[MDEV-31068] Reuse duplicate case conversion code in ctype-utf8.c and ctype-ucs2.c - Jira

Details

Type: Task
Status: Closed (View Workflow)
Priority: Critical
Resolution: Fixed
Fix Version/s: 11.1.1, 10.11.3, 11.0.2, 10.10.4
Component/s: Character Sets
Labels:
None

Description

Files ctype-utf8.c and ctype-ucs2.c have very similar pieces of the code:

static inline void

my_tolower_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].toupper;

static inline void

my_tolower_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].toupper;

static inline void

my_tolower_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))

    *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))

    *wc= page[*wc & 0xFF].toupper;

static inline void

my_tolower_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))

    *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  MY_UNICASE_CHARACTER *page;

  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))

    *wc= page[*wc & 0xFF].toupper;

static inline void

my_tolower_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  if (*wc <= uni_plane->maxchar)

    MY_UNICASE_CHARACTER *page;

    if ((page= uni_plane->page[(*wc >> 8)]))

      *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  if (*wc <= uni_plane->maxchar)

    MY_UNICASE_CHARACTER *page;

    if ((page= uni_plane->page[(*wc >> 8)]))

      *wc= page[*wc & 0xFF].toupper;

In order to simplify the patch for ~~MDEV-30577~~ lets move the repeatable code to a shared file ctype-unidata.h:

/* utf8mb3 and ucs2 share this code */

static inline void

my_tolower_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  const MY_UNICASE_CHARACTER *page;

  DBUG_ASSERT(*wc <= uni_plane->maxchar);

  if ((page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  const MY_UNICASE_CHARACTER *page;

  DBUG_ASSERT(*wc <= uni_plane->maxchar);

  if ((page= uni_plane->page[*wc >> 8]))

    *wc= page[*wc & 0xFF].toupper;

/* utf8mb4, utf16, utf32 share this code */

static inline void

my_tolower_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  if (*wc <= uni_plane->maxchar)

    const MY_UNICASE_CHARACTER *page;

    if ((page= uni_plane->page[(*wc >> 8)]))

      *wc= page[*wc & 0xFF].tolower;

static inline void

my_toupper_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)

  if (*wc <= uni_plane->maxchar)

    const MY_UNICASE_CHARACTER *page;

    if ((page= uni_plane->page[(*wc >> 8)]))

      *wc= page[*wc & 0xFF].toupper;

Attachments

Issue Links

blocks

MDEV-30577 Case folding for uca1400 collations is not up to date

Closed

relates to

MDEV-31069 Reuse duplicate char-to-weight conversion code in ctype-utf8.c and ctype-ucs2.c

Closed

Activity

Ascending order - Click to sort in descending order

Alexander Barkov created issue - 2023-04-18 02:39

Alexander Barkov made changes - 2023-04-18 02:39

Field	Original Value	New Value
Link		This issue blocks ~~MDEV-30577~~ [ ~~MDEV-30577~~ ]

Alexander Barkov made changes - 2023-04-18 02:40

Priority

Major [ 3 ]

Critical [ 2 ]

Alexander Barkov made changes - 2023-04-18 02:41

Description

Files ctype-utf8.c and ctype-ucs2.c have very similar pieces of the code:

{code:cpp}
static inline void
my_tolower_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].tolower;
  }
}

static inline void
my_toupper_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].toupper;
  }
}
{code}

In order to simplify the patch for ~~MDEV-30577~~ lets move the repeatable code to a shared file ctype-unidata.h:

{code:cpp}
static inline void
my_tolower_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  const MY_UNICASE_CHARACTER *page;
  DBUG_ASSERT(*wc <= uni_plane->maxchar);
  if ((page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  const MY_UNICASE_CHARACTER *page;
  DBUG_ASSERT(*wc <= uni_plane->maxchar);
  if ((page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    const MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].tolower;
  }
}

static inline void
my_toupper_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    const MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].toupper;
  }
}
{code}

Files ctype-utf8.c and ctype-ucs2.c have very similar pieces of the code:

{code:cpp}
static inline void
my_tolower_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf16(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf32(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((*wc <= uni_plane->maxchar) && (page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_ucs2(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_utf8mb3(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  MY_UNICASE_CHARACTER *page;
  if ((page= uni_plane->page[(*wc >> 8) & 0xFF]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
static inline void
my_tolower_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].tolower;
  }
}

static inline void
my_toupper_utf8mb4(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].toupper;
  }
}
{code}

In order to simplify the patch for ~~MDEV-30577~~ lets move the repeatable code to a shared file ctype-unidata.h:

{code:cpp}
/* utf8mb3 and ucs2 share this code */
static inline void
my_tolower_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  const MY_UNICASE_CHARACTER *page;
  DBUG_ASSERT(*wc <= uni_plane->maxchar);
  if ((page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].tolower;
}

static inline void
my_toupper_unicode_bmp(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  const MY_UNICASE_CHARACTER *page;
  DBUG_ASSERT(*wc <= uni_plane->maxchar);
  if ((page= uni_plane->page[*wc >> 8]))
    *wc= page[*wc & 0xFF].toupper;
}
{code}

{code:cpp}
/* utf8mb4, utf16, utf32 share this code */
static inline void
my_tolower_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    const MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].tolower;
  }
}

static inline void
my_toupper_unicode(MY_UNICASE_INFO *uni_plane, my_wc_t *wc)
{
  if (*wc <= uni_plane->maxchar)
  {
    const MY_UNICASE_CHARACTER *page;
    if ((page= uni_plane->page[(*wc >> 8)]))
      *wc= page[*wc & 0xFF].toupper;
  }
}
{code}

Alexander Barkov made changes - 2023-04-18 03:25

Status

Open [ 1 ]

In Progress [ 3 ]

Alexander Barkov made changes - 2023-04-18 03:25

issue.field.resolutiondate

2023-04-18 03:25:59.0

2023-04-18 03:25:59.38

Alexander Barkov made changes - 2023-04-18 03:26

Fix Version/s		10.10.4 [ 28522 ]
Fix Version/s		10.11.3 [ 28524 ]
Fix Version/s		11.1.1 [ 28704 ]
Fix Version/s		11.0.2 [ 28706 ]
Fix Version/s	10.10 [ 27530 ]
Resolution		Fixed [ 1 ]
Status	In Progress [ 3 ]	Closed [ 6 ]

Alexander Barkov made changes - 2023-04-18 05:36

Link

This issue relates to ~~MDEV-31069~~ [ ~~MDEV-31069~~ ]

People

Assignee:: Alexander Barkov

Reporter:: Alexander Barkov

Votes:: 0 Vote for this issue

Watchers:: 1 Start watching this issue

Dates

Created:: 2023-04-18 02:39

Updated:: 2023-04-18 05:36

Resolved:: 2023-04-18 03:25

Git Integration

Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.

MariaDB Server

Details

Description

Attachments

Issue Links

Activity

People

Dates

Git Integration