- clean up

- renamed safebuf to streambuf
- added arrays
This commit is contained in:
ari committed 2020-10-17 15:54:39 -04:00
1 parent c752346638
commit a03857d0f7
29 files changed
+280 -1381

No files matched your search

+145 -7
View File
@@ -1,21 +1,27 @@
from decimal import Decimal
import io
from decimal import Decimal
from typing import List
from unittest import TestCase
from pdfmajor.lexer import PDFLexer
from pdfmajor.lexer.exceptions import LexerError
from pdfmajor.lexer.number import parse_number
from pdfmajor.lexer.dict_and_hex import parse_double_angled_bracket, parse_hexstring
from pdfmajor.lexer.string import parse_string
from unittest import TestCase
from pdfmajor.lexer.name import parse_name
from pdfmajor.lexer.keyword import parse_keyword
from pdfmajor.safebufiterator import SafeBufferIt
from pdfmajor.safebufiterator import BufferStream
from pdfmajor.lexer.token import (
TDictVaue,
TDictValue,
Token,
TokenBoolean,
TokenComment,
TokenDecimal,
TokenDictionary,
TokenHexString,
TokenInteger,
TokenKeyword,
TokenName,
TokenString,
)
@@ -23,7 +29,7 @@ from pdfmajor.lexer.comment import parse_comment
def make_stream_iter(data: bytes, buf_size: int):
return SafeBufferIt(io.BytesIO(data), buf_size)
return BufferStream(io.BytesIO(data), buf_size)
class Basics(TestCase):
@@ -156,7 +162,7 @@ class Basics(TestCase):
token = parse_double_angled_bracket(buffer)
self.assertIsInstance(token, TokenDictionary)
self.assertEqual(token.start_loc, -1)
self.assertEqual(token.value, TDictVaue.OPEN)
self.assertEqual(token.value, TDictValue.OPEN)
self.assertEqual(token.end_loc, 2)
def test_parse_dd_to_hex(self):
@@ -190,3 +196,135 @@ class Basics(TestCase):
self.assertEqual(token.start_loc, -1)
self.assertEqual(token.value, 130)
self.assertEqual(token.end_loc, 3)
class Lexer(TestCase):
def run_test(self, raw: bytes, expected: List[Token]):
tokenizer = PDFLexer(io.BytesIO(raw))
for i, (etoken, token) in enumerate(zip(expected, tokenizer.iter_tokens())):
self.assertEqual(token, etoken, f"Failed at example {i}")
try:
next_token = next(tokenizer.iter_tokens())
self.assertIsNone(next_token)
except LexerError as e:
self.assertIsNotNone(e)
def test_case1(self):
self.run_test(
raw=br"""%!PS
begin end
" @ #
/a/BCD /Some_Name /foo#5f#6eaa
0 +1 -2 .5 1.234
(abc) () (abc ( def ) ghi)
(def\040\0\0404ghi) (bach\\slask) (foo\nbaa)
(this % is not a comment.)
(foo
baa)
(foo\
baa)
<> <20> < 40 4020 >
<abcd00
12345>
func/a/b{(c)do*}def
[ 1 (z) ! ]
<< /foo (bar) >>""",
expected=[
TokenComment(start_loc=0, end_loc=4, value=b"!PS"),
TokenKeyword(start_loc=5, end_loc=5, value=b"begin"),
TokenKeyword(start_loc=11, end_loc=3, value=b"end"),
TokenKeyword(start_loc=16, end_loc=1, value=b'"'),
TokenKeyword(start_loc=19, end_loc=1, value=b"@"),
TokenKeyword(start_loc=21, end_loc=1, value=b"#"),
TokenName(start_loc=24, end_loc=1, value="a"),
TokenName(start_loc=26, end_loc=3, value="BCD"),
TokenName(start_loc=31, end_loc=9, value="Some_Name"),
TokenName(start_loc=42, end_loc=11, value="foo_naa"),
TokenInteger(start_loc=54, end_loc=1, value=0),
TokenInteger(start_loc=56, end_loc=2, value=1),
TokenInteger(start_loc=59, end_loc=2, value=-2),
TokenDecimal(start_loc=62, end_loc=2, value=Decimal("0.5")),
TokenDecimal(start_loc=65, end_loc=5, value=Decimal("1.234")),
TokenString(start_loc=71, end_loc=5, value="abc"),
TokenString(start_loc=77, end_loc=2, value=""),
TokenString(start_loc=80, end_loc=17, value="abc ( def ) ghi"),
TokenString(start_loc=98, end_loc=19, value="def 4ghi"),
TokenString(start_loc=118, end_loc=13, value=r"bach\slask"),
TokenString(start_loc=132, end_loc=10, value="foo\nbaa"),
TokenString(
start_loc=143, end_loc=26, value="this % is not a comment."
