com.aowagie.text.pdf.BaseFont Maven / Gradle / Ivy
/*
* $Id: BaseFont.java 3869 2009-04-17 18:03:45Z blowagie $
*
* Copyright 2000-2006 by Paulo Soares.
*
* The contents of this file are subject to the Mozilla Public License Version 1.1
* (the "License"); you may not use this file except in compliance with the License.
* You may obtain a copy of the License at http://www.mozilla.org/MPL/
*
* Software distributed under the License is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
* for the specific language governing rights and limitations under the License.
*
* The Original Code is 'iText, a free JAVA-PDF library'.
*
* The Initial Developer of the Original Code is Bruno Lowagie. Portions created by
* the Initial Developer are Copyright (C) 1999, 2000, 2001, 2002 by Bruno Lowagie.
* All Rights Reserved.
* Co-Developer of the code is Paulo Soares. Portions created by the Co-Developer
* are Copyright (C) 2000, 2001, 2002 by Paulo Soares. All Rights Reserved.
*
* Contributor(s): all the names of the contributors are added in the source code
* where applicable.
*
* Alternatively, the contents of this file may be used under the terms of the
* LGPL license (the "GNU LIBRARY GENERAL PUBLIC LICENSE"), in which case the
* provisions of LGPL are applicable instead of those above. If you wish to
* allow use of your version of this file only under the terms of the LGPL
* License and not to allow others to use your version of this file under
* the MPL, indicate your decision by deleting the provisions above and
* replace them with the notice and other provisions required by the LGPL.
* If you do not delete the provisions above, a recipient may use your version
* of this file under either the MPL or the GNU LIBRARY GENERAL PUBLIC LICENSE.
*
* This library is free software; you can redistribute it and/or modify it
* under the terms of the MPL as stated above or under the terms of the GNU
* Library General Public License as published by the Free Software Foundation;
* either version 2 of the License, or any later version.
*
* This library is distributed in the hope that it will be useful, but WITHOUT
* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
* FOR A PARTICULAR PURPOSE. See the GNU Library general Public License for more
* details.
*
* If you didn't download this code from the following link, you should check if
* you aren't using an obsolete version:
* http://www.lowagie.com/iText/
*/
package com.aowagie.text.pdf;
import java.io.IOException;
import java.io.InputStream;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.StringTokenizer;
import com.aowagie.text.DocumentException;
/**
* Base class for the several font types supported
*
* @author Paulo Soares ([email protected])
*/
public abstract class BaseFont {
/** This is a possible value of a base 14 type 1 font */
public static final String COURIER = "Courier";
/** This is a possible value of a base 14 type 1 font */
public static final String COURIER_BOLD = "Courier-Bold";
/** This is a possible value of a base 14 type 1 font */
public static final String COURIER_OBLIQUE = "Courier-Oblique";
/** This is a possible value of a base 14 type 1 font */
public static final String COURIER_BOLDOBLIQUE = "Courier-BoldOblique";
/** This is a possible value of a base 14 type 1 font */
public static final String HELVETICA = "Helvetica";
/** This is a possible value of a base 14 type 1 font */
public static final String HELVETICA_BOLD = "Helvetica-Bold";
/** This is a possible value of a base 14 type 1 font */
public static final String HELVETICA_OBLIQUE = "Helvetica-Oblique";
/** This is a possible value of a base 14 type 1 font */
public static final String HELVETICA_BOLDOBLIQUE = "Helvetica-BoldOblique";
/** This is a possible value of a base 14 type 1 font */
public static final String SYMBOL = "Symbol";
/** This is a possible value of a base 14 type 1 font */
public static final String TIMES_ROMAN = "Times-Roman";
/** This is a possible value of a base 14 type 1 font */
public static final String TIMES_BOLD = "Times-Bold";
/** This is a possible value of a base 14 type 1 font */
public static final String TIMES_ITALIC = "Times-Italic";
/** This is a possible value of a base 14 type 1 font */
public static final String TIMES_BOLDITALIC = "Times-BoldItalic";
/** This is a possible value of a base 14 type 1 font */
public static final String ZAPFDINGBATS = "ZapfDingbats";
/** The maximum height above the baseline reached by glyphs in this
* font, excluding the height of glyphs for accented characters.
*/
public static final int ASCENT = 1;
/** The y coordinate of the top of flat capital letters, measured from
* the baseline.
*/
public static final int CAPHEIGHT = 2;
/** The maximum depth below the baseline reached by glyphs in this
* font. The value is a negative number.
*/
public static final int DESCENT = 3;
/** The angle, expressed in degrees counterclockwise from the vertical,
* of the dominant vertical strokes of the font. The value is
* negative for fonts that slope to the right, as almost all italic fonts do.
*/
public static final int ITALICANGLE = 4;
/** The lower left x glyph coordinate.
*/
public static final int BBOXLLX = 5;
/** The lower left y glyph coordinate.
*/
public static final int BBOXLLY = 6;
/** The upper right x glyph coordinate.
*/
public static final int BBOXURX = 7;
/** The upper right y glyph coordinate.
*/
public static final int BBOXURY = 8;
/** java.awt.Font property */
public static final int AWT_ASCENT = 9;
/** java.awt.Font property */
public static final int AWT_DESCENT = 10;
/** java.awt.Font property */
public static final int AWT_LEADING = 11;
/** java.awt.Font property */
public static final int AWT_MAXADVANCE = 12;
/**
* The underline position. Usually a negative value.
*/
public static final int UNDERLINE_POSITION = 13;
/**
* The underline thickness.
*/
public static final int UNDERLINE_THICKNESS = 14;
/**
* The strikethrough position.
