diff options
| -rw-r--r-- | docs/pango-sections.txt | 2 | ||||
| -rw-r--r-- | pango/break.c | 76 | ||||
| -rw-r--r-- | pango/pango-break.h | 12 | ||||
| -rw-r--r-- | pango/pango-layout.c | 82 |
4 files changed, 108 insertions, 64 deletions
diff --git a/docs/pango-sections.txt b/docs/pango-sections.txt index e20c917a..496db41a 100644 --- a/docs/pango-sections.txt +++ b/docs/pango-sections.txt @@ -23,6 +23,7 @@ pango_break pango_get_log_attrs pango_find_paragraph_boundary pango_default_break +pango_tailor_break PangoLogAttr <SUBSECTION> @@ -71,6 +72,7 @@ PANGO_CONTEXT_CLASS PANGO_IS_CONTEXT PANGO_IS_CONTEXT_CLASS PANGO_CONTEXT_GET_CLASS +PANGO_TYPE_ITEM <SUBSECTION Private> pango_context_get_type diff --git a/pango/break.c b/pango/break.c index 11a1e34a..8066fca0 100644 --- a/pango/break.c +++ b/pango/break.c @@ -146,13 +146,11 @@ typedef enum * @attrs: logical attributes to fill in * @attrs_len: size of the array passed as @attrs * - * This is the default break algorithm, used if no language - * engine overrides it. Normally you should use pango_break() - * instead. Unlike pango_break(), - * @analysis can be %NULL, but only do that if you know what - * you're doing. If you need an analysis to pass to pango_break(), - * you need to pango_itemize(). In most cases however you should - * simply use pango_get_log_attrs(). + * This is the default break algorithm. It applies Unicode + * rules without language-specific tailoring, therefore + * the @analyis argument is unused and can be %NULL. + * + * See pango_tailor_break() for language-specific breaks. **/ void pango_default_break (const gchar *text, @@ -1604,8 +1602,10 @@ tailor_break (const gchar *text, * @attrs_len: size of the array passed as @attrs * * Determines possible line, word, and character breaks - * for a string of Unicode text with a single analysis. For most - * purposes you may want to use pango_get_log_attrs(). + * for a string of Unicode text with a single analysis. + * For most purposes you may want to use pango_get_log_attrs(). + * + * Deprecated: 1.44: Use pango_default_break() and pango_tailor_break() */ void pango_break (const gchar *text, @@ -1721,6 +1721,43 @@ pango_find_paragraph_boundary (const gchar *text, *next_paragraph_start = start - text; } +/** + * pango_tailor_break: + * @text: text to process. Must be valid UTF-8 + * @length: length in bytes of @text + * @analysis: #PangoAnalysis structure from pango_itemize() for @text + * @log_attrs: (array length=attrs_len): array with one #PangoLogAttr + * per character in @text, plus one extra, to be filled in + * @attrs_len: length of @log_attrs array + * + * Apply language-specific tailoring to the breaks in + * @log_attrs, which are assumed to have been produced + * by pango_default_break(). + */ +void +pango_tailor_break (const char *text, + int length, + PangoAnalysis *analysis, + PangoLogAttr *log_attrs, + int log_attrs_len) +{ + PangoLogAttr *start = log_attrs; + PangoLogAttr attr_before = *start; + + if (tailor_break (text, length, analysis, log_attrs, log_attrs_len)) + { + /* if tailored, we enforce some of the attrs from before + * tailoring at the boundary + */ + + start->backspace_deletes_character = attr_before.backspace_deletes_character; + + start->is_line_break |= attr_before.is_line_break; + start->is_mandatory_break |= attr_before.is_mandatory_break; + start->is_cursor_position |= attr_before.is_cursor_position; + } +} + static int tailor_segment (const char *range_start, const char *range_end, @@ -1734,22 +1771,11 @@ tailor_segment (const char *range_start, chars_in_range = pango_utf8_strlen (range_start, range_end - range_start); - if (tailor_break (range_start, - range_end - range_start, - analysis, - start, - chars_in_range + 1)) - { - /* if tailored, we enforce some of the attrs from before tailoring at - * the boundary - */ - - start->backspace_deletes_character = attr_before.backspace_deletes_character; - - start->is_line_break |= attr_before.is_line_break; - start->is_mandatory_break |= attr_before.is_mandatory_break; - start->is_cursor_position |= attr_before.is_cursor_position; - } + pango_tailor_break (range_start, + range_end - range_start, + analysis, + start, + chars_in_range + 1); return chars_in_range; } diff --git a/pango/pango-break.h b/pango/pango-break.h index b035506e..4e1db0bc 100644 --- a/pango/pango-break.h +++ b/pango/pango-break.h @@ -100,10 +100,7 @@ struct _PangoLogAttr guint is_word_boundary : 1; }; -/* Determine information about cluster/word/line breaks in a string - * of Unicode text. - */ -PANGO_AVAILABLE_IN_ALL +PANGO_DEPRECATED_IN_1_44 void pango_break (const gchar *text, int length, PangoAnalysis *analysis, @@ -136,6 +133,13 @@ void pango_default_break (const gchar *text, PangoLogAttr *attrs, int attrs_len); +PANGO_AVAILABLE_IN_1_44 +void pango_tailor_break (const gchar *text, + int length, + PangoAnalysis *analysis, + PangoLogAttr *attrs, + int attrs_len); + G_END_DECLS #endif /* __PANGO_BREAK_H__ */ diff --git a/pango/pango-layout.c b/pango/pango-layout.c index e0725adc..21897e29 100644 --- a/pango/pango-layout.c +++ b/pango/pango-layout.c @@ -3958,51 +3958,61 @@ process_line (PangoLayout *layout, state->line_start_offset = state->start_offset; } +static gboolean +can_break_together (PangoAnalysis *analysis1, + PangoAnalysis *analysis2) +{ + return analysis1->level == analysis2->level && + analysis1->script == analysis2->script && + analysis1->language == analysis2->language; +} + static void get_items_log_attrs (const char *text, + int length, GList *items, PangoLogAttr *log_attrs, - int para_delimiter_len) + int log_attrs_len) { int offset = 0; int index = 0; + int num_bytes = 0; + int num_chars = 0; + PangoAnalysis analysis = { NULL }; + GList *l; - while (items) - { - PangoItem tmp_item = *(PangoItem *)items->data; - - /* Accumulate all the consecutive items that match in language - * characteristics, ignoring font, style tags, etc. - */ - while (items->next) - { - PangoItem *next_item = items->next->data; - - /* FIXME: Handle language tags */ - tmp_item.length += next_item->length; - tmp_item.num_chars += next_item->num_chars; + analysis.level = -1; - items = items->next; - } + pango_default_break (text, length, &analysis, log_attrs, log_attrs_len); - /* Break the paragraph delimiters with the last item */ - if (items->next == NULL) - { - tmp_item.num_chars += pango_utf8_strlen (text + index + tmp_item.length, para_delimiter_len); - tmp_item.length += para_delimiter_len; - } - - /* XXX This is wrong. we should call pango_default_break on the entire - * layout text and then tailor_break on each language change, like - * pango_get_log_attrs does. - */ - pango_break (text + index, tmp_item.length, &tmp_item.analysis, - log_attrs + offset, tmp_item.num_chars + 1); + for (l = items; l; l = l->next) + { + PangoItem *item = l->data; - offset += tmp_item.num_chars; - index += tmp_item.length; + if (l == items) + { + analysis = item->analysis; + index = item->offset; + offset = 0; + } - items = items->next; + if (can_break_together (&analysis, &item->analysis)) + { + num_bytes += item->length; + num_chars += item->num_chars; + } + else + { + pango_tailor_break (text + index, + num_bytes, + &analysis, + log_attrs + offset, + num_chars + 1); + + analysis = item->analysis; + index += num_bytes; + offset += num_chars; + } } } @@ -4243,9 +4253,11 @@ pango_layout_check_lines (PangoLayout *layout) if (no_break_attrs) apply_no_break_attributes (state.items, no_break_attrs); - get_items_log_attrs (start, state.items, + get_items_log_attrs (start, + delimiter_index + delim_len, + state.items, layout->log_attrs + start_offset, - delim_len); + layout->n_chars + 1 - start_offset); state.base_dir = base_dir; state.line_of_par = 1; |