[PATCH 2/4] dev/io: Add base64 decoder

Gedare Bloom gedare at rtems.org
Mon Jan 22 15:34:18 UTC 2024


This seems to be generically useful by itself. I'd probably be OK with
it, if it has a suitable (non device-related) home.

On Mon, Jan 15, 2024 at 2:46 AM Sebastian Huber
<sebastian.huber at embedded-brains.de> wrote:
>
> From: Matthew Joyce <matthew.joyce at embedded-brains.de>
>
> ---
>  cpukit/dev/iobase64decode.c                   | 166 ++++++++++++++++
>  cpukit/include/rtems/dev/io.h                 |  63 ++++++
>  spec/build/cpukit/librtemscpu.yml             |   1 +
>  .../build/testsuites/unit/unit-no-clock-0.yml |   1 +
>  testsuites/unit/tc-io-base64-decode.c         | 180 ++++++++++++++++++
>  5 files changed, 411 insertions(+)
>  create mode 100644 cpukit/dev/iobase64decode.c
>  create mode 100644 testsuites/unit/tc-io-base64-decode.c
>
> diff --git a/cpukit/dev/iobase64decode.c b/cpukit/dev/iobase64decode.c
> new file mode 100644
> index 0000000000..5d3fe1c3cb
> --- /dev/null
> +++ b/cpukit/dev/iobase64decode.c
> @@ -0,0 +1,166 @@
> +/* SPDX-License-Identifier: BSD-2-Clause */
> +
> +/**
> + * @file
> + *
> + * @ingroup RTEMSDeviceIO
> + *
> + * @brief This source file contains the implementation of
> + *   _IO_Base64_decode_initialize() and _IO_Base64_decode().
> + */
> +
> +/*
> + * Copyright (C) 2023 embedded brains GmbH & Co. KG
> + *
> + * Redistribution and use in source and binary forms, with or without
> + * modification, are permitted provided that the following conditions
> + * are met:
> + * 1. Redistributions of source code must retain the above copyright
> + *    notice, this list of conditions and the following disclaimer.
> + * 2. Redistributions in binary form must reproduce the above copyright
> + *    notice, this list of conditions and the following disclaimer in the
> + *    documentation and/or other materials provided with the distribution.
> + *
> + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
> + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
> + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
> + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
> + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
> + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
> + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
> + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
> + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
> + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
> + * POSSIBILITY OF SUCH DAMAGE.
> + */
> +
> +#include <rtems/dev/io.h>
> +
> +#define SPACE 253
> +
> +#define PAD 254
> +
> +#define INVALID 255
> +
> +const uint8_t _IO_Base64_decode_table[128] = {
> +    ['A'] = 0,       ['B'] = 1,       ['C'] = 2,       ['D'] = 3,
> +    ['E'] = 4,       ['F'] = 5,       ['G'] = 6,       ['H'] = 7,
> +    ['I'] = 8,       ['J'] = 9,       ['K'] = 10,      ['L'] = 11,
> +    ['M'] = 12,      ['N'] = 13,      ['O'] = 14,      ['P'] = 15,
> +    ['Q'] = 16,      ['R'] = 17,      ['S'] = 18,      ['T'] = 19,
> +    ['U'] = 20,      ['V'] = 21,      ['W'] = 22,      ['X'] = 23,
> +    ['Y'] = 24,      ['Z'] = 25,      ['a'] = 26,      ['b'] = 27,
> +    ['c'] = 28,      ['d'] = 29,      ['e'] = 30,      ['f'] = 31,
> +    ['g'] = 32,      ['h'] = 33,      ['i'] = 34,      ['j'] = 35,
> +    ['k'] = 36,      ['l'] = 37,      ['m'] = 38,      ['n'] = 39,
> +    ['o'] = 40,      ['p'] = 41,      ['q'] = 42,      ['r'] = 43,
> +    ['s'] = 44,      ['t'] = 45,      ['u'] = 46,      ['v'] = 47,
> +    ['w'] = 48,      ['x'] = 49,      ['y'] = 50,      ['z'] = 51,
> +    ['0'] = 52,      ['1'] = 53,      ['2'] = 54,      ['3'] = 55,
> +    ['4'] = 56,      ['5'] = 57,      ['6'] = 58,      ['7'] = 59,
> +    ['8'] = 60,      ['9'] = 61,      ['+'] = 62,      ['-'] = 62,
> +    ['/'] = 63,      ['_'] = 63,      ['='] = PAD,     [' '] = SPACE,
