pybase16384 0.3.5__cp38-cp38-macosx_11_0_arm64.whl → 0.3.7__cp38-cp38-macosx_11_0_arm64.whl

pybase16384/__init__.py

@@ -17,38 +17,68 @@ def _should_use_cffi() -> bool:
 
 if not _should_use_cffi():
     from pybase16384.backends.cython import (
+        DECBUFSZ,
+        ENCBUFSZ,
+        FLAG_DO_SUM_CHECK_FORCELY,
+        FLAG_NOHEADER,
+        FLAG_SUM_CHECK_ON_REMAIN,
         _decode,
         _decode_into,
+        _decode_into_safe,
+        _decode_safe,
         _encode,
         _encode_into,
+        _encode_into_safe,
+        _encode_safe,
         decode_fd,
+        decode_fd_detailed,
         decode_file,
+        decode_file_safe,
         decode_len,
         decode_local_file,
+        decode_local_file_detailed,
         encode_fd,
+        encode_fd_detailed,
         encode_file,
+        encode_file_safe,
         encode_len,
         encode_local_file,
+        encode_local_file_detailed,
         is_64bits,
     )
 else:
     from pybase16384.backends.cffi import (
+        DECBUFSZ,
+        ENCBUFSZ,
+        FLAG_DO_SUM_CHECK_FORCELY,
+        FLAG_NOHEADER,
+        FLAG_SUM_CHECK_ON_REMAIN,
         _decode,
         _decode_into,
+        _decode_into_safe,
+        _decode_safe,
         _encode,
         _encode_into,
+        _encode_into_safe,
+        _encode_safe,
         decode_fd,
+        decode_fd_detailed,
         decode_file,
+        decode_file_safe,
         decode_len,
         decode_local_file,
+        decode_local_file_detailed,
         encode_fd,
+        encode_fd_detailed,
         encode_file,
+        encode_file_safe,
         encode_len,
         encode_local_file,
+        encode_local_file_detailed,
         is_64bits,
     )
 
-__version__ = "0.3.5"
+__version__ = "0.3.7"
 
 
 def encode(data: bytes) -> bytes:
@@ -58,6 +88,13 @@ def encode(data: bytes) -> bytes:
     return out.getvalue()
 
 
+def encode_safe(data: bytes) -> bytes:
+    inp = BytesIO(data)
+    out = BytesIO()
+    encode_file_safe(inp, out, False, len(data) // 7)
+    return out.getvalue()
+
+
 def decode(data: bytes) -> bytes:
     inp = BytesIO(data)
     out = BytesIO()
@@ -65,6 +102,13 @@ def decode(data: bytes) -> bytes:
     return out.getvalue()
 
 
+def decode_safe(data: bytes) -> bytes:
+    inp = BytesIO(data)
+    out = BytesIO()
+    decode_file_safe(inp, out, len(data) // 8)
+    return out.getvalue()
+
+
 def encode_from_string(data: str, write_head: bool = False) -> bytes:
     bt = data.encode()
     inp = BytesIO(bt)
@@ -73,6 +117,14 @@ def encode_from_string(data: str, write_head: bool = False) -> bytes:
     return out.getvalue()
 
 
+def encode_from_string_safe(data: str, write_head: bool = False) -> bytes:
+    bt = data.encode()
+    inp = BytesIO(bt)
+    out = BytesIO()
+    encode_file_safe(inp, out, write_head, len(bt) // 7)
+    return out.getvalue()
+
+
 def encode_to_string(data: bytes) -> str:
     inp = BytesIO(data)
     out = BytesIO()
@@ -80,6 +132,13 @@ def encode_to_string(data: bytes) -> str:
     return out.getvalue().decode("utf-16-be")
 
 
+def encode_to_string_safe(data: bytes) -> str:
+    inp = BytesIO(data)
+    out = BytesIO()
+    encode_file_safe(inp, out, False, len(data) // 7)
+    return out.getvalue().decode("utf-16-be")
+
+
 def encode_string(data: str) -> str:
     data = data.encode()
     inp = BytesIO(data)
@@ -88,6 +147,14 @@ def encode_string(data: str) -> str:
     return out.getvalue().decode("utf-16-be")
 
