From 2a11d6812f703e01150c2aedbca7ec2a6590f254 Mon Sep 17 00:00:00 2001
From: Bert JW Regeer <bertjw@regeer.org>
Date: Mon, 23 Dec 2019 14:36:41 +0100
Subject: [PATCH] Add file with RFC7230 definitions/ABNF -> regex

Currently it is used for header name/fields only, but this should get
expanded as necessary.

Other stuff from other parts of waitress may move here too to create a
sort utility library of useful RFC standards based functions.
---
 waitress/rfc7230.py | 44 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)
 create mode 100644 waitress/rfc7230.py

diff --git a/waitress/rfc7230.py b/waitress/rfc7230.py
new file mode 100644
index 00000000..a9f047c8
--- /dev/null
+++ b/waitress/rfc7230.py
@@ -0,0 +1,44 @@
+"""
+This contains a bunch of RFC7230 definitions and regular expressions that are
+needed to properly parse HTTP messages.
+"""
+
+import re
+
+from .compat import tobytes
+
+WS = "[ \t]"
+OWS = WS + "{0,}?"
+RWS = WS + "{1,}?"
+BWS = OWS
+
+# RFC 7230 Section 3.2.6 "Field Value Components":
+# tchar          = "!" / "#" / "$" / "%" / "&" / "'" / "*"
+#                / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~"
+#                / DIGIT / ALPHA
+# obs-text      = %x80-FF
+TCHAR = r"[!#$%&'*+\-.^_`|~0-9A-Za-z]"
+OBS_TEXT = r"\x80-\xff"
+
+TOKEN = TCHAR + "{1,}"
+
+# RFC 5234 Appendix B.1 "Core Rules":
+# VCHAR         =  %x21-7E
+#                  ; visible (printing) characters
+VCHAR = r"\x21-\x7e"
+
+# header-field   = field-name ":" OWS field-value OWS
+# field-name     = token
+# field-value    = *( field-content / obs-fold )
+# field-content  = field-vchar [ 1*( SP / HTAB ) field-vchar ]
+# field-vchar    = VCHAR / obs-text
+
+FIELD_VCHAR = "[" + VCHAR + OBS_TEXT + "]"
+FIELD_CONTENT = FIELD_VCHAR + "(" + RWS + FIELD_VCHAR + "){0,}"
+FIELD_VALUE = "(" + FIELD_CONTENT + "){0,}"
+
+HEADER_FIELD = re.compile(
+    tobytes(
+        "^(?P<name>" + TOKEN + "):" + OWS + "(?P<value>" + FIELD_VALUE + ")" + OWS + "$"
+    )
+)