),
TokenString(start_loc=170, end_loc=9, value="foo\nbaa"),
TokenString(start_loc=180, end_loc=10, value="foobaa"),
TokenHexString(start_loc=191, end_loc=2, value=b""),
TokenHexString(start_loc=194, end_loc=4, value=b" "),
TokenHexString(start_loc=199, end_loc=11, value=b"@@ "),
TokenHexString(
start_loc=211, end_loc=14, value=b"\xab\xcd\x00\x124\x05"
),
TokenKeyword(start_loc=226, end_loc=4, value=b"func"),
TokenName(start_loc=231, end_loc=1, value="a"),
TokenName(start_loc=233, end_loc=1, value="b"),
TokenKeyword(start_loc=234, end_loc=1, value=b"{"),
],
)
def test_case2(self):
self.run_test(
raw=br"""10.32
0 +1 -2 .5 1.234
-1.234 990 """,
expected=[
TokenDecimal(start_loc=0, end_loc=5, value=Decimal("10.32")),
TokenInteger(start_loc=6, end_loc=1, value=0),
TokenInteger(start_loc=8, end_loc=2, value=1),
TokenInteger(start_loc=11, end_loc=2, value=-2),
TokenDecimal(start_loc=14, end_loc=2, value=Decimal("0.5")),
TokenDecimal(start_loc=17, end_loc=5, value=Decimal("1.234")),
TokenDecimal(start_loc=23, end_loc=6, value=Decimal("-1.234")),
TokenInteger(start_loc=30, end_loc=3, value=990),
],
)
def test_case3(self):
self.run_test(
raw=br"""/b{(c)do*}def""",
expected=[
TokenName(start_loc=0, end_loc=2, value="b"),
TokenKeyword(start_loc=2, end_loc=3, value=b"{"),
TokenString(start_loc=4, end_loc=5, value="c"),
],
)
def test_case4(self):
self.run_test(
raw=br"""%!PS
% This is some secret sauce
begin end
" @ #
/a/BCD /Some_Name /foo#5f#6eaa /bar#6a
/need /oh1oh
func/a/b
""",
expected=[
TokenComment(start_loc=0, end_loc=4, value=b"!PS"),
TokenComment(
start_loc=5, end_loc=27, value=b" This is some secret sauce"
),
TokenKeyword(start_loc=33, end_loc=5, value=b"begin"),
TokenKeyword(start_loc=39, end_loc=3, value=b"end"),
TokenKeyword(start_loc=44, end_loc=1, value=b'"'),
TokenKeyword(start_loc=47, end_loc=1, value=b"@"),
TokenKeyword(start_loc=49, end_loc=1, value=b"#"),
TokenName(start_loc=51, end_loc=2, value="a"),
TokenName(start_loc=53, end_loc=4, value="BCD"),
TokenName(start_loc=58, end_loc=10, value="Some_Name"),
TokenName(start_loc=69, end_loc=12, value="foo_naa"),
TokenName(start_loc=83, end_loc=7, value="barj"),
TokenName(start_loc=90, end_loc=5, value="need"),
TokenName(start_loc=96, end_loc=6, value="oh1oh"),
TokenKeyword(start_loc=103, end_loc=4, value=b"func"),
TokenName(start_loc=107, end_loc=3, value="a"),
TokenName(start_loc=109, end_loc=3, value="b"),
],
)
+4 -4
View File
@@ -1,12 +1,12 @@
import io
from unittest import TestCase
from pdfmajor.safebufiterator import EOF, SafeBufferIt
from pdfmajor.safebufiterator import EOF, BufferStream
class UseCase(TestCase):
def test_safebuff(self):
for bufsize in [2, 33, 65, 66, 100, 200]:
it = SafeBufferIt(
it = BufferStream(
io.BytesIO(
b"this is a lengthy comment that ends here\nso this is not reachable"
),
@@ -16,7 +16,7 @@ class UseCase(TestCase):
self.assertLessEqual(len(subbuf), it.buffer_size)
def test_itraises(self):
it = SafeBufferIt(io.BytesIO(b"this"))
it = BufferStream(io.BytesIO(b"this"))
list(it) # read everything
with self.assertRaises(EOF):
for _ in it:
@@ -24,7 +24,7 @@ class UseCase(TestCase):
def test_seekd(self):
for bufsize in [2, 33, 65, 66, 100, 200]:
it = SafeBufferIt(
it = BufferStream(
io.BytesIO(
b"this is a lengthy comment that ends here\nso this is not reachable"
),
-230
View File
@@ -1,230 +0,0 @@
import io
from random import randint
from decimal import Decimal
from typing import List
from pdfmajor.tokenizer.token_parsers.string import parse_string
from pdfmajor.tokenizer import PSTokenizer
from pdfmajor.tokenizer.token import (
Token,
TokenBoolean,
TokenComment,
TokenDecimal,
TokenInteger,
TokenKeyword,
TokenLiteral,
TokenString,
)
from pdfmajor.tokenizer.token_parsers.util import PInput, SafeBufferIt
from pdfmajor.tokenizer.token_parsers.comment import parse_comment
from pdfmajor.tokenizer.token_parsers.keyword import parse_keyword