 
+def encode_string_safe(data: str) -> str:
+    data = data.encode()
+    inp = BytesIO(data)
+    out = BytesIO()
+    encode_file_safe(inp, out, False, len(data) // 7)
+    return out.getvalue().decode("utf-16-be")
+
+
 def decode_from_bytes(data: bytes) -> str:
     inp = BytesIO(data)
     out = BytesIO()
@@ -95,6 +162,13 @@ def decode_from_bytes(data: bytes) -> str:
     return out.getvalue().decode()
 
 
+def decode_from_bytes_safe(data: bytes) -> str:
+    inp = BytesIO(data)
+    out = BytesIO()
+    decode_file_safe(inp, out, len(data) // 8)
+    return out.getvalue().decode()
+
+
 def decode_from_string(data: str) -> bytes:
     bt = data.encode("utf-16-be")
     inp = BytesIO(bt)
@@ -103,9 +177,25 @@ def decode_from_string(data: str) -> bytes:
     return out.getvalue()
 
 
+def decode_from_string_safe(data: str) -> bytes:
+    bt = data.encode("utf-16-be")
+    inp = BytesIO(bt)
+    out = BytesIO()
+    decode_file_safe(inp, out, len(bt) // 8)
+    return out.getvalue()
+
+
 def decode_string(data: str) -> str:
     bt = data.encode("utf-16-be")
     inp = BytesIO(bt)
     out = BytesIO()
     decode_file(inp, out, len(bt) // 8)
     return out.getvalue().decode()
+
+
+def decode_string_safe(data: str) -> str:
+    bt = data.encode("utf-16-be")
+    inp = BytesIO(bt)
+    out = BytesIO()
+    decode_file_safe(inp, out, len(bt) // 8)
+    return out.getvalue().decode()

pybase16384/backends/cffi/__init__.py

@@ -11,6 +11,12 @@ __version__ = "0.1.0"
 encode_len = lib.base16384_encode_len
 decode_len = lib.base16384_decode_len
 
+ENCBUFSZ = lib.get_encsize()
+DECBUFSZ = lib.get_decsize()
+FLAG_NOHEADER = lib.BASE16384_FLAG_NOHEADER_()
+FLAG_SUM_CHECK_ON_REMAIN = lib.BASE16384_FLAG_SUM_CHECK_ON_REMAIN_()
+FLAG_DO_SUM_CHECK_FORCELY = lib.BASE16384_FLAG_DO_SUM_CHECK_FORCELY_()
+
 
 # -----------------low level api------------------------------
 def _encode(data: bytes) -> bytes:
@@ -23,8 +29,22 @@ def _encode(data: bytes) -> bytes:
     return ffi.unpack(output_buf, count)
 
 
+def _encode_safe(data: bytes) -> bytes:
+    length = len(data)
+    output_size = encode_len(length)
+    output_buf = ffi.new(f"char[{output_size}]")
+    if output_buf == ffi.NULL:
+        raise MemoryError
+    count = lib.base16384_encode_safe(ffi.from_buffer(data), length, output_buf)
+    return ffi.unpack(output_buf, count)
+
+
 def _encode_into(data: bytes, out: bytearray) -> int:
-    return lib.base16384_encode(
+    return lib.base16384_encode(ffi.from_buffer(data), len(data), ffi.from_buffer(out))
+
+
+def _encode_into_safe(data: bytes, out: bytearray) -> int:
+    return lib.base16384_encode_safe(
         ffi.from_buffer(data), len(data), ffi.from_buffer(out)
     )
 
@@ -39,8 +59,22 @@ def _decode(data: bytes) -> bytes:
     return ffi.unpack(output_buf, count)
 
 
+def _decode_safe(data: bytes) -> bytes:
+    length = len(data)
+    output_size = decode_len(length, 0)
+    output_buf = ffi.new(f"char[{output_size}]")
+    if output_buf == ffi.NULL:
+        raise MemoryError
+    count = lib.base16384_decode_safe(ffi.from_buffer(data), length, output_buf)
+    return ffi.unpack(output_buf, count)
+
+
 def _decode_into(data: bytes, out: bytearray) -> int:
-    return lib.base16384_decode(
+    return lib.base16384_decode(ffi.from_buffer(data), len(data), ffi.from_buffer(out))
+
+
+def _decode_into_safe(data: bytes, out: bytearray) -> int:
+    return lib.base16384_decode_safe(
         ffi.from_buffer(data), len(data), ffi.from_buffer(out)
     )
 
@@ -92,9 +126,51 @@ def encode_file(input: IO, output: IO, write_head: bool = False, buf_rate: int =
                 input.seek(-size, 1)
                 continue
 
