可用的回归测试通过的标准版本
This commit is contained in:
@@ -1,550 +0,0 @@
|
||||
from stdint import *
|
||||
import zdeflate
|
||||
import zinflate
|
||||
import zchecksum
|
||||
import zdef
|
||||
import zhuff
|
||||
import stdlib
|
||||
import string
|
||||
import stdio
|
||||
import t, c
|
||||
|
||||
|
||||
# ============================================================
|
||||
# Constants - matching Python zlib module names and values
|
||||
# ============================================================
|
||||
|
||||
# Compression levels
|
||||
Z_NO_COMPRESSION: t.CDefine = 0
|
||||
Z_BEST_SPEED: t.CDefine = 1
|
||||
Z_BEST_COMPRESSION: t.CDefine = 9
|
||||
Z_DEFAULT_COMPRESSION: t.CDefine = (-1)
|
||||
|
||||
# Compression methods
|
||||
DEFLATED: t.CDefine = 8
|
||||
|
||||
# Flush modes
|
||||
Z_NO_FLUSH: t.CDefine = 0
|
||||
Z_PARTIAL_FLUSH: t.CDefine = 1
|
||||
Z_SYNC_FLUSH: t.CDefine = 2
|
||||
Z_FULL_FLUSH: t.CDefine = 3
|
||||
Z_FINISH: t.CDefine = 4
|
||||
Z_BLOCK: t.CDefine = 5
|
||||
Z_TREES: t.CDefine = 6
|
||||
|
||||
# Strategies
|
||||
Z_DEFAULT_STRATEGY: t.CDefine = 0
|
||||
Z_FILTERED: t.CDefine = 1
|
||||
Z_HUFFMAN_ONLY: t.CDefine = 2
|
||||
Z_RLE: t.CDefine = 3
|
||||
Z_FIXED: t.CDefine = 4
|
||||
|
||||
# Return codes
|
||||
Z_OK: t.CDefine = 0
|
||||
Z_STREAM_END: t.CDefine = 1
|
||||
Z_NEED_DICT: t.CDefine = 2
|
||||
Z_ERRNO: t.CDefine = (-1)
|
||||
Z_STREAM_ERROR: t.CDefine = (-2)
|
||||
Z_DATA_ERROR: t.CDefine = (-3)
|
||||
Z_MEM_ERROR: t.CDefine = (-4)
|
||||
Z_BUF_ERROR: t.CDefine = (-5)
|
||||
Z_VERSION_ERROR: t.CDefine = (-6)
|
||||
|
||||
# Window / Buffer constants
|
||||
MAX_WBITS: t.CDefine = 15
|
||||
DEF_BUF_SIZE: t.CDefine = 16384
|
||||
DEF_MEM_LEVEL: t.CDefine = 8
|
||||
|
||||
# Version
|
||||
ZLIB_VERSION: t.CDefine = "1.3.2"
|
||||
|
||||
pyzlib_error_msg: list[t.CChar, 512] = ""
|
||||
pyzlib_error_code_val: t.CInt = 0
|
||||
|
||||
# ============================================================
|
||||
# Structures
|
||||
# ============================================================
|
||||
@t.Object
|
||||
class Compress:
|
||||
stream: VOIDPTR
|
||||
is_initialized: t.CInt
|
||||
is_finished: t.CInt
|
||||
level: t.CInt
|
||||
method: t.CInt
|
||||
wbits: t.CInt
|
||||
memLevel: t.CInt
|
||||
strategy: t.CInt
|
||||
zdict: BYTEPTR
|
||||
zdict_len: t.CSizeT
|
||||
last_pos: t.CSizeT
|
||||
input_buf: BYTEPTR
|
||||
input_buf_len: t.CSizeT
|
||||
input_buf_cap: t.CSizeT
|
||||
header_written: t.CInt
|
||||
|
||||
def compress(self,
|
||||
data: BYTEPTR, data_len: t.CSizeT,
|
||||
out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "compress object not initialized")
|
||||
return None
|
||||
if self.is_finished:
|
||||
set_error(Z_STREAM_ERROR, "compress object already finished")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
if data_len == 0:
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
while self.input_buf_len + data_len > self.input_buf_cap:
|
||||
self.input_buf_cap *= 2
|
||||
new_buf: BYTEPTR = BYTEPTR(realloc(self.input_buf, self.input_buf_cap))
|
||||
if not new_buf:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
self.input_buf = new_buf
|
||||
memcpy(self.input_buf + self.input_buf_len, data, data_len)
|
||||
self.input_buf_len += data_len
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
|
||||
def flush(self, mode: t.CInt, out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "compress object not initialized")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
s: zdeflate.zdeflate_stream | t.CPtr = self.stream # 隐式类型转换
|
||||
if mode == Z_FINISH:
|
||||
if not self.header_written:
|
||||
self.header_written = 1
|
||||
input_data: BYTEPTR = self.input_buf
|
||||
input_len: t.CSizeT = self.input_buf_len
|
||||
if input_len == 0:
|
||||
s.writer.write_bits(1, 1)
|
||||
s.writer.write_bits(1, 2)
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
lit_tree.encode_symbol(256, c.Addr(s.writer))
|
||||
else:
|
||||
s.adler = zchecksum.zchecksum_adler32(input_data, input_len, 1)
|
||||
s.compress_block(input_data, input_len, 1)
|
||||
s.is_finished = 1
|
||||
if s.wbits >= 0:
|
||||
s.writer.align()
|
||||
adler: t.CUInt32T = s.adler
|
||||
s.writer.buf[s.writer.byte_pos + 0] = (adler >> 24) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 1] = (adler >> 16) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 2] = (adler >> 8) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 3] = (adler >> 0) & 0xFF
|
||||
s.writer.byte_pos += 4
|
||||
self.is_finished = 1
|
||||
total: t.CSizeT = s.writer.total()
|
||||
result: BYTEPTR = BYTEPTR(malloc(total))
|
||||
if result: memcpy(result, s.writer.buf, total)
|
||||
c.Set(c.Deref(out_len), total)
|
||||
free(self.input_buf)
|
||||
self.input_buf = None
|
||||
self.input_buf_len = 0
|
||||
self.input_buf_cap = 0
|
||||
return result
|
||||
elif mode == Z_SYNC_FLUSH or mode == Z_FULL_FLUSH:
|
||||
input_data: BYTEPTR = self.input_buf
|
||||
input_len: t.CSizeT = self.input_buf_len
|
||||
if input_len > 0:
|
||||
s.adler = zchecksum.zchecksum_adler32(input_data, input_len, s.adler)
|
||||
s.compress_block(input_data, input_len, 0)
|
||||
self.input_buf_len = 0
|
||||
s.writer.write_bits(0, 1)
|
||||
s.writer.write_bits(0, 2)
|
||||
s.writer.align()
|
||||
prev_pos: t.CSizeT = self.last_pos
|
||||
total: t.CSizeT = s.writer.total()
|
||||
delta: t.CSizeT = total - prev_pos
|
||||
self.last_pos = s.writer.byte_pos
|
||||
if delta == 0:
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
|
||||
result: BYTEPTR = BYTEPTR(malloc(delta))
|
||||
if result: memcpy(result, s.writer.buf + prev_pos, delta)
|
||||
c.Set(c.Deref(out_len), delta)
|
||||
return result
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
|
||||
def copy(self) -> Compress | t.CPtr:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "compress object not initialized")
|
||||
return None
|
||||
copy_obj: Compress | t.CPtr = calloc(1, Compress.__sizeof__())
|
||||
if not copy_obj:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
s: zdeflate.zdeflate_stream | t.CPtr = self.stream # 隐式类型转换
|
||||
copy_obj.stream = s.copy()
|
||||
if not copy_obj.stream:
|
||||
free(copy_obj)
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
copy_obj.is_initialized = 1
|
||||
copy_obj.is_finished = self.is_finished
|
||||
copy_obj.level = self.level
|
||||
copy_obj.method = self.method
|
||||
copy_obj.wbits = self.wbits
|
||||
copy_obj.memLevel = self.memLevel
|
||||
copy_obj.strategy = self.strategy
|
||||
copy_obj.last_pos = self.last_pos
|
||||
copy_obj.header_written = self.header_written
|
||||
if self.input_buf and self.input_buf_len > 0:
|
||||
copy_obj.input_buf = BYTEPTR(calloc(1, self.input_buf_cap))
|
||||
if copy_obj.input_buf:
|
||||
memcpy(copy_obj.input_buf, self.input_buf, self.input_buf_len)
|
||||
copy_obj.input_buf_cap = self.input_buf_cap
|
||||
copy_obj.input_buf_len = self.input_buf_len
|
||||
else:
|
||||
copy_obj.input_buf = BYTEPTR(calloc(1, 256))
|
||||
copy_obj.input_buf_cap = 256
|
||||
copy_obj.input_buf_len = 0
|
||||
if self.zdict and self.zdict_len > 0:
|
||||
copy_obj.zdict = clone_bytes(self.zdict, self.zdict_len)
|
||||
copy_obj.zdict_len = self.zdict_len
|
||||
return copy_obj
|
||||
|
||||
def delete(self):
|
||||
if self.is_initialized and self.stream:
|
||||
s: zdeflate.zdeflate_stream | t.CPtr = self.stream # 隐式类型转换
|
||||
s.destroy()
|
||||
free(self.zdict)
|
||||
free(self.input_buf)
|
||||
free(self)
|
||||
|
||||
def __del__(self):
|
||||
self.delete()
|
||||
|
||||
|
||||
@t.Object
|
||||
class Decompress:
|
||||
stream: VOIDPTR
|
||||
is_initialized: t.CInt
|
||||
eof: t.CInt
|
||||
wbits: t.CInt
|
||||
zdict: BYTEPTR
|
||||
zdict_len: t.CSizeT
|
||||
_unused_data: BYTEPTR
|
||||
_unused_data_len: t.CSizeT
|
||||
_unused_data_cap: t.CSizeT
|
||||
_unconsumed_tail: BYTEPTR
|
||||
_unconsumed_tail_len: t.CSizeT
|
||||
_unconsumed_tail_cap: t.CSizeT
|
||||
input_buf: BYTEPTR
|
||||
input_buf_len: t.CSizeT
|
||||
input_buf_cap: t.CSizeT
|
||||
|
||||
def decompress(self, data: BYTEPTR, data_len: t.CSizeT,
|
||||
max_length: t.CSizeT, out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "decompress object not initialized")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
if self.eof:
|
||||
if data and data_len > 0:
|
||||
append_bytes(c.Addr(self._unused_data), c.Addr(self._unused_data_len),
|
||||
c.Addr(self._unused_data_cap), data, data_len)
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
self._unconsumed_tail_len = 0
|
||||
if data and data_len > 0:
|
||||
append_bytes(c.Addr(self.input_buf), c.Addr(self.input_buf_len),
|
||||
c.Addr(self.input_buf_cap), data, data_len)
|
||||
if self.input_buf_len == 0:
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
s: zinflate.zinflate_stream | t.CPtr = zinflate.zinflate_create(self.wbits)
|
||||
if not s:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
if self.zdict and self.zdict_len > 0:
|
||||
s.set_dictionary(self.zdict, self.zdict_len)
|
||||
out: t.CUInt8T | t.CPtr = None
|
||||
err: t.CInt = s.decompress(self.input_buf, self.input_buf_len,
|
||||
max_length, c.Addr(out), out_len)
|
||||
if err != 0:
|
||||
s.destroy()
|
||||
set_error(Z_DATA_ERROR, "decompression failed")
|
||||
return None
|
||||
if s.is_finished:
|
||||
self.eof = 1
|
||||
if s.input_pos < s.input_len:
|
||||
append_bytes(c.Addr(self._unused_data), c.Addr(self._unused_data_len),
|
||||
c.Addr(self._unused_data_cap),
|
||||
s.input_data + s.input_pos,
|
||||
s.input_len - s.input_pos)
|
||||
self.input_buf_len = 0
|
||||
else:
|
||||
consumed: t.CSizeT = s.input_pos
|
||||
if consumed < self.input_buf_len:
|
||||
remaining: t.CSizeT = self.input_buf_len - consumed
|
||||
memmove(self.input_buf, self.input_buf + consumed, remaining)
|
||||
self.input_buf_len = remaining
|
||||
else:
|
||||
self.input_buf_len = 0
|
||||
s.destroy()
|
||||
return out
|
||||
|
||||
def flush(self, length: t.CSizeT, out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "decompress object not initialized")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
|
||||
def copy(self) -> Decompress | t.CPtr:
|
||||
if not self.is_initialized:
|
||||
set_error(Z_STREAM_ERROR, "decompress object not initialized")
|
||||
return None
|
||||
copy_obj: Decompress | t.CPtr = calloc(1, Decompress.__sizeof__())
|
||||
if not copy_obj:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
copy_obj.is_initialized = 1
|
||||
copy_obj.eof = self.eof
|
||||
copy_obj.wbits = self.wbits
|
||||
if self.zdict and self.zdict_len > 0:
|
||||
copy_obj.zdict = clone_bytes(self.zdict, self.zdict_len)
|
||||
copy_obj.zdict_len = self.zdict_len
|
||||
if self._unused_data and self._unused_data_len > 0:
|
||||
copy_obj._unused_data = clone_bytes(self._unused_data, self._unused_data_len)
|
||||
copy_obj._unused_data_len = self._unused_data_len
|
||||
copy_obj._unused_data_cap = self._unused_data_len
|
||||
if self._unconsumed_tail and self._unconsumed_tail_len > 0:
|
||||
copy_obj._unconsumed_tail = clone_bytes(self._unconsumed_tail, self._unconsumed_tail_len)
|
||||
copy_obj._unconsumed_tail_len = self._unconsumed_tail_len
|
||||
copy_obj._unconsumed_tail_cap = self._unconsumed_tail_len
|
||||
if self.input_buf and self.input_buf_len > 0:
|
||||
copy_obj.input_buf = clone_bytes(self.input_buf, self.input_buf_len)
|
||||
copy_obj.input_buf_len = self.input_buf_len
|
||||
copy_obj.input_buf_cap = self.input_buf_cap
|
||||
else:
|
||||
copy_obj.input_buf = BYTEPTR(calloc(1, 256))
|
||||
copy_obj.input_buf_cap = 256
|
||||
copy_obj.input_buf_len = 0
|
||||
return copy_obj
|
||||
|
||||
def delete(self):
|
||||
free(self.zdict)
|
||||
free(self._unused_data)
|
||||
free(self._unconsumed_tail)
|
||||
free(self.input_buf)
|
||||
free(self)
|
||||
|
||||
def unused_data(self, length: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
#if not self:
|
||||
# if length:
|
||||
# c.Set(c.Deref(length), 0)
|
||||
# return None
|
||||
if length:
|
||||
c.Set(c.Deref(length), self._unused_data_len)
|
||||
return self._unused_data
|
||||
|
||||
def unconsumed_tail(self, length: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
#if not self:
|
||||
# if length:
|
||||
# c.Set(c.Deref(length), 0)
|
||||
# return None
|
||||
if length:
|
||||
c.Set(c.Deref(length), self._unconsumed_tail_len)
|
||||
return self._unconsumed_tail
|
||||
|
||||
# def eof(self) -> t.CInt:
|
||||
# # if not obj: return 0
|
||||
# return self.eof
|
||||
|
||||
|
||||
|
||||
def set_error(code: int, msg: str):
|
||||
global pyzlib_error_code_val, pyzlib_error_msg
|
||||
pyzlib_error_code_val = code
|
||||
if msg:
|
||||
strncpy(pyzlib_error_msg, msg, pyzlib_error_msg.__sizeof__() - 1)
|
||||
pyzlib_error_msg[pyzlib_error_msg.__sizeof__() - 1] = '\0'
|
||||
else:
|
||||
pyzlib_error_msg[0] = '\0'
|
||||
|
||||
def clone_bytes(src: BYTEPTR, length: t.CSizeT) -> BYTEPTR:
|
||||
if not src or length == 0: return None
|
||||
dst: BYTEPTR = BYTEPTR(malloc(length))
|
||||
if dst: memcpy(dst, src, length)
|
||||
return dst
|
||||
|
||||
def append_bytes(buf: BYTE | t.CPtr[t.CPtr], length: t.CSizeT | t.CPtr, cap: t.CSizeT | t.CPtr,
|
||||
data: BYTEPTR, data_len: t.CSizeT) -> t.CInt:
|
||||
if not data or data_len == 0: return 0
|
||||
needed: t.CSizeT = c.Deref(length) + data_len
|
||||
if needed > c.Deref(cap):
|
||||
new_cap: t.CSizeT = c.Deref(cap) * 2
|
||||
if new_cap < needed: new_cap = needed
|
||||
new_buf: BYTEPTR = BYTEPTR(realloc(c.Deref(buf), new_cap))
|
||||
if not new_buf: return -1
|
||||
c.Set(c.Deref(buf), new_buf)
|
||||
c.Set(c.Deref(cap), new_cap)
|
||||
memcpy(c.Deref(buf) + c.Deref(length), data, data_len)
|
||||
c.Set(c.Deref(length), c.Deref(length) + data_len)
|
||||
return 0
|
||||
|
||||
|
||||
def shrink_to_fit(buf: BYTEPTR, length: t.CSizeT) -> BYTEPTR:
|
||||
if length == 0:
|
||||
free(buf)
|
||||
return BYTEPTR(calloc(1, 1))
|
||||
result: BYTEPTR = BYTEPTR(realloc(buf, length))
|
||||
return result if result else buf
|
||||
|
||||
|
||||
def runtime_version() -> str:
|
||||
return ZLIB_VERSION
|
||||
|
||||
def get_error() -> str:
|
||||
return pyzlib_error_msg
|
||||
|
||||
def get_error_code() -> t.CInt:
|
||||
return pyzlib_error_code_val
|
||||
|
||||
def clear_error():
|
||||
global pyzlib_error_msg, pyzlib_error_code_val
|
||||
pyzlib_error_msg[0] = '\0'
|
||||
pyzlib_error_code_val = 0
|
||||
|
||||
def compress(data: BYTEPTR, data_len: t.CSizeT,
|
||||
level: t.CInt, wbits: t.CInt, out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not data and data_len > 0:
|
||||
set_error(Z_STREAM_ERROR, "data is None but data_len > 0")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
actual_wbits: t.CInt = wbits
|
||||
if wbits > MAX_WBITS:
|
||||
actual_wbits = -MAX_WBITS
|
||||
raw_result: UINT8PTR = zdeflate.zdeflate_one_shot(data, data_len, level, actual_wbits, out_len)
|
||||
if not raw_result:
|
||||
set_error(Z_DATA_ERROR, "compression failed")
|
||||
return None
|
||||
if wbits > MAX_WBITS:
|
||||
gzip_len: t.CSizeT = 10 + c.Deref(out_len) + 8
|
||||
gzip_buf: BYTEPTR = BYTEPTR(malloc(gzip_len))
|
||||
if not gzip_buf:
|
||||
free(raw_result)
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
pos: t.CSizeT = 0
|
||||
gzip_buf[pos + 0] = 0x1F
|
||||
gzip_buf[pos + 1] = 0x8B
|
||||
gzip_buf[pos + 2] = 0x08
|
||||
gzip_buf[pos + 3] = 0x00
|
||||
gzip_buf[pos + 4] = 0x00
|
||||
gzip_buf[pos + 5] = 0x00
|
||||
gzip_buf[pos + 6] = 0x00
|
||||
gzip_buf[pos + 7] = 0x00
|
||||
gzip_buf[pos + 8] = 0x00
|
||||
gzip_buf[pos + 9] = 0xFF
|
||||
pos += 10
|
||||
|
||||
memcpy(gzip_buf + pos, raw_result, c.Deref(out_len))
|
||||
pos += c.Deref(out_len)
|
||||
free(raw_result)
|
||||
|
||||
crc: t.CUInt32T = zchecksum.zchecksum_crc32(data, data_len, 0)
|
||||
gzip_buf[pos + 0] = (crc) & 0xFF
|
||||
gzip_buf[pos + 1] = (crc >> 8) & 0xFF
|
||||
gzip_buf[pos + 2] = (crc >> 16) & 0xFF
|
||||
gzip_buf[pos + 3] = (crc >> 24) & 0xFF
|
||||
gzip_buf[pos + 4] = (data_len) & 0xFF
|
||||
gzip_buf[pos + 5] = (data_len >> 8) & 0xFF
|
||||
gzip_buf[pos + 6] = (data_len >> 16) & 0xFF
|
||||
gzip_buf[pos + 7] = (data_len >> 24) & 0xFF
|
||||
pos += 8
|
||||
c.Set(c.Deref(out_len), pos)
|
||||
return shrink_to_fit(gzip_buf, pos)
|
||||
return shrink_to_fit(raw_result, c.Deref(out_len))
|
||||
|
||||
|
||||
def decompress(data: BYTEPTR, data_len: t.CSizeT,
|
||||
wbits: t.CInt, bufsize: t.CSizeT, out_len: t.CSizeT | t.CPtr) -> BYTEPTR:
|
||||
if not data and data_len > 0:
|
||||
set_error(Z_STREAM_ERROR, "data is None but data_len > 0")
|
||||
