diff --git a/src/HTMLRenderer/text.cc b/src/HTMLRenderer/text.cc index aa1d022..6870648 100644 --- a/src/HTMLRenderer/text.cc +++ b/src/HTMLRenderer/text.cc @@ -78,10 +78,10 @@ void HTMLRenderer::drawString(GfxState * state, GooString * s) * This is by standard * however some PDF will use ' ' as a normal encoding slot * such that it will be mapped to other unicodes - * In that case, when sapce_as_offset is on, we will simply ignore that character... + * In that case, when space_as_offset is on, we will simply ignore that character... * * Checking mapped unicode may or may not work - * There are always ugly PDF files with no usefull info at all. + * There are always ugly PDF files with no useful info at all. */ is_space = true; ++nSpaces; @@ -113,11 +113,13 @@ void HTMLRenderer::drawString(GfxState * state, GooString * s) html_text_page.get_cur_line()->append_unicodes(&uu, 1); /* * In PDF, word_space is appended if (n == 1 and *p = ' ') - * but in HTML, word_space is appended if (uu == ' ') + * but in HTML, word_space is appended if (uu == ' ' || 0xa0) */ - int space_count = (is_space ? 1 : 0) - (uu == ' ' ? 1 : 0); + int space_count = (is_space ? 1 : 0) - ((uu == ' ' || uu == 0xa0) ? 1 : 0); if(space_count != 0) + { html_text_page.get_cur_line()->append_offset(cur_word_space * draw_text_scale * space_count); + } } } diff --git a/src/util/ffw.c b/src/util/ffw.c index 99fcb25..1be092c 100644 --- a/src/util/ffw.c +++ b/src/util/ffw.c @@ -262,6 +262,7 @@ void ffw_cidflatten(void) /* * There is no check if a glyph with the same unicode exists! + * TODO: let FontForge fill in the standard glyph name <- or maybe this might cause collision? */ void ffw_add_empty_char(int32_t unicode, int width) {