def test_parser_edge_cases():
# Our parser implementation is general enough that
# some of its branches are not being exercised by our regular tests,
# so I had to come up with these contrived examples to test them.
p = many(rfc7230.tchar) > named(u'p')
p1 = '1' * p > named(u'p1')
p2 = '11' * p * skip('\n') > named(u'p2')
assert parse(p1 | p2, b'11abc') == (u'1', [u'1', u'a', u'b', u'c'])
assert parse(p1 | p2, b'11abc\n') == (u'11', [u'a', u'b', u'c'])
p = recursive() > named(u'p')
p.rec = (rfc7230.tchar * p | subst(None) << empty)
assert parse(p, b'abc') == (u'a', (u'b', (u'c', None)))
p = literal('ab') > named(u'p')
p0 = subst(u'') << empty | p > named(u'p0')
p1 = 'xab' * p0 > named(u'p1')
p2 = 'x' * string(p0) * '!' > named(u'p2')
assert parse(p1 | p2, b'xabab') == (u'xab', u'ab')
assert parse(p1 | p2, b'xabab!') == (u'x', u'abab', u'!')
p = empty | literal('a') > named(u'p')
p0 = p * 'x' > named(u'x')
assert parse(p0, b'x') == u'x'
def test_parser_edge_cases():
# Our parser implementation is general enough that
# some of its branches are not being exercised by our regular tests,
# so I had to come up with these contrived examples to test them.
p = many(rfc7230.tchar) > named(u'p')
p1 = '1' * p > named(u'p1')
p2 = '11' * p * skip('\n') > named(u'p2')
assert parse(p1 | p2, b'11abc') == (u'1', [u'1', u'a', u'b', u'c'])
assert parse(p1 | p2, b'11abc\n') == (u'11', [u'a', u'b', u'c'])
p = recursive() > named(u'p')
p.rec = (rfc7230.tchar * p | subst(None) << empty)
assert parse(p, b'abc') == (u'a', (u'b', (u'c', None)))
p = literal('ab') > named(u'p')
p0 = subst(u'') << empty | p > named(u'p0')
p1 = 'xab' * p0 > named(u'p1')
p2 = 'x' * string(p0) * '!' > named(u'p2')
assert parse(p1 | p2, b'xabab') == (u'xab', u'ab')
assert parse(p1 | p2, b'xabab!') == (u'x', u'abab', u'!')
p = empty | literal('a') > named(u'p')
p0 = p * 'x' > named(u'x')
assert parse(p0, b'x') == u'x'
can_complain,
fill_names,
maybe,
octet,
octet_range,
pivot,
string,
subst,
)
from httpolice.structure import EntityTag
from httpolice.syntax.common import DQUOTE
from httpolice.syntax.rfc7230 import comma_list1, obs_text
from httpolice.syntax.rfc7231 import HTTP_date
weak = subst(True) << octet(0x57) * octet(0x2F) > auto
etagc = octet(0x21) | octet_range(0x23, 0x7E) | obs_text > auto
@can_complain
def _no_backslashes(complain, s):
if u'\\' in s:
complain(1119)
return s
opaque_tag = _no_backslashes << DQUOTE + string(etagc) + DQUOTE > auto
entity_tag = EntityTag << maybe(weak, False) * opaque_tag > pivot
ETag = entity_tag > pivot
Last_Modified = HTTP_date > pivot
If_Match = '*' | comma_list1(entity_tag) > pivot
# -*- coding: utf-8; -*-
from httpolice.citation import RFC
from httpolice.parse import (auto, can_complain, fill_names, maybe, octet,
octet_range, pivot, string, subst)
from httpolice.structure import EntityTag
from httpolice.syntax.common import DQUOTE
from httpolice.syntax.rfc7230 import comma_list1, obs_text
from httpolice.syntax.rfc7231 import HTTP_date
weak = subst(True) << octet(0x57) * octet(0x2F) > auto
etagc = octet(0x21) | octet_range(0x23, 0x7E) | obs_text > auto
@can_complain
def _no_backslashes(complain, s):
if u'\\' in s:
complain(1119)
return s
opaque_tag = _no_backslashes << DQUOTE + string(etagc) + DQUOTE > auto
entity_tag = EntityTag << maybe(weak, False) * opaque_tag > pivot
ETag = entity_tag > pivot
Last_Modified = HTTP_date > pivot
If_Match = '*' | comma_list1(entity_tag) > pivot
If_None_Match = '*' | comma_list1(entity_tag) > pivot
If_Modified_Since = HTTP_date > pivot
If_Unmodified_Since = HTTP_date > pivot
# -*- coding: utf-8; -*-
from httpolice.parse import (fill_names, literal, maybe, pivot, skip, string1,