return None
|
||||
if not out_len:
|
||||
set_error(Z_STREAM_ERROR, "out_len is None")
|
||||
return None
|
||||
result: UINT8PTR = zinflate.zinflate_one_shot(data, data_len, wbits, bufsize, out_len)
|
||||
if not result:
|
||||
set_error(Z_DATA_ERROR, "decompression failed")
|
||||
return None
|
||||
return shrink_to_fit(result, c.Deref(out_len))
|
||||
|
||||
def compressobj(level: t.CInt, method: t.CInt, wbits: t.CInt,
|
||||
memLevel: t.CInt, strategy: t.CInt,
|
||||
zdict: BYTEPTR, zdict_len: t.CSizeT) -> Compress | t.CPtr:
|
||||
obj: Compress | t.CPtr = calloc(1, Compress.__sizeof__())
|
||||
if not obj:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
actual_wbits: t.CInt = wbits
|
||||
if wbits > MAX_WBITS: actual_wbits = wbits - 16
|
||||
s: zdeflate.zdeflate_stream | t.CPtr = zdeflate.zdeflate_create(level, actual_wbits, memLevel, strategy)
|
||||
if not s:
|
||||
free(obj)
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
obj.stream = s
|
||||
obj.is_initialized = 1
|
||||
obj.level = level
|
||||
obj.method = method
|
||||
obj.wbits = wbits
|
||||
obj.memLevel = memLevel
|
||||
obj.strategy = strategy
|
||||
obj.input_buf = BYTEPTR(calloc(1, 256))
|
||||
obj.input_buf_cap = 256
|
||||
obj.input_buf_len = 0
|
||||
obj.header_written = 0
|
||||
if zdict and zdict_len > 0:
|
||||
s.set_dictionary(zdict, zdict_len)
|
||||
obj.zdict = clone_bytes(zdict, zdict_len)
|
||||
obj.zdict_len = zdict_len
|
||||
return obj
|
||||
|
||||
|
||||
def decompressobj(wbits: t.CInt, zdict: BYTEPTR, zdict_len: t.CSizeT) -> Decompress | t.CPtr:
|
||||
obj: Decompress | t.CPtr = calloc(1, Decompress.__sizeof__())
|
||||
if not obj:
|
||||
set_error(Z_MEM_ERROR, "out of memory")
|
||||
return None
|
||||
actual_wbits: t.CInt = wbits
|
||||
if wbits > MAX_WBITS: actual_wbits = wbits - 16
|
||||
obj.stream = None
|
||||
obj.is_initialized = 1
|
||||
obj.wbits = actual_wbits
|
||||
if zdict and zdict_len > 0:
|
||||
obj.zdict = clone_bytes(zdict, zdict_len)
|
||||
obj.zdict_len = zdict_len
|
||||
obj.input_buf = BYTEPTR(calloc(1, 256))
|
||||
obj.input_buf_cap = 256
|
||||
obj.input_buf_len = 0
|
||||
return obj
|
||||
|
||||
def zlib_adler32(data: BYTEPTR, length: t.CSizeT, value: t.CUnsignedLong) -> t.CUnsignedLong:
|
||||
return zchecksum.zchecksum_adler32(data, length, UINT32(value))
|
||||
|
||||
def zlib_crc32(data: BYTEPTR, length: t.CSizeT, value: t.CUnsignedLong) -> t.CUnsignedLong:
|
||||
return zchecksum.zchecksum_crc32(data, length, UINT32(value))
|
||||
|
||||
|
||||
@@ -1,81 +0,0 @@
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <stdint.h>
|
||||
#include "pyzlib.h"
|
||||
|
||||
int main(void)
|
||||
{
|
||||
printf("=== Basic Self-Implemented zlib Test ===\n\n");
|
||||
|
||||
printf("1. Adler-32:\n");
|
||||
unsigned long a = zlib_adler32((const unsigned char *)"Hello", 5, 1);
|
||||
printf(" adler32(\"Hello\") = 0x%08lX\n", a);
|
||||
|
||||
unsigned long a2 = zlib_adler32((const unsigned char *)"Hel", 3, 1);
|
||||
a2 = zlib_adler32((const unsigned char *)"lo", 2, a2);
|
||||
printf(" incremental = 0x%08lX %s\n", a2, a == a2 ? "MATCH" : "MISMATCH");
|
||||
|
||||
printf("\n2. CRC-32:\n");
|
||||
unsigned long c = zlib_crc32((const unsigned char *)"Hello", 5, 0);
|
||||
printf(" crc32(\"Hello\") = 0x%08lX\n", c);
|
||||
|
||||
unsigned long c2 = zlib_crc32((const unsigned char *)"Hel", 3, 0);
|
||||
c2 = zlib_crc32((const unsigned char *)"lo", 2, c2);
|
||||
printf(" incremental = 0x%08lX %s\n", c2, c == c2 ? "MATCH" : "MISMATCH");
|
||||
|
||||
printf("\n3. Compress (zlib format):\n");
|
||||
const char *input = "Hello, World!";
|
||||
size_t comp_len = 0;
|
||||
unsigned char *comp = zlib_compress((const unsigned char *)input, strlen(input),
|
||||
Z_DEFAULT_COMPRESSION, MAX_WBITS, &comp_len);
|
||||
if (comp) {
|
||||
printf(" Input: %zu bytes\n", strlen(input));
|
||||
printf(" Output: %zu bytes\n", comp_len);
|
||||
printf(" Hex: ");
|
||||
for (size_t i = 0; i < comp_len && i < 40; i++) printf("%02X ", comp[i]);
|
||||
printf("\n");
|
||||
|
||||
printf("\n4. Decompress (zlib format):\n");
|
||||
size_t dec_len = 0;
|
||||
unsigned char *dec = zlib_decompress(comp, comp_len, MAX_WBITS, DEF_BUF_SIZE, &dec_len);
|
||||
if (dec) {
|
||||
printf(" Output: %zu bytes\n", dec_len);
|
||||
printf(" Content: \"%.*s\"\n", (int)dec_len, (char *)dec);
|
||||
printf(" Match: %s\n", dec_len == strlen(input) && memcmp(dec, input, dec_len) == 0 ? "YES" : "NO");
|
||||
free(dec);
|
||||
} else {
|
||||
printf(" FAILED! Error: %s (code %d)\n", zlib_get_error(), zlib_get_error_code());
|
||||
}
|
||||
free(comp);
|
||||
} else {
|
||||
printf(" FAILED! Error: %s (code %d)\n", zlib_get_error(), zlib_get_error_code());
|
||||
}
|
||||
|
||||
printf("\n5. Compress (raw deflate):\n");
|
||||
comp_len = 0;
|
||||
comp = zlib_compress((const unsigned char *)input, strlen(input),
|
||||
Z_DEFAULT_COMPRESSION, -MAX_WBITS, &comp_len);
|
||||
if (comp) {
|
||||
printf(" Output: %zu bytes\n", comp_len);
|
||||
printf(" Hex: ");
|
||||
for (size_t i = 0; i < comp_len && i < 40; i++) printf("%02X ", comp[i]);
|
||||
printf("\n");
|
||||
|
||||
size_t dec_len = 0;
|
||||
unsigned char *dec = zlib_decompress(comp, comp_len, -MAX_WBITS, DEF_BUF_SIZE, &dec_len);
|
||||
if (dec) {
|
||||
printf(" Decompress: \"%.*s\" %s\n", (int)dec_len, (char *)dec,
|
||||
dec_len == strlen(input) && memcmp(dec, input, dec_len) == 0 ? "OK" : "FAIL");
|
||||
free(dec);
|
||||
} else {
|
||||
printf(" Decompress FAILED! Error: %s\n", zlib_get_error());
|
||||
}
|
||||
free(comp);
|
||||
} else {
|
||||
printf(" FAILED! Error: %s (code %d)\n", zlib_get_error(), zlib_get_error_code());
|
||||
}
|
||||
|
||||
printf("\nDone.\n");
|
||||
return 0;
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
#include <stdint.h>
|
||||
#include "pyzlib.h"
|
||||
|
||||
int main(void)
|
||||
{
|
||||
printf("=== Full Self-Implemented zlib Test ===\n\n");
|
||||
|
||||
printf("1. Adler-32:\n");
|
||||
unsigned long a = zlib_adler32((const unsigned char *)"Hello", 5, 1);
|
||||
printf(" adler32(\"Hello\") = 0x%08lX\n", a);
|
||||
|
||||
printf("\n2. CRC-32:\n");
|
||||
unsigned long c = zlib_crc32((const unsigned char *)"Hello", 5, 0);
|
||||
printf(" crc32(\"Hello\") = 0x%08lX\n", c);
|
||||
|
||||
const char *tests[] = {"Hello", "Hello, World!", "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", ""};
|
||||
int levels[] = {2, 6, 0};
|
||||
|
||||
for (int t = 0; t < 4; t++) {
|
||||
const char *input = tests[t];
|
||||
size_t input_len = strlen(input);
|
||||
printf("\n--- Test: \"%s\" (%zu bytes) ---\n", input_len > 0 ? input : "(empty)", input_len);
|
||||
|
||||
for (int li = 0; li < 3; li++) {
|
||||
int level = levels[li];
|
||||
printf("\n Level %d, zlib format:\n", level);
|
||||
size_t comp_len = 0;
|
||||
unsigned char *comp = zlib_compress((const unsigned char *)input, input_len,
|
||||
level, MAX_WBITS, &comp_len);
|
||||
if (comp) {
|
||||
printf(" Compressed: %zu bytes\n", comp_len);
|
||||
size_t dec_len = 0;
|
||||
unsigned char *dec = zlib_decompress(comp, comp_len, MAX_WBITS, DEF_BUF_SIZE, &dec_len);
|
||||
if (dec) {
|
||||
int match = (dec_len == input_len && memcmp(dec, input, dec_len) == 0);
|
||||
printf(" Decompressed: %zu bytes, Match: %s\n", dec_len, match ? "YES" : "NO");
|
||||
if (!match) {
|
||||
printf(" Expected: \"%s\"\n", input_len > 0 ? input : "(empty)");
|
||||
printf(" Got: \"%.*s\"\n", (int)dec_len, (char *)dec);
|
||||
}
|
||||
free(dec);
|
||||
} else {
|
||||
printf(" Decompress FAILED: %s (code %d)\n", zlib_get_error(), zlib_get_error_code());
|
||||
}
|
||||
free(comp);
|
||||
} else {
|
||||
printf(" Compress FAILED: %s (code %d)\n", zlib_get_error(), zlib_get_error_code());
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
printf("\n3. Raw deflate:\n");
|
||||
{
|
||||
const char *input = "Raw deflate test";
|
||||
size_t comp_len = 0;
|
||||
unsigned char *comp = zlib_compress((const unsigned char *)input, strlen(input),
|
||||
Z_DEFAULT_COMPRESSION, -MAX_WBITS, &comp_len);
|
||||
if (comp) {
|
||||
printf(" Compressed: %zu bytes\n", comp_len);
|
||||
size_t dec_len = 0;
|
||||
unsigned char *dec = zlib_decompress(comp, comp_len, -MAX_WBITS, DEF_BUF_SIZE, &dec_len);
|
||||
if (dec) {
|
||||
printf(" Decompressed: \"%.*s\" %s\n", (int)dec_len, (char *)dec,
|
||||
dec_len == strlen(input) && memcmp(dec, input, dec_len) == 0 ? "OK" : "FAIL");
|
||||
free(dec);
|
||||
} else {
|
||||
printf(" Decompress FAILED: %s\n", zlib_get_error());
|
||||
}
|
||||
free(comp);
|
||||
}
|
||||
}
|
||||
|
||||
printf("\nDone.\n");
|
||||
return 0;
|
||||
}
|
||||
@@ -1,880 +0,0 @@
|
||||
from stdint import *
|
||||
import pyzlib
|
||||
import zhuff
|
||||
import zdef
|
||||
from stdio import printf, strlen
|
||||
import stdlib
|
||||
import string
|
||||
import t, c
|
||||
|
||||
|
||||
test_passed: t.CInt = 0
|
||||
test_failed: t.CInt = 0
|
||||
def check(name: str, condition: t.CInt, detail: str):
|
||||
global test_passed, test_failed
|
||||
if condition:
|
||||
test_passed += 1
|
||||
printf(" [PASS] %s\n", name)
|
||||
else:
|
||||
test_failed += 1
|
||||
printf(" [FAIL] %s -- %s\n", name, detail if detail else "")
|
||||
|
||||
def print_hex_test(data: BYTEPTR, length: t.CSizeT, max_show: t.CSizeT):
|
||||
if not data or length == 0:
|
||||
printf("(empty)\n")
|
||||
return
|
||||
show: t.CSizeT = max_show if (length > max_show) else length
|
||||
i: t.CSizeT
|
||||
for i in range(show):
|
||||
printf("%02X ", data[i])
|
||||
if length > max_show:
|
||||
printf("... (%zu bytes total)", length)
|
||||
printf("\n")
|
||||
|
||||
def print_separator():
|
||||
printf(" -------------------------------------------\n")
|
||||
|
||||
def section_header(title: str):
|
||||
printf("\n+-- %s --+\n", title)
|
||||
|
||||
def section_footer():
|
||||
printf("+--------------------------------------------+\n")
|
||||
|
||||
# ============================================================
|
||||
|
||||
def test_version():
|
||||
section_header("Version Info")
|
||||
|
||||
ver: str = pyzlib.ZLIB_VERSION
|
||||
printf(" ZLIB_VERSION (compile-time) : %s\n", ver)
|
||||
check("ZLIB_VERSION not None", ver != None, "version string is None")
|
||||
check("ZLIB_VERSION not empty", strlen(ver) > 0, "version string is empty")
|
||||
|
||||
runtime_ver: str = pyzlib.runtime_version()
|
||||
printf(" ZLIB_RUNTIME_VERSION : %s\n", runtime_ver)
|
||||
check("runtime version not None", runtime_ver != None, "runtime version is None")
|
||||
check("runtime version not empty", strlen(runtime_ver) > 0, "runtime version is empty")
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_compress_decompress():
|
||||
section_header("compress / decompress")
|
||||
|
||||
inp: str = "Hello, World! This is a test of zlib compression and decompression."
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input (%zu bytes): \"%s\"\n", input_length, inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("compress returns non-None", compressed != None, "compress returned None")
|
||||
check("compress output size > 0", out_length > 0, "compressed size is 0")
|
||||
check("compress output produced", out_length > 0, "compressed size is 0")
|
||||
|
||||
printf(" Compressed (%zu bytes): ", out_length)
|
||||
print_hex_test(compressed, out_length, 32)
|
||||
printf(" Compression ratio: %.1f%%\n", t.CDouble(out_length) / input_length * 100)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
decompressed: BYTEPTR = pyzlib.decompress(compressed, out_length,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("decompress returns non-None", decompressed != None, "decompress returned None")
|
||||
check("decompress output size matches input", dec_length == input_length, "size mismatch")
|
||||
check("decompress output matches input", decompressed != None and memcmp(decompressed, inp, input_length) == 0, "content mismatch")
|
||||
|
||||
printf(" Decompressed (%zu bytes): \"%.*s\"\n", dec_length, t.CInt(dec_length), str(decompressed))
|
||||
|
||||
free(compressed)
|
||||
free(decompressed)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_compress_levels():
|
||||
section_header("Compression Levels")
|
||||
|
||||
inp: str = "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input (%zu bytes): \"%s\"\n", input_length, inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
|
||||
c0: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_NO_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("level 0 compress OK", c0 != None, "level 0 returned None")
|
||||
printf(" Level 0 (Z_NO_COMPRESSION): %zu bytes\n", out_length)
|
||||
length0: t.CSizeT = out_length
|
||||
free(c0)
|
||||
|
||||
c1: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_BEST_SPEED, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("level 1 compress OK", c1 != None, "level 1 returned None")
|
||||
printf(" Level 1 (Z_BEST_SPEED) : %zu bytes\n", out_length)
|
||||
free(c1)
|
||||
|
||||
c6: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("level -1 compress OK", c6 != None, "default level returned None")
|
||||
printf(" Level -1 (DEFAULT) : %zu bytes\n", out_length)
|
||||
free(c6)
|
||||
|
||||
c9: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_BEST_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("level 9 compress OK", c9 != None, "level 9 returned None")
|
||||
printf(" Level 9 (Z_BEST_COMPRESS) : %zu bytes\n", out_length)
|
||||
length9: t.CSizeT = out_length
|
||||
free(c9)
|
||||
|
||||
check("level 9 smaller than level 0", length9 < length0, "level 9 not smaller than level 0")
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_compressobj():
|
||||
section_header("compressobj (incremental)")
|
||||
|
||||
part1: str = "Hello, "
|
||||
part2: str = "World!"
|
||||
printf(" Part 1: \"%s\"\n", part1)
|
||||
printf(" Part 2: \"%s\"\n", part2)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
|
||||
s: pyzlib.Compress | t.CPtr = pyzlib.compressobj(pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.DEFLATED,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_MEM_LEVEL,
|
||||
pyzlib.Z_DEFAULT_STRATEGY, None, 0)
|
||||
check("compressobj created", s != None, "compressobj is None")
|
||||
print_separator()
|
||||
|
||||
c1: BYTEPTR = s.compress(BYTEPTR(part1), strlen(part1), c.Addr(out_length))
|
||||
check("compress part1 OK", c1 != None, "compress part1 failed")
|
||||
printf(" Compressed part1: %zu bytes -> ", out_length)
|
||||
print_hex_test(c1, out_length, 16)
|
||||
total: t.CSizeT = out_length
|
||||
all: BYTEPTR = None
|
||||
if out_length > 0 and c1:
|
||||
all = BYTEPTR(malloc(total))
|
||||
if all: memcpy(all, c1, out_length)
|
||||
free(c1)
|
||||
|
||||
c2: BYTEPTR = s.compress(BYTEPTR(part2),
|
||||
strlen(part2), c.Addr(out_length))
|
||||
check("compress part2 OK", c2 != None, "compress part2 failed")
|
||||
printf(" Compressed part2: %zu bytes -> ", out_length)
|
||||
print_hex_test(c2, out_length, 16)
|
||||
if out_length > 0 and c2:
|
||||
new_all: BYTEPTR = BYTEPTR(realloc(all, total + out_length))
|
||||
if new_all:
|
||||
all = new_all
|
||||
memcpy(all + total, c2, out_length)
|
||||
total += out_length
|
||||
free(c2)
|
||||
|
||||
c3: BYTEPTR = s.flush(pyzlib.Z_FINISH, c.Addr(out_length))
|
||||
check("flush (Z_FINISH) OK", c3 != None, "flush failed")
|
||||
printf(" Flush result : %zu bytes -> ", out_length)
|
||||
print_hex_test(c3, out_length, 16)
|
||||
if out_length > 0 and c3:
|
||||
new_all: BYTEPTR = realloc(all, total + out_length) # éå¼è½¬æ¢
|
||||
if new_all:
|
||||
all = new_all
|
||||
memcpy(all + total, c3, out_length)
|
||||
total += out_length
|
||||
free(c3)
|
||||
|
||||
print_separator()
|
||||
|
||||
if all:
|
||||
dec_length: t.CSizeT = 0
|
||||
dec: BYTEPTR = pyzlib.decompress(all, total, pyzlib.MAX_WBITS, pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("decompress incremental OK", dec != None, "decompress returned None")
|
||||
check("decompress incremental size", dec != None and dec_length == strlen(part1) + strlen(part2), "size mismatch")
|
||||
check("decompress incremental content",
|
||||
dec != None and memcmp(dec, "Hello, World!", dec_length) == 0, "content mismatch")
|
||||
if dec:
|
||||
printf(" Decompressed: \"%.*s\" (%zu bytes)\n", t.CInt(dec_length), str(dec), dec_length)
|
||||
free(dec)
|
||||
free(all)
|
||||
|
||||
s.delete() # del s
|
||||
section_footer()
|
||||
|
||||
def test_decompressobj():
|
||||
section_header("decompressobj")
|
||||
|
||||
inp: str = "Test data for decompress object."