from pdfmajor.tokenizer.token_parsers.number import parse_number
from pdfmajor.tokenizer.token_parsers.literal import parse_literal
from unittest import TestCase
def make_stream_iter(data: bytes, initpos: int = 0):
stream = io.BytesIO(data)
def iter_stream(buf_size: int):
while True:
bufpos = stream.tell()
buf = stream.read(buf_size)
if buf:
yield PInput(initpos + bufpos, buf)
else:
raise "EOF"
return iter_stream
class Util(TestCase):
def test_safebuff(self):
it = SafeBufferIt(
b"this is a lengthy comment that ends here\nso this is not reachable"
)
for subbuf in it.into_iter():
self.assertEqual(len(subbuf), len(it.buffer) - it.skip)
it.skip += randint(1, max(len(subbuf) - 1, 1))
class Individual(TestCase):
def test_comment_parser_ln(self):
for buf_size in [2, 7, 11, 150]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(
b"this is a lengthy comment that ends here\nso this is not reachable"
)
token = parse_comment(0, iter_stream(buf_size))
self.assertIsInstance(token, TokenComment)
self.assertEqual(token.pos, 0)
self.assertEqual(
token.value, b"this is a lengthy comment that ends here"
)
self.assertEqual(token.size, 40)
def test_comment_parser_lr(self):
for buf_size in [2, 7, 11, 150]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(
b"this is a lengthy comment that ends here\rso this is not reachable"
)
token = parse_comment(0, iter_stream(buf_size))
self.assertIsInstance(token, TokenComment)
self.assertEqual(token.pos, 0)
self.assertEqual(
token.value, b"this is a lengthy comment that ends here"
)
self.assertEqual(token.size, 40)
def test_parse_float(self):
for buf_size in [2, 3, 50]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(b"0.120 + some invalid text", 5)
token = parse_number(5, iter_stream(buf_size))
self.assertIsInstance(token, TokenDecimal)
self.assertEqual(token.pos, 5)
self.assertEqual(token.value, Decimal("0.12"))
self.assertEqual(token.size, 5)
def test_parse_int(self):
for buf_size in [2, 3, 50]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(
b"130 oh oh 0.12 + some invalid text", 11
)
token = parse_number(11, iter_stream(buf_size))
self.assertIsInstance(token, TokenInteger)
self.assertEqual(token.pos, 11)
self.assertEqual(token.value, 130)
self.assertEqual(token.size, 3)
def test_parse_bool_true(self):
for buf_size in [2, 3, 50]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(b"true and the american way", 91)
token = parse_keyword(91, iter_stream(buf_size))
self.assertIsInstance(token, TokenBoolean)
self.assertEqual(token.pos, 91)
self.assertEqual(token.value, True)
self.assertEqual(token.size, 4)
def test_parse_bool_false(self):
for buf_size in [2, 3, 50]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(b"false mannn", 33)
token = parse_keyword(33, iter_stream(buf_size))
self.assertIsInstance(token, TokenBoolean)
self.assertEqual(token.pos, 33)
self.assertEqual(token.value, False)
self.assertEqual(token.size, 5)
def test_parse_literals(self):
for buf_size in [2, 3, 50, 444]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(br"Some_Name /foo#5f#xbaa", 71)
token = parse_literal(71, iter_stream(buf_size))
self.assertIsInstance(token, TokenLiteral)
self.assertEqual(token.pos, 71)
self.assertEqual(token.value, "Some_Name")
self.assertEqual(token.size, 9)
def test_parse_literals_hex(self):
for buf_size in [2]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(br"foo#5faa ", 71)
token = parse_literal(71, iter_stream(buf_size))
self.assertIsInstance(token, TokenLiteral)
self.assertEqual(token.pos, 71)
self.assertEqual(token.value, "foo_aa")
self.assertEqual(token.size, 8)
def test_parse_string(self):
for buf_size in [2]:
with self.subTest(buf_size=buf_size):
iter_stream = make_stream_iter(br"this % is not a comment.) ", 88)
token = parse_string(88, iter_stream(buf_size))
self.assertIsInstance(token, TokenString)
self.assertEqual(token.pos, 88)
self.assertEqual(token.value, "this % is not a comment.")