subst)
from httpolice.syntax.common import DIGIT
from httpolice.syntax.rfc7230 import RWS, comma_list
notice_id = int << string1(DIGIT) > pivot
resp = subst(True) << literal('resp') > pivot
HTTPolice_Silence = comma_list(notice_id * maybe(skip(RWS) * resp)) > pivot
fill_names(globals(), citation=None)
(lambda x: (None, x)) << skip('-') * suffix_length > pivot
byte_range_set = comma_list1(byte_range_spec | suffix_byte_range_spec) > auto
byte_ranges_specifier = RangeSpecifier << (bytes_unit * skip('=') *
byte_range_set) > pivot
other_range_set = string1(VCHAR) > auto
other_ranges_specifier = RangeSpecifier << (other_range_unit * skip('=') *
other_range_set) > pivot
Range = byte_ranges_specifier | other_ranges_specifier > pivot
byte_range = first_byte_pos * skip('-') * last_byte_pos > auto
complete_length = int << string1(DIGIT) > auto
byte_range_resp = (byte_range * skip('/') *
(complete_length | subst(None) << literal('*'))) > pivot
unsatisfied_range = ((subst(None) << literal('*/')) * complete_length) > pivot
@can_complain
def _well_formed2(complain, r):
bounds, complete = r.range
if bounds is not None:
first, last = bounds
if (last < first) or ((complete is not None) and (complete <= last)):
complain(1148)
return r
byte_content_range = _well_formed2 << (
subtype = token > pivot
media_type = Parametrized << (
(_check_media_type << (MediaType << type_ + '/' + subtype)) *
(MultiDict << many(skip(OWS * ';' * OWS) * parameter()))) > pivot
content_coding = ContentCoding << token > pivot
product_version = token > pivot
product = Versioned << (
(ProductName << token) * maybe(skip('/') * product_version)) > pivot
User_Agent = product % many(
skip(RWS) * (product | comment(include_parens=False))) > pivot
Server = product % many(skip(RWS) *
(product | comment(include_parens=False))) > pivot
day_name = (subst(0) << octet(0x4D) * octet(0x6F) * octet(0x6E)
| subst(1) << octet(0x54) * octet(0x75) * octet(0x65)
| subst(2) << octet(0x57) * octet(0x65) * octet(0x64)
| subst(3) << octet(0x54) * octet(0x68) * octet(0x75)
| subst(4) << octet(0x46) * octet(0x72) * octet(0x69)
| subst(5) << octet(0x53) * octet(0x61) * octet(0x74)
| subst(6) << octet(0x53) * octet(0x75) * octet(0x6E)) > pivot
@can_complain
def _to_date(complain, d, m, y):
try:
return date(y, m, d)
except ValueError:
complain(1222, date=u'%d-%02d-%02d' % (y, m, d))
return Unavailable
# -*- coding: utf-8; -*-
from httpolice.citation import RFC
from httpolice.parse import (auto, can_complain, fill_names, literal, maybe,
pivot, skip, string, string1, subst)
from httpolice.structure import ContentRange, RangeSpecifier, RangeUnit
from httpolice.syntax.common import CHAR, DIGIT, SP, VCHAR
from httpolice.syntax.rfc7230 import comma_list1, token__excluding
from httpolice.syntax.rfc7231 import HTTP_date
from httpolice.syntax.rfc7232 import entity_tag
bytes_unit = RangeUnit << literal('bytes') > auto
other_range_unit = RangeUnit << token__excluding(['bytes']) > auto
range_unit = bytes_unit | other_range_unit > pivot
acceptable_ranges = (subst([]) << literal('none')
| comma_list1(range_unit)) > pivot
Accept_Ranges = acceptable_ranges > pivot
@can_complain
def _well_formed1(complain, first, last):
if (last is not None) and (first > last):
complain(1133)
return (first, last)
first_byte_pos = int << string1(DIGIT) > auto
last_byte_pos = int << string1(DIGIT) > auto
byte_range_spec = _well_formed1 << (first_byte_pos * skip('-') *
maybe(last_byte_pos)) > pivot
def comma_list1(element):
return _collect_elements << (many(subst(None) << ',' * OWS) + (
(lambda x: [x]) << group(element)) + many(