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input (%zu bytes): \"%s\"\n", input_length, inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
printf(" Compressed: %zu bytes\n", out_length)
|
||||
|
||||
d: pyzlib.Decompress | t.CPtr = pyzlib.decompressobj(pyzlib.MAX_WBITS, None, 0)
|
||||
check("decompressobj created", d != None, "decompressobj is None")
|
||||
check("eof is false initially", d.eof == 0, "eof should be 0")
|
||||
print_separator()
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
dec: BYTEPTR = d.decompress(compressed, out_length,
|
||||
0, c.Addr(dec_length))
|
||||
check("decompress OK", dec != None, "decompress returned None")
|
||||
check("decompress size", dec != None and dec_length == input_length, "size mismatch")
|
||||
check("decompress content", dec != None and memcmp(dec, inp, input_length) == 0, "content mismatch")
|
||||
check("eof is true after stream end", d.eof == 1, "eof should be 1")
|
||||
|
||||
if dec:
|
||||
printf(" Decompressed (%zu bytes): \"%.*s\"\n", dec_length, t.CInt(dec_length), str(dec))
|
||||
printf(" eof = %d\n", d.eof)
|
||||
|
||||
free(dec)
|
||||
free(compressed)
|
||||
d.delete()
|
||||
section_footer()
|
||||
|
||||
def test_decompressobj_max_lengthgth():
|
||||
section_header("decompressobj max_lengthgth")
|
||||
|
||||
inp: str = "This is a longer string for testing max_lengthgth parameter in decompress."
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input (%zu bytes): \"%s\"\n", input_length, inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
printf(" Compressed: %zu bytes\n", out_length)
|
||||
|
||||
d: pyzlib.Decompress | t.CPtr = pyzlib.decompressobj(pyzlib.MAX_WBITS, None, 0)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
dec: BYTEPTR = d.decompress(compressed, out_length,
|
||||
10, c.Addr(dec_length))
|
||||
check("max_lengthgth decompress OK", dec != None, "decompress returned None")
|
||||
check("max_lengthgth limits output", dec_length <= 10, "output exceeded max_lengthgth")
|
||||
if dec:
|
||||
printf(" First chunk (max_lengthgth=10): \"%.*s\" (%zu bytes)\n", t.CInt(dec_length), str(dec), dec_length)
|
||||
free(dec)
|
||||
|
||||
tail_length: t.CSizeT = 0
|
||||
tail: BYTEPTR = d.unconsumed_tail(c.Addr(tail_length))
|
||||
check("unconsumed_tail exists", tail != None or tail_length == 0, "no unconsumed_tail")
|
||||
printf(" unconsumed_tail: %zu bytes remaining\n", tail_length)
|
||||
printf(" eof = %d\n", d.eof)
|
||||
|
||||
if tail_length > 0 or not d.eof:
|
||||
flushed: BYTEPTR = d.flush(pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("flush after max_lengthgth OK", flushed != None, "flush returned None")
|
||||
if flushed: printf(" Flushed (%zu bytes): \"%.*s\"\n", dec_length, t.CInt(dec_length), str(flushed))
|
||||
free(flushed)
|
||||
|
||||
printf(" eof after flush = %d\n", d.eof)
|
||||
|
||||
d.delete()
|
||||
free(compressed)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_decompressobj_unused_data():
|
||||
section_header("decompressobj unused_data")
|
||||
|
||||
inp: str = "Hello"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input: \"%s\"\n", inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
printf(" Compressed: %zu bytes\n", out_length)
|
||||
|
||||
total_length: t.CSizeT = out_length + 5
|
||||
combined: BYTEPTR = BYTEPTR(malloc(total_length))
|
||||
memcpy(combined, compressed, out_length)
|
||||
memcpy(combined + out_length, "EXTRA", 5)
|
||||
free(compressed)
|
||||
printf(" Combined (compressed + \"EXTRA\"): %zu bytes\n", total_length)
|
||||
|
||||
d: pyzlib.Decompress | t.CPtr = pyzlib.decompressobj(pyzlib.MAX_WBITS, None, 0)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
dec: BYTEPTR = d.decompress(combined, total_length,
|
||||
0, c.Addr(dec_length))
|
||||
check("decompress with extra OK", dec != None, "decompress returned None")
|
||||
check("eof after stream end", d.eof == 1, "eof should be 1")
|
||||
if dec:
|
||||
printf(" Decompressed: \"%.*s\" (%zu bytes)\n", t.CInt(dec_length), str(dec), dec_length)
|
||||
free(dec)
|
||||
|
||||
print_separator()
|
||||
|
||||
unused_length: t.CSizeT = 0
|
||||
unused: BYTEPTR = d.unused_data(c.Addr(unused_length))
|
||||
check("unused_data exists", unused != None and unused_length > 0, "no unused_data")
|
||||
check("unused_data is EXTRA", unused_length == 5 and unused != None and memcmp(unused, "EXTRA", 5) == 0, "unused_data mismatch")
|
||||
printf(" unused_data (%zu bytes): \"%.*s\"\n", unused_length, t.CInt(unused_length), str(unused))
|
||||
|
||||
d.delete()
|
||||
free(combined)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_compress_copy():
|
||||
section_header("Compress.copy()")
|
||||
|
||||
inp: str = "Copy test data for compress object"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input: \"%s\"\n", inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
c1: pyzlib.Compress | t.CPtr = pyzlib.compressobj(pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.DEFLATED,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_MEM_LEVEL,
|
||||
pyzlib.Z_DEFAULT_STRATEGY, None, 0)
|
||||
c1.compress(BYTEPTR(inp), input_length, c.Addr(out_length))
|
||||
printf(" Original: compressed %zu bytes so far\n", out_length)
|
||||
|
||||
c2: pyzlib.Compress | t.CPtr = c1.copy()
|
||||
check("compress copy OK", c2 != None, "copy returned None")
|
||||
printf(" Copied compressobj\n")
|
||||
|
||||
print_separator()
|
||||
|
||||
f1: BYTEPTR = c1.flush(pyzlib.Z_FINISH, c.Addr(out_length))
|
||||
check("flush original OK", f1 != None, "flush original failed")
|
||||
length1: t.CSizeT = out_length
|
||||
printf(" Original flush: %zu bytes\n", length1)
|
||||
free(f1)
|
||||
|
||||
f2: BYTEPTR = c2.flush(pyzlib.Z_FINISH, c.Addr(out_length))
|
||||
check("flush copy OK", f2 != None, "flush copy failed")
|
||||
length2: t.CSizeT = out_length
|
||||
printf(" Copy flush : %zu bytes\n", length2)
|
||||
free(f2)
|
||||
|
||||
check("copy produces same output", length1 == length2, "output lengthgths differ")
|
||||
|
||||
c1.delete()
|
||||
c2.delete()
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_decompress_copy():
|
||||
section_header("Decompress.copy()")
|
||||
|
||||
inp: str = "Decompress copy test"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input: \"%s\"\n", inp)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
printf(" Compressed: %zu bytes\n", out_length)
|
||||
|
||||
d1: pyzlib.Decompress | t.CPtr = pyzlib.decompressobj(pyzlib.MAX_WBITS, None, 0)
|
||||
d2: pyzlib.Decompress | t.CPtr = d1.copy()
|
||||
check("decompress copy OK", d2 != None, "copy returned None")
|
||||
printf(" Copied decompressobj\n")
|
||||
|
||||
print_separator()
|
||||
|
||||
dec_length1: t.CSizeT = 0
|
||||
dec_length2: t.CSizeT = 0
|
||||
dec1: BYTEPTR = d1.decompress(compressed, out_length,
|
||||
0, c.Addr(dec_length1))
|
||||
dec2: BYTEPTR = d2.decompress(compressed, out_length,
|
||||
0, c.Addr(dec_length2))
|
||||
check("decompress copy output size", dec_length1 == dec_length2, "sizes differ")
|
||||
check("decompress copy output content",
|
||||
dec1 and dec2 and memcmp(dec1, dec2, dec_length1) == 0, "content differs")
|
||||
|
||||
if dec1: printf(" Original: \"%.*s\" (%zu bytes)\n", t.CInt(dec_length1), str(dec1), dec_length1)
|
||||
if dec2: printf(" Copy : \"%.*s\" (%zu bytes)\n", t.CInt(dec_length2), str(dec2), dec_length2)
|
||||
|
||||
free(dec1)
|
||||
free(dec2)
|
||||
free(compressed)
|
||||
d1.delete()
|
||||
d2.delete()
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_adler32():
|
||||
section_header("adler32 checksum")
|
||||
|
||||
checksum: ULONG = pyzlib.zlib_adler32(BYTEPTR("Hello"), 5, 1)
|
||||
printf(" adler32(\"Hello\") = 0x%08lX (%lu)\n", checksum, checksum)
|
||||
check("adler32 returns non-zero", checksum != 0, "checksum is 0")
|
||||
|
||||
incremental: ULONG = pyzlib.zlib_adler32(BYTEPTR("Hel"), 3, 1)
|
||||
printf(" adler32(\"Hel\") = 0x%08lX\n", incremental)
|
||||
incremental = pyzlib.zlib_adler32(BYTEPTR("lo"), 2, incremental)
|
||||
printf(" adler32(\"lo\", prev) = 0x%08lX\n", incremental)
|
||||
check("adler32 incremental matches one-shot", incremental == checksum, "incremental != one-shot")
|
||||
printf(" Incremental == One-shot: %s\n", incremental == "YES" if checksum else "NO")
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_crc32():
|
||||
section_header("crc32 checksum")
|
||||
|
||||
checksum: ULONG = pyzlib.zlib_crc32(BYTEPTR("Hello"), 5, 0)
|
||||
printf(" crc32(\"Hello\") = 0x%08lX (%lu)\n", checksum, checksum)
|
||||
check("crc32 returns non-zero", checksum != 0, "checksum is 0")
|
||||
|
||||
incremental: ULONG = pyzlib.zlib_crc32(BYTEPTR("Hel"), 3, 0)
|
||||
printf(" crc32(\"Hel\") = 0x%08lX\n", incremental)
|
||||
incremental = pyzlib.zlib_crc32(BYTEPTR("lo"), 2, incremental)
|
||||
printf(" crc32(\"lo\", prev) = 0x%08lX\n", incremental)
|
||||
check("crc32 incremental matches one-shot", incremental == checksum, "incremental != one-shot")
|
||||
printf(" Incremental == One-shot: %s\n", incremental == "YES" if checksum else "NO")
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_empty_compress():
|
||||
section_header("Empty data compress/decompress")
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(""), 0,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.MAX_WBITS, c.Addr(out_length))
|
||||
check("compress empty OK", compressed != None, "compress empty returned None")
|
||||
check("compress empty output > 0", out_length > 0, "compressed empty is 0 bytes")
|
||||
printf(" Empty input compressed: %zu bytes (header only)\n", out_length)
|
||||
printf(" Hex: ")
|
||||
print_hex_test(compressed, out_length, 32)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
decompressed: BYTEPTR = pyzlib.decompress(compressed, out_length,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("decompress empty OK", decompressed != None, "decompress empty returned None")
|
||||
check("decompress empty size == 0", dec_length == 0, "decompressed size != 0")
|
||||
printf(" Decompressed back: %zu bytes\n", dec_length)
|
||||
|
||||
free(compressed)
|
||||
free(decompressed)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_gzip_format():
|
||||
section_header("Gzip format (wbits=31)")
|
||||
|
||||
inp: str = "Gzip format test"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input: \"%s\" (%zu bytes)\n", inp, input_length)
|
||||
|
||||
gzip_wbits: t.CInt = pyzlib.MAX_WBITS + 16
|
||||
printf(" wbits = MAX_WBITS + 16 = %d\n", gzip_wbits)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, gzip_wbits, c.Addr(out_length))
|
||||
check("gzip compress OK", compressed != None, "gzip compress failed")
|
||||
printf(" Gzip compressed: %zu bytes\n", out_length)
|
||||
printf(" Header bytes: ")
|
||||
print_hex_test(compressed, 4 if (out_length > 4) else out_length, 4)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
decompressed: BYTEPTR = pyzlib.decompress(compressed, out_length,
|
||||
gzip_wbits, pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("gzip decompress OK", decompressed != None, "gzip decompress failed")
|
||||
check("gzip decompress size", dec_length == input_length, "size mismatch")
|
||||
check("gzip decompress content", decompressed != None and memcmp(decompressed, inp, input_length) == 0, "content mismatch")
|
||||
if decompressed: printf(" Decompressed: \"%.*s\" (%zu bytes)\n", t.CInt(dec_length), str(decompressed), dec_length)
|
||||
|
||||
free(compressed)
|
||||
free(decompressed)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_raw_deflate():
|
||||
section_header("Raw deflate (wbits=-15)")
|
||||
|
||||
inp: str = "Raw deflate test"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Input: \"%s\" (%zu bytes)\n", inp, input_length)
|
||||
|
||||
raw_wbits: t.CInt = -pyzlib.MAX_WBITS
|
||||
printf(" wbits = -MAX_WBITS = %d\n", raw_wbits)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
compressed: BYTEPTR = pyzlib.compress(BYTEPTR(inp), input_length,
|
||||
pyzlib.Z_DEFAULT_COMPRESSION, raw_wbits, c.Addr(out_length))
|
||||
check("raw deflate compress OK", compressed != None, "raw deflate compress failed")
|
||||
printf(" Raw compressed: %zu bytes\n", out_length)
|
||||
|
||||
dec_length: t.CSizeT = 0
|
||||
decompressed: BYTEPTR = pyzlib.decompress(compressed, out_length,
|
||||
raw_wbits, pyzlib.DEF_BUF_SIZE, c.Addr(dec_length))
|
||||
check("raw deflate decompress OK", decompressed != None, "raw deflate decompress failed")
|
||||
check("raw deflate decompress size", dec_length == input_length, "size mismatch")
|
||||
check("raw deflate decompress content", decompressed != None and memcmp(decompressed, inp, input_length) == 0, "content mismatch")
|
||||
if decompressed: printf(" Decompressed: \"%.*s\" (%zu bytes)\n", t.CInt(dec_length), str(decompressed), dec_length)
|
||||
|
||||
free(compressed)
|
||||
free(decompressed)
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_zdict():
|
||||
section_header("Preset dictionary (zdict)")
|
||||
|
||||
dict: str = "common words that appear frequently in the data"
|
||||
inp: str = "common words that appear frequently"
|
||||
input_length: t.CSizeT = strlen(inp)
|
||||
printf(" Dictionary: \"%s\" (%zu bytes)\n", dict, strlen(dict))
|
||||
printf(" Input : \"%s\" (%zu bytes)\n", inp, input_length)
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
s: pyzlib.Compress | t.CPtr = pyzlib.compressobj(pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.DEFLATED,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_MEM_LEVEL,
|
||||
pyzlib.Z_DEFAULT_STRATEGY,
|
||||
BYTEPTR(dict), strlen(dict))
|
||||
check("compressobj with zdict OK", s != None, "compressobj with zdict failed")
|
||||
|
||||
comp_data: BYTEPTR = s.compress(BYTEPTR(inp),
|
||||
input_length, c.Addr(out_length))
|
||||
check("compress with zdict OK", comp_data != None, "compress with zdict failed")
|
||||
printf(" Compressed with zdict: %zu bytes\n", out_length)
|
||||
free(comp_data)
|
||||
|
||||
flush_data: BYTEPTR = s.flush(pyzlib.Z_FINISH, c.Addr(out_length))
|
||||
check("flush with zdict OK", flush_data != None, "flush with zdict failed")
|
||||
printf(" Flush: %zu bytes\n", out_length)
|
||||
free(flush_data)
|
||||
|
||||
print_separator()
|
||||
|
||||
total_comp: t.CSizeT = out_length
|
||||
s.delete()
|
||||
|
||||
s = pyzlib.compressobj(pyzlib.Z_DEFAULT_COMPRESSION, pyzlib.DEFLATED,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_MEM_LEVEL,
|
||||
pyzlib.Z_DEFAULT_STRATEGY, None, 0)
|
||||
comp_no_dict: BYTEPTR = s.compress(BYTEPTR(inp),
|
||||
input_length, c.Addr(out_length))
|
||||
flush_no_dict: BYTEPTR = s.flush(pyzlib.Z_FINISH, c.Addr(out_length))
|
||||
no_dict_total: t.CSizeT = 0
|
||||
if comp_no_dict: no_dict_total += out_length
|
||||
if flush_no_dict: no_dict_total += out_length
|
||||
printf(" Without zdict: ~%zu bytes compressed\n", no_dict_total)
|
||||
printf(" With zdict : %zu bytes compressed\n", total_comp)
|
||||
free(comp_no_dict)
|
||||
free(flush_no_dict)
|
||||
s.delete()
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_huffman_tree():
|
||||
section_header("Huffman tree construction")
|
||||
|
||||
printf(" --- Fixed Huffman tree (literal/lengthgth) ---\n")
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
|
||||
check("fixed lit tree count == 288", lit_tree.count == 288, "count mismatch")
|
||||
check("fixed lit tree max_bits == 9", lit_tree.max_bits == 9, "max_bits mismatch")
|
||||
|
||||
has_8bit: t.CInt = 0
|
||||
has_9bit: t.CInt = 0
|
||||
has_7bit: t.CInt = 0
|
||||
for i in range(143 + 1):
|
||||
if lit_tree.codes[i].bits == 8:
|
||||
has_8bit += 1
|
||||
for i in range(143 + 1, 255 + 1):
|
||||
if lit_tree.codes[i].bits == 9:
|
||||
has_9bit += 1
|
||||
for i in range(255 + 1, 279 + 1):
|
||||
if lit_tree.codes[i].bits == 7:
|
||||
has_7bit += 1
|
||||
check("0-143 are 8-bit", has_8bit == 144, "wrong count")
|
||||
check("144-255 are 9-bit", has_9bit == 112, "wrong count")
|
||||
check("256-279 are 7-bit", has_7bit == 24, "wrong count")
|
||||
|
||||
printf(" 0-143: %d codes with 8 bits\n", has_8bit)
|
||||
printf(" 144-255: %d codes with 9 bits\n", has_9bit)
|
||||
printf(" 256-279: %d codes with 7 bits\n", has_7bit)
|
||||
printf(" 280-287: 8-bit (end-of-block 256 = 7-bit)\n")
|
||||
printf(" EOB (256): code=0x%X, bits=%d\n", lit_tree.codes[256].code, lit_tree.codes[256].bits)
|
||||
|
||||
printf("\n --- Fixed Huffman tree (distance) ---\n")
|
||||
dist_tree = zhuff.zhuff_tree()
|
||||
dist_tree.build_fixed_dist_tree()
|
||||
|
||||
check("fixed dist tree count == 32", dist_tree.count == 32, "count mismatch")
|
||||
check("fixed dist tree max_bits == 5", dist_tree.max_bits == 5, "max_bits mismatch")
|
||||
|
||||
all_5bit: t.CInt = 1
|
||||
for i in range(32):
|
||||
if dist_tree.codes[i].bits != 5:
|
||||
all_5bit = 0
|
||||
check("all 32 distance codes are 5-bit", all_5bit, "not all 5-bit")
|
||||
printf(" All 32 distance codes: 5 bits each\n")
|
||||
|
||||
printf("\n --- Dynamic Huffman tree from frequencies ---\n")
|
||||
freqs: list[t.CInt, 256]
|
||||
memset(freqs, 0, freqs.__sizeof__())
|
||||
freqs['A'] = 50
|
||||
freqs['B'] = 25
|
||||
freqs['C'] = 12
|
||||
freqs['D'] = 6
|
||||
freqs['E'] = 3
|
||||
freqs['F'] = 1
|
||||
freqs[256] = 1
|
||||
|
||||
dyn_tree = zhuff.zhuff_tree()
|
||||
dyn_tree.build_codes(freqs, 257, 15)
|
||||
|
||||
check("dynamic tree count == 257", dyn_tree.count == 257, "count mismatch")
|
||||
check("dynamic tree max_bits <= 15", dyn_tree.max_bits <= 15, "max_bits overflow")
|
||||
|
||||
total_codes: t.CInt = 0
|
||||
max_length_found: t.CInt = 0
|
||||
for i in range(257):
|
||||
if dyn_tree.codes[i].bits > 0:
|
||||
total_codes += 1
|
||||
if dyn_tree.codes[i].bits > max_length_found:
|
||||
max_length_found = dyn_tree.codes[i].bits
|
||||
check("dynamic tree has 7 active codes", total_codes == 7, "wrong active count")
|
||||
check("dynamic tree max code lengthgth found <= 15", max_length_found <= 15, "code lengthgth overflow")
|
||||
|
||||
printf(" Frequencies: A=50, B=25, C=12, D=6, E=3, F=1, EOB=1\n")
|
||||
printf(" Active codes: %d, max code lengthgth: %d\n", total_codes, max_length_found)
|
||||
|
||||
printf(" Code assignments:\n")
|
||||
names: list[str, None] = ["A","B","C","D","E","F"] # ä¸ä¸ªé½æ?char* çæ°ç»?
