1/*
2 * Copyright (c) 2008, 2009 Google Inc. All rights reserved.
3 *
4 * Redistribution and use in source and binary forms, with or without
5 * modification, are permitted provided that the following conditions are
6 * met:
7 *
8 *     * Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 *     * Redistributions in binary form must reproduce the above
11 * copyright notice, this list of conditions and the following disclaimer
12 * in the documentation and/or other materials provided with the
13 * distribution.
14 *     * Neither the name of Google Inc. nor the names of its
15 * contributors may be used to endorse or promote products derived from
16 * this software without specific prior written permission.
17 *
18 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
19 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
20 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
21 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
22 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
23 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
24 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
25 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
26 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
27 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
28 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 */
30
31#include "config.h"
32#include <windows.h>
33#include <vector>
34
35#include "Font.h"
36#include "GlyphPageTreeNode.h"
37#include "PlatformBridge.h"
38#include "SimpleFontData.h"
39#include "SystemInfo.h"
40#include "UniscribeHelperTextRun.h"
41
42namespace WebCore {
43
44// Fills one page of font data pointers with 0 to indicate that there
45// are no glyphs for the characters.
46static void fillEmptyGlyphs(GlyphPage* page)
47{
48    for (int i = 0; i < GlyphPage::size; ++i)
49        page->setGlyphDataForIndex(i, 0, 0);
50}
51
52// Lazily initializes space glyph
53static Glyph initSpaceGlyph(HDC dc, Glyph* spaceGlyph)
54{
55    if (*spaceGlyph)
56        return *spaceGlyph;
57
58    static wchar_t space = ' ';
59    GetGlyphIndices(dc, &space, 1, spaceGlyph, 0);
60    return *spaceGlyph;
61}
62
63// Fills |length| glyphs starting at |offset| in a |page| in the Basic
64// Multilingual Plane (<= U+FFFF). The input buffer size should be the
65// same as |length|. We can use the standard Windows GDI functions here.
66// Returns true if any glyphs were found.
67static bool fillBMPGlyphs(unsigned offset,
68                          unsigned length,
69                          UChar* buffer,
70                          GlyphPage* page,
71                          const SimpleFontData* fontData,
72                          bool recurse)
73{
74    HDC dc = GetDC((HWND)0);
75    HGDIOBJ oldFont = SelectObject(dc, fontData->platformData().hfont());
76
77    TEXTMETRIC tm = {0};
78    if (!GetTextMetrics(dc, &tm)) {
79        SelectObject(dc, oldFont);
80        ReleaseDC(0, dc);
81
82        if (recurse) {
83            if (PlatformBridge::ensureFontLoaded(fontData->platformData().hfont()))
84                return fillBMPGlyphs(offset, length, buffer, page, fontData, false);
85
86            fillEmptyGlyphs(page);
87            return false;
88        } else {
89            // FIXME: Handle gracefully the error if this call also fails.
90            // See http://crbug.com/6401
91            LOG_ERROR("Unable to get the text metrics after second attempt");
92            fillEmptyGlyphs(page);
93            return false;
94        }
95    }
96
97    // FIXME: GetGlyphIndices() sets each item of localGlyphBuffer[]
98    // with the one of the values listed below.
99    //  * With the GGI_MARK_NONEXISTING_GLYPHS flag
100    //    + If the font has a glyph available for the character,
101    //      localGlyphBuffer[i] > 0x0.
102    //    + If the font does not have glyphs available for the character,
103    //      localGlyphBuffer[i] = 0x1F (TrueType Collection?) or
104    //                            0xFFFF (OpenType?).
105    //  * Without the GGI_MARK_NONEXISTING_GLYPHS flag
106    //    + If the font has a glyph available for the character,
107    //      localGlyphBuffer[i] > 0x0.
108    //    + If the font does not have glyphs available for the character,
109    //      localGlyphBuffer[i] = 0x80.
110    //      (Windows automatically assigns the glyph for a box character to
111    //      prevent ExtTextOut() from returning errors.)
112    // To avoid from hurting the rendering performance, this code just
113    // tells WebKit whether or not the all glyph indices for the given
114    // characters are 0x80 (i.e. a possibly-invalid glyph) and let it
115    // use alternative fonts for the characters.
116    // Although this may cause a problem, it seems to work fine as far as I
117    // have tested. (Obviously, I need more tests.)
118    WORD localGlyphBuffer[GlyphPage::size];
119
120    // FIXME: I find some Chinese characters can not be correctly displayed
121    // when call GetGlyphIndices without flag GGI_MARK_NONEXISTING_GLYPHS,
122    // because the corresponding glyph index is set as 0x20 when current font
123    // does not have glyphs available for the character. According a blog post
124    // http://blogs.msdn.com/michkap/archive/2006/06/28/649791.aspx
125    // I think we should switch to the way about calling GetGlyphIndices with
126    // flag GGI_MARK_NONEXISTING_GLYPHS, it should be OK according the
127    // description of MSDN.