skip(OWS * ',') * maybe(skip(OWS) * element))) > named(
u'1#rule', RFC(7230, section=u'7'))
# As updated by RFC 6874
ZoneID = string1(unreserved | pct_encoded) > pivot
IPv6addrz = IPv6address + '%25' + ZoneID > pivot
IP_literal = '[' + (IPv6address | IPv6addrz | IPvFuture) + ']' > pivot
reg_name = string(unreserved | sub_delims | pct_encoded) > pivot
host = IP_literal | IPv4address | reg_name > pivot
port = string(DIGIT) > pivot
authority = maybe_str(userinfo + '@') + host + maybe_str(':' + port) > pivot
path_abempty = string('/' + segment) > auto
path_absolute = '/' + maybe_str(segment_nz + string('/' + segment)) > auto
path_noscheme = segment_nz_nc + string('/' + segment) > auto
path_rootless = segment_nz + string('/' + segment) > auto
path_empty = subst(u'') << empty > auto
hier_part = ('//' + authority + path_abempty | path_absolute | path_rootless
| path_empty) > pivot
query = string(pchar | '/' | '?') > pivot
fragment = string(pchar | '/' | '?') > pivot
absolute_URI = scheme + ':' + hier_part + maybe_str('?' + query) > pivot
relative_part = ('//' + authority + path_abempty | path_absolute
| path_noscheme | path_empty) > pivot
URI = (scheme + ':' + hier_part + maybe_str('?' + query) +
maybe_str('#' + fragment)) > pivot
from httpolice.citation import RFC
from httpolice.parse import (auto, can_complain, fill_names, literal, maybe,
pivot, skip, string, string1, subst)
from httpolice.structure import ContentRange, RangeSpecifier, RangeUnit
from httpolice.syntax.common import CHAR, DIGIT, SP, VCHAR
from httpolice.syntax.rfc7230 import comma_list1, token__excluding
from httpolice.syntax.rfc7231 import HTTP_date
from httpolice.syntax.rfc7232 import entity_tag
bytes_unit = RangeUnit << literal('bytes') > auto
other_range_unit = RangeUnit << token__excluding(['bytes']) > auto
range_unit = bytes_unit | other_range_unit > pivot
acceptable_ranges = (
subst([]) << literal('none') |
comma_list1(range_unit)) > pivot
Accept_Ranges = acceptable_ranges > pivot
@can_complain
def _well_formed1(complain, first, last):
if (last is not None) and (first > last):
complain(1133)
return (first, last)
first_byte_pos = int << string1(DIGIT) > auto
last_byte_pos = int << string1(DIGIT) > auto
byte_range_spec = _well_formed1 << (first_byte_pos * skip('-') *
maybe(last_byte_pos)) > pivot
suffix_length = int << string1(DIGIT) > auto
def comma_list(element):
return _collect_elements << maybe(
(subst([None, None]) << literal(',') |
(lambda x: [x]) << group(element)) +
many(skip(OWS * ',') * maybe(skip(OWS) * element))) > named(
u'#rule', RFC(7230, section=(7, )))
# -*- coding: utf-8; -*-
from httpolice.parse import (
fill_names,
literal,
maybe,
pivot,
skip,
string1,
subst,
)
from httpolice.syntax.common import DIGIT
from httpolice.syntax.rfc7230 import RWS, comma_list
notice_id = int << string1(DIGIT) > pivot
resp = subst(True) << literal('resp') > pivot
HTTPolice_Silence = comma_list(notice_id * maybe(skip(RWS) * resp)) > pivot
fill_names(globals(), citation=None)
def comma_list(element):
return _collect_elements << maybe(
(subst([None, None]) << literal(',') |
(lambda x: [x]) << group(element)) +
many(skip(OWS * ',') * maybe(skip(OWS) * element))
) > named(u'#rule', RFC(7230, section=(7,)))
correct_encoded_id += pct_encode(c, safe='').upper()
if encoded_id != correct_encoded_id:
complain(1256, actual=encoded_id, correct=correct_encoded_id)
return decoded_id
protocol_id = _check_protocol_id << token > pivot
@can_complain
def _check_alt_authority(complain, value):
return parse(value, maybe_str(uri_host) + ':' + port, complain, 1257,
authority=value)
alt_authority = _check_alt_authority << quoted_string > pivot
