Doctests for the fundamental element classes

Test fixture initialization:

>>> from dragonfly import *
>>> from dragonfly.test import ElementTester

Alternative element class

Usage:

>>> alt = Alternative([Literal("hello"), Literal("goodbye")])
>>> test_alt = ElementTester(alt)
>>> test_alt.recognize("hello")
u'hello'
>>> test_alt.recognize("goodbye")
u'goodbye'
>>> test_alt.recognize("hi")
RecognitionFailure

Literal element class

Usage:

>>> literal = Literal("hello")
>>> test_literal = ElementTester(literal)
>>> test_literal.recognize("hello")
u'hello'
>>> test_literal.recognize("goodbye")
RecognitionFailure

Literal with non-ASCII characters:

>>> # We do not test the confusing Python 3 "bytes" type here.
>>> string = "touch\xe9"
>>> literal = Literal(string)
>>> test_literal = ElementTester(literal)
>>> test_literal.recognize(string)
u'touch\xe9'

>>> string = u"touch\xe9"
>>> literal = Literal(string)
>>> test_literal = ElementTester(literal)
>>> test_literal.recognize(string)
u'touch\xe9'

Quoted words usage:

>>> # Quoted words are not joined in the 'words' property.
>>> # Also, double quotes are not present.
>>> literal = Literal('this is a "quoted string" example')
>>> literal.words
['this', 'is', 'a', 'quoted', 'string', 'example']
>>> # The 'words_ext' property shows quoted words as single items.
>>> # Double quotes are also not present.
>>> literal.words_ext
['this', 'is', 'a', 'quoted string', 'example']
>>> # Rules with quoted words can be recognized.
>>> test_literal = ElementTester(literal)
>>> test_literal.recognize('this is a quoted string example')
u'this is a quoted string example'

Incomplete quotes:

>>> # Incomplete quotes are left in.
>>> literal = Literal('"quote')
>>> literal.words
['"quote']
>>> literal.words_ext
['"quote']
>>> literal = Literal('a "quoted string" example plus "quote')
>>> literal.words
['a', 'quoted', 'string', 'example', 'plus', '"quote']
>>> literal.words_ext
['a', 'quoted string', 'example', 'plus', '"quote']

Non-default quote constructor arguments:

>>> literal = Literal(u'this is a «quoted string» example',
...                   quote_start_str=u'«', quote_end_str=u'»')
>>> literal.words
['this', 'is', 'a', 'quoted', 'string', 'example']
>>> literal.words_ext
['this', 'is', 'a', 'quoted string', 'example']
>>> test_literal = ElementTester(literal)
>>> test_literal.recognize('this is a quoted string example')
u'this is a quoted string example'

>>> # Quotes are left in if specified.
>>> literal = Literal('"quoted string"', strip_quote_strs=False)
>>> literal.words
['"quoted', 'string"']
>>> literal.words_ext
['"quoted string"']

Optional element class

Usage:

>>> seq = Sequence([Literal("hello"), Optional(Literal("there"))])
>>> test_seq = ElementTester(seq)
>>> # Optional parts of the sequence can be left out.
>>> test_seq.recognize("hello")
[u'hello', None]
>>> test_seq.recognize("hello there")
[u'hello', u'there']
>>> test_seq.recognize("goodbye")
RecognitionFailure

Sequence element class

Basic usage:

>>> seq = Sequence([Literal("hello"), Literal("world")])
>>> test_seq = ElementTester(seq)
>>> test_seq.recognize("hello world")
[u'hello', u'world']
>>> test_seq.recognize("hello universe")
RecognitionFailure

Constructor arguments:

>>> c1, c2 = Literal("hello"), Literal("world")
>>> len(Sequence(children=[c1, c2]).children)
2
>>> Sequence(children=[c1, c2], name="sequence_test").name
'sequence_test'
>>> Sequence([c1, c2], "sequence_test").name
'sequence_test'
>>> Sequence("invalid_children_type")
Traceback (most recent call last):
  ...
TypeError: children object must contain only <class 'dragonfly.grammar.elements_basic.ElementBase'> types.  (Received ('i', 'n', 'v', 'a', 'l', 'i', 'd', '_', 'c', 'h', 'i', 'l', 'd', 'r', 'e', 'n', '_', 't', 'y', 'p', 'e'))

