(new_cwd, os.path.basename(filepath))
+              ) as in_file:
+            extend(self.__read_args_from_files(
+              shlex.split(in_file.read(), comments=True),
+              new_cwd,
+            ))
+        except IOError, exc:
+          self.error(str(exc))
+      else:
+        append(arg)
+    return new_args
+
+  def parse_known_args(self, args=None, namespace=None):
+    if args is None:
+      args = sys.argv[1:]
+    else:
+      args = list(args)
+    args = self.__read_args_from_files(args, os.getcwd())
+    return super(ShlexArgumentParser, self).parse_known_args(args=args,
+      namespace=namespace)
+
 def _asMonthString(timestamp):
   dt, _ = timestamp.split(' ')
   _, month, year = dt.split(':', 1)[0].split('/')
@@ -1060,8 +1104,8 @@ json.decoder.scanstring = _scanstring
 
 def main():
   global abs_file_container
-  parser = argparse.ArgumentParser(description='Compute Apdex out of '
-    'apache-style log files')
+  parser = ShlexArgumentParser(description='Compute Apdex out of '
+    'apache-style log files', fromfile_prefix_chars='@')
   parser.add_argument('logfile', nargs='*',
     help='Log files to process. Use - for stdin.')
   parser.add_argument('-l', '--logformat',