|
||||
syms: list[t.CInt, None] = ['A','B','C','D','E','F']
|
||||
for i in range(6):
|
||||
printf(" '%s' (freq=%3d): code=0x%04X, bits=%d\n",
|
||||
names[i], freqs[syms[i]],
|
||||
dyn_tree.codes[syms[i]].code,
|
||||
dyn_tree.codes[syms[i]].bits)
|
||||
|
||||
|
||||
|
||||
printf(" EOB (freq=1): code=0x%04X, bits=%d\n",
|
||||
dyn_tree.codes[256].code, dyn_tree.codes[256].bits)
|
||||
a_bits: t.CInt = dyn_tree.codes['A'].bits
|
||||
f_bits: t.CInt = dyn_tree.codes['F'].bits
|
||||
check("high-freq symbol has shorter code", a_bits <= f_bits, "A should be <= F")
|
||||
printf(" High-freq 'A' (%d bits) <= low-freq 'F' (%d bits): %s\n",
|
||||
a_bits, f_bits, "YES" if (a_bits <= f_bits) else "NO")
|
||||
|
||||
printf("\n --- Huffman encode/decode roundtrip ---\n")
|
||||
freqs: list[t.CInt, 288]
|
||||
memset(freqs, 0, freqs.__sizeof__())
|
||||
freqs['H'] = 10
|
||||
freqs['e'] = 8
|
||||
freqs['l'] = 20
|
||||
freqs['o'] = 8
|
||||
freqs[' '] = 5
|
||||
freqs['W'] = 3
|
||||
freqs['r'] = 5
|
||||
freqs['d'] = 3
|
||||
freqs['!'] = 2
|
||||
freqs[256] = 1
|
||||
|
||||
enc_tree = zhuff.zhuff_tree()
|
||||
enc_tree.build_codes(freqs, 257, 15)
|
||||
|
||||
dec_tree = zhuff.zhuff_decode_tree()
|
||||
dec_tree.build_decode_tree(c.Addr(enc_tree))
|
||||
|
||||
writer = zdef.zbit_writer()
|
||||
|
||||
symbols: list[t.CInt, 16]
|
||||
symbols[0] = 'H'; symbols[1] = 'e'; symbols[2] = 'l'; symbols[3] = 'l'
|
||||
symbols[4] = 'o'; symbols[5] = ' '; symbols[6] = 'W'; symbols[7] = 'o'
|
||||
symbols[8] = 'r'; symbols[9] = 'l'; symbols[10] = 'd'; symbols[11] = '!'
|
||||
symbols[12] = 256
|
||||
num_symbols: t.CInt = 13
|
||||
|
||||
for i in range(num_symbols):
|
||||
enc_tree.encode_symbol(symbols[i], c.Addr(writer))
|
||||
|
||||
reader = zdef.zbit_reader()
|
||||
reader.buf = writer.buf
|
||||
reader.length = writer.byte_pos + (1 if (writer.bit_pos > 0) else 0)
|
||||
reader.byte_pos = 0
|
||||
reader.bit_pos = 0
|
||||
|
||||
roundtrip_ok: t.CInt = 1
|
||||
for i in range(num_symbols):
|
||||
decoded: t.CInt = dec_tree.decode_symbol(c.Addr(reader))
|
||||
if decoded != symbols[i]:
|
||||
roundtrip_ok = 0
|
||||
printf(" MISMATCH at symbol %d: expected %d, got %d\n", i, symbols[i], decoded)
|
||||
break
|
||||
check("encode/decode roundtrip matches", roundtrip_ok, "roundtrip failed")
|
||||
printf(" Encoded %d symbols into %zu bytes, decoded all correctly\n",
|
||||
num_symbols, writer.byte_pos + (1 if (writer.bit_pos > 0) else 0))
|
||||
|
||||
free(writer.buf)
|
||||
|
||||
printf("\n --- Overflow handling (many symbols, limited max_bits) ---\n")
|
||||
freqs: list[t.CInt, 256]
|
||||
for i in range(256):
|
||||
freqs[i] = 1
|
||||
freqs[256] = 1
|
||||
|
||||
tree = zhuff.zhuff_tree()
|
||||
tree.build_codes(freqs, 257, 15)
|
||||
|
||||
overflow_ok: t.CInt = 1
|
||||
for i in range(257):
|
||||
if tree.codes[i].bits > 15 or tree.codes[i].bits < 0:
|
||||
overflow_ok = 0
|
||||
break
|
||||
check("all code lengthgths <= 15 with 257 symbols", overflow_ok, "code lengthgth overflow")
|
||||
|
||||
bl_count: list[t.CInt, 16] = [0]
|
||||
for i in range(257):
|
||||
if tree.codes[i].bits > 0:
|
||||
bl_count[tree.codes[i].bits] += 1
|
||||
printf(" Code lengthgth distribution (257 equal-freq symbols, max_bits=15):\n")
|
||||
for i in range(1, 15 + 1):
|
||||
if bl_count[i] > 0:
|
||||
printf(" %2d bits: %3d codes\n", i, bl_count[i])
|
||||
|
||||
dt = zhuff.zhuff_decode_tree()
|
||||
dt.build_decode_tree(c.Addr(tree))
|
||||
|
||||
w = zdef.zbit_writer()
|
||||
tree.encode_symbol(0, c.Addr(w))
|
||||
tree.encode_symbol(128, c.Addr(w))
|
||||
tree.encode_symbol(255, c.Addr(w))
|
||||
tree.encode_symbol(256, c.Addr(w))
|
||||
|
||||
r = zdef.zbit_reader()
|
||||
r.buf = w.buf
|
||||
r.length = w.byte_pos + (1 if (w.bit_pos > 0) else 0)
|
||||
r.byte_pos = 0
|
||||
r.bit_pos = 0
|
||||
|
||||
s0: t.CInt = dt.decode_symbol(c.Addr(r))
|
||||
s1: t.CInt = dt.decode_symbol(c.Addr(r))
|
||||
s2: t.CInt = dt.decode_symbol(c.Addr(r))
|
||||
s3: t.CInt = dt.decode_symbol(c.Addr(r))
|
||||
|
||||
check("overflow roundtrip: symbol 0", s0 == 0, "decode mismatch")
|
||||
check("overflow roundtrip: symbol 128", s1 == 128, "decode mismatch")
|
||||
check("overflow roundtrip: symbol 255", s2 == 255, "decode mismatch")
|
||||
check("overflow roundtrip: symbol 256 (EOB)", s3 == 256, "decode mismatch")
|
||||
|
||||
free(w.buf)
|
||||
|
||||
|
||||
printf("\n --- Kraft inequality check ---\n")
|
||||
freqs: list[t.CInt, 256]
|
||||
memset(freqs, 0, freqs.__sizeof__())
|
||||
freqs['X'] = 100
|
||||
freqs['Y'] = 50
|
||||
freqs['Z'] = 25
|
||||
freqs[256] = 1
|
||||
|
||||
tree = zhuff.zhuff_tree()
|
||||
tree.build_codes(freqs, 257, 15)
|
||||
|
||||
kraft_sum: t.CDouble = 0.0
|
||||
for i in range(257):
|
||||
if tree.codes[i].bits > 0:
|
||||
kraft_sum += 1.0 / (t.CDouble(ULONGLONG(1) << tree.codes[i].bits))
|
||||
check("Kraft inequality: sum <= 1.0", kraft_sum <= 1.0 + 1e-9, "Kraft violated")
|
||||
printf(" Kraft sum = %.10f (must be <= 1.0)\n", kraft_sum)
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_error_handling():
|
||||
section_header("Error handling")
|
||||
|
||||
out_length: t.CSizeT = 0
|
||||
printf(" Trying to decompress garbage data...\n")
|
||||
|
||||
result: BYTEPTR = pyzlib.decompress(BYTEPTR("garbage"), 7,
|
||||
pyzlib.MAX_WBITS, pyzlib.DEF_BUF_SIZE, c.Addr(out_length))
|
||||
check("decompress garbage returns None", result == None, "should have returned None")
|
||||
printf(" Error code : %d\n", pyzlib.get_error_code())
|
||||
printf(" Error message: \"%s\"\n", pyzlib.get_error())
|
||||
check("error message set", strlen(pyzlib.get_error()) > 0, "error message is empty")
|
||||
check("error code is set", pyzlib.get_error_code() != pyzlib.Z_OK, "error code is Z_OK")
|
||||
|
||||
print_separator()
|
||||
|
||||
pyzlib.clear_error()
|
||||
check("clear error clears code", pyzlib.get_error_code() == 0, "error code not cleared")
|
||||
printf(" After clear: code=%d, msg=\"%s\"\n", pyzlib.get_error_code(), pyzlib.get_error())
|
||||
|
||||
section_footer()
|
||||
|
||||
|
||||
def test_constants():
|
||||
section_header("Constants")
|
||||
printf(" Compression Levels:\n")
|
||||
printf(" Z_NO_COMPRESSION = %d\n", pyzlib.Z_NO_COMPRESSION)
|
||||
printf(" Z_BEST_SPEED = %d\n", pyzlib.Z_BEST_SPEED)
|
||||
printf(" Z_BEST_COMPRESSION = %d\n", pyzlib.Z_BEST_COMPRESSION)
|
||||
printf(" Z_DEFAULT_COMPRESSION = %d\n", pyzlib.Z_DEFAULT_COMPRESSION)
|
||||
printf(" Methods:\n")
|
||||
printf(" DEFLATED = %d\n", pyzlib.DEFLATED)
|
||||
printf(" Flush Modes:\n")
|
||||
printf(" Z_NO_FLUSH = %d\n", pyzlib.Z_NO_FLUSH)
|
||||
printf(" Z_PARTIAL_FLUSH = %d\n", pyzlib.Z_PARTIAL_FLUSH)
|
||||
printf(" Z_SYNC_FLUSH = %d\n", pyzlib.Z_SYNC_FLUSH)
|
||||
printf(" Z_FULL_FLUSH = %d\n", pyzlib.Z_FULL_FLUSH)
|
||||
printf(" Z_FINISH = %d\n", pyzlib.Z_FINISH)
|
||||
printf(" Z_BLOCK = %d\n", pyzlib.Z_BLOCK)
|
||||
printf(" Z_TREES = %d\n", pyzlib.Z_TREES)
|
||||
printf(" Strategies:\n")
|
||||
printf(" Z_DEFAULT_STRATEGY = %d\n", pyzlib.Z_DEFAULT_STRATEGY)
|
||||
printf(" Z_FILTERED = %d\n", pyzlib.Z_FILTERED)
|
||||
printf(" Z_HUFFMAN_ONLY = %d\n", pyzlib.Z_HUFFMAN_ONLY)
|
||||
printf(" Z_RLE = %d\n", pyzlib.Z_RLE)
|
||||
printf(" Z_FIXED = %d\n", pyzlib.Z_FIXED)
|
||||
printf(" Return Codes:\n")
|
||||
printf(" Z_OK = %d\n", pyzlib.Z_OK)
|
||||
printf(" Z_STREAM_END = %d\n", pyzlib.Z_STREAM_END)
|
||||
printf(" Z_NEED_DICT = %d\n", pyzlib.Z_NEED_DICT)
|
||||
printf(" Z_ERRNO = %d\n", pyzlib.Z_ERRNO)
|
||||
printf(" Z_STREAM_ERROR = %d\n", pyzlib.Z_STREAM_ERROR)
|
||||
printf(" Z_DATA_ERROR = %d\n", pyzlib.Z_DATA_ERROR)
|
||||
printf(" Z_MEM_ERROR = %d\n", pyzlib.Z_MEM_ERROR)
|
||||
printf(" Z_BUF_ERROR = %d\n", pyzlib.Z_BUF_ERROR)
|
||||
printf(" Z_VERSION_ERROR = %d\n", pyzlib.Z_VERSION_ERROR)
|
||||
printf(" Window / Buffer:\n")
|
||||
printf(" MAX_WBITS = %d\n", pyzlib.MAX_WBITS)
|
||||
printf(" DEF_BUF_SIZE = %d\n", pyzlib.DEF_BUF_SIZE)
|
||||
printf(" DEF_MEM_LEVEL = %d\n", pyzlib.DEF_MEM_LEVEL)
|
||||
section_footer()
|
||||
|
||||
|
||||
def main123456() -> t.CInt:
|
||||
printf("==============================================\n")
|
||||
printf(" Python zlib - C Implementation Tests\n")
|
||||
printf("==============================================\n")
|
||||
|
||||
test_constants()
|
||||
test_version()
|
||||
test_compress_decompress()
|
||||
test_compress_levels()
|
||||
test_compressobj()
|
||||
test_decompressobj()
|
||||
test_decompressobj_max_lengthgth()
|
||||
test_decompressobj_unused_data()
|
||||
test_compress_copy()
|
||||
test_decompress_copy()
|
||||
test_adler32()
|
||||
test_crc32()
|
||||
test_empty_compress()
|
||||
test_gzip_format()
|
||||
test_raw_deflate()
|
||||
test_zdict()
|
||||
test_huffman_tree()
|
||||
test_error_handling()
|
||||
|
||||
printf("\n==============================================\n")
|
||||
printf(" Results: %d passed, %d failed\n", test_passed, test_failed)
|
||||
printf("==============================================\n")
|
||||
|
||||
return 1 if test_failed > 0 else 0
|
||||
|
||||
@@ -1,90 +0,0 @@
|
||||
from stdint import *
|
||||
import t, c
|
||||
|
||||
|
||||
def zchecksum_adler32(data: UINT8PTR, length: t.CSizeT, init: UINT32) -> t.CUInt32T:
|
||||
a: t.CUInt32T = (init >> 0) & 0xFFFF
|
||||
b: t.CUInt32T = (init >> 16) & 0xFFFF
|
||||
if not data or length == 0:
|
||||
return init
|
||||
i: t.CSizeT
|
||||
for i in range(length):
|
||||
a = (a + data[i]) % 65521
|
||||
b = (b + a) % 65521
|
||||
return (b << 16) | a
|
||||
|
||||
crc32_table: list[t.CUInt32T, 256] = [
|
||||
0x00000000, 0x77073096, 0xEE0E612C, 0x990951BA,
|
||||
0x076DC419, 0x706AF48F, 0xE963A535, 0x9E6495A3,
|
||||
0x0EDB8832, 0x79DCB8A4, 0xE0D5E91E, 0x97D2D988,
|
||||
0x09B64C2B, 0x7EB17CBD, 0xE7B82D07, 0x90BF1D91,
|
||||
0x1DB71064, 0x6AB020F2, 0xF3B97148, 0x84BE41DE,
|
||||
0x1ADAD47D, 0x6DDDE4EB, 0xF4D4B551, 0x83D385C7,
|
||||
0x136C9856, 0x646BA8C0, 0xFD62F97A, 0x8A65C9EC,
|
||||
0x14015C4F, 0x63066CD9, 0xFA0F3D63, 0x8D080DF5,
|
||||
0x3B6E20C8, 0x4C69105E, 0xD56041E4, 0xA2677172,
|
||||
0x3C03E4D1, 0x4B04D447, 0xD20D85FD, 0xA50AB56B,
|
||||
0x35B5A8FA, 0x42B2986C, 0xDBBBBBD6, 0xACBCCB40,
|
||||
0x32D86CE3, 0x45DF5C75, 0xDCD60DCF, 0xABD13D59,
|
||||
0x26D930AC, 0x51DE003A, 0xC8D75180, 0xBFD06116,
|
||||
0x21B4F4B5, 0x56B3C423, 0xCFBA9599, 0xB8BDA50F,
|
||||
0x2802B89E, 0x5F058808, 0xC60CD9B2, 0xB10BE924,
|
||||
0x2F6F7C87, 0x58684C11, 0xC1611DAB, 0xB6662D3D,
|
||||
0x76DC4190, 0x01DB7106, 0x98D220BC, 0xEFD5102A,
|
||||
0x71B18589, 0x06B6B51F, 0x9FBFE4A5, 0xE8B8D433,
|
||||
0x7807C9A2, 0x0F00F934, 0x9609A88E, 0xE10E9818,
|
||||
0x7F6A0DBB, 0x086D3D2D, 0x91646C97, 0xE6635C01,
|
||||
0x6B6B51F4, 0x1C6C6162, 0x856530D8, 0xF262004E,
|
||||
0x6C0695ED, 0x1B01A57B, 0x8208F4C1, 0xF50FC457,
|
||||
0x65B0D9C6, 0x12B7E950, 0x8BBEB8EA, 0xFCB9887C,
|
||||
0x62DD1DDF, 0x15DA2D49, 0x8CD37CF3, 0xFBD44C65,
|
||||
0x4DB26158, 0x3AB551CE, 0xA3BC0074, 0xD4BB30E2,
|
||||
0x4ADFA541, 0x3DD895D7, 0xA4D1C46D, 0xD3D6F4FB,
|
||||
0x4369E96A, 0x346ED9FC, 0xAD678846, 0xDA60B8D0,
|
||||
0x44042D73, 0x33031DE5, 0xAA0A4C5F, 0xDD0D7D49,
|
||||
0x5005713C, 0x270241AA, 0xBE0B1010, 0xC90C2086,
|
||||
0x5768B525, 0x206F85B3, 0xB966D409, 0xCE61E49F,
|
||||
0x5EDEF90E, 0x29D9C998, 0xB0D09822, 0xC7D7A8B4,
|
||||
0x59B33D17, 0x2EB40D81, 0xB7BD5C3B, 0xC0BA6CAD,
|
||||
0xEDB88320, 0x9ABFB3B6, 0x03B6E20C, 0x74B1D29A,
|
||||
0xEAD54739, 0x9DD277AF, 0x04DB2615, 0x73DC1683,
|
||||
0xE3630B12, 0x94643B84, 0x0D6D6A3E, 0x7A6A5AA8,
|
||||
0xE40ECF0B, 0x9309FF9D, 0x0A00AE27, 0x7D079EB1,
|
||||
0xF00F9344, 0x8708A3D2, 0x1E01F268, 0x6906C2FE,
|
||||
0xF762575D, 0x806567CB, 0x196C3671, 0x6E6B06E7,
|
||||
0xFED41B76, 0x89D32BE0, 0x10DA7A5A, 0x67DD4ACC,
|
||||
0xF9B9DF6F, 0x8EBEEFF9, 0x17B7BE43, 0x60B08ED5,
|
||||
0xD6D6A3E8, 0xA1D1937E, 0x38D8C2C4, 0x4FDFF252,
|
||||
0xD1BB67F1, 0xA6BC5767, 0x3FB506DD, 0x48B2364B,
|
||||
0xD80D2BDA, 0xAF0A1B4C, 0x36034AF6, 0x41047A60,
|
||||
0xDF60EFC3, 0xA867DF55, 0x316E8EEF, 0x4669BE79,
|
||||
0xCB61B38C, 0xBC66831A, 0x256FD2A0, 0x5268E236,
|
||||
0xCC0C7795, 0xBB0B4703, 0x220216B9, 0x5505262F,
|
||||
0xC5BA3BBE, 0xB2BD0B28, 0x2BB45A92, 0x5CB36A04,
|
||||
0xC2D7FFA7, 0xB5D0CF31, 0x2CD99E8B, 0x5BDEAE1D,
|
||||
0x9B64C2B0, 0xEC63F226, 0x756AA39C, 0x026D930A,
|
||||
0x9C0906A9, 0xEB0E363F, 0x72076785, 0x05005713,
|
||||
0x95BF4A82, 0xE2B87A14, 0x7BB12BAE, 0x0CB61B38,
|
||||
0x92D28E9B, 0xE5D5BE0D, 0x7CDCEFB7, 0x0BDBDF21,
|
||||
0x86D3D2D4, 0xF1D4E242, 0x68DDB3F8, 0x1FDA836E,
|
||||
0x81BE16CD, 0xF6B9265B, 0x6FB077E1, 0x18B74777,
|
||||
0x88085AE6, 0xFF0F6A70, 0x66063BCA, 0x11010B5C,
|
||||
0x8F659EFF, 0xF862AE69, 0x616BFFD3, 0x166CCF45,
|
||||
0xA00AE278, 0xD70DD2EE, 0x4E048354, 0x3903B3C2,
|
||||
0xA7672661, 0xD06016F7, 0x4969474D, 0x3E6E77DB,
|
||||
0xAED16A4A, 0xD9D65ADC, 0x40DF0B66, 0x37D83BF0,
|
||||
0xA9BCAE53, 0xDede9EC5, 0x47D7897F, 0x30D0B8E9,
|
||||
0xBDDA8B1C, 0xCADD8B8A, 0x53D3903A, 0x24D4C2AC,
|
||||
0xBAD03605, 0xCDD70693, 0x54DE5729, 0x23D967BF,
|
||||
0xB3667A2E, 0xC4614AB8, 0x5D681B02, 0x2A6F2B94,
|
||||
0xB40BBE37, 0xC30C8EA1, 0x5A05DF1B, 0x2D02EF8D
|
||||
]
|
||||
|
||||
def zchecksum_crc32(data: UINT8PTR, length: t.CSizeT, init: t.CUInt32T) -> t.CUInt32T:
|
||||
crc: t.CUInt32T = init ^ 0xFFFFFFFF
|
||||
if not data or length == 0: return init
|
||||
i: t.CSizeT
|
||||
for i in range(length):
|
||||
crc = crc32_table[(crc ^ data[i]) & 0xFF] ^ (crc >> 8)
|
||||
return crc ^ 0xFFFFFFFF
|
||||
|
||||
@@ -1,269 +0,0 @@
|
||||