> +    ['\t'] = SPACE,  ['\n'] = SPACE,  ['\v'] = SPACE,  ['\f'] = SPACE,
> +    ['\r'] = SPACE,  [0] = INVALID,   [1] = INVALID,   [2] = INVALID,
> +    [3] = INVALID,   [4] = INVALID,   [5] = INVALID,   [6] = INVALID,
> +    [7] = INVALID,   [8] = INVALID,   [14] = INVALID,  [15] = INVALID,
> +    [16] = INVALID,  [17] = INVALID,  [18] = INVALID,  [19] = INVALID,
> +    [20] = INVALID,  [21] = INVALID,  [22] = INVALID,  [23] = INVALID,
> +    [24] = INVALID,  [25] = INVALID,  [26] = INVALID,  [27] = INVALID,
> +    [28] = INVALID,  [29] = INVALID,  [30] = INVALID,  [31] = INVALID,
> +    [33] = INVALID,  [34] = INVALID,  [35] = INVALID,  [36] = INVALID,
> +    [37] = INVALID,  [38] = INVALID,  [39] = INVALID,  [40] = INVALID,
> +    [41] = INVALID,  [42] = INVALID,  [44] = INVALID,  [46] = INVALID,
> +    [58] = INVALID,  [59] = INVALID,  [60] = INVALID,  [62] = INVALID,
> +    [63] = INVALID,  [64] = INVALID,  [91] = INVALID,  [92] = INVALID,
> +    [93] = INVALID,  [94] = INVALID,  [96] = INVALID,  [123] = INVALID,
> +    [124] = INVALID, [125] = INVALID, [126] = INVALID, [127] = INVALID};
> +
> +void _IO_Base64_decode_initialize(IO_Base64_decode_control* self,
> +                                  uint8_t* target,
> +                                  size_t target_size) {
> +  self->state = IO_BASE64_DECODE_STATE_0;
> +  self->target = target;
> +  self->target_end = target + target_size;
> +}
> +
> +IO_Base64_decode_status _IO_Base64_decode(IO_Base64_decode_control* self,
> +                                          char ch) {
> +  uint8_t decoded_ch;
> +  uint8_t next_ch;
> +  uint8_t* target;
> +  const uint8_t* target_end;
> +  IO_Base64_decode_state next_state;
> +
> +  if ((unsigned char)ch >= 128) {
> +    return IO_BASE64_DECODE_INVALID_INPUT;
> +  }
> +
> +  decoded_ch = _IO_Base64_decode_table[(unsigned char)ch];
> +
> +  if (decoded_ch == SPACE) {
> +    return IO_BASE64_DECODE_SUCCESS;
> +  }
> +
> +  target = self->target;
> +
> +  if (decoded_ch == PAD) {
> +    self->target_end = target;
> +    return IO_BASE64_DECODE_SUCCESS;
> +  }
> +
> +  if (decoded_ch == INVALID) {
> +    return IO_BASE64_DECODE_INVALID_INPUT;
> +  }
> +
> +  target_end = self->target_end;
> +
> +  if (target == target_end) {
> +    return IO_BASE64_DECODE_OVERFLOW;
> +  }
> +
> +  switch (self->state) {
> +    case IO_BASE64_DECODE_STATE_0:
> +      *target = decoded_ch << 2;
> +      next_state = IO_BASE64_DECODE_STATE_1;
> +      break;
> +
> +    case IO_BASE64_DECODE_STATE_1:
> +      *target |= decoded_ch >> 4;
> +      next_ch = (decoded_ch & 0x0fU) << 4;
> +      ++target;
> +
> +      if (target != target_end) {
> +        *target = next_ch;
> +      } else if (next_ch != 0) {
> +        return IO_BASE64_DECODE_OVERFLOW;
> +      }
> +
> +      next_state = IO_BASE64_DECODE_STATE_2;
> +      break;
> +
> +    case IO_BASE64_DECODE_STATE_2:
> +      *target |= decoded_ch >> 2;
> +      next_ch = (decoded_ch & 0x03U) << 6;
> +      ++target;
> +
> +      if (target != target_end) {
> +        *target = next_ch;
> +      } else if (next_ch != 0) {
> +        return IO_BASE64_DECODE_OVERFLOW;
> +      }
> +
> +      next_state = IO_BASE64_DECODE_STATE_3;
> +      break;
> +
> +    default: /* IO_BASE64_DECODE_STATE_3 */
> +      *target |= decoded_ch;
> +      ++target;
> +      next_state = IO_BASE64_DECODE_STATE_0;
> +      break;
> +  }
> +
> +  self->state = next_state;
> +  self->target = target;
> +  return IO_BASE64_DECODE_SUCCESS;
> +}
> diff --git a/cpukit/include/rtems/dev/io.h b/cpukit/include/rtems/dev/io.h
> index efe0a32d0a..9b724e5df8 100644
> --- a/cpukit/include/rtems/dev/io.h
> +++ b/cpukit/include/rtems/dev/io.h
> @@ -178,6 +178,69 @@ int _IO_Base64url(
>    int          wordlen
>  );
>
> +/**
> + * @brief Represents the base64 decoder state.
