diff --git a/src/Makefile.am b/src/Makefile.am index 9311ab7c9..f53505ed4 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -67,6 +67,7 @@ HBSOURCES += \ hb-ot-shape-complex-arabic-table.hh \ hb-ot-shape-complex-indic.cc \ hb-ot-shape-complex-indic-machine.hh \ + hb-ot-shape-complex-indic-private.hh \ hb-ot-shape-complex-indic-table.hh \ hb-ot-shape-complex-misc.cc \ hb-ot-shape-complex-private.hh \ @@ -169,13 +170,17 @@ hb-ot-shape-complex-indic-machine.hh: hb-ot-shape-complex-indic-machine.rl $(AM_V_GEN)$(top_srcdir)/missing --run ragel -e -F1 -o "$@.tmp" "$<" && \ mv "$@.tmp" "$@" || ( $(RM) "$@.tmp" && false ) -noinst_PROGRAMS = main +noinst_PROGRAMS = main indic bin_PROGRAMS = main_SOURCES = main.cc main_CPPFLAGS = $(HBCFLAGS) main_LDADD = libharfbuzz.la $(HBLIBS) +indic_SOURCES = indic.cc +indic_CPPFLAGS = $(HBCFLAGS) +indic_LDADD = libharfbuzz.la $(HBLIBS) + dist_check_SCRIPTS = \ check-c-linkage-decls.sh \ check-header-guards.sh \ diff --git a/src/hb-ot-shape-complex-indic-private.hh b/src/hb-ot-shape-complex-indic-private.hh new file mode 100644 index 000000000..b7bce172c --- /dev/null +++ b/src/hb-ot-shape-complex-indic-private.hh @@ -0,0 +1,264 @@ +/* + * Copyright © 2012 Google, Inc. + * + * This is part of HarfBuzz, a text shaping library. + * + * Permission is hereby granted, without written agreement and without + * license or royalty fees, to use, copy, modify, and distribute this + * software and its documentation for any purpose, provided that the + * above copyright notice and the following two paragraphs appear in + * all copies of this software. + * + * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR + * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES + * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN + * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH + * DAMAGE. + * + * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, + * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS + * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO + * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. + * + * Google Author(s): Behdad Esfahbod + */ + +#ifndef HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH +#define HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH + +#include "hb-private.hh" + + +#include "hb-ot-shape-complex-private.hh" + + +/* buffer var allocations */ +#define indic_category() complex_var_persistent_u8_0() /* indic_category_t */ +#define indic_position() complex_var_persistent_u8_1() /* indic_matra_category_t */ + +#define INDIC_TABLE_ELEMENT_TYPE uint8_t + +/* Cateories used in the OpenType spec: + * https://www.microsoft.com/typography/otfntdev/devanot/shaping.aspx + */ +/* Note: This enum is duplicated in the -machine.rl source file. + * Not sure how to avoid duplication. */ +enum indic_category_t { + OT_X = 0, + OT_C, + OT_Ra, /* Not explicitly listed in the OT spec, but used in the grammar. */ + OT_V, + OT_N, + OT_H, + OT_ZWNJ, + OT_ZWJ, + OT_M, + OT_SM, + OT_VD, + OT_A, + OT_NBSP +}; + +/* Visual positions in a syllable from left to right. */ +enum indic_position_t { + POS_PRE = 1, + POS_BASE = 2, + POS_ABOVE = 3, + POS_BELOW = 4, + POS_POST = 5 +}; + +/* Categories used in IndicSyllabicCategory.txt from UCD */ +/* The assignments are guesswork */ +enum indic_syllabic_category_t { + INDIC_SYLLABIC_CATEGORY_OTHER = OT_X, + + INDIC_SYLLABIC_CATEGORY_AVAGRAHA = OT_X, + INDIC_SYLLABIC_CATEGORY_BINDU = OT_SM, + INDIC_SYLLABIC_CATEGORY_CONSONANT = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_DEAD = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_FINAL = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_HEAD_LETTER = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_MEDIAL = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_PLACEHOLDER = OT_NBSP, + INDIC_SYLLABIC_CATEGORY_CONSONANT_SUBJOINED = OT_C, + INDIC_SYLLABIC_CATEGORY_CONSONANT_REPHA = OT_C, + INDIC_SYLLABIC_CATEGORY_MODIFYING_LETTER = OT_X, + INDIC_SYLLABIC_CATEGORY_NUKTA = OT_N, + INDIC_SYLLABIC_CATEGORY_REGISTER_SHIFTER = OT_X, + INDIC_SYLLABIC_CATEGORY_TONE_LETTER = OT_X, + INDIC_SYLLABIC_CATEGORY_TONE_MARK = OT_X, + INDIC_SYLLABIC_CATEGORY_VIRAMA = OT_H, + INDIC_SYLLABIC_CATEGORY_VISARGA = OT_SM, + INDIC_SYLLABIC_CATEGORY_VOWEL = OT_V, + INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT = OT_M, + INDIC_SYLLABIC_CATEGORY_VOWEL_INDEPENDENT = OT_V +}; + +/* Categories used in IndicSMatraCategory.txt from UCD */ +enum indic_matra_category_t { + INDIC_MATRA_CATEGORY_NOT_APPLICABLE = POS_BASE, + + INDIC_MATRA_CATEGORY_LEFT = POS_PRE - 1, /* Move *before* existing "pre" chars */ + INDIC_MATRA_CATEGORY_TOP = POS_ABOVE, + INDIC_MATRA_CATEGORY_BOTTOM = POS_BELOW, + INDIC_MATRA_CATEGORY_RIGHT = POS_POST, + + /* We don't really care much about these since we decompose them + * in the generic pre-shaping layer. They will only be used if + * the font does not cover the decomposition. In which case, we + * define these as aliases to the place we want the split-matra + * glyph to show up. Quite arbitrary. + * + * TODO: There are some split matras without Unicode decompositions. + * We have to figure out what to do with them. + */ + INDIC_MATRA_CATEGORY_BOTTOM_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_BOTTOM, + INDIC_MATRA_CATEGORY_LEFT_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_LEFT, + INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM = 8 | INDIC_MATRA_CATEGORY_BOTTOM, + INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_BOTTOM, + INDIC_MATRA_CATEGORY_TOP_AND_LEFT = 8 | INDIC_MATRA_CATEGORY_LEFT, + INDIC_MATRA_CATEGORY_TOP_AND_LEFT_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_LEFT, + INDIC_MATRA_CATEGORY_TOP_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_RIGHT, + + INDIC_MATRA_CATEGORY_INVISIBLE = INDIC_MATRA_CATEGORY_NOT_APPLICABLE, + INDIC_MATRA_CATEGORY_OVERSTRUCK = INDIC_MATRA_CATEGORY_NOT_APPLICABLE, + INDIC_MATRA_CATEGORY_VISUAL_ORDER_LEFT = INDIC_MATRA_CATEGORY_NOT_APPLICABLE +}; + +/* Note: We use ASSERT_STATIC_EXPR_ZERO() instead of ASSERT_STATIC_EXPR() and the comma operation + * because gcc fails to optimize the latter and fills the table in at runtime. */ +#define INDIC_COMBINE_CATEGORIES(S,M) \ + (ASSERT_STATIC_EXPR_ZERO (M == INDIC_MATRA_CATEGORY_NOT_APPLICABLE || (S == INDIC_SYLLABIC_CATEGORY_VIRAMA || S == INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT)) + \ + ASSERT_STATIC_EXPR_ZERO (S < 16 && M < 16) + \ + ((M << 4) | S)) + + +#include "hb-ot-shape-complex-indic-table.hh" + +/* XXX + * This is a hack for now. We should: + * 1. Move this data into the main Indic table, + * and/or + * 2. Probe font lookups to determine consonant positions. + */ +static const struct consonant_position_t { + hb_codepoint_t u; + indic_position_t position; +} consonant_positions[] = { + {0x0930, POS_BELOW}, + {0x09AC, POS_BELOW}, + {0x09AF, POS_POST}, + {0x09B0, POS_BELOW}, + {0x09F0, POS_BELOW}, + {0x0A2F, POS_POST}, + {0x0A30, POS_BELOW}, + {0x0A35, POS_BELOW}, + {0x0A39, POS_BELOW}, + {0x0AB0, POS_BELOW}, + {0x0B24, POS_BELOW}, + {0x0B28, POS_BELOW}, + {0x0B2C, POS_BELOW}, + {0x0B2D, POS_BELOW}, + {0x0B2E, POS_BELOW}, + {0x0B2F, POS_POST}, + {0x0B30, POS_BELOW}, + {0x0B32, POS_BELOW}, + {0x0B33, POS_BELOW}, + {0x0B5F, POS_POST}, + {0x0B71, POS_BELOW}, + {0x0C15, POS_BELOW}, + {0x0C16, POS_BELOW}, + {0x0C17, POS_BELOW}, + {0x0C18, POS_BELOW}, + {0x0C19, POS_BELOW}, + {0x0C1A, POS_BELOW}, + {0x0C1B, POS_BELOW}, + {0x0C1C, POS_BELOW}, + {0x0C1D, POS_BELOW}, + {0x0C1E, POS_BELOW}, + {0x0C1F, POS_BELOW}, + {0x0C20, POS_BELOW}, + {0x0C21, POS_BELOW}, + {0x0C22, POS_BELOW}, + {0x0C23, POS_BELOW}, + {0x0C24, POS_BELOW}, + {0x0C25, POS_BELOW}, + {0x0C26, POS_BELOW}, + {0x0C27, POS_BELOW}, + {0x0C28, POS_BELOW}, + {0x0C2A, POS_BELOW}, + {0x0C2B, POS_BELOW}, + {0x0C2C, POS_BELOW}, + {0x0C2D, POS_BELOW}, + {0x0C2E, POS_BELOW}, + {0x0C2F, POS_BELOW}, + {0x0C30, POS_BELOW}, + {0x0C32, POS_BELOW}, + {0x0C33, POS_BELOW}, + {0x0C35, POS_BELOW}, + {0x0C36, POS_BELOW}, + {0x0C37, POS_BELOW}, + {0x0C38, POS_BELOW}, + {0x0C39, POS_BELOW}, + {0x0C95, POS_BELOW}, + {0x0C96, POS_BELOW}, + {0x0C97, POS_BELOW}, + {0x0C98, POS_BELOW}, + {0x0C99, POS_BELOW}, + {0x0C9A, POS_BELOW}, + {0x0C9B, POS_BELOW}, + {0x0C9C, POS_BELOW}, + {0x0C9D, POS_BELOW}, + {0x0C9E, POS_BELOW}, + {0x0C9F, POS_BELOW}, + {0x0CA0, POS_BELOW}, + {0x0CA1, POS_BELOW}, + {0x0CA2, POS_BELOW}, + {0x0CA3, POS_BELOW}, + {0x0CA4, POS_BELOW}, + {0x0CA5, POS_BELOW}, + {0x0CA6, POS_BELOW}, + {0x0CA7, POS_BELOW}, + {0x0CA8, POS_BELOW}, + {0x0CAA, POS_BELOW}, + {0x0CAB, POS_BELOW}, + {0x0CAC, POS_BELOW}, + {0x0CAD, POS_BELOW}, + {0x0CAE, POS_BELOW}, + {0x0CAF, POS_BELOW}, + {0x0CB0, POS_BELOW}, + {0x0CB2, POS_BELOW}, + {0x0CB3, POS_BELOW}, + {0x0CB5, POS_BELOW}, + {0x0CB6, POS_BELOW}, + {0x0CB7, POS_BELOW}, + {0x0CB8, POS_BELOW}, + {0x0CB9, POS_BELOW}, + {0x0CDE, POS_BELOW}, + {0x0D2F, POS_POST}, + {0x0D30, POS_POST}, + {0x0D32, POS_BELOW}, + {0x0D35, POS_POST}, +}; + +/* XXX + * This is a hack for now. We should move this data into the main Indic table. + */ +static const hb_codepoint_t ra_chars[] = { + 0x0930, /* Devanagari */ + 0x09B0, /* Bengali */ + 0x09F0, /* Bengali */ +//0x09F1, /* Bengali */ +//0x0A30, /* Gurmukhi */ + 0x0AB0, /* Gujarati */ + 0x0B30, /* Oriya */ +//0x0BB0, /* Tamil */ +//0x0C30, /* Telugu */ + 0x0CB0, /* Kannada */ +//0x0D30, /* Malayalam */ +}; + + +#endif /* HB_OT_SHAPE_COMPLEX_INDIC_PRIVATE_HH */ diff --git a/src/hb-ot-shape-complex-indic.cc b/src/hb-ot-shape-complex-indic.cc index 06b8d4869..09222f715 100644 --- a/src/hb-ot-shape-complex-indic.cc +++ b/src/hb-ot-shape-complex-indic.cc @@ -24,235 +24,8 @@ * Google Author(s): Behdad Esfahbod */ -#include "hb-ot-shape-complex-private.hh" - - -/* buffer var allocations */ -#define indic_category() complex_var_persistent_u8_0() /* indic_category_t */ -#define indic_position() complex_var_persistent_u8_1() /* indic_matra_category_t */ - -#define INDIC_TABLE_ELEMENT_TYPE uint8_t - -/* Cateories used in the OpenType spec: - * https://www.microsoft.com/typography/otfntdev/devanot/shaping.aspx - */ -/* Note: This enum is duplicated in the -machine.rl source file. - * Not sure how to avoid duplication. */ -enum indic_category_t { - OT_X = 0, - OT_C, - OT_Ra, /* Not explicitly listed in the OT spec, but used in the grammar. */ - OT_V, - OT_N, - OT_H, - OT_ZWNJ, - OT_ZWJ, - OT_M, - OT_SM, - OT_VD, - OT_A, - OT_NBSP -}; - -/* Visual positions in a syllable from left to right. */ -enum indic_position_t { - POS_PRE = 1, - POS_BASE = 2, - POS_ABOVE = 3, - POS_BELOW = 4, - POS_POST = 5 -}; - -/* Categories used in IndicSyllabicCategory.txt from UCD */ -/* The assignments are guesswork */ -enum indic_syllabic_category_t { - INDIC_SYLLABIC_CATEGORY_OTHER = OT_X, - - INDIC_SYLLABIC_CATEGORY_AVAGRAHA = OT_X, - INDIC_SYLLABIC_CATEGORY_BINDU = OT_SM, - INDIC_SYLLABIC_CATEGORY_CONSONANT = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_DEAD = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_FINAL = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_HEAD_LETTER = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_MEDIAL = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_PLACEHOLDER = OT_NBSP, - INDIC_SYLLABIC_CATEGORY_CONSONANT_SUBJOINED = OT_C, - INDIC_SYLLABIC_CATEGORY_CONSONANT_REPHA = OT_C, - INDIC_SYLLABIC_CATEGORY_MODIFYING_LETTER = OT_X, - INDIC_SYLLABIC_CATEGORY_NUKTA = OT_N, - INDIC_SYLLABIC_CATEGORY_REGISTER_SHIFTER = OT_X, - INDIC_SYLLABIC_CATEGORY_TONE_LETTER = OT_X, - INDIC_SYLLABIC_CATEGORY_TONE_MARK = OT_X, - INDIC_SYLLABIC_CATEGORY_VIRAMA = OT_H, - INDIC_SYLLABIC_CATEGORY_VISARGA = OT_SM, - INDIC_SYLLABIC_CATEGORY_VOWEL = OT_V, - INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT = OT_M, - INDIC_SYLLABIC_CATEGORY_VOWEL_INDEPENDENT = OT_V -}; - -/* Categories used in IndicSMatraCategory.txt from UCD */ -enum indic_matra_category_t { - INDIC_MATRA_CATEGORY_NOT_APPLICABLE = POS_BASE, - - INDIC_MATRA_CATEGORY_LEFT = POS_PRE - 1, /* Move *before* existing "pre" chars */ - INDIC_MATRA_CATEGORY_TOP = POS_ABOVE, - INDIC_MATRA_CATEGORY_BOTTOM = POS_BELOW, - INDIC_MATRA_CATEGORY_RIGHT = POS_POST, - - /* We don't really care much about these since we decompose them - * in the generic pre-shaping layer. They will only be used if - * the font does not cover the decomposition. In which case, we - * define these as aliases to the place we want the split-matra - * glyph to show up. Quite