-        count = lib.base16384_encode(
-            ffi.from_buffer(chunk), size, output_buf
+        count = lib.base16384_encode(ffi.from_buffer(chunk), size, output_buf)
+        output.write(ffi.unpack(output_buf, count))
+        if size < 7:
+            break
+
+
+def encode_file_safe(
+    input: IO, output: IO, write_head: bool = False, buf_rate: int = 10
+):
+    if not _check_file(input):
+        raise TypeError(
+            "input except a file-like object, got %s" % type(input).__name__
+        )
+    if not _check_file(output):
+        raise TypeError(
+            "output except a file-like object, got %s" % type(input).__name__
         )
+    if buf_rate <= 0:
+        buf_rate = 1
+    if write_head:
+        output.write(b"\xfe\xff")
+
+    current_buf_len: int = buf_rate * 7  # 一次读取这么多字节
+    output_size: int = encode_len(current_buf_len)  # 因为encode_len不是单调的 safe不用加16
+    output_buf = ffi.new(f"char[{output_size}]")
+    if output_buf == ffi.NULL:
+        raise MemoryError
+    first_check: int = 1  # 检查一次就行了 怎么可能出现第一次读出来是bytes 以后又变卦了的对象呢 不会吧不会吧
+    while True:
+        chunk = input.read(current_buf_len)
+        if first_check:
+            first_check = 0
+            if not isinstance(chunk, bytes):
+                raise TypeError(
+                    f"input must be a file-like rb object, got {type(input).__name__}"
+                )
+        size = len(chunk)
+        if size < current_buf_len:  # 数据不够了 要减小一次读取的量
+            if buf_rate > 1:  # 重新设置一次读取的大小 重新设置流的位置 当然要是已经是一次读取7字节了 那就不能再变小了 直接encode吧
+                buf_rate = buf_rate // 2
+                current_buf_len = buf_rate * 7
+                input.seek(-size, 1)
+                continue
+
+        count = lib.base16384_encode_safe(ffi.from_buffer(chunk), size, output_buf)
         output.write(ffi.unpack(output_buf, count))
         if size < 7:
             break
@@ -146,9 +222,57 @@ def decode_file(input: IO, output: IO, buf_rate: int = 10):
             else:
                 input.seek(-2, 1)
 
-        count = lib.base16384_decode(
-            ffi.from_buffer(chunk), size, output_buf
+        count = lib.base16384_decode(ffi.from_buffer(chunk), size, output_buf)
+        output.write(ffi.unpack(output_buf, count))
+
+
+def decode_file_safe(input: IO, output: IO, buf_rate: int = 10):
+    if not _check_file(input):
+        raise TypeError(
+            "input except a file-like object, got %s" % type(input).__name__
         )
+    if not _check_file(output):
+        raise TypeError(
+            "output except a file-like object, got %s" % type(output).__name__
+        )
+    if buf_rate <= 0:
+        buf_rate = 1
+
+    chunk = input.read(1)  # type: bytes
+    if not isinstance(chunk, bytes):
+        raise TypeError(
+            f"input must be a file-like rb object, got {type(input).__name__}"
+        )
+    if chunk == b"\xfe":  # 去头
+        input.read(1)
+    else:
+        input.seek(0, 0)  # 没有头 回到开头
+
+    current_buf_len: int = buf_rate * 8
+    output_size: int = decode_len(current_buf_len, 0)
+    output_buf = ffi.new(f"char[{output_size}]")
+    if output_buf == ffi.NULL:
+        raise MemoryError
+    while True:
+        chunk = input.read(current_buf_len)  # 8的倍数
+        size = len(chunk)
+        if size == 0:
+            break
+        if size < current_buf_len:  # 长度不够了
+            if buf_rate > 1:  # 还能继续变小
+                buf_rate = buf_rate // 2  # 重新设置一次读取的大小
+                current_buf_len = buf_rate * 8
+                input.seek(-size, 1)
+                continue
+        tmp = input.read(2)  # type: bytes
+        if len(tmp) == 2:
+            if tmp[0] == 61:  # = stream完了   一次解码8n+2个字节
+                chunk += tmp
+                size += 2
+            else:
+                input.seek(-2, 1)
+
+        count = lib.base16384_decode_safe(ffi.from_buffer(chunk), size, output_buf)
         output.write(ffi.unpack(output_buf, count))
 