> + */
> +typedef enum {
> +  IO_BASE64_DECODE_STATE_0,
> +  IO_BASE64_DECODE_STATE_1,
> +  IO_BASE64_DECODE_STATE_2,
> +  IO_BASE64_DECODE_STATE_3
> +} IO_Base64_decode_state;
> +
> +/**
> + * @brief Contains the base64 decoder control.
> + */
> +typedef struct {
> +  IO_Base64_decode_state state;
> +  uint8_t *target;
> +  const uint8_t *target_end;
> +} IO_Base64_decode_control;
> +
> +/**
> + * @brief Maps a 7-bit character to the 6-bit integer as defined by the base64
> + *   or base64url encoding or a special value.
> + */
> +extern const uint8_t _IO_Base64_decode_table[ 128 ];
> +
> +/**
> + * @brief Initializes the base64 decoder.
> + *
> + * @param[out] self is the base64 decoder control to initialize.
> + *
> + * @param[out] target is the base address of the target area for decoding.
> + *
> + * @param target_size is the size in bytes of the target area for decoding.
> + */
> +void _IO_Base64_decode_initialize(
> +  IO_Base64_decode_control *self,
> +  uint8_t                  *target,
> +  size_t                    target_size
> +);
> +
> +/**
> + * @brief Represents the base64 decoder status.
> + */
> +typedef enum {
> +  IO_BASE64_DECODE_SUCCESS,
> +  IO_BASE64_DECODE_OVERFLOW,
> +  IO_BASE64_DECODE_INVALID_INPUT
> +} IO_Base64_decode_status;
> +
> +/**
> + * @brief Decodes the character.
> + *
> + * The decoder accepts base64 and base64url encodings.  White space is ignored.
> + *
> + * @param[in, out] self is the base64 decoder control.
> + *
> + * @param ch is the character to decode.
> + */
> +IO_Base64_decode_status _IO_Base64_decode(
> +  IO_Base64_decode_control *self,
> +  char                      ch
> +);
> +
>  /**
>   * @brief Issues a couple of no-operation instructions.
