Optimize is_diacritic
is_diacritic is used for pretty much all of the text display sizing Timing here matters, so important it uses a priority buffer in order to cut down on searching the database. This patch is verified functionally equlivalent but saves 80 bytes and is ~20% faster for the clip+ I tested it on Several things were done: Optimizing the copy loop for the priority buffer Globbing some operations together but the main one has to do with the database structure its self -- Rather than a bit packed struct it combines the lower 8 bytes into one value [info] and uses flags to access the values instead Change-Id: I475073419b647639ef1eeaa3b9213e4e1067c0db
This commit is contained in:
parent
dd40c46d50
commit
b977b77fb7
1 changed files with 29 additions and 20 deletions
|
@ -10,6 +10,7 @@
|
|||
* Copyright (C) 2009 Phinitnun Chanasabaeng
|
||||
* Initial work
|
||||
* Copyright (C) 2009 Tomer Shalev
|
||||
* Copyright (C) 2018 William Wilgus
|
||||
*
|
||||
* Rockbox diacritic positioning
|
||||
* Based on initial work by Phinitnun Chanasabaeng
|
||||
|
@ -27,19 +28,23 @@
|
|||
#include "system.h"
|
||||
|
||||
#define DIAC_NUM_RANGES (ARRAYLEN(diac_ranges))
|
||||
#define DIAC_RTL (1 << 7)
|
||||
#define DIAC_CNT (0xFF ^ DIAC_RTL)
|
||||
|
||||
/* Each diac_range_ struct defines a Unicode range that begins with
|
||||
* N diacritic characters, and continues with non-diacritic characters up to the
|
||||
* base of the next item in the array */
|
||||
* base of the next item in the array, [info] packs RTL status and the count of
|
||||
* diacritic chars after [base]. RTL occupies the MSB and CNT the (7) lower bits
|
||||
*/
|
||||
|
||||
struct diac_range
|
||||
{
|
||||
unsigned base : 16;
|
||||
unsigned num_diacritics : 7;
|
||||
unsigned is_rtl : 1;
|
||||
uint16_t base;
|
||||
uint8_t info; /* [RTL:1 CNT:7] */
|
||||
};
|
||||
|
||||
#define DIAC_RANGE_ENTRY(first_diac, first_non_diac, is_rtl) \
|
||||
{ first_diac, first_non_diac - first_diac, is_rtl }
|
||||
{ first_diac, ((first_non_diac - first_diac) & DIAC_CNT) | (is_rtl * DIAC_RTL)}
|
||||
|
||||
/* Sorted by Unicode value */
|
||||
static const struct diac_range diac_ranges[] =
|
||||
|
@ -190,22 +195,28 @@ static const struct diac_range diac_ranges[] =
|
|||
|
||||
#define MRU_MAX_LEN 32
|
||||
|
||||
static unsigned short mru_len = 0;
|
||||
static unsigned short diacritic_mru[MRU_MAX_LEN];
|
||||
|
||||
bool is_diacritic(const unsigned short char_code, bool *is_rtl)
|
||||
{
|
||||
unsigned short mru, i;
|
||||
static uint8_t mru_len = 0;
|
||||
static uint8_t diacritic_mru[MRU_MAX_LEN];
|
||||
|
||||
uint8_t i, itmp;
|
||||
uint8_t info, mru;
|
||||
|
||||
const struct diac_range *diac;
|
||||
|
||||
/* Search in MRU */
|
||||
for (mru = 0; mru < mru_len; mru++)
|
||||
for (mru = 0, i = 0; mru < mru_len; mru++)
|
||||
{
|
||||
|
||||
/* Items shifted >> 1 */
|
||||
itmp = i;
|
||||
i = diacritic_mru[mru];
|
||||
diacritic_mru[mru] = itmp;
|
||||
|
||||
/* Found in MRU */
|
||||
if (diac_ranges[i].base <= char_code &&
|
||||
char_code < diac_ranges[i + 1].base)
|
||||
if ((char_code >= (diac = &diac_ranges[i])->base)
|
||||
&& (char_code < (++diac)->base))
|
||||
{
|
||||
goto Found;
|
||||
}
|
||||
|
@ -219,23 +230,21 @@ bool is_diacritic(const unsigned short char_code, bool *is_rtl)
|
|||
break;
|
||||
}
|
||||
|
||||
/* Add MRU entry, or overwrite LRU if MRU array is full */
|
||||
/* Add MRU entry */
|
||||
if (mru_len < MRU_MAX_LEN)
|
||||
mru_len++;
|
||||
else
|
||||
mru--;
|
||||
|
||||
Found:
|
||||
|
||||
/* Promote MRU item to top of MRU */
|
||||
for ( ; mru > 0; mru--)
|
||||
diacritic_mru[mru] = diacritic_mru[mru - 1];
|
||||
diacritic_mru[0] = i;
|
||||
|
||||
diac = &diac_ranges[i];
|
||||
info = diac->info;
|
||||
|
||||
/* Update RTL */
|
||||
if (is_rtl)
|
||||
*is_rtl = diac->is_rtl;
|
||||
*is_rtl = ((DIAC_RTL & info) == DIAC_RTL);
|
||||
|
||||
return (char_code < diac->base + diac->num_diacritics);
|
||||
}
|
||||
return (char_code < diac->base + (info & DIAC_CNT));
|
||||
}
|
Loading…
Reference in a new issue