from stdint import *
|
||||
import stddef
|
||||
import string
|
||||
import stdlib
|
||||
import t, c
|
||||
|
||||
|
||||
# ============================================================
|
||||
# DEFLATE / zlib constants
|
||||
# ============================================================
|
||||
ZDEFLATE_WINDOW_BITS: t.CDefine = 15
|
||||
ZDEFLATE_WINDOW_SIZE: t.CDefine = (1 << ZDEFLATE_WINDOW_BITS)
|
||||
ZDEFLATE_WINDOW_MASK: t.CDefine = (ZDEFLATE_WINDOW_SIZE - 1)
|
||||
|
||||
ZDEFLATE_MIN_MATCH: t.CDefine = 3
|
||||
ZDEFLATE_MAX_MATCH: t.CDefine = 258
|
||||
|
||||
ZDEFLATE_HASH_BITS: t.CDefine = 15
|
||||
ZDEFLATE_HASH_SIZE: t.CDefine = (1 << ZDEFLATE_HASH_BITS)
|
||||
ZDEFLATE_HASH_MASK: t.CDefine = (ZDEFLATE_HASH_SIZE - 1)
|
||||
|
||||
ZDEFLATE_MAX_CODES: t.CDefine = 288
|
||||
ZDEFLATE_MAX_DIST_CODES: t.CDefine = 32
|
||||
ZDEFLATE_MAX_BITS: t.CDefine = 15
|
||||
ZDEFLATE_MAX_CODELEN_CODES: t.CDefine = 19
|
||||
|
||||
ZDEFLATE_LIT_COUNT: t.CDefine = 286
|
||||
ZDEFLATE_DIST_COUNT: t.CDefine = 30
|
||||
|
||||
ZDEFLATE_LEN_SYMBOLS_BASE: t.CDefine = 257
|
||||
ZDEFLATE_END_OF_BLOCK: t.CDefine = 256
|
||||
|
||||
# ============================================================
|
||||
# Length / Distance extra bits tables (RFC 1951)
|
||||
# ============================================================
|
||||
zdeflate_len_extra_bits: list[t.CInt, 29] = [
|
||||
0, 0, 0, 0, 0, 0, 0, 0,
|
||||
1, 1, 1, 1,
|
||||
2, 2, 2, 2,
|
||||
3, 3, 3, 3,
|
||||
4, 4, 4, 4,
|
||||
5, 5, 5, 5,
|
||||
0
|
||||
]
|
||||
|
||||
zdeflate_len_base: list[t.CInt, 29] = [
|
||||
3, 4, 5, 6, 7, 8, 9, 10,
|
||||
11, 13, 15, 17,
|
||||
19, 23, 27, 31,
|
||||
35, 43, 51, 59,
|
||||
67, 83, 99, 115,
|
||||
131, 163, 195, 227,
|
||||
258
|
||||
]
|
||||
|
||||
zdeflate_dist_extra_bits: list[t.CInt, 30] = [
|
||||
0, 0, 0, 0,
|
||||
1, 1,
|
||||
2, 2,
|
||||
3, 3,
|
||||
4, 4,
|
||||
5, 5,
|
||||
6, 6,
|
||||
7, 7,
|
||||
8, 8,
|
||||
9, 9,
|
||||
10, 10,
|
||||
11, 11,
|
||||
12, 12,
|
||||
13, 13
|
||||
]
|
||||
|
||||
zdeflate_dist_base: list[t.CInt, 30] = [
|
||||
1, 2, 3, 4,
|
||||
5, 7,
|
||||
9, 13,
|
||||
17, 25,
|
||||
33, 49,
|
||||
65, 97,
|
||||
129, 193,
|
||||
257, 385,
|
||||
513, 769,
|
||||
1025, 1537,
|
||||
2049, 3073,
|
||||
4097, 6145,
|
||||
8193, 12289,
|
||||
16385, 24577
|
||||
]
|
||||
|
||||
zdeflate_codelen_order: list[int, 19] = [
|
||||
16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15
|
||||
]
|
||||
|
||||
# ============================================================
|
||||
# Bit writer - writes bits LSB first into a byte buffer
|
||||
# ============================================================
|
||||
@t.Object
|
||||
class zbit_writer:
|
||||
buf: BYTE | t.CPtr
|
||||
cap: t.CSizeT
|
||||
byte_pos: t.CSizeT
|
||||
bit_pos: t.CInt
|
||||
|
||||
def __init__(self):
|
||||
self.cap = 4096
|
||||
self.buf = BYTEPTR(malloc(self.cap))
|
||||
memset(self.buf, 0, self.cap)
|
||||
self.byte_pos = 0
|
||||
self.bit_pos = 0
|
||||
|
||||
def ensure(self, need: t.CSizeT):
|
||||
while self.byte_pos + need + 4 >= self.cap:
|
||||
new_cap: t.CSizeT = self.cap * 2
|
||||
new_buf: BYTE | t.CPtr = BYTEPTR(realloc(self.buf, new_cap))
|
||||
if not new_buf: return
|
||||
memset(new_buf + self.cap, 0, new_cap - self.cap)
|
||||
self.buf = new_buf
|
||||
self.cap = new_cap
|
||||
|
||||
def write_bits(self, value: UINT, nbits: t.CInt):
|
||||
self.ensure((nbits + 7) / 8 + 1)
|
||||
for i in range(nbits):
|
||||
if value & (UINT(1) << i):
|
||||
self.buf[self.byte_pos] |= (UINT(1) << self.bit_pos)
|
||||
self.bit_pos += 1
|
||||
if self.bit_pos == 8:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
|
||||
def write_bits_rev(self, code: UINT, nbits: t.CInt):
|
||||
self.ensure((nbits + 7) / 8 + 1)
|
||||
for i in range(nbits - 1, -1, -1):
|
||||
if code & (UINT(1) << i):
|
||||
self.buf[self.byte_pos] |= (UINT(1) << self.bit_pos)
|
||||
self.bit_pos += 1
|
||||
if self.bit_pos == 8:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
|
||||
def stored_block(self, data: UINT8PTR, length: t.CSizeT, final: t.CInt):
|
||||
self.write_bits(1 if final else 0, 1)
|
||||
self.write_bits(0, 2)
|
||||
self.align()
|
||||
|
||||
pos: t.CSizeT = 0
|
||||
while pos < length:
|
||||
block_length: t.CSizeT = length - pos
|
||||
if block_length > 65535:
|
||||
block_length = 65535
|
||||
n: t.CUInt16T = t.CUInt16T(block_length)
|
||||
ncomp: t.CUInt16T = ~n
|
||||
self.write_bits(n, 16)
|
||||
self.write_bits(ncomp, 16)
|
||||
|
||||
self.ensure(block_length)
|
||||
i: t.CSizeT = 0
|
||||
for i in range(block_length):
|
||||
self.buf[self.byte_pos] = data[pos + i]
|
||||
self.byte_pos += 1
|
||||
pos += block_length
|
||||
if pos < length:
|
||||
self.write_bits(0, 1)
|
||||
self.write_bits(0, 2)
|
||||
self.align()
|
||||
|
||||
def write_zlib_header(self, wbits: t.CInt, level: t.CInt):
|
||||
cinfo: t.CInt = wbits - 8
|
||||
if cinfo < 1: cinfo = 1
|
||||
if cinfo > 7: cinfo = 7
|
||||
cmf: t.CUnsignedChar = t.CUnsignedChar((cinfo << 4) | 8)
|
||||
|
||||
flevel: t.CInt = 0
|
||||
if level >= 5: flevel = 3
|
||||
elif level >= 3: flevel = 1
|
||||
|
||||
flg: t.CUnsignedChar = t.CUnsignedChar(flevel << 6)
|
||||
flg |= t.CUnsignedChar(31 - (cmf * 256 + flg) % 31)
|
||||
|
||||
self.buf[self.byte_pos + 0] = cmf
|
||||
self.buf[self.byte_pos + 1] = flg
|
||||
self.byte_pos += 2
|
||||
|
||||
|
||||
def write_gzip_header(self):
|
||||
self.buf[self.byte_pos + 0] = 0x1F
|
||||
self.buf[self.byte_pos + 1] = 0x8B
|
||||
self.buf[self.byte_pos + 2] = 0x08
|
||||
self.buf[self.byte_pos + 3] = 0x00
|
||||
self.buf[self.byte_pos + 4] = 0x00
|
||||
self.buf[self.byte_pos + 5] = 0x00
|
||||
self.buf[self.byte_pos + 6] = 0x00
|
||||
self.buf[self.byte_pos + 7] = 0x00
|
||||
self.buf[self.byte_pos + 8] = 0x00
|
||||
self.buf[self.byte_pos + 9] = 0xFF
|
||||
self.byte_pos += 10
|
||||
|
||||
def align(self):
|
||||
if self.bit_pos > 0:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
|
||||
def total(self) -> t.CSizeT:
|
||||
return self.byte_pos + (1 if (self.bit_pos > 0) else 0)
|
||||
|
||||
def free(self):
|
||||
free(self.buf)
|
||||
self.buf = None
|
||||
self.cap = 0
|
||||
self.byte_pos = 0
|
||||
self.bit_pos = 0
|
||||
|
||||
def __del__(self):
|
||||
self.free()
|
||||
|
||||
# ============================================================
|
||||
# Bit reader - reads bits LSB first from a byte buffer
|
||||
# ============================================================
|
||||
@t.Object
|
||||
class zbit_reader:
|
||||
buf: BYTE | t.CPtr
|
||||
length: t.CSizeT
|
||||
byte_pos: t.CSizeT
|
||||
bit_pos: t.CInt
|
||||
|
||||
def init(self, buf: BYTE | t.CPtr, length: t.CSizeT):
|
||||
self.buf = buf
|
||||
self.length = length
|
||||
self.byte_pos = 0
|
||||
self.bit_pos = 0
|
||||
|
||||
def read_bits(self, nbits: t.CInt, out: UINTPTR) -> t.CInt:
|
||||
val: UINT = 0
|
||||
for i in range(nbits):
|
||||
if self.byte_pos >= self.length: return -1
|
||||
if self.buf[self.byte_pos] & (UINT(1) << self.bit_pos):
|
||||
val |= (UINT(1) << i)
|
||||
self.bit_pos += 1
|
||||
if self.bit_pos == 8:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
c.Set(c.Deref(out), val)
|
||||
return 0
|
||||
|
||||
def read_bits_rev(self, nbits: t.CInt, out: UINTPTR) -> t.CInt:
|
||||
val: UINT = 0
|
||||
for i in range(nbits - 1, 0, -1):
|
||||
if self.byte_pos >= self.length: return -1
|
||||
if self.buf[self.byte_pos] & (UINT(1) << self.bit_pos):
|
||||
val |= (UINT(1) << i)
|
||||
self.bit_pos += 1
|
||||
if self.bit_pos == 8:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
c.Set(c.Deref(out), val)
|
||||
return 0
|
||||
|
||||
def align(self):
|
||||
if self.bit_pos > 0:
|
||||
self.bit_pos = 0
|
||||
self.byte_pos += 1
|
||||
|
||||
# ============================================================
|
||||
# Memory allocation helper for bare metal
|
||||
# Can be replaced with custom allocator
|
||||
# ============================================================
|
||||
def zdef_alloc(n: t.CSizeT) -> VOIDPTR:
|
||||
return malloc(n)
|
||||
def zdef_free(p: VOIDPTR):
|
||||
free(p)
|
||||
@@ -1,431 +0,0 @@
|
||||
from stdint import *
|
||||
import zchecksum
|
||||
import zhuff
|
||||
import zdef
|
||||
import string
|
||||
import stdio
|
||||
import t, c
|
||||
|
||||
|
||||
@t.Object
|
||||
class zdeflate_stream:
|
||||
writer: zdef.zbit_writer
|
||||
window: BYTE | t.CPtr
|
||||
window_pos: t.CInt
|
||||
window_size: t.CInt
|
||||
hash_head: t.CInt | t.CPtr
|
||||
hash_prev: t.CInt | t.CPtr
|
||||
level: t.CInt
|
||||
strategy: t.CInt
|
||||
wbits: t.CInt
|
||||
is_finished: t.CInt
|
||||
adler: t.CUInt32T
|
||||
|
||||
def find_match(self, data: BYTE | t.CPtr, data_len: t.CSizeT,
|
||||
pos: t.CSizeT, best_dist: t.CInt | t.CPtr) -> t.CInt:
|
||||
if pos + 2 >= data_len: return 0
|
||||
h: t.CUnsignedInt = zdeflate_hash(data + pos)
|
||||
chain: t.CInt = self.hash_head[h]
|
||||
best_len: t.CInt = zdef.ZDEFLATE_MIN_MATCH - 1
|
||||
c.Set(c.Deref(best_dist), 0)
|
||||
max_chain: t.CInt = 128 if (self.level >= 5) else (32 if (self.level >= 2) else 4)
|
||||
limit: t.CInt = (1 << self.wbits) if (self.wbits > 0) else zdef.ZDEFLATE_WINDOW_SIZE
|
||||
if limit > zdef.ZDEFLATE_WINDOW_SIZE:
|
||||
limit = zdef.ZDEFLATE_WINDOW_SIZE
|
||||
attempts: t.CInt = 0
|
||||
while chain >= 0 and attempts < max_chain:
|
||||
dist: t.CInt = t.CInt(pos) - chain
|
||||
if dist <= 0 or dist > limit: break
|
||||
match_len: t.CInt = 0
|
||||
max_len: t.CInt = t.CInt(data_len - pos)
|
||||
if max_len > zdef.ZDEFLATE_MAX_MATCH:
|
||||
max_len = zdef.ZDEFLATE_MAX_MATCH
|
||||
while match_len < max_len and data[pos + match_len] == data[chain + match_len]:
|
||||
match_len += 1
|
||||
if match_len > best_len:
|
||||
best_len = match_len
|
||||
c.Set(c.Deref(best_dist), dist)
|
||||
if best_len >= zdef.ZDEFLATE_MAX_MATCH: break
|
||||
chain = self.hash_prev[chain & zdef.ZDEFLATE_WINDOW_MASK]
|
||||
if chain <= t.CInt(pos) - limit: break
|
||||
attempts += 1
|
||||
return best_len if (best_len >= zdef.ZDEFLATE_MIN_MATCH) else 0
|
||||
|
||||
def update_hash(self, data: BYTE | t.CPtr, pos: t.CSizeT):
|
||||
if pos + 2 < pos: return
|
||||
h: t.CUnsignedInt = zdeflate_hash(data + pos)
|
||||
self.hash_prev[pos & zdef.ZDEFLATE_WINDOW_MASK] = self.hash_head[h]
|
||||
self.hash_head[h] = t.CInt(pos)
|
||||
|
||||
def write_fixed_block(self, data: UINT8PTR, length: t.CSizeT, final: t.CInt):
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
dist_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
dist_tree.build_fixed_dist_tree()
|
||||
self.writer.write_bits(1 if final else 0, 1)
|
||||
self.writer.write_bits(1, 2)
|
||||
pos: t.CSizeT = 0
|
||||
while pos < length:
|
||||
best_dist: t.CInt = 0
|
||||
match_len: t.CInt = 0
|
||||
if self.level > 0 and pos + 2 < length:
|
||||
match_len = self.find_match(data, length, pos, c.Addr(best_dist))
|
||||
if match_len >= zdef.ZDEFLATE_MIN_MATCH:
|
||||
len_sym: t.CInt = zdeflate_len_to_symbol(match_len)
|
||||
lit_tree.encode_symbol(len_sym, c.Addr(self.writer))
|
||||
extra: t.CInt = zdef.zdeflate_len_extra_bits[len_sym - 257]
|
||||
if extra > 0:
|
||||
self.writer.write_bits(match_len - zdef.zdeflate_len_base[len_sym - 257], extra)
|
||||
dist_sym: t.CInt = zdeflate_dist_to_symbol(best_dist)
|
||||
dist_tree.encode_symbol(dist_sym, c.Addr(self.writer))
|
||||
extra = zdef.zdeflate_dist_extra_bits[dist_sym]
|
||||
if extra > 0:
|
||||
self.writer.write_bits(best_dist - zdef.zdeflate_dist_base[dist_sym], extra)
|
||||
for i in range(match_len):
|
||||
self.update_hash(data, pos + i)
|
||||
pos += match_len
|
||||
else:
|
||||
lit_tree.encode_symbol(data[pos], c.Addr(self.writer))
|
||||
self.update_hash(data, pos)
|
||||
pos += 1
|
||||
lit_tree.encode_symbol(256, c.Addr(self.writer))
|
||||
|
||||
def encode_block_data(self, data: UINT8PTR, length: t.CSizeT,
|
||||
lit_tree: zhuff.zhuff_tree | t.CPtr, dist_tree: zhuff.zhuff_tree | t.CPtr):
|
||||
pos: t.CSizeT = 0
|
||||
while pos < length:
|
||||
best_dist: t.CInt = 0
|
||||
match_len: t.CInt = 0
|
||||
if self.level > 0 and pos + 2 < length:
|
||||
match_len = self.find_match(data, length, pos, c.Addr(best_dist))
|
||||
if match_len >= zdef.ZDEFLATE_MIN_MATCH:
|
||||
len_sym: t.CInt = zdeflate_len_to_symbol(match_len)
|
||||
lit_tree.encode_symbol(len_sym, c.Addr(self.writer))
|
||||
extra: t.CInt = zdef.zdeflate_len_extra_bits[len_sym - 257]
|
||||
if extra > 0:
|
||||
self.writer.write_bits(match_len - zdef.zdeflate_len_base[len_sym - 257], extra)
|
||||
dist_sym: t.CInt = zdeflate_dist_to_symbol(best_dist)
|
||||
dist_tree.encode_symbol(dist_sym, c.Addr(self.writer))
|
||||
extra = zdef.zdeflate_dist_extra_bits[dist_sym]
|
||||
if extra > 0:
|
||||
self.writer.write_bits(best_dist - zdef.zdeflate_dist_base[dist_sym], extra)
|
||||
for j in range(match_len):
|
||||
self.update_hash(data, pos + j)
|
||||
pos += match_len
|
||||
else:
|
||||
lit_tree.encode_symbol(data[pos], c.Addr(self.writer))
|
||||
self.update_hash(data, pos)
|
||||
pos += 1
|
||||
lit_tree.encode_symbol(256, c.Addr(self.writer))
|
||||
|
||||
|
||||
def write_dynamic_block(self, data: UINT8PTR, length: t.CSizeT, final: t.CInt):
|
||||
lit_freqs: list[t.CInt, 288]
|
||||
dist_freqs: list[t.CInt, 32]
|
||||
zdeflate_count_freqs(lit_freqs, dist_freqs, self, data, length)
|
||||
hlit: t.CInt = 286
|
||||
while hlit > 257 and lit_freqs[hlit - 1] == 0: hlit -= 1
|
||||
hdist: t.CInt = 30
|
||||
while hdist > 1 and dist_freqs[hdist - 1] == 0: hdist -= 1
