diff options
author | ElOraiby <wael.eloraiby@gmail.com> | 2015-05-31 17:44:04 -0400 |
---|---|---|
committer | ElOraiby <wael.eloraiby@gmail.com> | 2015-05-31 17:44:04 -0400 |
commit | 2b81c17e2ca89df942479f51b8057ac713abe2d4 (patch) | |
tree | 3c1e482212fdb4da136650bbcd0bf3b83472fd52 | |
parent | d6621fd7ef1971a4d5bc3e2dba97c83d6eff43f5 (diff) |
optimized code (one pass no conditional switch)
-rw-r--r-- | arabic_test.txt | 1 | ||||
-rw-r--r-- | arabtype.c | 117 | ||||
-rw-r--r-- | mainwindow.cpp | 4 |
3 files changed, 66 insertions, 56 deletions
diff --git a/arabic_test.txt b/arabic_test.txt index 2d49a48..5c014fa 100644 --- a/arabic_test.txt +++ b/arabic_test.txt @@ -10,3 +10,4 @@ زاهداً في ما سيأتي ناسياً ما قد مضى أعطني الناي وغني وانسى داء ودواء إنما الناس سطورٌ كتبت لكن بماء + @@ -33,58 +33,59 @@ #define UNICODE_LAM 0x644 + // // 0: isolated form // 1: ending form // 2: beginning form (if 0, it's a cutting type) // 3: middle form // -typedef uint32_t char_form_t[4]; +typedef uint32_t char_form_t[4]; typedef char_form_t all_form_t[2]; static all_form_t arabic_forms_b[] = { - { {0xFE80, 0xFE80, 0, 0}, {0, 0, 0, 0} }, // hamza (0) - { {0xFE81, 0xFE82, 0, 0}, {0, 0, 0xFEF5, 0xFEF6} }, // 2alif madda (1) - { {0xFE83, 0xFE84, 0, 0}, {0, 0, 0xFEF7, 0xFEF8} }, // 2alif hamza (2) - { {0xFE85, 0xFE86, 0, 0}, {0, 0, 0, 0} }, // waw hamza (3) - { {0xFE87, 0xFE88, 0, 0}, {0, 0, 0xFEF9, 0xFEFA} }, // 2alif hamza maksoura (4) - { {0xFE89, 0xFE8A, 0xFE8B, 0xFE8C}, {0, 0, 0, 0} }, // 2alif maqsoura hamza (5) - { {0xFE8D, 0xFE8E, 0, 0}, {0, 0, 0xFEFB, 0xFEFC} }, // 2alif (6) - { {0xFE8F, 0xFE90, 0xFE91, 0xFE92}, {0, 0, 0, 0} }, // ba2 (7) - { {0xFE93, 0xFE94, 0, 0}, {0, 0, 0, 0} }, // ta2 marbouta (8) - { {0xFE95, 0xFE96, 0xFE97, 0xFE98}, {0, 0, 0, 0} }, // ta2 (9) - { {0xFE99, 0xFE9A, 0xFE9B, 0xFE9C}, {0, 0, 0, 0} }, // tha2 (10) - { {0xFE9D, 0xFE9E, 0xFE9F, 0xFEA0}, {0, 0, 0, 0} }, // jim (11) - { {0xFEA1, 0xFEA2, 0xFEA3, 0xFEA4}, {0, 0, 0, 0} }, // 7a2 (12) - { {0xFEA5, 0xFEA6, 0xFEA7, 0xFEA8}, {0, 0, 0, 0} }, // kha2 (13) - { {0xFEA9, 0xFEAA, 0, 0}, {0, 0, 0, 0} }, // dal (14) - { {0xFEAB, 0xFEAC, 0, 0}, {0, 0, 0, 0} }, // dhal (15) - { {0xFEAD, 0xFEAE, 0, 0}, {0, 0, 0, 