arbitrary. - * - * TODO: There are some split matras without Unicode decompositions. - * We have to figure out what to do with them. - */ - INDIC_MATRA_CATEGORY_BOTTOM_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_BOTTOM, - INDIC_MATRA_CATEGORY_LEFT_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_LEFT, - INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM = 8 | INDIC_MATRA_CATEGORY_BOTTOM, - INDIC_MATRA_CATEGORY_TOP_AND_BOTTOM_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_BOTTOM, - INDIC_MATRA_CATEGORY_TOP_AND_LEFT = 8 | INDIC_MATRA_CATEGORY_LEFT, - INDIC_MATRA_CATEGORY_TOP_AND_LEFT_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_LEFT, - INDIC_MATRA_CATEGORY_TOP_AND_RIGHT = 8 | INDIC_MATRA_CATEGORY_RIGHT, - - INDIC_MATRA_CATEGORY_INVISIBLE = INDIC_MATRA_CATEGORY_NOT_APPLICABLE, - INDIC_MATRA_CATEGORY_OVERSTRUCK = INDIC_MATRA_CATEGORY_NOT_APPLICABLE, - INDIC_MATRA_CATEGORY_VISUAL_ORDER_LEFT = INDIC_MATRA_CATEGORY_NOT_APPLICABLE -}; - -/* Note: We use ASSERT_STATIC_EXPR_ZERO() instead of ASSERT_STATIC_EXPR() and the comma operation - * because gcc fails to optimize the latter and fills the table in at runtime. */ -#define INDIC_COMBINE_CATEGORIES(S,M) \ - (ASSERT_STATIC_EXPR_ZERO (M == INDIC_MATRA_CATEGORY_NOT_APPLICABLE || (S == INDIC_SYLLABIC_CATEGORY_VIRAMA || S == INDIC_SYLLABIC_CATEGORY_VOWEL_DEPENDENT)) + \ - ASSERT_STATIC_EXPR_ZERO (S < 16 && M < 16) + \ - ((M << 4) | S)) - -#include "hb-ot-shape-complex-indic-table.hh" - -/* XXX - * This is a hack for now. We should: - * 1. Move this data into the main Indic table, - * and/or - * 2. Probe font lookups to determine consonant positions. - */ -static const struct consonant_position_t { - hb_codepoint_t u; - indic_position_t position; -} consonant_positions[] = { - {0x0930, POS_BELOW}, - {0x09AC, POS_BELOW}, - {0x09AF, POS_POST}, - {0x09B0, POS_BELOW}, - {0x09F0, POS_BELOW}, - {0x0A2F, POS_POST}, - {0x0A30, POS_BELOW}, - {0x0A35, POS_BELOW}, - {0x0A39, POS_BELOW}, - {0x0AB0, POS_BELOW}, - {0x0B24, POS_BELOW}, - {0x0B28, POS_BELOW}, - {0x0B2C, POS_BELOW}, - {0x0B2D, POS_BELOW}, - {0x0B2E, POS_BELOW}, - {0x0B2F, POS_POST}, - {0x0B30, POS_BELOW}, - {0x0B32, POS_BELOW}, - {0x0B33, POS_BELOW}, - {0x0B5F, POS_POST}, - {0x0B71, POS_BELOW}, - {0x0C15, POS_BELOW}, - {0x0C16, POS_BELOW}, - {0x0C17, POS_BELOW}, - {0x0C18, POS_BELOW}, - {0x0C19, POS_BELOW}, - {0x0C1A, POS_BELOW}, - {0x0C1B, POS_BELOW}, - {0x0C1C, POS_BELOW}, - {0x0C1D, POS_BELOW}, - {0x0C1E, POS_BELOW}, - {0x0C1F, POS_BELOW}, - {0x0C20, POS_BELOW}, - {0x0C21, POS_BELOW}, - {0x0C22, POS_BELOW}, - {0x0C23, POS_BELOW}, - {0x0C24, POS_BELOW}, - {0x0C25, POS_BELOW}, - {0x0C26, POS_BELOW}, - {0x0C27, POS_BELOW}, - {0x0C28, POS_BELOW}, - {0x0C2A, POS_BELOW}, - {0x0C2B, POS_BELOW}, - {0x0C2C, POS_BELOW}, - {0x0C2D, POS_BELOW}, - {0x0C2E, POS_BELOW}, - {0x0C2F, POS_BELOW}, - {0x0C30, POS_BELOW}, - {0x0C32, POS_BELOW}, - {0x0C33, POS_BELOW}, - {0x0C35, POS_BELOW}, - {0x0C36, POS_BELOW}, - {0x0C37, POS_BELOW}, - {0x0C38, POS_BELOW}, - {0x0C39, POS_BELOW}, - {0x0C95, POS_BELOW}, - {0x0C96, POS_BELOW}, - {0x0C97, POS_BELOW}, - {0x0C98, POS_BELOW}, - {0x0C99, POS_BELOW}, - {0x0C9A, POS_BELOW}, - {0x0C9B, POS_BELOW}, - {0x0C9C, POS_BELOW}, - {0x0C9D, POS_BELOW}, - {0x0C9E, POS_BELOW}, - {0x0C9F, POS_BELOW}, - {0x0CA0, POS_BELOW}, - {0x0CA1, POS_BELOW}, - {0x0CA2, POS_BELOW}, - {0x0CA3, POS_BELOW}, - {0x0CA4, POS_BELOW}, - {0x0CA5, POS_BELOW}, - {0x0CA6, POS_BELOW}, - {0x0CA7, POS_BELOW}, - {0x0CA8, POS_BELOW}, - {0x0CAA, POS_BELOW}, - {0x0CAB, POS_BELOW}, - {0x0CAC, POS_BELOW}, - {0x0CAD, POS_BELOW}, - {0x0CAE, POS_BELOW}, - {0x0CAF, POS_BELOW}, - {0x0CB0, POS_BELOW}, - {0x0CB2, POS_BELOW}, - {0x0CB3, POS_BELOW}, - {0x0CB5, POS_BELOW}, - {0x0CB6, POS_BELOW}, - {0x0CB7, POS_BELOW}, - {0x0CB8, POS_BELOW}, - {0x0CB9, POS_BELOW}, - {0x0CDE, POS_BELOW}, - {0x0D2F, POS_POST}, - {0x0D30, POS_POST}, - {0x0D32, POS_BELOW}, - {0x0D35, POS_POST}, -}; - -/* XXX - * This is a hack for now. We should move this data into the main Indic table. - */ -static const hb_codepoint_t ra_chars[] = { - 0x0930, /* Devanagari */ - 0x09B0, /* Bengali */ - 0x09F0, /* Bengali */ -//0x09F1, /* Bengali */ -//0x0A30, /* Gurmukhi */ - 0x0AB0, /* Gujarati */ - 0x0B30, /* Oriya */ -//0x0BB0, /* Tamil */ -//0x0C30, /* Telugu */ - 0x0CB0, /* Kannada */ -//0x0D30, /* Malayalam */ -}; +#include "hb-ot-shape-complex-indic-private.hh" static int compare_codepoint (const void *pa, const void *pb) diff --git a/src/indic.cc b/src/indic.cc new file mode 100644 index 000000000..6c6e05a9b --- /dev/null +++ b/src/indic.cc @@ -0,0 +1,46 @@ +/* + * Copyright © 2012 Google, Inc. + * + * This is part of HarfBuzz, a text shaping library. + * + * Permission is hereby granted, without written agreement and without + * license or royalty fees, to use, copy, modify, and distribute this + * software and its documentation for any purpose, provided that the + * above copyright notice and the following two paragraphs appear in + * all copies of this software. + * + * IN NO EVENT SHALL THE COPYRIGHT HOLDER BE LIABLE TO ANY PARTY FOR + * DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES + * ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN + * IF THE COPYRIGHT HOLDER HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH + * DAMAGE. + * + * THE COPYRIGHT HOLDER SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, + * BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND + * FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED HEREUNDER IS + * ON AN "AS IS" BASIS, AND THE COPYRIGHT HOLDER HAS NO OBLIGATION TO + * PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS. + * + * Google Author(s): Behdad Esfahbod + */ + +#include "hb-ot-shape-complex-indic-private.hh" + +int +main (void) +{ + hb_unicode_funcs_t *funcs = hb_unicode_funcs_get_default (); + + printf ("There are split matras without a Unicode decomposition:\n"); + for (hb_codepoint_t u = 0; u < 0x110000; u++) + { + unsigned int type = get_indic_categories (u); + + unsigned int category = type & 0x0F; + unsigned int position = type >> 4; + + hb_codepoint_t a, b; + if ((position & 8) && !hb_unicode_decompose (funcs, u, &a, &b)) + printf ("U+%04X\n", u); + } +}