Merge branch 'master' into stdin-reader

.dockerignore

@@ -0,0 +1,57 @@
+# Byte-compiled / optimized / DLL files
+**/__pycache__/
+**/*.py[cod]
+
+# C extensions
+***/*.so
+
+# Distribution / packaging
+.Python
+env/
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+*.egg-info/
+.installed.cfg
+*.egg
+
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+cover
+
+# Translations
+**/*.mo
+**/*.pot
+
+# PyBuilder
+target/
+
+venv/
+venv3/
+.idea/
+
+# 100% unnecessary for docker image
+.*
+*.md
+docs
+rpm
+Dockerfile

.editorconfig

@@ -0,0 +1,19 @@
+root = true
+
+[*]
+end_of_file = lf
+insert_final_newline = true
+
+[*.{py,j2}]
+charset = utf-8
+
+[*.py]
+indent_style = space
+indent_size = 4
+
+[Makefile]
+indent_style = tab
+
+[.travis.yml]
+indent_style = space
+indent_size = 2

Dockerfile

@@ -0,0 +1,9 @@
+FROM python:2.7-alpine
+
+ADD . /src
+
+WORKDIR /src
+
+RUN python2 setup.py install
+
+ENTRYPOINT ["gixy"]

docs/en/plugins/addheaderredefinition.md

@@ -1,6 +1,6 @@
 # [add_header_redefinition] Redefining of response headers by  "add_header" directive
 
-Unfortunately, many people doesn't know how the inheritance of directives works. Most often this leads to misuse of the `add_header` directive while trying to add a new response header on the nested level.
+Unfortunately, many people don't know how the inheritance of directives works. Most often this leads to misuse of the `add_header` directive while trying to add a new response header on the nested level.
 This feature is mentioned in Nginx [docs](http://nginx.org/en/docs/http/ngx_http_headers_module.html#add_header):
 > There could be several `add_header` directives. These directives are inherited from the previous level if and only if there are no `add_header` directives defined on the current level.
 

docs/en/plugins/hostspoofing.md

@@ -10,7 +10,7 @@ Spoofing of this header, may leads to a variety of problems, from phishing to SS
 Most of the time it's a result of using `$http_host` variable instead of `$host`.
 
 And they are quite different:
-  * `$http` - host in this order of precedence: host name from the request line, or host name from the “Host” request header field, or the server name matching a request;
+  * `$host` - host in this order of precedence: host name from the request line, or host name from the “Host” request header field, or the server name matching a request;
   * `$http_host` - "Host" request header.
 
 Config sample:
@@ -29,4 +29,4 @@ Luckily, all is quite obvious:
 
 ## Additional info
   * [Host of Troubles Vulnerabilities](https://hostoftroubles.com/)
-  * [Practical HTTP Host header attacks](http://www.skeletonscribe.net/2013/05/practical-http-host-header-attacks.html)
+  * [Practical HTTP Host header attacks](http://www.skeletonscribe.net/2013/05/practical-http-host-header-attacks.html)

gixy/directives/block.py

@@ -141,7 +141,7 @@ class IfBlock(Block):
             # if ($request_method = POST)
             self.variable, self.operand, self.value = args
         else:
-            raise Exception('Unknown "if" definition')
+            raise Exception('Unknown "if" definition, args: {}'.format(repr(args)))
 
     def __str__(self):
         return '{} ({}) {{'.format(self.name, ' '.join(self.args))

gixy/parser/raw_parser.py

@@ -1,4 +1,3 @@
-import re
 import logging
 from cached_property import cached_property
 
@@ -31,9 +30,7 @@ class RawParser(object):
         """
         Returns the parsed tree.
         """
-        # Temporary, dirty hack :(
-        content = self._if_fixer.sub('\\1) )\\2', data)
-        return self.script.parseString(content, parseAll=True)
+        return self.script.parseString(data, parseAll=True)
 
     @cached_property
     def script(self):
@@ -59,10 +56,15 @@ class RawParser(object):
             Keyword("=") |
             Keyword("~*") | Keyword("~") |
             (Literal("-") + (Literal("f") | Literal("d") | Literal("e") | Literal("x")))))
-        condition = (
+        condition_body = (
             (if_modifier + Optional(space) + value) |
             (variable + Optional(space + if_modifier + Optional(space) + value))
         )
+        # This ugly workaround needed to parse unquoted regex with nested parentheses
+        # pyparsing.nestedExpr doesn't work in some rare cases like: ($http_user_agent ~* \( )
+        # so we capture all content between parentheses and then parse it:)
+        # TODO(buglloc): may be use something better?
+        condition = Regex(r'\(.*\)').setParseAction(lambda s, l, t: condition_body.parseString(t[0][1:-1]))
 
         # rules
         include = (
@@ -112,9 +114,7 @@ class RawParser(object):
 
         if_block << (
             Keyword("if") +
-            Suppress("(") +
             Group(condition) +
-            Suppress(")") +
             Group(
                 left_bracket +
                 Optional(sub_block) +

tests/parser/test_raw_parser.py

@@ -225,11 +225,14 @@ if (!-e "/var/data/$dataset") {
     return 503;
 }
 
-if ($https_or_slb = (by_slb|https)) {
+if ($https_or_slb = (by_\(sl\)b|https)) {
 }
 
 if ($host ~* (lori|rage2)\.yandex\.(ru|ua|com|com\.tr)) {
     set $x_frame_options ALLOW;
+}
+
+if ($request_filename ~* ^.*?/(\d+_)([^/]+)$) {
 }
         '''
 
@@ -253,11 +256,13 @@ if ($host ~* (lori|rage2)\.yandex\.(ru|ua|com|com\.tr)) {
         ['if', ['!-e', '/var/data/$dataset'], [
             ['return', '503']
         ]],
-        ['if', ['$https_or_slb', '=', '(by_slb|https)'], [
+        ['if', ['$https_or_slb', '=', '(by_\(sl\)b|https)'], [
         ]],
         ['if', ['$host', '~*', '(lori|rage2)\.yandex\.(ru|ua|com|com\.tr)'], [
             ['set', '$x_frame_options', 'ALLOW']
         ]],
+        ['if', ['$request_filename', '~*', '^.*?/(\d+_)([^/]+)$'], [
+        ]]
     ]
 
     assert_config(config, expected)