0} }, // ra2 (16) - { {0xFEAF, 0xFEB0, 0, 0}, {0, 0, 0, 0} }, // zayn (17) - { {0xFEB1, 0xFEB2, 0xFEB3, 0xFEB4}, {0, 0, 0, 0} }, // syn (18) - { {0xFEB5, 0xFEB6, 0xFEB7, 0xFEB8}, {0, 0, 0, 0} }, // shin (19) - { {0xFEB9, 0xFEBA, 0xFEBB, 0xFEBC}, {0, 0, 0, 0} }, // sad (20) - { {0xFEBD, 0xFEBE, 0xFEBF, 0xFEC0}, {0, 0, 0, 0} }, // dad (21) - { {0xFEC1, 0xFEC2, 0xFEC3, 0xFEC4}, {0, 0, 0, 0} }, // tah (22) - { {0xFEC5, 0xFEC6, 0xFEC7, 0xFEC8}, {0, 0, 0, 0} }, // thah (23) - { {0xFEC9, 0xFECA, 0xFECB, 0xFECC}, {0, 0, 0, 0} }, // 3ayn (24) - { {0xFECD, 0xFECE, 0xFECF, 0xFED0}, {0, 0, 0, 0} }, // ghayn (25) - { { 0, 0, 0, 0}, {0, 0, 0, 0} }, // (26) - { { 0, 0, 0, 0}, {0, 0, 0, 0} }, // (27) - { { 0, 0, 0, 0}, {0, 0, 0, 0} }, // (28) - { { 0, 0, 0, 0}, {0, 0, 0, 0} }, // (29) - { { 0, 0, 0, 0}, {0, 0, 0, 0} }, // (30) - { { 0x640, 0x640, 0x640, 0x640}, {0, 0, 0, 0} }, // wasla (31) - { {0xFED1, 0xFED2, 0xFED3, 0xFED4}, {0, 0, 0, 0} }, // fa2 (32) - { {0xFED5, 0xFED6, 0xFED7, 0xFED8}, {0, 0, 0, 0} }, // qaf (33) - { {0xFED9, 0xFEDA, 0xFEDB, 0xFEDC}, {0, 0, 0, 0} }, // kaf (34) - { {0xFEDD, 0xFEDE, 0xFEDF, 0xFEE0}, {0, 0, 0, 0} }, // lam (35) - { {0xFEE1, 0xFEE2, 0xFEE3, 0xFEE4}, {0, 0, 0, 0} }, // mim (36) - { {0xFEE5, 0xFEE6, 0xFEE7, 0xFEE8}, {0, 0, 0, 0} }, // noon (37) - { {0xFEE9, 0xFEEA, 0xFEEB, 0xFEEC}, {0, 0, 0, 0} }, // ha2 (38) - { {0xFEED, 0xFEEE, 0, 0}, {0, 0, 0, 0} }, // waw (39) - { {0xFEFF, 0xFEF0, 0, 0}, {0, 0, 0, 0} }, // 2alif maksoura (40) - { {0xFEF1, 0xFEF2, 0xFEF3, 0xFEF4}, {0, 0, 0, 0} }, // ya2 (41) + { {0xFE80, 0xFE80, 0, 0}, {-1, -1, 0, 0} }, // hamza (0) + { {0xFE81, 0xFE82, 0, 0}, {-1, -1, 0xFEF5, 0xFEF6} }, // 2alif madda (1) + { {0xFE83, 0xFE84, 0, 0}, {-1, -1, 0xFEF7, 0xFEF8} }, // 2alif hamza (2) + { {0xFE85, 0xFE86, 0, 0}, {-1, -1, 0, 0} }, // waw hamza (3) + { {0xFE87, 0xFE88, 0, 0}, {-1, -1, 0xFEF9, 0xFEFA} }, // 2alif hamza maksoura (4) + { {0xFE89, 0xFE8A, 0xFE8B, 0xFE8C}, {-1, -1, 0, 0} }, // 2alif maqsoura hamza (5) + { {0xFE8D, 0xFE8E, 0, 0}, {-1, -1, 0xFEFB, 0xFEFC} }, // 2alif (6) + { {0xFE8F, 0xFE90, 0xFE91, 0xFE92}, {-1, -1, 0, 0} }, // ba2 (7) + { {0xFE93, 0xFE94, 0, 0}, {-1, -1, 0, 0} }, // ta2 marbouta (8) + { {0xFE95, 0xFE96, 0xFE97, 