Lines
100 %
Functions
Branches
/*
* Copyright © 2004 Red Hat, Inc.
*
* Permission to use, copy, modify, distribute, and sell this software
* and its documentation for any purpose is hereby granted without
* fee, provided that the above copyright notice appear in all copies
* and that both that copyright notice and this permission notice
* appear in supporting documentation, and that the name of
* Red Hat, Inc. not be used in advertising or publicity pertaining to
* distribution of the software without specific, written prior
* permission. Red Hat, Inc. makes no representations about the
* suitability of this software for any purpose. It is provided "as
* is" without express or implied warranty.
* RED HAT, INC. DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS
* SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
* FITNESS, IN NO EVENT SHALL RED HAT, INC. BE LIABLE FOR ANY SPECIAL,
* INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER
* RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
* OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR
* IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
* Author: Carl D. Worth <cworth@cworth.org>
*/
/* Bug history
* 2004-11-03 Steve Chaplin <stevech1097@yahoo.com.au>
* Reported bug on mailing list:
* From: Steve Chaplin <stevech1097@yahoo.com.au>
* To: cairo@cairographics.org
* Date: Thu, 04 Nov 2004 00:00:17 +0800
* Subject: [cairo] Rotated text bug on drawable target
* The attached file draws text rotated 90 degrees first to a PNG file and
* then to a drawable. The PNG file looks fine, the text on the drawable is
* unreadable.
* Steve
* 2004-11-03 Carl Worth <cworth@cworth.org>
* Looks like the major problems with this bug appeared in the great
* font rework between 0.1.23 and 0.2.0. And it looks like we need
* to fix the regression test suite to test the xlib target (since
* the bug does not show up in the png backend).
* Hmm... Actually, things don't look perfect even in the PNG
* output. Look at how that 'o' moves around. It's particularly off
* in the case where it's rotated by PI.
* And I'm still not sure about what to do for test cases with
* text--a new version of freetype will change everything. We may
* need to add a simple backend for stroked fonts and add a simple
* builtin font to cairo for pixel-perfect tests with text.
* 2005-08-23
* It appears that the worst placement and glyph selection problems
* have now been resolved. In the past some letters were noticeably
* of a different size at some rotations, and there was a lot of
* drift away from the baseline. These problems do not appear
* anymore.
* Another thing that helps is that we now have font options which
* we can use to disable hinting in order to get more repeatable
* results. I'm doing that in this test now.
* There are still some subtle positioning problems which I'm
* assuming are due to the lack of finer-than-whole-pixel glyph
* positioning. I'm generating a reference image now by replacing
* cairo_show_text with cairo_text_path; cairo_fill. This will let
* us look more closely at the remaining positioning problems. (In
* particular, I want to make sure we're rounding as well as
* possible).
* 2007-02-21
* Seems like all the "bugs" have been fixed and all remaining is
* missing support for subpixel glyph positioning. Removing from
* XFAIL now.
#include "cairo-test.h"
#define WIDTH 150
#define HEIGHT 150
#define NUM_TEXT 20
#define TEXT_SIZE 12
/* Draw the word cairo at NUM_TEXT different angles.
* We separate the circle into quadrants to reduce
* numerical errors i.e. so each quarter is pixel-aligned.
static void
draw_quadrant (cairo_t *cr,
const char *text,
const cairo_text_extents_t *extents,
const cairo_matrix_t *transform,
int x_off, int y_off)
{
int i;
for (i = 0; i < NUM_TEXT/4; i++) {
cairo_save (cr);
cairo_rotate (cr, 2*M_PI*i/NUM_TEXT);
cairo_transform (cr, transform);
cairo_set_line_width (cr, 1.0);
cairo_rectangle (cr, x_off - 0.5, y_off - 0.5, extents->width + 1, extents->height + 1);
cairo_set_source_rgb (cr, 1, 0, 0);
cairo_stroke (cr);
cairo_move_to (cr, x_off - extents->x_bearing, y_off - extents->y_bearing);
cairo_set_source_rgb (cr, 0, 0, 0);
#if CAIRO_TEST_GENERATE_REFERENCE_IMAGE
cairo_text_path (cr, text);
cairo_fill (cr);
#else
cairo_show_text (cr, text);
#endif
cairo_restore (cr);
}
static cairo_test_status_t
draw (cairo_t *cr, int width, int height)
cairo_text_extents_t extents;
cairo_font_options_t *font_options;
const char text[] = "cairo";
int x_off, y_off;
cairo_matrix_t m;
/* paint white so we don't need separate ref images for
* RGB24 and ARGB32 */
cairo_set_source_rgb (cr, 1., 1., 1.);
cairo_paint (cr);
cairo_select_font_face (cr, CAIRO_TEST_FONT_FAMILY " Sans",
CAIRO_FONT_SLANT_NORMAL,
CAIRO_FONT_WEIGHT_NORMAL);
cairo_set_font_size (cr, TEXT_SIZE);
font_options = cairo_font_options_create ();
cairo_get_font_options (cr, font_options);
cairo_font_options_set_hint_metrics (font_options, CAIRO_HINT_METRICS_OFF);
cairo_set_font_options (cr, font_options);
cairo_font_options_destroy (font_options);
cairo_translate (cr, WIDTH/2.0, HEIGHT/2.0);
cairo_text_extents (cr, text, &extents);
if (NUM_TEXT == 1) {
x_off = y_off = 0;
} else {
y_off = - floor (0.5 + extents.height / 2.0);
x_off = floor (0.5 + (extents.height+1) / (2 * tan (M_PI/NUM_TEXT)));
cairo_matrix_init_identity (&m);
draw_quadrant (cr, text, &extents, &m, x_off, y_off);
cairo_matrix_init (&m, 0, 1, -1, 0, 0, 0);
cairo_scale (cr, -1, -1);
return CAIRO_TEST_SUCCESS;
CAIRO_TEST (text_rotate,
"Tests show_text under various rotations",
"text, transform", /* keywords */
NULL, /* requirements */
WIDTH, HEIGHT,
NULL, draw)