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
dist_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_codes(lit_freqs, hlit, zdef.ZDEFLATE_MAX_BITS)
|
||||
dist_tree.build_codes(dist_freqs, hdist, zdef.ZDEFLATE_MAX_BITS)
|
||||
all_lengths: list[t.CInt, 288 + 32]
|
||||
for i in range(hlit): all_lengths[i] = lit_tree.codes[i].bits
|
||||
for i in range(hdist): all_lengths[hlit + i] = dist_tree.codes[i].bits
|
||||
total_lengths: t.CInt = hlit + hdist
|
||||
cl_freqs: list[t.CInt, 19]
|
||||
zdeflate_count_cl_freqs(all_lengths, total_lengths, cl_freqs)
|
||||
cl_tree = zhuff.zhuff_tree()
|
||||
cl_tree.build_codes(cl_freqs, 19, 7)
|
||||
hclen: int = 19
|
||||
while hclen > 4 and cl_tree.codes[zdef.zdeflate_codelen_order[hclen - 1]].bits == 0: hclen -= 1
|
||||
self.writer.write_bits(1 if final else 0, 1)
|
||||
self.writer.write_bits(2, 2)
|
||||
self.writer.write_bits(hlit - 257, 5)
|
||||
self.writer.write_bits(hdist - 1, 5)
|
||||
self.writer.write_bits(hclen - 4, 4)
|
||||
for j in range(hclen):
|
||||
self.writer.write_bits(cl_tree.codes[zdef.zdeflate_codelen_order[j]].bits, 3)
|
||||
zdeflate_write_cl_encoded(all_lengths, total_lengths, c.Addr(self.writer), c.Addr(cl_tree))
|
||||
self.encode_block_data(data, length, c.Addr(lit_tree), c.Addr(dist_tree))
|
||||
|
||||
def compress_block(self, data: UINT8PTR, length: t.CSizeT, final: t.CInt):
|
||||
if self.level == 0:
|
||||
self.write_stored_block(data, length, final)
|
||||
elif self.strategy == 4 or length < 128:
|
||||
self.write_fixed_block(data, length, final)
|
||||
else:
|
||||
self.write_dynamic_block(data, length, final)
|
||||
|
||||
def write_stored_block(self, data: UINT8PTR, length: t.CSizeT, final: t.CInt):
|
||||
self.writer.stored_block(data, length, final)
|
||||
|
||||
def compress(self, data: UINT8PTR, length: t.CSizeT,
|
||||
out: t.CUInt8T | t.CPtr[t.CPtr], out_len: t.CSizeT | t.CPtr) -> t.CInt:
|
||||
if self.is_finished: return -2
|
||||
if not data or length == 0:
|
||||
c.Set(c.Deref(out), None)
|
||||
c.Set(c.Deref(out_len), 0)
|
||||
return 0
|
||||
self.adler = zchecksum.zchecksum_adler32(data, length, self.adler)
|
||||
memset(self.hash_head, -1, zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__())
|
||||
memset(self.hash_prev, -1, zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__())
|
||||
self.compress_block(data, length, 0)
|
||||
c.Set(c.Deref(out_len), self.writer.total())
|
||||
c.Set(c.Deref(out), UINT8PTR(zdef.zdef_alloc(c.Deref(out_len))))
|
||||
if not c.Deref(out): return -4
|
||||
memcpy(c.Deref(out), self.writer.buf, c.Deref(out_len))
|
||||
return 0
|
||||
|
||||
def flush(self, mode: t.CInt, out: t.CUInt8T | t.CPtr[t.CPtr], out_len: t.CSizeT | t.CPtr) -> t.CInt:
|
||||
if mode == 4:
|
||||
if not self.is_finished:
|
||||
if self.level == 0:
|
||||
self.writer.write_bits(1, 1)
|
||||
self.writer.write_bits(0, 2)
|
||||
self.writer.align()
|
||||
else:
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
self.writer.write_bits(1, 1)
|
||||
self.writer.write_bits(1, 2)
|
||||
lit_tree.encode_symbol(256, c.Addr(self.writer))
|
||||
self.is_finished = 1
|
||||
if self.wbits >= 0:
|
||||
self.writer.align()
|
||||
adler: t.CUInt32T = self.adler
|
||||
self.writer.buf[self.writer.byte_pos + 0] = (adler >> 24) & 0xFF
|
||||
self.writer.buf[self.writer.byte_pos + 1] = (adler >> 16) & 0xFF
|
||||
self.writer.buf[self.writer.byte_pos + 2] = (adler >> 8) & 0xFF
|
||||
self.writer.buf[self.writer.byte_pos + 3] = (adler >> 0) & 0xFF
|
||||
self.writer.byte_pos += 4
|
||||
elif mode == 2 or mode == 3:
|
||||
self.writer.write_bits(0, 1)
|
||||
self.writer.write_bits(0, 2)
|
||||
self.writer.align()
|
||||
c.Set(c.Deref(out_len), self.writer.total())
|
||||
c.Set(c.Deref(out), UINT8PTR(zdef.zdef_alloc(c.Deref(out_len))))
|
||||
if not c.Deref(out): return -4
|
||||
memcpy(c.Deref(out), self.writer.buf, c.Deref(out_len))
|
||||
return 0
|
||||
|
||||
def set_dictionary(self, dict: UINT8PTR, length: t.CSizeT) -> t.CInt:
|
||||
if not dict: return -2
|
||||
use: t.CSizeT = length
|
||||
if use > zdef.ZDEFLATE_WINDOW_SIZE:
|
||||
dict += length - zdef.ZDEFLATE_WINDOW_SIZE
|
||||
use = zdef.ZDEFLATE_WINDOW_SIZE
|
||||
memcpy(self.window, dict, use)
|
||||
self.window_size = t.CInt(use)
|
||||
self.window_pos = t.CInt(use)
|
||||
self.adler = zchecksum.zchecksum_adler32(dict, length, self.adler)
|
||||
return 0
|
||||
|
||||
def copy(self) -> zdeflate_stream | t.CPtr:
|
||||
z: zdeflate_stream | t.CPtr = zdef.zdef_alloc(zdeflate_stream.__sizeof__())
|
||||
if not z: return None
|
||||
memcpy(z, self, zdeflate_stream.__sizeof__())
|
||||
z.writer.buf = BYTEPTR(zdef.zdef_alloc(self.writer.cap))
|
||||
if not z.writer.buf:
|
||||
zdef.zdef_free(z)
|
||||
return None
|
||||
memcpy(z.writer.buf, self.writer.buf, self.writer.cap)
|
||||
z.window = BYTEPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE))
|
||||
z.hash_head = INTPTR(zdef.zdef_alloc(zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__()))
|
||||
z.hash_prev = INTPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__()))
|
||||
if not z.window or not z.hash_head or not z.hash_prev:
|
||||
if z.writer.buf: zdef.zdef_free(z.writer.buf)
|
||||
zdef.zdef_free(z)
|
||||
return None
|
||||
memcpy(z.window, self.window, zdef.ZDEFLATE_WINDOW_SIZE)
|
||||
memcpy(z.hash_head, self.hash_head, zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__())
|
||||
memcpy(z.hash_prev, self.hash_prev, zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__())
|
||||
return z
|
||||
|
||||
def destroy(self):
|
||||
if self.writer.buf: zdef.zdef_free(self.writer.buf)
|
||||
if self.window: zdef.zdef_free(self.window)
|
||||
if self.hash_head: zdef.zdef_free(self.hash_head)
|
||||
if self.hash_prev: zdef.zdef_free(self.hash_prev)
|
||||
zdef.zdef_free(self)
|
||||
|
||||
|
||||
|
||||
|
||||
def zdeflate_count_freqs(lit_freqs: INTPTR, dist_freqs: INTPTR,
|
||||
s: zdeflate_stream | t.CPtr, data: UINT8PTR, length: t.CSizeT):
|
||||
memset(lit_freqs, 0, 288 * int.__sizeof__())
|
||||
memset(dist_freqs, 0, 32 * int.__sizeof__())
|
||||
pos: t.CSizeT = 0
|
||||
while pos < length:
|
||||
best_dist: t.CInt = 0
|
||||
match_len: t.CInt = 0
|
||||
if s.level > 0 and pos + 2 < length:
|
||||
match_len = zdeflate_stream.find_match(s, data, length, pos, c.Addr(best_dist))
|
||||
if match_len >= zdef.ZDEFLATE_MIN_MATCH:
|
||||
len_sym: t.CInt = zdeflate_len_to_symbol(match_len)
|
||||
lit_freqs[len_sym] += 1
|
||||
dist_freqs[zdeflate_dist_to_symbol(best_dist)] += 1
|
||||
for i in range(match_len):
|
||||
zdeflate_stream.update_hash(s, data, pos + i)
|
||||
pos += match_len
|
||||
else:
|
||||
lit_freqs[data[pos]] += 1
|
||||
zdeflate_stream.update_hash(s, data, pos)
|
||||
pos += 1
|
||||
lit_freqs[256] = 1
|
||||
|
||||
|
||||
|
||||
def zdeflate_hash(p: BYTE | t.CPtr) -> t.CUnsignedInt:
|
||||
return (t.CUnsignedInt(p[0]) ^ (t.CUnsignedInt(p[1]) << 5) ^ (t.CUnsignedInt(p[2]) << 10)) & zdef.ZDEFLATE_HASH_MASK
|
||||
|
||||
|
||||
def zdeflate_len_to_symbol(length: t.CInt) -> t.CInt:
|
||||
for i in range(29):
|
||||
if length <= zdef.zdeflate_len_base[i] + (1 << zdef.zdeflate_len_extra_bits[i]) - 1:
|
||||
return 257 + i
|
||||
return 285
|
||||
|
||||
def zdeflate_dist_to_symbol(dist: t.CInt) -> t.CInt:
|
||||
for i in range(30):
|
||||
if dist <= zdef.zdeflate_dist_base[i] + (1 << zdef.zdeflate_dist_extra_bits[i]) - 1:
|
||||
return i
|
||||
return 29
|
||||
|
||||
def zdeflate_count_cl_freqs(all_lengths: INTPTR, total: t.CInt, cl_freqs: INTPTR):
|
||||
memset(cl_freqs, 0, 19 * int.__sizeof__())
|
||||
i: t.CInt = 0
|
||||
while i < total:
|
||||
if all_lengths[i] == 0:
|
||||
run: t.CInt = 0
|
||||
while i + run < total and all_lengths[i + run] == 0: run += 1
|
||||
while run > 0:
|
||||
if run >= 11:
|
||||
count: t.CInt = run
|
||||
if count > 138: count = 138
|
||||
cl_freqs[18] += 1
|
||||
run -= count
|
||||
elif run >= 3:
|
||||
count: t.CInt = run
|
||||
if count > 10: count = 10
|
||||
cl_freqs[17] += 1
|
||||
run -= count
|
||||
else:
|
||||
cl_freqs[0] += 1
|
||||
run -= 1
|
||||
while i < total and all_lengths[i] == 0: i += 1
|
||||
else:
|
||||
cl_freqs[all_lengths[i]] += 1
|
||||
val: t.CInt = all_lengths[i]
|
||||
i += 1
|
||||
run: t.CInt = 0
|
||||
while i + run < total and all_lengths[i + run] == val: run += 1
|
||||
while run >= 3:
|
||||
count: t.CInt = run
|
||||
if count > 6: count = 6
|
||||
cl_freqs[16] += 1
|
||||
run -= count
|
||||
i += run
|
||||
|
||||
def zdeflate_write_cl_encoded(all_lengths: INTPTR, total: t.CInt,
|
||||
w: zdef.zbit_writer | t.CPtr, cl_tree: zhuff.zhuff_tree | t.CPtr):
|
||||
i: t.CInt = 0
|
||||
while i < total:
|
||||
if all_lengths[i] == 0:
|
||||
run: t.CInt = 0
|
||||
while i + run < total and all_lengths[i + run] == 0: run += 1
|
||||
while run > 0:
|
||||
if run >= 11:
|
||||
count: t.CInt = run
|
||||
if count > 138: count = 138
|
||||
cl_tree.encode_symbol(18, w)
|
||||
w.write_bits(count - 11, 7)
|
||||
run -= count
|
||||
elif run >= 3:
|
||||
count: t.CInt = run
|
||||
if count > 10: count = 10
|
||||
cl_tree.encode_symbol(17, w)
|
||||
w.write_bits(count - 3, 3)
|
||||
run -= count
|
||||
else:
|
||||
cl_tree.encode_symbol(0, w)
|
||||
run -= 1
|
||||
while i < total and all_lengths[i] == 0: i += 1
|
||||
else:
|
||||
cl_tree.encode_symbol(all_lengths[i], w)
|
||||
val: t.CInt = all_lengths[i]
|
||||
i += 1
|
||||
run: t.CInt = 0
|
||||
while i + run < total and all_lengths[i + run] == val: run += 1
|
||||
while run >= 3:
|
||||
count: t.CInt = run
|
||||
if count > 6: count = 6
|
||||
cl_tree.encode_symbol(16, w)
|
||||
w.write_bits(count - 3, 2)
|
||||
run -= count
|
||||
i += run
|
||||
|
||||
|
||||
|
||||
def zdeflate_create(level: t.CInt, wbits: t.CInt, mem_level: t.CInt, strategy: t.CInt) -> zdeflate_stream | t.CPtr:
|
||||
s: zdeflate_stream | t.CPtr = zdef.zdef_alloc(zdeflate_stream.__sizeof__())
|
||||
if not s: return None
|
||||
memset(s, 0, zdeflate_stream.__sizeof__())
|
||||
s.writer = zdef.zbit_writer()
|
||||
s.window = BYTEPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE))
|
||||
s.hash_head = INTPTR(zdef.zdef_alloc(zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__()))
|
||||
s.hash_prev = INTPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__()))
|
||||
if not s.window or not s.hash_head or not s.hash_prev:
|
||||
s.destroy()
|
||||
return None
|
||||
memset(s.hash_head, -1, zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__())
|
||||
memset(s.hash_prev, -1, zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__())
|
||||
s.level = level
|
||||
s.wbits = wbits
|
||||
s.strategy = strategy
|
||||
s.is_finished = 0
|
||||
s.adler = 1
|
||||
if wbits > 0:
|
||||
s.writer.write_zlib_header(wbits, level)
|
||||
elif wbits == 0:
|
||||
s.writer.write_zlib_header(15, level)
|
||||
return s
|
||||
|
||||
|
||||
def zdeflate_one_shot(data: UINT8PTR, length: t.CSizeT,
|
||||
level: t.CInt, wbits: t.CInt, out_len: t.CSizeT | t.CPtr) -> t.CUInt8T | t.CPtr:
|
||||
s: zdeflate_stream | t.CPtr = zdeflate_create(level, wbits, 8, 0)
|
||||
if not s: return None
|
||||
|
||||
memset(s.hash_head, -1, zdef.ZDEFLATE_HASH_SIZE * int.__sizeof__())
|
||||
memset(s.hash_prev, -1, zdef.ZDEFLATE_WINDOW_SIZE * int.__sizeof__())
|
||||
|
||||
s.adler = zchecksum.zchecksum_adler32(data, length, 1)
|
||||
|
||||
if length == 0:
|
||||
s.writer.write_bits(1, 1)
|
||||
s.writer.write_bits(1, 2)
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
lit_tree.encode_symbol(256, c.Addr(s.writer))
|
||||
else:
|
||||
s.compress_block(data, length, 1)
|
||||
|
||||
s.is_finished = 1
|
||||
|
||||
if wbits >= 0:
|
||||
s.writer.align()
|
||||
adler: t.CUInt32T = s.adler
|
||||
s.writer.buf[s.writer.byte_pos + 0] = (adler >> 24) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 1] = (adler >> 16) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 2] = (adler >> 8) & 0xFF
|
||||
s.writer.buf[s.writer.byte_pos + 3] = (adler >> 0) & 0xFF
|
||||
s.writer.byte_pos += 4
|
||||
c.Set(c.Deref(out_len), s.writer.total())
|
||||
result: UINT8PTR = UINT8PTR(zdef.zdef_alloc(c.Deref(out_len)))
|
||||
if result: memcpy(result, s.writer.buf, c.Deref(out_len))
|
||||
s.destroy()
|
||||
return result
|
||||
@@ -1,353 +0,0 @@
|
||||
#include <string.h>
|
||||
from stdint import *
|
||||
import zdef
|
||||
import t, c
|
||||
|
||||
|
||||
ZHUFF_MAX_CODES: t.CDefine = 288
|
||||
ZHUFF_MAX_BITS: t.CDefine = 15
|
||||
|
||||
class zhuff_code:
|
||||
code: UINT
|
||||
bits: t.CInt
|
||||
|
||||
@t.Object
|
||||
class zhuff_tree:
|
||||
codes: list[zhuff_code, ZHUFF_MAX_CODES]
|
||||
count: t.CInt
|
||||
max_bits: t.CInt
|
||||
|
||||
def __init__(self):
|
||||
pass
|
||||
|
||||
def build_codes(self, freqs: INTPTR, count: t.CInt, max_bits: t.CInt):
|
||||
lengths: list[t.CInt, ZHUFF_MAX_CODES]
|
||||
bl_count: list[t.CInt, ZHUFF_MAX_BITS + 1]
|
||||
next_code: list[UINT, ZHUFF_MAX_BITS + 1]
|
||||
self.count = count
|
||||
self.max_bits = max_bits
|
||||
self.build_code_lengths(lengths, freqs, count, max_bits)
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
for i in range(count):
|
||||
if lengths[i] > 0:
|
||||
bl_count[lengths[i]] += 1
|
||||
code: t.CUnsignedInt = 0
|
||||
next_code[0] = 0
|
||||
for bits in range(1, max_bits + 1):
|
||||
code = (code + bl_count[bits - 1]) << 1
|
||||
next_code[bits] = code
|
||||
for i in range(count):
|
||||
self.codes[i].bits = lengths[i]
|
||||
if lengths[i] > 0:
|
||||