0xFE98}, {-1, -1, 0, 0} }, // ta2 (9) + { {0xFE99, 0xFE9A, 0xFE9B, 0xFE9C}, {-1, -1, 0, 0} }, // tha2 (10) + { {0xFE9D, 0xFE9E, 0xFE9F, 0xFEA0}, {-1, -1, 0, 0} }, // jim (11) + { {0xFEA1, 0xFEA2, 0xFEA3, 0xFEA4}, {-1, -1, 0, 0} }, // 7a2 (12) + { {0xFEA5, 0xFEA6, 0xFEA7, 0xFEA8}, {-1, -1, 0, 0} }, // kha2 (13) + { {0xFEA9, 0xFEAA, 0, 0}, {-1, -1, 0, 0} }, // dal (14) + { {0xFEAB, 0xFEAC, 0, 0}, {-1, -1, 0, 0} }, // dhal (15) + { {0xFEAD, 0xFEAE, 0, 0}, {-1, -1, 0, 0} }, // ra2 (16) + { {0xFEAF, 0xFEB0, 0, 0}, {-1, -1, 0, 0} }, // zayn (17) + { {0xFEB1, 0xFEB2, 0xFEB3, 0xFEB4}, {-1, -1, 0, 0} }, // syn (18) + { {0xFEB5, 0xFEB6, 0xFEB7, 0xFEB8}, {-1, -1, 0, 0} }, // shin (19) + { {0xFEB9, 0xFEBA, 0xFEBB, 0xFEBC}, {-1, -1, 0, 0} }, // sad (20) + { {0xFEBD, 0xFEBE, 0xFEBF, 0xFEC0}, {-1, -1, 0, 0} }, // dad (21) + { {0xFEC1, 0xFEC2, 0xFEC3, 0xFEC4}, {-1, -1, 0, 0} }, // tah (22) + { {0xFEC5, 0xFEC6, 0xFEC7, 0xFEC8}, {-1, -1, 0, 0} }, // thah (23) + { {0xFEC9, 0xFECA, 0xFECB, 0xFECC}, {-1, -1, 0, 0} }, // 3ayn (24) + { {0xFECD, 0xFECE, 0xFECF, 0xFED0}, {-1, -1, 0, 0} }, // ghayn (25) + { { 0, 0, 0, 0}, {-1, -1, 0, 0} }, // (26) + { { 0, 0, 0, 0}, {-1, -1, 0, 0} }, // (27) + { { 0, 0, 0, 0}, {-1, -1, 0, 0} }, // (28) + { { 0, 0, 0, 0}, {-1, -1, 0, 0} }, // (29) + { { 0, 0, 0, 0}, {-1, -1, 0, 0} }, // (30) + { { 0x640, 0x640, 0x640, 0x640}, {-1, -1, 0, 0} }, // wasla (31) + { {0xFED1, 0xFED2, 0xFED3, 0xFED4}, {-1, -1, 0, 0} }, // fa2 (32) + { {0xFED5, 0xFED6, 0xFED7, 0xFED8}, {-1, -1, 0, 0} }, // qaf (33) + { {0xFED9, 0xFEDA, 0xFEDB, 0xFEDC}, {-1, -1, 0, 0} }, // kaf (34) + { {0xFEDD, 0xFEDE, 0xFEDF, 0xFEE0}, {-1, -1, 0, 0} }, // lam (35) + { {0xFEE1, 0xFEE2, 0xFEE3, 0xFEE4}, {-1, -1, 0, 0} }, // mim (36) + { {0xFEE5, 0xFEE6, 0xFEE7, 0xFEE8}, {-1, -1, 0, 0} }, // noon (37) + { {0xFEE9, 0xFEEA, 0xFEEB, 0xFEEC}, {-1, -1, 0, 0} }, // ha2 (38) + { {0xFEED, 0xFEEE, 0, 0}, {-1, -1, 0, 0} }, // waw (39) + { {0xFEFF, 0xFEF0, 0, 0}, {-1, -1, 0, 0} }, // 2alif maksoura (40) + { {0xFEF1, 0xFEF2, 0xFEF3, 0xFEF4}, {-1, -1, 0, 0} }, // ya2 (41) }; @@ -96,31 +97,39 @@ static inline bool is_lam_alef(uint32_t cp, uint32_t next) { return cp == UNICOD static inline bool is_alef_prev_lam(uint32_t