 
@@ -176,13 +300,23 @@ def err_to_str(ret) -> str:
         return "base16384_err_open_input_file"
     elif ret == lib.base16384_err_map_input_file:
         return "base16384_err_map_input_file"
+    elif ret == lib.base16384_err_read_file:
+        return "base16384_err_read_file"
+    elif ret == lib.base16384_err_invalid_file_name:
+        return "base16384_err_invalid_file_name"
+    elif ret == lib.base16384_err_invalid_file_name:
+        return "base16384_err_invalid_file_name"
+    elif ret == lib.base16384_err_invalid_commandline_parameter:
+        return "base16384_err_invalid_commandline_parameter"
+    elif ret == lib.base16384_err_invalid_decoding_checksum:
+        return "base16384_err_invalid_decoding_checksum"
 
 
 def encode_local_file(inp, out) -> None:
     inp_name: bytes = ensure_bytes(inp)
     out_name: bytes = ensure_bytes(out)
-    encbuf = ffi.new(f"char[{lib.get_encsize()}]")
-    decbuf = ffi.new(f"char[{lib.get_decsize()}]")
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
     ret = lib.base16384_encode_file(
         ffi.from_buffer(inp_name), ffi.from_buffer(out_name), encbuf, decbuf
     )
@@ -193,8 +327,8 @@ def encode_local_file(inp, out) -> None:
 def decode_local_file(inp, out) -> None:
     inp_name: bytes = ensure_bytes(inp)
     out_name: bytes = ensure_bytes(out)
-    encbuf = ffi.new(f"char[{lib.get_encsize()}]")
-    decbuf = ffi.new(f"char[{lib.get_decsize()}]")
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
     ret = lib.base16384_decode_file(
         ffi.from_buffer(inp_name), ffi.from_buffer(out_name), encbuf, decbuf
     )
@@ -203,16 +337,57 @@ def decode_local_file(inp, out) -> None:
 
 
 def encode_fd(inp: int, out: int) -> None:
-    encbuf = ffi.new(f"char[{lib.get_encsize()}]")
-    decbuf = ffi.new(f"char[{lib.get_decsize()}]")
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
     ret = lib.base16384_encode_fd(inp, out, encbuf, decbuf)
     if ret != lib.base16384_err_ok:
         raise ValueError(err_to_str(ret))
 
 
 def decode_fd(inp: int, out: int) -> None:
-    encbuf = ffi.new(f"char[{lib.get_encsize()}]")
-    decbuf = ffi.new(f"char[{lib.get_decsize()}]")
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
     ret = lib.base16384_decode_fd(inp, out, encbuf, decbuf)
     if ret != lib.base16384_err_ok:
         raise ValueError(err_to_str(ret))
+
+
+# detail
+def encode_local_file_detailed(inp, out, flag: int) -> None:
+    inp_name: bytes = ensure_bytes(inp)
+    out_name: bytes = ensure_bytes(out)
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
+    ret = lib.base16384_encode_file_detailed(
+        ffi.from_buffer(inp_name), ffi.from_buffer(out_name), encbuf, decbuf, flag
+    )
+    if ret != lib.base16384_err_ok:
+        raise ValueError(err_to_str(ret))
+
+
+def decode_local_file_detailed(inp, out, flag: int) -> None:
+    inp_name: bytes = ensure_bytes(inp)
+    out_name: bytes = ensure_bytes(out)
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
+    ret = lib.base16384_decode_file_detailed(
+        ffi.from_buffer(inp_name), ffi.from_buffer(out_name), encbuf, decbuf, flag
+    )
+    if ret != lib.base16384_err_ok:
+        raise ValueError(err_to_str(ret))
+
+
+def encode_fd_detailed(inp: int, out: int, flag: int) -> None:
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
+    ret = lib.base16384_encode_fd_detailed(inp, out, encbuf, decbuf, flag)
+    if ret != lib.base16384_err_ok:
+        raise ValueError(err_to_str(ret))
+
+
+def decode_fd_detailed(inp: int, out: int, flag: int) -> None:
+    encbuf = ffi.new(f"char[{ENCBUFSZ}]")
+    decbuf = ffi.new(f"char[{DECBUFSZ}]")
+    ret = lib.base16384_decode_fd_detailed(inp, out, encbuf, decbuf, flag)
+    if ret != lib.base16384_err_ok:
+        raise ValueError(err_to_str(ret))

pybase16384/backends/cffi/build.py

@@ -26,6 +26,7 @@ if sys.byteorder != "little":
 
 if CPUBIT == 64:
     macro_base.append(("CPUBIT64", None))
+    macro_base.append(("IS_64BIT_PROCESSOR", None))
 else:
     macro_base.append(("CPUBIT32", None))
 