>   *
> diff --git a/spec/build/cpukit/librtemscpu.yml b/spec/build/cpukit/librtemscpu.yml
> index 9202c31715..56e5243673 100644
> --- a/spec/build/cpukit/librtemscpu.yml
> +++ b/spec/build/cpukit/librtemscpu.yml
> @@ -540,6 +540,7 @@ source:
>  - cpukit/dev/i2c/ti-tmp112.c
>  - cpukit/dev/i2c/xilinx-axi-i2c.c
>  - cpukit/dev/iobase64.c
> +- cpukit/dev/iobase64decode.c
>  - cpukit/dev/ioprintf.c
>  - cpukit/dev/iorelax.c
>  - cpukit/dev/iovprintf.c
> diff --git a/spec/build/testsuites/unit/unit-no-clock-0.yml b/spec/build/testsuites/unit/unit-no-clock-0.yml
> index 5d15bf52b7..be5cafe990 100644
> --- a/spec/build/testsuites/unit/unit-no-clock-0.yml
> +++ b/spec/build/testsuites/unit/unit-no-clock-0.yml
> @@ -13,6 +13,7 @@ links: []
>  source:
>  - testsuites/unit/tc-compiler-builtins.c
>  - testsuites/unit/tc-config.c
> +- testsuites/unit/tc-io-base64-decode.c
>  - testsuites/unit/tc-misaligned-builtin-memcpy.c
>  - testsuites/unit/tc-score-msgq.c
>  - testsuites/unit/tc-score-rbtree.c
> diff --git a/testsuites/unit/tc-io-base64-decode.c b/testsuites/unit/tc-io-base64-decode.c
> new file mode 100644
> index 0000000000..b09390960f
> --- /dev/null
> +++ b/testsuites/unit/tc-io-base64-decode.c
> @@ -0,0 +1,180 @@
> +/* SPDX-License-Identifier: BSD-2-Clause */
> +
> +/*
> + * Copyright (C) 2023 embedded brains GmbH & Co. KG
> + *
> + * Redistribution and use in source and binary forms, with or without
> + * modification, are permitted provided that the following conditions
> + * are met:
> + * 1. Redistributions of source code must retain the above copyright
> + *    notice, this list of conditions and the following disclaimer.
> + * 2. Redistributions in binary form must reproduce the above copyright
> + *    notice, this list of conditions and the following disclaimer in the
> + *    documentation and/or other materials provided with the distribution.
> + *
> + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
> + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
> + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
> + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
> + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
> + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
> + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
> + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
> + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
> + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
> + * POSSIBILITY OF SUCH DAMAGE.
> + */
> +
> +#include <rtems/dev/io.h>
> +
> +#include <rtems/test.h>
> +
> +typedef struct {
> +  IO_Base64_decode_control base;
> +  uint8_t buf[64];
> +} test_control;
> +
> +static int test_run(test_control* self, const char* payload) {
> +  size_t len = strlen(payload);
> +
> +  for (size_t i = 0; i < len; ++i) {
> +    int rv = _IO_Base64_decode(&self->base, payload[i]);
> +
> +    if (rv != 0) {
> +      return rv;
> +    }
> +  }
> +
> +  return 0;
> +}
> +
> +static int test_payload(test_control* self, const char* payload) {
> +  memset(self->buf, 0xff, sizeof(self->buf));
> +  _IO_Base64_decode_initialize(&self->base, &self->buf[0], sizeof(self->buf));
> +  return test_run(self, payload);
> +}
> +
> +T_TEST_CASE(IOBase64Decode) {
> +  int rv;
> +  test_control instance;
> +  test_control* self = &instance;
> +
> +  rv = test_payload(self, "POOL");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->base.target, &self->buf[3]);
> +  const uint8_t expected1[] = {0x3c, 0xe3, 0x8b, 0xff};
> +  T_eq_mem(&self->buf[0], expected1, sizeof(expected1));
> +
> +  rv = test_payload(self, "ABCDEFGH");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->base.target, &self->buf[6]);
> +  const uint8_t expected2[] = {0x00, 0x10, 0x83, 0x10, 0x51, 0x87, 0xff};
> +  T_eq_mem(&self->buf[0], expected2, sizeof(expected2));
> +
> +  /* Non-base64 character results in an error */
> +  rv = test_payload(self, "PO*OL");
> +  T_eq_int(rv, IO_BASE64_DECODE_INVALID_INPUT);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_2);
> +  T_eq_ptr(self->base.target, &self->buf[1]);
> +  const uint8_t expected3[] = {0x3c};