128    // Also according to Jungshik and Hironori's suggestion and modification
129    // we treat turetype and raster Font as different way when windows version
130    // is less than Vista.
131    GetGlyphIndices(dc, buffer, length, localGlyphBuffer, GGI_MARK_NONEXISTING_GLYPHS);
132
133    // Copy the output to the GlyphPage
134    bool haveGlyphs = false;
135    int invalidGlyph = 0xFFFF;
136    const DWORD cffTableTag = 0x20464643; // 4-byte identifier for OpenType CFF table ('CFF ').
137    if ((windowsVersion() < WindowsVista) && !(tm.tmPitchAndFamily & TMPF_TRUETYPE) && (GetFontData(dc, cffTableTag, 0, 0, 0) == GDI_ERROR))
138        invalidGlyph = 0x1F;
139
140    Glyph spaceGlyph = 0;  // Glyph for a space. Lazily filled.
141
142    for (unsigned i = 0; i < length; i++) {
143        UChar c = buffer[i];
144        Glyph glyph = localGlyphBuffer[i];
145        const SimpleFontData* glyphFontData = fontData;
146        // When this character should be a space, we ignore whatever the font
147        // says and use a space. Otherwise, if fonts don't map one of these
148        // space or zero width glyphs, we will get a box.
149        if (Font::treatAsSpace(c)) {
150            // Hard code the glyph indices for characters that should be
151            // treated like spaces.
152            glyph = initSpaceGlyph(dc, &spaceGlyph);
153        } else if (glyph == invalidGlyph) {
154            // WebKit expects both the glyph index and FontData
155            // pointer to be 0 if the glyph is not present
156            glyph = 0;
157            glyphFontData = 0;
158        } else
159            haveGlyphs = true;
160        page->setGlyphDataForCharacter(offset + i, glyph, glyphFontData);
161    }
162
163    SelectObject(dc, oldFont);
164    ReleaseDC(0, dc);
165    return haveGlyphs;
166}
167
168// For non-BMP characters, each is two words (UTF-16) and the input buffer
169// size is 2 * |length|. Since GDI doesn't know how to handle non-BMP
170// characters, we must use Uniscribe to tell us the glyph indices.
171//
172// We don't want to call this in the case of "regular" characters since some
173// fonts may not have the correct combining rules for accents. See the notes
174// at the bottom of ScriptGetCMap. We can't use ScriptGetCMap, though, since
175// it doesn't seem to support UTF-16, despite what this blog post says:
176//   http://blogs.msdn.com/michkap/archive/2006/06/29/650680.aspx
177//
178// So we fire up the full Uniscribe doohicky, give it our string, and it will
179// correctly handle the UTF-16 for us. The hard part is taking this and getting
180// the glyph indices back out that correspond to the correct input characters,
181// since they may be missing.
182//
183// Returns true if any glyphs were found.
184static bool fillNonBMPGlyphs(unsigned offset,
185                             unsigned length,
186                             UChar* buffer,
187                             GlyphPage* page,
188                             const SimpleFontData* fontData)
189{
190    bool haveGlyphs = false;
191
192    UniscribeHelperTextRun state(buffer, length * 2, false,
193                                 fontData->platformData().hfont(),
194                                 fontData->platformData().scriptCache(),
195                                 fontData->platformData().scriptFontProperties());
196    state.setInhibitLigate(true);
197    state.setDisableFontFallback(true);
198    state.init();
199
200    for (unsigned i = 0; i < length; i++) {
201        // Each character in this input buffer is a surrogate pair, which
202        // consists of two UChars. So, the offset for its i-th character is
203        // (i * 2).
204        WORD glyph = state.firstGlyphForCharacter(i * 2);
205        if (glyph) {
206            haveGlyphs = true;
207            page->setGlyphDataForIndex(offset + i, glyph, fontData);
208        } else
209            // Clear both glyph and fontData fields.
210            page->setGlyphDataForIndex(offset + i, 0, 0);
211    }
212    return haveGlyphs;
213}
214
215// We're supposed to return true if there are any glyphs in the range
216// specified by |offset| and |length| in  our font,
217// false if there are none.
218bool GlyphPage::fill(unsigned offset, unsigned length, UChar* characterBuffer,
219                     unsigned bufferLength, const SimpleFontData* fontData)
220{
221    // We have to handle BMP and non-BMP characters differently.
222    // FIXME: Add assertions to make sure that buffer is entirely in BMP
223    // or entirely in non-BMP.
224    if (bufferLength == length)
225        return fillBMPGlyphs(offset, length, characterBuffer, this, fontData, true);
226
227    if (bufferLength == 2 * length) {
228        // A non-BMP input buffer will be twice as long as output glyph buffer
229        // because each character in the non-BMP input buffer will be
230        // represented by a surrogate pair (two UChar's).
231        return fillNonBMPGlyphs(offset, length, characterBuffer, this, fontData);
232    }
233
234    ASSERT_NOT_REACHED();
235    return false;
236}
237
238}  // namespace WebCore
239