Repetition element class

Basic usage:

>>> # Repetition is given a dragonfly element, in this case a Sequence.
>>> seq = Sequence([Literal("hello"), Literal("world")])
>>> # Specify min and max values to allow more than one repetition.
>>> rep = Repetition(seq, min=1, max=16, optimize=False)
>>> test_rep = ElementTester(rep)
>>> test_rep.recognize("hello world")
[[u'hello', u'world']]
>>> test_rep.recognize("hello world hello world")
[[u'hello', u'world'], [u'hello', u'world']]
>>> # Incomplete recognitions result in recognition failure.
>>> test_rep.recognize("hello universe")
RecognitionFailure
>>> test_rep.recognize("hello world hello universe")
RecognitionFailure
>>> # Too many recognitions also result in recognition failure.
>>> test_rep.recognize(" ".join(["hello world"] * 17))
RecognitionFailure
>>> # Using the 'optimize' argument:
rep = Repetition(seq, min=1, max=16, optimize=True)
>>> test_rep = ElementTester(rep)
>>> test_rep.recognize("hello world")
[[u'hello', u'world']]
>>> test_rep.recognize("hello world hello world")
[[u'hello', u'world'], [u'hello', u'world']]

Exact number of repetitions:

>>> seq = Sequence([Literal("hello"), Literal("world")])
>>> rep = Repetition(seq, min=3, max=None, optimize=False)
>>> test_rep = ElementTester(rep)
>>> test_rep.recognize("hello world")
RecognitionFailure
>>> test_rep.recognize("hello world hello world")
RecognitionFailure
>>> test_rep.recognize("hello world hello world hello world")
[[u'hello', u'world'], [u'hello', u'world'], [u'hello', u'world']]
>>> test_rep.recognize("hello world hello world hello world hello world")
RecognitionFailure

min must be less than max:

>>> rep = Repetition(Literal("hello"), min=3, max=3, optimize=False)
Traceback (most recent call last):
  ...
AssertionError: min must be less than max

Modifier element class

Basic usage:

>>> # Repetition is given a dragonfly element, in this case an IntegerRef.
>>> i = IntegerRef("n", 1, 10)
>>> # Specify min and max values to allow more than one repetition.
>>> rep = Repetition(i, min=1, max=16, optimize=False)
>>> # Modify the element to format the output
>>> mod = Modifier(rep, lambda rep: ", ".join(map(str, rep)))
>>> test_rep = ElementTester(mod)
>>> test_rep.recognize("one two three four")
'1, 2, 3, 4'

RuleRef element class

Basic usage:

>>> # Define a simple private CompoundRule and reference it.
>>> greet = CompoundRule(name="greet", spec="greetings", exported=False)
>>> ref = RuleRef(rule=greet, name="greet")
>>> test_ref = ElementTester(ref)
>>> test_ref.recognize("greetings")
u'greetings'
>>> test_ref.recognize("hello")
RecognitionFailure

Empty element class

Usage:

>>> empty = Empty()
>>> test_empty = ElementTester(empty)
>>> test_empty.recognize("hello")
RecognitionFailure
>>> empty_seq = Sequence([Literal("hello"), Empty(),
...                       Literal("goodbye")])
>>> test_empty = ElementTester(empty_seq)
>>> test_empty.recognize("hello goodbye")
[u'hello', True, u'goodbye']
>>> test_empty.recognize("hello empty goodbye")
RecognitionFailure
>>> empty_seq = Sequence([Literal("hello"),
...                       Alternative([Literal("gorgeous"),
...                                    Empty()])])
>>> test_empty = ElementTester(empty_seq)
>>> test_empty.recognize("hello")
[u'hello', True]
>>> test_empty.recognize("hello gorgeous")
[u'hello', u'gorgeous']

Impossible element class

Usage:

>>> impossible = Impossible()
>>> test_impossible = ElementTester(impossible)
>>> test_impossible.recognize("hello")
RecognitionFailure
>>> test_impossible.recognize("")
RecognitionFailure
>>> impossible_seq = Sequence([Literal("hello"), Impossible(),
...                       Literal("goodbye")])
>>> test_impossible = ElementTester(impossible_seq)
>>> test_impossible.recognize("hello goodbye")
RecognitionFailure
>>> test_impossible.recognize("hello empty goodbye")
RecognitionFailure