> +  T_eq_mem(&self->buf[0], expected3, sizeof(expected3));
> +
> +  /* Other non-base64 character results in an error */
> +  rv = test_payload(self, "PO\x80OL");
> +  T_eq_int(rv, IO_BASE64_DECODE_INVALID_INPUT);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_2);
> +  T_eq_ptr(self->base.target, &self->buf[1]);
> +  T_eq_mem(&self->buf[0], expected3, sizeof(expected3));
> +
> +  /* Space characters should be ignored */
> +  rv = test_payload(self, "P OOL");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->base.target, &self->buf[3]);
> +  const uint8_t expected4[] = {0x3c, 0xe3, 0x8b, 0xff};
> +  T_eq_mem(&self->buf[0], expected4, sizeof(expected4));
> +
> +  /* Handle pad characters */
> +  rv = test_payload(self, "POOL==");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->base.target, &self->buf[3]);
> +  T_eq_ptr(self->base.target, self->base.target_end);
> +  const uint8_t expected5[] = {0x3c, 0xe3, 0x8b, 0xff};
> +  T_eq_mem(&self->buf[0], expected5, sizeof(expected5));
> +
> +  /* If characters come after pad characters, an error results */
> +  rv = test_payload(self, "POOL==xy");
> +  T_eq_int(rv, IO_BASE64_DECODE_OVERFLOW);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_ptr(self->base.target, &self->buf[3]);
> +  T_eq_ptr(self->base.target, self->base.target_end);
> +  const uint8_t expected6[] = {0x3c, 0xe3, 0x8b, 0xff};
> +  T_eq_mem(&self->buf[0], expected6, sizeof(expected6));
> +
> +  rv = test_payload(self, "POOLPOOL");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->base.target, &self->buf[6]);
> +  const uint8_t expected7[] = {0x3c, 0xe3, 0x8b, 0x3c, 0xe3, 0x8b, 0xff};
> +  T_eq_mem(&self->buf[0], expected7, sizeof(expected7));
> +
> +  /*
> +   * Test valid payload with series of target sizes. All target sizes
> +   * less than three are invalid for the given payload and will result
> +   * in an error.
> +   */
> +  const uint8_t expected9[] = {0x3c, 0xe3, 0x8b, 0xff};
> +
> +  for (size_t i = 0; i < 4; ++i) {
> +    memset(&self->buf[0], 0xff, sizeof(self->buf));
> +    _IO_Base64_decode_initialize(&self->base, &self->buf[0], i);
> +    rv = test_run(self, "POOL");
> +
> +    if (i < 3) {
> +      T_eq_int(rv, IO_BASE64_DECODE_OVERFLOW);
> +      T_eq_int(self->base.state, i);
> +      T_ne_ptr(self->base.target, &self->buf[3]);
> +      T_ne_mem(&self->buf[0], expected9, sizeof(expected9));
> +    } else {
> +      T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +      T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_0);
> +      T_eq_ptr(self->base.target, &self->buf[3]);
> +      T_eq_mem(&self->buf[0], expected9, sizeof(expected9));
> +    }
> +  }
> +
> +  /* No overflow in state 1 */
> +  memset(&self->buf[0], 0xff, sizeof(self->buf));
> +  _IO_Base64_decode_initialize(&self->base, &self->buf[0], 1);
> +  rv = test_run(self, "AA");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_2);
> +  T_eq_ptr(self->base.target, &self->buf[1]);
> +  const uint8_t expected10[] = {0x00, 0xff};
> +  T_eq_mem(&self->buf[0], expected10, sizeof(expected10));
> +
> +  /* No overflow in state 2 */
> +  memset(&self->buf[0], 0xff, sizeof(self->buf));
> +  _IO_Base64_decode_initialize(&self->base, &self->buf[0], 2);
> +  rv = test_run(self, "AAA");
> +  T_eq_int(rv, IO_BASE64_DECODE_SUCCESS);
> +  T_eq_int(self->base.state, IO_BASE64_DECODE_STATE_3);
> +  T_eq_ptr(self->base.target, &self->buf[2]);
> +  const uint8_t expected11[] = {0x00, 0x00, 0xff};
> +  T_eq_mem(&self->buf[0], expected11, sizeof(expected11));
> +}
> +
> +T_TEST_CASE(IOBase64DecodeInitialize) {
> +  IO_Base64_decode_control instance;
> +  IO_Base64_decode_control* self = &instance;
> +  uint8_t buf[1];
> +
> +  memset(self, 0xff, sizeof(*self));
> +  _IO_Base64_decode_initialize(self, buf, sizeof(buf));
> +  T_eq_int(self->state, IO_BASE64_DECODE_STATE_0);
> +  T_eq_ptr(self->target, &buf[0]);
> +  T_eq_ptr(self->target_end, &buf[1]);
> +}
> --
> 2.35.3
>
> _______________________________________________
> devel mailing list
> devel at rtems.org
> http://lists.rtems.org/mailman/listinfo/devel


More information about the devel mailing list