self.assertEqual(token.size, 24)
class Tokenizer(TestCase):
def run_test(self, raw: bytes, expected: List[Token]):
tokenizer = PSTokenizer(io.BytesIO(raw))
for i, (etoken, token) in enumerate(zip(expected, tokenizer.iter_tokens())):
self.assertEqual(token, etoken, f"Failed at example {i}")
def test_case1(self):
tokenizer = PSTokenizer(
io.BytesIO(
br"""%!PS
begin end
" @ #
/a/BCD /Some_Name /foo#5f#6eaa
0 +1 -2 .5 1.234
(abc) () (abc ( def ) ghi)
(def\040\0\0404ghi) (bach\\slask) (foo\nbaa)
(this % is not a comment.)
(foo
baa)
(foo\
baa)
<> <20> < 40 4020 >
<abcd00
12345>
func/a/b{(c)do*}def
[ 1 (z) ! ]
<< /foo (bar) >>"""
)
)
for token in tokenizer.iter_tokens():
print(token)
def test_case2(self):
self.run_test(
raw=br"""10.32
0 +1 -2 .5 1.234
-1.234 990 """,
expected=[
TokenDecimal(pos=0, size=5, value=Decimal("10.32")),
TokenInteger(pos=6, size=1, value=0),
TokenInteger(pos=8, size=2, value=1),
TokenInteger(pos=11, size=2, value=-2),
TokenDecimal(pos=14, size=2, value=Decimal("0.5")),
TokenDecimal(pos=17, size=5, value=Decimal("1.234")),
TokenDecimal(pos=23, size=6, value=Decimal("-1.234")),
TokenInteger(pos=30, size=3, value=990),
],
)
def test_case_words1(self):
self.run_test(
raw=br"""%!PS
% This is some secret sauce
begin end
" @ #
/a/BCD /Some_Name /foo#5f#6eaa /bar#6a
/need /oh1oh
func/a/b
""",
expected=[
TokenComment(pos=0, size=4, value=b"!PS"),
TokenComment(pos=5, size=27, value=b" This is some secret sauce"),
TokenKeyword(pos=33, size=5, value=b"begin"),
TokenKeyword(pos=39, size=3, value=b"end"),
TokenKeyword(pos=44, size=1, value=b'"'),
TokenKeyword(pos=47, size=1, value=b"@"),
TokenKeyword(pos=49, size=1, value=b"#"),
TokenLiteral(pos=52, size=1, value="a"),
TokenLiteral(pos=54, size=3, value="BCD"),
TokenLiteral(pos=59, size=9, value="Some_Name"),
TokenLiteral(pos=70, size=11, value="foo_naa"),
TokenLiteral(pos=83, size=6, value="barj"),
TokenLiteral(pos=91, size=4, value="need"),
TokenLiteral(pos=97, size=5, value="oh1oh"),
TokenKeyword(pos=103, size=4, value=b"func"),
TokenLiteral(pos=108, size=1, value="a"),
TokenLiteral(pos=110, size=1, value="b"),
],
)