Fix #96 Add EBNF grammar for ROS2 names in self-testing python script

articles/ebnf_names.py

@@ -0,0 +1,207 @@
+#! /usr/bin/env python
+
+# PoC Ros2Names parsing using EBNF and Grako (install from pypi)
+# Tested with grako 3.14.0
+#
+
+
+## TODO:
+# maximum length + example
+
+## ambiguous texts
+# "balanced" curly braces may imply nesting?
+# url scheme only 'rostopic' and 'rosservice' allowed?
+
+
+import re
+import grako
+
+
+POSITIVES_QUALIFIED = [
+    "/x", "/foo", "/abc123", "/_foo", "/Foo", "/BAR",
+    "/foo/bar", "/x/y/z",
+    "/_foo/bar", "/_f_o_o/bar",
+    "rosservice:///foo", "rostopic:///foo/bar"
+    ]
+
+POSITIVES_UNQUALIFIED = [
+    "x", "foo", "abc123", "_foo", "Foo", "BAR",
+    "x/y/z",
+    "foo/bar",
+     "{foo}", "{foo}_bar", "foo/{ping}/bar",
+     "foo{x1}bar{x2}baz", "{x1}bar{x2}baz{top}",
+     "foo{x1}/bar/{x2}baz", "{x1}/bar/{x2}",
+     "~", "~/foo", "~/foo/x/_123",
+     "_foo/bar", "_f_o_o/bar",
+     "rosservice://foo", "rostopic://foo/bar", "rostopic://~/_bar/foo{x}",
+     "~/{foo}",
+     "foo{bar__baz}", "foo{_baz_}"
+    ]
+
+POSITIVES = POSITIVES_QUALIFIED + POSITIVES_UNQUALIFIED
+
+NEGATIVES_QUALIFIED = [
+    # empty
+    "/", "//",
+    # starts with numeric
+    "/foo/123bar", "/foo/123/bar",
+    # ends with underscore
+    "/foo/_/bar", "/foo_/bar", "/_/bar", "_", "_/_bar"
+    # contains illegal char
+    "foo bar", "foo^bar", "foo!bar",
+    # missing double slashes
+    "rosservice:/foo", "rostopic:/foo/bar", "rostopic:foo"
+    ## bad tilde
+    "~~", "/foo/~/",
+    ]
+
+NEGATIVES_UNQUALIFIED = [
+    # empty
+    "",
+    "123abc", "123",
+    "__foo", "foo__bar", "foo__",
+    "foo_", "foo_/bar",  "foo/_/bar",
+    "foo//bar",
+    "foo/",
+    "{", "}", "}{", "foo{}", "foo{bar{baz}}", "foo{}bar{}baz"
+    "~foo", "foo~", "foo~/bar", "foo/~bar", "/~", "foo/~/bar"
+    ]
+
+NEGATIVES = NEGATIVES_QUALIFIED + NEGATIVES_UNQUALIFIED
+
+
+
+## Short EBNF Grako introduction
+# [] optional element
+# {} repetition (zero or more)
+# repetitions are greedy
+# (* *) comment
+# | alternative  (order matters (for grako at least)), first are tried first)
+# '' token
+# / / regex
+GRAMMAR = """
+
+uriname
+    =
+    uriname_qualified
+    |
+    [ urischeme ] pathname $
+    ;
+
+uriname_qualified
+    =
+    [ urischeme ] pathname_qualified $
+    ;
+
+urischeme
+    =
+    namecharp '://'
+    ;
+
+pathname
+    =
+    ['/' | '~/'] nametoken '/' { nametoken '/' } nametoken
+    |
+    ['/' | '~/'] singlename
+    |
+    '~'
+    ;
+
+pathname_qualified
+    =
+    '/' nametoken_qualified '/' { nametoken_qualified '/' } nametoken_qualified
+    |
+    '/' singlename_qualified
+    ;
+
+(* must not start with digit, must not finish with _ **)
+nametoken
+    =
+    substitution { substitution } [ nametoken_qualified ]
+    |
+    nametoken_qualified
+    ;
+
+(* must not start with digit finish with _ *)
+nametoken_qualified
+    =
+    /[_A-Za-z][_A-Za-z0-9]*[A-Za-z0-9]/
+    |
+    /[A-Za-z]/
+    ;
+
+(* must not start with digit, nor finish with _ *)
+singlename
+    =
+    [ /[_A-Za-z][_A-Za-z0-9]*/ ] substitution { substitution } [ /[_A-Za-z0-9]*[A-Za-z0-9]/ ]
+    |
+    singlename_qualified
+    ;
+
+(* must not start with digit, nor finish with _ *)
+singlename_qualified
+    =
+    /[_A-Za-z][_A-Za-z0-9]*[A-Za-z0-9]/
+    |
+    /[A-Za-z]/
+    ;
+
+substitution
+    =
+    namecharx '{' namecharp '}'
+    ;
+
+namecharp
+    =
+    /[_A-Za-z0-9]+/
+    ;
+
+namecharx
+    =
+    /[_A-Za-z0-9]*/
+    ;
+
+"""
+
+MODEL = grako.genmodel("model", GRAMMAR)
+
+
+def parse_qualified(s):
+    ast = MODEL.parse(s, "uriname_qualified")
+    if '__' in s:
+        raise SyntaxError("Must not contain consecutive underlines: '" + s + "'")
+    return ast
+
+def resolve(s):
+    return re.sub(r'\{[a-zA-Z0-9_]+\}', 'xxx', s)
+
+def parse(s):
+    ast = MODEL.parse(s, "uriname")
+    if '__' in resolve(s):
+        raise SyntaxError("Must not contain consecutive underlines: '" + resolve(s) + "'")
+    return ast
+
+
+def test_main():
+    for s in POSITIVES:
+        print("ACCEPTED: %s \t as %s " % (s, parse(s)))
+    for s in POSITIVES_QUALIFIED:
+        print("ACCEPTED QUALIFIED: %s \t as %s " % (s, parse_qualified(s)))
+
+    print('=' * 40)
+
+    for s in NEGATIVES:
+        try:
+            raise AssertionError("ERROR, SHOULD REFUSE: %s \t\t but was %s" % (s, parse(s)))
+        except (grako.exceptions.FailedParse, SyntaxError):
+            print("REJECTED: " + s)
+
+    for s in NEGATIVES_QUALIFIED + POSITIVES_UNQUALIFIED:
+        try:
+            raise AssertionError("ERROR, SHOULD REFUSE: %s \t\t but was %s" % (s, parse_qualified(s)))
+        except (grako.exceptions.FailedParse, SyntaxError):
+            print("REJECTED QUALIFIED: " + s)
+
+
+if __name__ == '__main__':
+    test_main()