alternative = protocol_id * skip('=') * alt_authority > pivot
parameter = ((AltSvcParam << token) *
skip('=') * (token | quoted_string)) > pivot
alt_value = Parametrized << (
alternative *
(MultiDict << many(skip(OWS * ';' * OWS) * parameter))) > pivot
Alt_Svc = clear | comma_list1(alt_value) > pivot
ma = delta_seconds > pivot
persist = subst(True) << literal('1') > pivot
Alt_Used = uri_host + maybe_str(':' + port) > pivot
fill_names(globals(), RFC(7838))
def comma_list1(element):
return _collect_elements << (
many(subst(None) << ',' * OWS) +
((lambda x: [x]) << group(element)) +
many(skip(OWS * ',') * maybe(skip(OWS) * element))
) > named(u'1#rule', RFC(7230, section=(7,)))
protocol_id = _check_protocol_id << token > pivot
@can_complain
def _check_alt_authority(complain, value):
return parse(value,
maybe_str(uri_host) + ':' + port,
complain,
1257,
authority=value)
alt_authority = _check_alt_authority << quoted_string > pivot
alternative = protocol_id * skip('=') * alt_authority > pivot
parameter = ((AltSvcParam << token) * skip('=') *
(token | quoted_string)) > pivot
alt_value = Parametrized << (
alternative *
(MultiDict << many(skip(OWS * ';' * OWS) * parameter))) > pivot
Alt_Svc = clear | comma_list1(alt_value) > pivot
ma = delta_seconds > pivot
persist = subst(True) << literal('1') > pivot
Alt_Used = uri_host + maybe_str(':' + port) > pivot
fill_names(globals(), RFC(7838))
subtype = token > pivot
media_type = Parametrized << (
(MediaType << type_ + '/' + subtype) *
(MultiDict << many(skip(OWS * ';' * OWS) * parameter()))) > pivot
content_coding = ContentCoding << token > pivot
product_version = token > pivot
product = Versioned << ((ProductName << token) *
maybe(skip('/') * product_version)) > pivot
User_Agent = product % many(skip(RWS) *
(product | comment(include_parens=False))) > pivot
Server = product % many(skip(RWS) *
(product | comment(include_parens=False))) > pivot
day_name = (subst(0) << octet(0x4D) * octet(0x6F) * octet(0x6E) |
subst(1) << octet(0x54) * octet(0x75) * octet(0x65) |
subst(2) << octet(0x57) * octet(0x65) * octet(0x64) |
subst(3) << octet(0x54) * octet(0x68) * octet(0x75) |
subst(4) << octet(0x46) * octet(0x72) * octet(0x69) |
subst(5) << octet(0x53) * octet(0x61) * octet(0x74) |
subst(6) << octet(0x53) * octet(0x75) * octet(0x6E)) > pivot
@can_complain
def _to_date(complain, d, m, y):
try:
return date(y, m, d)
except ValueError:
complain(1222, date=u'%d-%02d-%02d' % (y, m, d))
return Unavailable
# As updated by RFC 6874
ZoneID = string1(unreserved | pct_encoded) > pivot
IPv6addrz = IPv6address + '%25' + ZoneID > pivot
IP_literal = '[' + (IPv6address | IPv6addrz | IPvFuture) + ']' > pivot
reg_name = string(unreserved | sub_delims | pct_encoded) > pivot
host = IP_literal | IPv4address | reg_name > pivot
port = string(DIGIT) > pivot
authority = maybe_str(userinfo + '@') + host + maybe_str(':' + port) > pivot
path_abempty = string('/' + segment) > auto
path_absolute = '/' + maybe_str(segment_nz + string('/' + segment)) > auto
path_noscheme = segment_nz_nc + string('/' + segment) > auto
path_rootless = segment_nz + string('/' + segment) > auto
path_empty = subst(u'') << empty > auto
hier_part = ('//' + authority + path_abempty |
path_absolute | path_rootless | path_empty) > pivot
query = string(pchar | '/' | '?') > pivot
fragment = string(pchar | '/' | '?') > pivot
absolute_URI = scheme + ':' + hier_part + maybe_str('?' + query) > pivot
relative_part = ('//' + authority + path_abempty |
path_absolute | path_noscheme | path_empty) > pivot
URI = (scheme + ':' + hier_part +
maybe_str('?' + query) + maybe_str('#' + fragment)) > pivot