@@ -41,6 +42,10 @@ enum base16384_err_t {
 	base16384_err_write_file,
 	base16384_err_open_input_file,
 	base16384_err_map_input_file,
+	base16384_err_read_file,
+    base16384_err_invalid_file_name,
+    base16384_err_invalid_commandline_parameter,
+    base16384_err_invalid_decoding_checksum
 };
 // base16384_err_t is the return value of base16384_en/decode_file
 typedef enum base16384_err_t base16384_err_t;
@@ -67,11 +72,29 @@ base16384_err_t base16384_decode_fp(FILE* input, FILE* output, char* encbuf, cha
 //    encbuf & decbuf must be no less than BASE16384_ENCBUFSZ & BASE16384_DECBUFSZ
 base16384_err_t base16384_decode_fd(int input, int output, char* encbuf, char* decbuf);
 
+int base16384_encode_unsafe(const char * data, int dlen, char * buf);
+int base16384_decode_unsafe(const char * data, int dlen, char * buf);
+int base16384_encode_safe(const char * data, int dlen, char * buf);
+int base16384_decode_safe(const char * data, int dlen, char * buf);
+
+base16384_err_t base16384_encode_file_detailed(const char* input, const char* output, char* encbuf, char* decbuf, int flag);
+base16384_err_t base16384_decode_file_detailed(const char* input, const char* output, char* encbuf, char* decbuf, int flag);
+base16384_err_t base16384_encode_fd_detailed(int input, int output, char* encbuf, char* decbuf, int flag);
+base16384_err_t base16384_decode_fd_detailed(int input, int output, char* encbuf, char* decbuf, int flag);
+base16384_err_t base16384_encode_fp_detailed(FILE* input, FILE* output, char* encbuf, char* decbuf, int flag);
+base16384_err_t base16384_decode_fp_detailed(FILE* input, FILE* output, char* encbuf, char* decbuf, int flag);
+
 int32_t pybase16384_64bits();
 
 int get_encsize();
 
 int get_decsize();
+
+int BASE16384_FLAG_NOHEADER_();
+
+int BASE16384_FLAG_SUM_CHECK_ON_REMAIN_();
+
+int BASE16384_FLAG_DO_SUM_CHECK_FORCELY_();
     """
 )
 
@@ -79,9 +102,9 @@ source = """
 #include "base16384.h"
 
 #ifdef CPUBIT32
-#define pybase16384_64bits() 0
+#define pybase16384_64bits() (0)
 #else
-#define pybase16384_64bits() 1
+#define pybase16384_64bits() (1)
 #endif
 
 int get_encsize()
@@ -93,6 +116,21 @@ int get_decsize()
 {
     return BASE16384_DECBUFSZ;
 }
+
+int BASE16384_FLAG_NOHEADER_()
+{
+    return BASE16384_FLAG_NOHEADER;
+}
+
+int BASE16384_FLAG_SUM_CHECK_ON_REMAIN_()
+{
+    return BASE16384_FLAG_SUM_CHECK_ON_REMAIN;
+}
+
+int BASE16384_FLAG_DO_SUM_CHECK_FORCELY_()
+{
+    return BASE16384_FLAG_DO_SUM_CHECK_FORCELY;
+}
 """
 
 ffibuilder.set_source(

pybase16384/backends/cython/__init__.py

@@ -2,17 +2,32 @@
 Copyright (c) 2008-2021 synodriver <synodriver@gmail.com>
 """
 from pybase16384.backends.cython._core import (
+    DECBUFSZ,
+    ENCBUFSZ,
+    FLAG_DO_SUM_CHECK_FORCELY,
+    FLAG_NOHEADER,
+    FLAG_SUM_CHECK_ON_REMAIN,
     _decode,
     _decode_into,
+    _decode_into_safe,
+    _decode_safe,
     _encode,
     _encode_into,
+    _encode_into_safe,
+    _encode_safe,
     decode_fd,
+    decode_fd_detailed,
     decode_file,
+    decode_file_safe,
     decode_len,
     decode_local_file,
+    decode_local_file_detailed,
     encode_fd,
+    encode_fd_detailed,
     encode_file,
+    encode_file_safe,
     encode_len,
     encode_local_file,
+    encode_local_file_detailed,
     is_64bits,
 )