self.codes[i].code = next_code[lengths[i]]
|
||||
next_code[lengths[i]] += 1
|
||||
else:
|
||||
self.codes[i].code = 0
|
||||
|
||||
def build_fixed_lit_tree(self):
|
||||
lengths: list[t.CInt, 288]
|
||||
bl_count: list[t.CInt, 16]
|
||||
next_code: list[t.CUnsignedInt, 16]
|
||||
self.get_fixed_lit_lengths(lengths)
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
for i in range(288):
|
||||
if lengths[i] > 0:
|
||||
bl_count[lengths[i]] += 1
|
||||
code: t.CUnsignedInt = 0
|
||||
next_code[0] = 0
|
||||
for bits in range(1, 9 + 1):
|
||||
code = (code + bl_count[bits - 1]) << 1
|
||||
next_code[bits] = code
|
||||
self.count = 288
|
||||
self.max_bits = 9
|
||||
for i in range(288):
|
||||
self.codes[i].bits = lengths[i]
|
||||
if lengths[i] > 0:
|
||||
self.codes[i].code = next_code[lengths[i]]
|
||||
next_code[lengths[i]] += 1
|
||||
else:
|
||||
self.codes[i].code = 0
|
||||
|
||||
def build_fixed_dist_tree(self):
|
||||
lengths: list[t.CInt, 32]
|
||||
bl_count: list[t.CInt, 16]
|
||||
next_code: list[t.CUnsignedInt, 16]
|
||||
self.get_fixed_dist_lengths(lengths)
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
for i in range(32):
|
||||
if lengths[i] > 0:
|
||||
bl_count[lengths[i]] += 1
|
||||
code: t.CUnsignedInt = 0
|
||||
next_code[0] = 0
|
||||
for bits in range(1, 5 + 1):
|
||||
code = (code + bl_count[bits - 1]) << 1
|
||||
next_code[bits] = code
|
||||
self.count = 32
|
||||
self.max_bits = 5
|
||||
for i in range(32):
|
||||
self.codes[i].bits = lengths[i]
|
||||
if lengths[i] > 0:
|
||||
self.codes[i].code = next_code[lengths[i]]
|
||||
next_code[lengths[i]] += 1
|
||||
else:
|
||||
self.codes[i].code = 0
|
||||
|
||||
def encode_symbol(self, symbol: int, writer: zdef.zbit_writer | t.CPtr):
|
||||
if symbol < 0 or symbol >= self.count:
|
||||
return
|
||||
if self.codes[symbol].bits == 0:
|
||||
return
|
||||
writer.write_bits_rev(self.codes[symbol].code, self.codes[symbol].bits)
|
||||
|
||||
|
||||
def build_code_lengths(self, lengths: t.CInt | t.CPtr, freqs: t.CInt | t.CPtr, count: t.CInt, max_bits: t.CInt):
|
||||
bl_count: list[t.CInt, ZHUFF_MAX_BITS + 2]
|
||||
sort_count: t.CInt = 0
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
memset(lengths, 0, int.__sizeof__() * count)
|
||||
for i in range(count):
|
||||
if freqs[i] > 0:
|
||||
sort_count += 1
|
||||
if sort_count == 0: return
|
||||
if sort_count == 1:
|
||||
for i in range(count):
|
||||
if freqs[i] > 0:
|
||||
lengths[i] = 1
|
||||
return
|
||||
items: t.CInt | t.CPtr = INTPTR(zdef.zdef_alloc(sort_count * int.__sizeof__()))
|
||||
item_freqs: t.CInt | t.CPtr = INTPTR(zdef.zdef_alloc(sort_count * int.__sizeof__()))
|
||||
idx: t.CInt = 0
|
||||
for i in range(count):
|
||||
if freqs[i] > 0:
|
||||
items[idx] = i
|
||||
item_freqs[idx] = freqs[i]
|
||||
idx += 1
|
||||
for i in range(sort_count):
|
||||
key_item: t.CInt = items[i]
|
||||
key_freq: t.CInt = item_freqs[i]
|
||||
j: t.CInt = i - 1
|
||||
while j >= 0 and item_freqs[j] > key_freq:
|
||||
items[j + 1] = items[j]
|
||||
item_freqs[j + 1] = item_freqs[j]
|
||||
j -= 1
|
||||
items[j + 1] = key_item
|
||||
item_freqs[j + 1] = key_freq
|
||||
parent: INTPTR = INTPTR(zdef.zdef_alloc((sort_count * 2) * int.__sizeof__()))
|
||||
for i in range(sort_count * 2):
|
||||
parent[i] = -1
|
||||
heap: INTPTR = INTPTR(zdef.zdef_alloc((sort_count + 1) * int.__sizeof__()))
|
||||
heap_size: t.CInt = 0
|
||||
for i in range(sort_count):
|
||||
heap[heap_size] = i
|
||||
heap_size += 1
|
||||
pos: t.CInt = heap_size - 1
|
||||
while pos > 0:
|
||||
par: t.CInt = (pos - 1) / 2
|
||||
if item_freqs[heap[par]] > item_freqs[heap[pos]]:
|
||||
tmp: t.CInt = heap[par]
|
||||
heap[par] = heap[pos]
|
||||
heap[pos] = tmp
|
||||
pos = par
|
||||
else: break
|
||||
combined_freq: INTPTR = INTPTR(zdef.zdef_alloc((sort_count * 2) * int.__sizeof__()))
|
||||
for i in range(sort_count):
|
||||
combined_freq[i] = item_freqs[i]
|
||||
internal: t.CInt = sort_count
|
||||
while heap_size > 1:
|
||||
a: t.CInt = heap[0]
|
||||
heap[0] = heap[heap_size - 1]
|
||||
heap_size -= 1
|
||||
pos: t.CInt = 0
|
||||
while True:
|
||||
left: t.CInt = 2 * pos + 1
|
||||
right: t.CInt = 2 * pos + 2
|
||||
smallest: t.CInt = pos
|
||||
if left < heap_size and combined_freq[heap[left]] < combined_freq[heap[smallest]]:
|
||||
smallest = left
|
||||
if right < heap_size and combined_freq[heap[right]] < combined_freq[heap[smallest]]:
|
||||
smallest = right
|
||||
if smallest != pos:
|
||||
tmp: t.CInt = heap[pos]
|
||||
heap[pos] = heap[smallest]
|
||||
heap[smallest] = tmp
|
||||
pos = smallest
|
||||
else: break
|
||||
b: t.CInt = heap[0]
|
||||
heap[0] = heap[heap_size - 1]
|
||||
heap_size -= 1
|
||||
pos = 0
|
||||
while True:
|
||||
left: t.CInt = 2 * pos + 1
|
||||
right: t.CInt = 2 * pos + 2
|
||||
smallest: t.CInt = pos
|
||||
if left < heap_size and combined_freq[heap[left]] < combined_freq[heap[smallest]]:
|
||||
smallest = left
|
||||
if right < heap_size and combined_freq[heap[right]] < combined_freq[heap[smallest]]:
|
||||
smallest = right
|
||||
if smallest != pos:
|
||||
tmp: t.CInt = heap[pos]
|
||||
heap[pos] = heap[smallest]
|
||||
heap[smallest] = tmp
|
||||
pos = smallest
|
||||
else: break
|
||||
if internal >= sort_count * 2 - 1: break
|
||||
combined_freq[internal] = combined_freq[a] + combined_freq[b]
|
||||
parent[a] = internal
|
||||
parent[b] = internal
|
||||
heap[heap_size] = internal
|
||||
heap_size += 1
|
||||
pos = heap_size - 1
|
||||
while pos > 0:
|
||||
par: t.CInt = (pos - 1) / 2
|
||||
if combined_freq[heap[par]] > combined_freq[heap[pos]]:
|
||||
tmp: t.CInt = heap[par]
|
||||
heap[par] = heap[pos]
|
||||
heap[pos] = tmp
|
||||
pos = par
|
||||
else: break
|
||||
internal += 1
|
||||
for i in range(sort_count):
|
||||
node: t.CInt = i
|
||||
length: t.CInt = 0
|
||||
while parent[node] >= 0:
|
||||
length += 1
|
||||
node = parent[node]
|
||||
lengths[items[i]] = length
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
for i in range(count):
|
||||
if lengths[i] > 0:
|
||||
bl_count[lengths[i]] += 1
|
||||
overflow: t.CInt = 0
|
||||
for i in range(count):
|
||||
if lengths[i] > max_bits:
|
||||
overflow += 1
|
||||
bl_count[lengths[i]] -= 1
|
||||
lengths[i] = max_bits
|
||||
bl_count[max_bits] += 1
|
||||
while overflow > 0:
|
||||
bits: t.CInt = max_bits - 1
|
||||
while bits > 0 and bl_count[bits] == 0:
|
||||
bits -= 1
|
||||
if bits == 0: break
|
||||
bl_count[bits] -= 1
|
||||
bl_count[bits + 1] += 2
|
||||
bl_count[max_bits] -= 1
|
||||
overflow -= 2
|
||||
sorted_items: INTPTR = INTPTR(zdef.zdef_alloc(sort_count * int.__sizeof__()))
|
||||
si: t.CInt = 0
|
||||
for i in range(count):
|
||||
if freqs[i] > 0:
|
||||
sorted_items[si] = i
|
||||
si += 1
|
||||
for i in range(si - 1):
|
||||
for j in range(i + 1, si):
|
||||
if lengths[sorted_items[i]] < lengths[sorted_items[j]]:
|
||||
tmp: t.CInt = sorted_items[i]
|
||||
sorted_items[i] = sorted_items[j]
|
||||
sorted_items[j] = tmp
|
||||
sidx: t.CInt = 0
|
||||
for bits in range(max_bits, 1, -1):
|
||||
n: t.CInt = bl_count[bits]
|
||||
while n > 0 and sidx < si:
|
||||
lengths[sorted_items[sidx]] = bits
|
||||
sidx += 1
|
||||
n -= 1
|
||||
zdef.zdef_free(sorted_items)
|
||||
zdef.zdef_free(heap)
|
||||
zdef.zdef_free(parent)
|
||||
zdef.zdef_free(combined_freq)
|
||||
zdef.zdef_free(item_freqs)
|
||||
zdef.zdef_free(items)
|
||||
|
||||
def get_fixed_lit_lengths(self, lengths: INTPTR):
|
||||
for i in range( 0, 143 + 1): lengths[i] = 8
|
||||
for i in range(143 + 1, 255 + 1): lengths[i] = 9
|
||||
for i in range(255 + 1, 279 + 1): lengths[i] = 7
|
||||
for i in range(279 + 1, 287 + 1): lengths[i] = 8
|
||||
|
||||
def get_fixed_dist_lengths(self, lengths: INTPTR):
|
||||
for i in range(32):
|
||||
lengths[i] = 5
|
||||
|
||||
def build_tree_from_lengths(self, lengths: INTPTR, count: t.CInt, max_bits: t.CInt):
|
||||
bl_count: list[t.CInt, 16]
|
||||
next_code: list[t.CUnsignedInt, 16]
|
||||
|
||||
memset(bl_count, 0, bl_count.__sizeof__())
|
||||
for i in range(count):
|
||||
if lengths[i] > 0:
|
||||
bl_count[lengths[i]] += 1
|
||||
|
||||
code: UINT = 0
|
||||
next_code[0] = 0
|
||||
for bits in range(1, max_bits + 1):
|
||||
code = (code + bl_count[bits - 1]) << 1
|
||||
next_code[bits] = code
|
||||
|
||||
self.count = count
|
||||
self.max_bits = max_bits
|
||||
for i in range(count):
|
||||
self.codes[i].bits = lengths[i]
|
||||
if lengths[i] > 0:
|
||||
self.codes[i].code = next_code[lengths[i]]
|
||||
next_code[lengths[i]] += 1
|
||||
else:
|
||||
self.codes[i].code = 0
|
||||
|
||||
|
||||
class zhuff_decode_node:
|
||||
children: list[t.CInt, 2]
|
||||
symbol: t.CInt
|
||||
|
||||
|
||||
@t.Object
|
||||
class zhuff_decode_tree:
|
||||
nodes: list[zhuff_decode_node, 2 * ZHUFF_MAX_CODES]
|
||||
node_count: t.CInt
|
||||
root: t.CInt
|
||||
|
||||
def __init__(self):
|
||||
pass
|
||||
|
||||
def build_decode_tree(self, ht: zhuff_tree | t.CPtr):
|
||||
self.node_count = 1
|
||||
self.root = 0
|
||||
self.nodes[0].children[0] = -1
|
||||
self.nodes[0].children[1] = -1
|
||||
self.nodes[0].symbol = -1
|
||||
for i in range(ht.count):
|
||||
if ht.codes[i].bits == 0: continue
|
||||
node: int = 0
|
||||
for bit in range(ht.codes[i].bits - 1, -1, -1):
|
||||
dir: t.CInt = (ht.codes[i].code >> bit) & 1
|
||||
if bit > 0:
|
||||
# Internal node: traverse or create
|
||||
if self.nodes[node].children[dir] == -1:
|
||||
new_node: int = self.node_count
|
||||
self.node_count += 1
|
||||
self.nodes[new_node].children[0] = -1
|
||||
self.nodes[new_node].children[1] = -1
|
||||
self.nodes[new_node].symbol = -1
|
||||
self.nodes[node].children[dir] = new_node
|
||||
node = self.nodes[node].children[dir]
|
||||
else:
|
||||
# Leaf: bit 0 - set symbol on the child
|
||||
if self.nodes[node].children[dir] == -1:
|
||||
leaf: int = self.node_count
|
||||
self.node_count += 1
|
||||
self.nodes[leaf].children[0] = -1
|
||||
self.nodes[leaf].children[1] = -1
|
||||
self.nodes[leaf].symbol = i
|
||||
self.nodes[node].children[dir] = leaf
|
||||
else:
|
||||
self.nodes[self.nodes[node].children[dir]].symbol = i
|
||||
|
||||
def decode_symbol(self, reader: zdef.zbit_reader | t.CPtr) -> t.CInt:
|
||||
node: int = self.root
|
||||
while self.nodes[node].symbol == -1:
|
||||
bit: t.CUnsignedInt
|
||||
if reader.read_bits(1, c.Addr(bit)) != 0: return -1
|
||||
dir: t.CInt = t.CInt(bit)
|
||||
if self.nodes[node].children[dir] == -1: return -1
|
||||
node = self.nodes[node].children[dir]
|
||||
return self.nodes[node].symbol
|
||||
@@ -1,454 +0,0 @@
|
||||
from stdint import *
|
||||
import zinflate
|
||||
import zchecksum
|
||||
import zdef
|
||||
import zhuff
|
||||
import string
|
||||
import t, c
|
||||
|
||||
|
||||
@t.Object
|
||||
class zinflate_stream:
|
||||
reader: zdef.zbit_reader
|
||||
|
||||
window: BYTEPTR
|
||||
window_pos: t.CInt
|
||||
window_size: t.CInt
|
||||
|
||||
wbits: t.CInt
|
||||
is_finished: t.CInt
|
||||
is_initialized: t.CInt
|
||||
|
||||
header_parsed: t.CInt
|
||||
is_gzip: t.CInt
|
||||
|
||||
adler: t.CUInt32T
|
||||
crc: t.CUInt32T
|
||||
|
||||
output: BYTEPTR
|
||||
output_len: t.CSizeT
|
||||
output_cap: t.CSizeT
|
||||
|
||||
input_data: t.CConst | BYTEPTR
|
||||
input_len: t.CSizeT
|
||||
input_pos: t.CSizeT
|
||||
|
||||
max_length: t.CSizeT
|
||||
|
||||
def __init__(self):
|
||||
pass
|
||||
|
||||
def output_byte(self, byte: BYTE):
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: return
|
||||
|
||||
if self.output_len >= self.output_cap:
|
||||
new_cap: t.CSizeT = self.output_cap * 2
|
||||
if new_cap < 256: new_cap = 256
|
||||
new_buf: BYTEPTR = BYTEPTR(zdef.zdef_alloc(new_cap))
|
||||
if not new_buf: return
|
||||
memcpy(new_buf, self.output, self.output_len)
|
||||
zdef.zdef_free(self.output)
|
||||
self.output = new_buf
|
||||
self.output_cap = new_cap
|
||||
self.output[self.output_len] = byte
|
||||
self.output_len += 1
|
||||
self.window[self.window_pos & (zdef.ZDEFLATE_WINDOW_SIZE - 1)] = byte
|
||||
self.window_pos += 1
|
||||
|
||||
|
||||
def parse_header(self) -> t.CInt:
|
||||
if self.header_parsed: return 0
|
||||
if self.wbits < 0:
|
||||
self.header_parsed = 1
|
||||
self.is_gzip = 0
|
||||
return 0
|
||||
if self.input_pos + 2 > self.input_len: return -3
|
||||
b0: BYTE = self.input_data[self.input_pos]
|
||||
b1: BYTE = self.input_data[self.input_pos + 1]
|
||||
if b0 == 0x1F and b1 == 0x8B:
|
||||
self.is_gzip = 1
|
||||
if self.input_pos + 10 > self.input_len: return -3
|
||||
self.input_pos += 10
|
||||
flg: BYTE = self.input_data[self.input_pos - 6]
|
||||
if flg & 0x04:
|
||||
if self.input_pos + 2 > self.input_len: return -3
|
||||
xlen: UINT = self.input_data[self.input_pos] | (self.input_data[self.input_pos + 1] << 8)
|
||||
self.input_pos += 2 + xlen
|
||||
if flg & 0x08:
|
||||
while self.input_pos < self.input_len and self.input_data[self.input_pos] != 0: self.input_pos += 1
|
||||
self.input_pos += 1
|
||||
if flg & 0x10:
|
||||
while self.input_pos < self.input_len and self.input_data[self.input_pos] != 0: self.input_pos += 1
|
||||
self.input_pos += 1
|
||||
if flg & 0x02:
|
||||
self.input_pos += 2
|
||||
if self.input_pos > self.input_len: return -3
|
||||
else:
|
||||
self.is_gzip = 0
|
||||
if (b0 * 256 + b1) % 31 != 0: return -3
|
||||
cm: t.CInt = b0 & 0x0F
|
||||
if cm != 8: return -3
|
||||
self.input_pos += 2
|
||||
self.header_parsed = 1
|
||||
return 0
|
||||
|
||||
|
||||
def read_bits_from_input(self, nbits: t.CInt, out: UINTPTR) -> t.CInt:
|