prev, uint32_t cp) { return prev == UNICODE_LAM && is_arabic_letter(cp) && arabic_forms_b[cp - ARABIC_LETTER_START][1][INITIAL] != 0; } -static inline bool is_linking_type(uint32_t cp) { - if( is_arabic_letter(cp) && ( arabic_forms_b[cp - ARABIC_LETTER_START][0][INITIAL] || arabic_forms_b[cp - ARABIC_LETTER_START][0][MEDIAL] ) ) { - return true; - } else { - return false; - } -} + +static inline bool is_linking_type(uint32_t cp) { return is_arabic_letter(cp) && arabic_forms_b[cp - ARABIC_LETTER_START][0][MEDIAL] != 0; } uint32_t get_presentation_form_b(uint32_t prev, uint32_t next, uint32_t cp) { + if( !is_arabic_letter(cp) ) { return cp; /* not an Arabic letter */ } - if( is_lam_alef(cp, next) ) { + bool is_la = is_lam_alef(cp, next); + bool is_apl = is_alef_prev_lam(prev, cp); + + bool is_lapl = is_la | is_apl; + +#ifdef CLEAR_CODE + if( is_lapl ) { uint32_t index = (is_linking_type(cp) << 1) | is_linking_type(prev); - return arabic_forms_b[next - ARABIC_LETTER_START][1][index]; + return arabic_forms_b[next - ARABIC_LETTER_START][is_lapl][index]; } else { - if( is_alef_prev_lam(prev, cp) ) { + if( is_apl ) { return -1; // skip previously processed lam alef } else { uint32_t index = ((is_arabic_letter(next) & is_linking_type(cp)) << 1) | is_linking_type(prev); - return arabic_forms_b[cp - ARABIC_LETTER_START][0][index]; + return arabic_forms_b[cp - ARABIC_LETTER_START][is_lapl][index]; } } +#else + // optimized code + uint32_t index = (((is_lapl | is_arabic_letter(next)) & is_linking_type(cp)) << 1) | is_linking_type(prev); + uint32_t ref = next * is_la + cp * (1 - is_la) - ARABIC_LETTER_START; + return arabic_forms_b[ref][is_lapl][index]; +#endif } diff --git a/mainwindow.cpp b/mainwindow.cpp index 5173fda..14a2871 100644 --- a/mainwindow.cpp +++ b/mainwindow.cpp @@ -58,7 +58,7 @@ MainWindow::MainWindow(QWidget *parent) : ui__->setupUi(this); int width = 480; - int height = 320; + int height = 392; int font_size = 24; std::ifstream ifs("arabic_test.txt"); @@ -117,7 +117,7 @@ MainWindow::MainWindow(QWidget *parent) : { uint ch = get_arabic_form(arabic_cp, idx); if( ch == 0xA ) { - line += font_size; + line += font_size + 5; col = width - font_size; continue; } |