||||
val: UINT = 0
|
||||
for i in range(nbits):
|
||||
if self.input_pos >= self.input_len: return -3
|
||||
if self.input_data[self.input_pos] & (UINT(1) << (self.reader.bit_pos)):
|
||||
val |= (UINT(1) << i)
|
||||
self.reader.bit_pos += 1
|
||||
if self.reader.bit_pos == 8:
|
||||
self.reader.bit_pos = 0
|
||||
self.input_pos += 1
|
||||
c.Set(c.Deref(out), val)
|
||||
return 0
|
||||
|
||||
|
||||
def read_huffman(self, dt: zhuff.zhuff_decode_tree | t.CPtr) -> t.CInt:
|
||||
node: int = dt.root
|
||||
while dt.nodes[node].symbol == -1:
|
||||
bit: UINT
|
||||
if self.read_bits_from_input(1, c.Addr(bit)) != 0: return -1
|
||||
dir: int = t.CInt(bit)
|
||||
if dt.nodes[node].children[dir] == -1: return -3
|
||||
node = dt.nodes[node].children[dir]
|
||||
return dt.nodes[node].symbol
|
||||
|
||||
def inflate_block_stored(self) -> t.CInt:
|
||||
if self.reader.bit_pos != 0:
|
||||
self.reader.bit_pos = 0
|
||||
self.input_pos += 1
|
||||
if self.input_pos + 4 > self.input_len: return -3
|
||||
length: UINT = self.input_data[self.input_pos] | (self.input_data[self.input_pos + 1] << 8)
|
||||
nlen: UINT = self.input_data[self.input_pos + 2] | (self.input_data[self.input_pos + 3] << 8)
|
||||
self.input_pos += 4
|
||||
if (length ^ nlen) != 0xFFFF: return -3
|
||||
if self.input_pos + length > self.input_len: return -3
|
||||
i: t.CUnsignedInt
|
||||
for i in range(length):
|
||||
self.output_byte(self.input_data[self.input_pos])
|
||||
self.input_pos += 1
|
||||
return 0
|
||||
|
||||
|
||||
def inflate_block_fixed(self) -> t.CInt:
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
dist_tree = zhuff.zhuff_tree()
|
||||
lit_tree.build_fixed_lit_tree()
|
||||
dist_tree.build_fixed_dist_tree()
|
||||
|
||||
lit_dt: zhuff.zhuff_decode_tree | t.CPtr = zdef.zdef_alloc(zhuff.zhuff_decode_tree.__sizeof__())
|
||||
dist_dt: zhuff.zhuff_decode_tree | t.CPtr = zdef.zdef_alloc(zhuff.zhuff_decode_tree.__sizeof__())
|
||||
if not lit_dt or not dist_dt:
|
||||
if lit_dt: zdef.zdef_free(lit_dt)
|
||||
if dist_dt: zdef.zdef_free(dist_dt)
|
||||
return -4
|
||||
memset(lit_dt, 0, zhuff.zhuff_decode_tree.__sizeof__())
|
||||
memset(dist_dt, 0, zhuff.zhuff_decode_tree.__sizeof__())
|
||||
lit_dt.build_decode_tree(c.Addr(lit_tree))
|
||||
dist_dt.build_decode_tree(c.Addr(dist_tree))
|
||||
|
||||
result: t.CInt = 0
|
||||
while True:
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: break
|
||||
|
||||
sym: t.CInt = self.read_huffman(lit_dt)
|
||||
if sym < 0:
|
||||
result = -3
|
||||
break
|
||||
if sym == 256: break
|
||||
if sym < 256:
|
||||
self.output_byte(BYTE(sym))
|
||||
else:
|
||||
len_idx: t.CInt = sym - 257
|
||||
if len_idx < 0 or len_idx >= 29:
|
||||
result = -3
|
||||
break
|
||||
length: t.CInt = zdef.zdeflate_len_base[len_idx]
|
||||
extra: t.CInt = zdef.zdeflate_len_extra_bits[len_idx]
|
||||
if extra > 0:
|
||||
extra_val: UINT
|
||||
if self.read_bits_from_input(extra, c.Addr(extra_val)) != 0:
|
||||
result = -3
|
||||
break
|
||||
length += extra_val
|
||||
|
||||
dist_sym: t.CInt = self.read_huffman(dist_dt)
|
||||
if dist_sym < 0 or dist_sym >= 30:
|
||||
result = -3
|
||||
break
|
||||
dist: t.CInt = zdef.zdeflate_dist_base[dist_sym]
|
||||
extra = zdef.zdeflate_dist_extra_bits[dist_sym]
|
||||
if extra > 0:
|
||||
extra_val: UINT
|
||||
if self.read_bits_from_input(extra, c.Addr(extra_val)) != 0:
|
||||
result = -3
|
||||
break
|
||||
dist += extra_val
|
||||
|
||||
for i in range(length):
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: break
|
||||
src_pos: t.CInt = (self.window_pos - dist) & (zdef.ZDEFLATE_WINDOW_SIZE - 1)
|
||||
byte: BYTE = self.window[src_pos]
|
||||
self.output_byte(byte)
|
||||
|
||||
zdef.zdef_free(lit_dt)
|
||||
zdef.zdef_free(dist_dt)
|
||||
return result
|
||||
|
||||
|
||||
def inflate_block_dynamic(self) -> t.CInt:
|
||||
hlit_val: UINT
|
||||
hdist_val: UINT
|
||||
hclen_val: UINT
|
||||
if self.read_bits_from_input(5, c.Addr(hlit_val)) != 0: return -3
|
||||
if self.read_bits_from_input(5, c.Addr(hdist_val)) != 0: return -3
|
||||
if self.read_bits_from_input(4, c.Addr(hclen_val)) != 0: return -3
|
||||
|
||||
hlit: t.CInt = t.CInt(hlit_val) + 257
|
||||
hdist: t.CInt = t.CInt(hdist_val) + 1
|
||||
hclen: t.CInt = t.CInt(hclen_val) + 4
|
||||
|
||||
cl_lengths: list[t.CInt, 19]
|
||||
memset(cl_lengths, 0, cl_lengths.__sizeof__())
|
||||
for i in range(hclen):
|
||||
len_val: UINT
|
||||
if self.read_bits_from_input(3, c.Addr(len_val)) != 0: return -3
|
||||
cl_lengths[zdef.zdeflate_codelen_order[i]] = t.CInt(len_val)
|
||||
|
||||
cl_tree = zhuff.zhuff_tree()
|
||||
cl_tree.build_tree_from_lengths(cl_lengths, 19, 7)
|
||||
|
||||
cl_dt = zhuff.zhuff_decode_tree()
|
||||
cl_dt.build_decode_tree(c.Addr(cl_tree))
|
||||
|
||||
total: t.CInt = hlit + hdist
|
||||
all_lengths: t.CInt | t.CPtr = zdef.zdef_alloc(total * int.__sizeof__())
|
||||
if not all_lengths: return -4
|
||||
|
||||
idx: t.CInt = 0
|
||||
while idx < total:
|
||||
sym: t.CInt = self.read_huffman(c.Addr(cl_dt))
|
||||
if sym < 0 or sym > 18:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
if sym < 16:
|
||||
all_lengths[idx] = sym
|
||||
idx += 1
|
||||
elif sym == 16:
|
||||
rep: UINT
|
||||
if self.read_bits_from_input(2, c.Addr(rep)) != 0:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
rep += 3
|
||||
if idx == 0 or idx + rep > total:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
prev: t.CInt = all_lengths[idx - 1]
|
||||
i: t.CUnsignedInt
|
||||
for i in range(rep):
|
||||
all_lengths[idx] = prev
|
||||
idx += 1
|
||||
elif sym == 17:
|
||||
rep: t.CUnsignedInt
|
||||
if self.read_bits_from_input(3, c.Addr(rep)) != 0:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
rep += 3
|
||||
if idx + rep > total:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
i: t.CUnsignedInt
|
||||
for i in range(rep):
|
||||
all_lengths[idx] = 0
|
||||
idx += 1
|
||||
else:
|
||||
rep: t.CUnsignedInt
|
||||
if self.read_bits_from_input(7, c.Addr(rep)) != 0:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
rep += 11
|
||||
if idx + rep > total:
|
||||
zdef.zdef_free(all_lengths)
|
||||
return -3
|
||||
i: t.CUnsignedInt
|
||||
for i in range(rep):
|
||||
all_lengths[idx] = 0
|
||||
idx += 1
|
||||
|
||||
lit_tree = zhuff.zhuff_tree()
|
||||
dist_tree = zhuff.zhuff_tree()
|
||||
|
||||
lit_lengths: list[t.CInt, 288]
|
||||
memset(lit_lengths, 0, lit_lengths.__sizeof__())
|
||||
for i in range(hlit):
|
||||
lit_lengths[i] = all_lengths[i]
|
||||
lit_tree.build_tree_from_lengths(lit_lengths, hlit, zdef.ZDEFLATE_MAX_BITS)
|
||||
|
||||
dist_lengths: list[t.CInt, 32]
|
||||
memset(dist_lengths, 0, dist_lengths.__sizeof__())
|
||||
for i in range(hdist):
|
||||
dist_lengths[i] = all_lengths[hlit + i]
|
||||
dist_tree.build_tree_from_lengths(dist_lengths, hdist, zdef.ZDEFLATE_MAX_BITS)
|
||||
|
||||
zdef.zdef_free(all_lengths)
|
||||
|
||||
lit_dt = zhuff.zhuff_decode_tree()
|
||||
dist_dt = zhuff.zhuff_decode_tree()
|
||||
lit_dt.build_decode_tree(c.Addr(lit_tree))
|
||||
dist_dt.build_decode_tree(c.Addr(dist_tree))
|
||||
|
||||
while True:
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: return 0
|
||||
sym: t.CInt = self.read_huffman(c.Addr(lit_dt))
|
||||
if sym < 0: return -3
|
||||
if sym == 256: return 0
|
||||
if sym < 256:
|
||||
self.output_byte(BYTE(sym))
|
||||
else:
|
||||
len_idx: t.CInt = sym - 257
|
||||
if len_idx < 0 or len_idx >= 29: return -3
|
||||
length: t.CInt = zdef.zdeflate_len_base[len_idx]
|
||||
extra: t.CInt = zdef.zdeflate_len_extra_bits[len_idx]
|
||||
if extra > 0:
|
||||
extra_val: UINT
|
||||
if self.read_bits_from_input(extra, c.Addr(extra_val)) != 0: return -3
|
||||
length += extra_val
|
||||
dist_sym: t.CInt = self.read_huffman(c.Addr(dist_dt))
|
||||
if dist_sym < 0 or dist_sym >= 30: return -3
|
||||
dist: t.CInt = zdef.zdeflate_dist_base[dist_sym]
|
||||
extra = zdef.zdeflate_dist_extra_bits[dist_sym]
|
||||
if extra > 0:
|
||||
extra_val: UINT
|
||||
if self.read_bits_from_input(extra, c.Addr(extra_val)) != 0: return -3
|
||||
dist += extra_val
|
||||
for i in range(length):
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: return 0
|
||||
src_pos: t.CInt = (self.window_pos - dist) & (zdef.ZDEFLATE_WINDOW_SIZE - 1)
|
||||
byte: BYTE = self.window[src_pos]
|
||||
self.output_byte(byte)
|
||||
|
||||
def inflate_stream(self) -> t.CInt:
|
||||
err: t.CInt = self.parse_header()
|
||||
if err != 0: return err
|
||||
while not self.is_finished:
|
||||
bfinal_val: UINT
|
||||
btype_val: UINT
|
||||
if self.read_bits_from_input(1, c.Addr(bfinal_val)) != 0: return -3
|
||||
if self.read_bits_from_input(2, c.Addr(btype_val)) != 0: return -3
|
||||
bfinal: t.CInt = t.CInt(bfinal_val)
|
||||
btype: t.CInt = t.CInt(btype_val)
|
||||
if btype == 0:
|
||||
err = self.inflate_block_stored()
|
||||
elif btype == 1:
|
||||
err = self.inflate_block_fixed()
|
||||
elif btype == 2:
|
||||
err = self.inflate_block_dynamic()
|
||||
else:
|
||||
return -3
|
||||
if err != 0: return err
|
||||
if bfinal: self.is_finished = 1
|
||||
if self.max_length > 0 and self.output_len >= self.max_length: return 0
|
||||
if self.is_gzip:
|
||||
if self.reader.bit_pos != 0:
|
||||
self.reader.bit_pos = 0
|
||||
self.input_pos += 1
|
||||
if self.input_pos + 8 <= self.input_len:
|
||||
self.crc = ((t.CUInt32T(self.input_data[self.input_pos + 0]) << 0) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 1]) << 8) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 2]) << 16) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 3]) << 24))
|
||||
self.input_pos += 8
|
||||
elif self.wbits >= 0:
|
||||
if self.reader.bit_pos != 0:
|
||||
self.reader.bit_pos = 0
|
||||
self.input_pos += 1
|
||||
if self.input_pos + 4 <= self.input_len:
|
||||
self.adler = ((t.CUInt32T(self.input_data[self.input_pos + 0]) << 24) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 1]) << 16) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 2]) << 8) |
|
||||
(t.CUInt32T(self.input_data[self.input_pos + 3]) << 0))
|
||||
self.input_pos += 4
|
||||
return 0
|
||||
|
||||
def decompress(self, data: UINT8PTR, length: t.CSizeT,
|
||||
max_length: t.CSizeT, out: t.CUInt8T | t.CPtr[t.CPtr], out_len: t.CSizeT | t.CPtr) -> t.CInt:
|
||||
if not self.is_initialized: return -2
|
||||
self.input_data = data
|
||||
self.input_len = length
|
||||
self.input_pos = 0
|
||||
self.reader.bit_pos = 0
|
||||
self.output_len = 0
|
||||
self.max_length = max_length
|
||||
err: t.CInt = self.inflate_stream()
|
||||
if err != 0: return err
|
||||
c.Set(c.Deref(out_len), self.output_len)
|
||||
if self.output_len == 0:
|
||||
c.Set(c.Deref(out), UINT8PTR(zdef.zdef_alloc(1)))
|
||||
else:
|
||||
dst: UINT8PTR = UINT8PTR(zdef.zdef_alloc(self.output_len))
|
||||
c.Set(c.Deref(out), dst)
|
||||
if dst:
|
||||
memcpy(dst, self.output, self.output_len)
|
||||
return 0
|
||||
|
||||
def set_dictionary(self, dict: UINT8PTR, length: t.CSizeT) -> t.CInt:
|
||||
if not dict: return -2
|
||||
use: t.CSizeT = length
|
||||
if use > zdef.ZDEFLATE_WINDOW_SIZE:
|
||||
dict += length - zdef.ZDEFLATE_WINDOW_SIZE
|
||||
use = zdef.ZDEFLATE_WINDOW_SIZE
|
||||
memcpy(self.window, dict, use)
|
||||
self.window_pos = t.CInt(use)
|
||||
return 0
|
||||
|
||||
def copy(self) -> zinflate_stream | t.CPtr:
|
||||
z: zinflate_stream | t.CPtr = zdef.zdef_alloc(zinflate_stream.__sizeof__())
|
||||
if not z: return None
|
||||
memcpy(z, self, zinflate_stream.__sizeof__())
|
||||
z.window = BYTEPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE))
|
||||
z.output = BYTEPTR(zdef.zdef_alloc(self.output_cap))
|
||||
if not z.window or not z.output:
|
||||
zinflate_stream.destroy(z)
|
||||
return None
|
||||
memcpy(z.window, self.window, zdef.ZDEFLATE_WINDOW_SIZE)
|
||||
memcpy(z.output, self.output, self.output_cap)
|
||||
return z
|
||||
|
||||
def destroy(self):
|
||||
if not self: return
|
||||
if self.window: zdef.zdef_free(self.window)
|
||||
if self.output: zdef.zdef_free(self.output)
|
||||
zdef.zdef_free(self)
|
||||
|
||||
|
||||
def zinflate_one_shot(data: UINT8PTR, length: t.CSizeT,
|
||||
wbits: t.CInt, bufsize: t.CSizeT, out_len: t.CSizeT | t.CPtr) -> t.CUInt8T | t.CPtr:
|
||||
s: zinflate_stream | t.CPtr = zinflate_create(wbits)
|
||||
if not s: return None
|
||||
out: t.CUInt8T | t.CPtr = None
|
||||
err: t.CInt = s.decompress(data, length, 0, c.Addr(out), out_len)
|
||||
s.destroy()
|
||||
if err != 0: return None
|
||||
return out
|
||||
|
||||
|
||||
def zinflate_create(wbits: t.CInt) -> zinflate_stream | t.CPtr:
|
||||
s: zinflate_stream | t.CPtr = zdef.zdef_alloc(zinflate_stream.__sizeof__())
|
||||
if not s: return None
|
||||
memset(s, 0, zinflate_stream.__sizeof__())
|
||||
s.wbits = wbits
|
||||
s.is_initialized = 1
|
||||
s.adler = 1
|
||||
s.window = BYTEPTR(zdef.zdef_alloc(zdef.ZDEFLATE_WINDOW_SIZE))
|
||||
s.output = BYTEPTR(zdef.zdef_alloc(256))
|
||||
s.output_cap = 256
|
||||
s.output_len = 0
|
||||
s.max_length = 0
|
||||
if not s.window or not s.output:
|
||||
s.destroy()
|
||||
return None
|
||||
return s
|
||||
@@ -2,7 +2,9 @@ from stdint import *
|
||||
import zlib.pyzlib as pyzlib
|
||||
import zlib.zhuff as zhuff
|
||||
import zlib.zdef as zdef
|
||||
from stdio import printf, strlen
|
||||
from stdio import printf
|
||||
from string import strlen, memcmp, memcpy, memset
|
||||
from stdlib import malloc, free, realloc
|
||||
import stdlib
|
||||
import string
|
||||
import mpool
|
||||
|
||||
Reference in New Issue
Block a user