*/
public static final int STRIKETHROUGH_POSITION = 15;
/**
* The strikethrough thickness.
*/
public static final int STRIKETHROUGH_THICKNESS = 16;
/**
* The recommended vertical size for subscripts for this font.
*/
public static final int SUBSCRIPT_SIZE = 17;
/**
* The recommended vertical offset from the baseline for subscripts for this font. Usually a negative value.
*/
public static final int SUBSCRIPT_OFFSET = 18;
/**
* The recommended vertical size for superscripts for this font.
*/
public static final int SUPERSCRIPT_SIZE = 19;
/**
* The recommended vertical offset from the baseline for superscripts for this font.
*/
public static final int SUPERSCRIPT_OFFSET = 20;
/** The font is Type 1.
*/
public static final int FONT_TYPE_T1 = 0;
/** The font is True Type with a standard encoding.
*/
public static final int FONT_TYPE_TT = 1;
/** The font is CJK.
*/
public static final int FONT_TYPE_CJK = 2;
/** The font is True Type with a Unicode encoding.
*/
public static final int FONT_TYPE_TTUNI = 3;
/** A font already inside the document.
*/
public static final int FONT_TYPE_DOCUMENT = 4;
/** A Type3 font.
*/
public static final int FONT_TYPE_T3 = 5;
/** The Unicode encoding with horizontal writing.
*/
public static final String IDENTITY_H = "Identity-H";
/** The Unicode encoding with vertical writing.
*/
public static final String IDENTITY_V = "Identity-V";
/** A possible encoding. */
public static final String CP1250 = "Cp1250";
/** A possible encoding. */
public static final String CP1252 = "Cp1252";
/** A possible encoding. */
public static final String CP1257 = "Cp1257";
/** A possible encoding. */
public static final String WINANSI = "Cp1252";
/** A possible encoding. */
public static final String MACROMAN = "MacRoman";
public static final int[] CHAR_RANGE_LATIN = {0, 0x17f, 0x2000, 0x206f, 0x20a0, 0x20cf, 0xfb00, 0xfb06};
public static final int[] CHAR_RANGE_ARABIC = {0, 0x7f, 0x0600, 0x067f, 0x20a0, 0x20cf, 0xfb50, 0xfbff, 0xfe70, 0xfeff};
public static final int[] CHAR_RANGE_HEBREW = {0, 0x7f, 0x0590, 0x05ff, 0x20a0, 0x20cf, 0xfb1d, 0xfb4f};
public static final int[] CHAR_RANGE_CYRILLIC = {0, 0x7f, 0x0400, 0x052f, 0x2000, 0x206f, 0x20a0, 0x20cf};
/** if the font has to be embedded */
public static final boolean EMBEDDED = true;
/** if the font doesn't have to be embedded */
public static final boolean NOT_EMBEDDED = false;
/** if the font has to be cached */
public static final boolean CACHED = true;
/** if the font doesn't have to be cached */
public static final boolean NOT_CACHED = false;
/** The path to the font resources. */
public static final String RESOURCE_PATH = "com/lowagie/text/pdf/fonts/";
/** The fake CID code that represents a newline. */
public static final char CID_NEWLINE = '\u7fff';
protected ArrayList subsetRanges;
/** The font type.
*/
int fontType;
/** a not defined character in a custom PDF encoding */
public static final String notdef = ".notdef";
/** table of characters widths for this encoding */
protected int widths[] = new int[256];
/** encoding names */
protected String differences[] = new String[256];
/** same as differences but with the unicode codes */
protected char unicodeDifferences[] = new char[256];
protected int charBBoxes[][] = new int[256][];
/** encoding used with this font */
protected String encoding;
/** true if the font is to be embedded in the PDF */
protected boolean embedded;
/**
* The compression level for the font stream.
* @since 2.1.3
*/
protected int compressionLevel = PdfStream.DEFAULT_COMPRESSION;
/**
* true if the font must use its built in encoding. In that case the
* encoding is only used to map a char to the position inside
* the font, not to the expected char name.
*/
protected boolean fontSpecific = true;
/** cache for the fonts already used. */
protected static HashMap fontCache = new HashMap();
/** list of the 14 built in fonts. */
protected static final HashMap BuiltinFonts14 = new HashMap();
/** Forces the output of the width array. Only matters for the 14
* built-in fonts.
*/
protected boolean forceWidthsOutput = false;
/** Converts char directly to byte
* by casting.
*/
protected boolean directTextToByte = false;
/** Indicates if all the glyphs and widths for that particular
* encoding should be included in the document.
*/
protected boolean subset = true;
protected boolean fastWinansi = false;
/**
* Custom encodings use this map to key the Unicode character
* to the single byte code.
*/
protected IntHashtable specialMap;
static {
BuiltinFonts14.put(COURIER, PdfName.COURIER);
BuiltinFonts14.put(COURIER_BOLD, PdfName.COURIER_BOLD);
BuiltinFonts14.put(COURIER_BOLDOBLIQUE, PdfName.COURIER_BOLDOBLIQUE);
BuiltinFonts14.put(COURIER_OBLIQUE, PdfName.COURIER_OBLIQUE);
BuiltinFonts14.put(HELVETICA, PdfName.HELVETICA);
BuiltinFonts14.put(HELVETICA_BOLD, PdfName.HELVETICA_BOLD);
BuiltinFonts14.put(HELVETICA_BOLDOBLIQUE, PdfName.HELVETICA_BOLDOBLIQUE);
BuiltinFonts14.put(HELVETICA_OBLIQUE, PdfName.HELVETICA_OBLIQUE);
BuiltinFonts14.put(SYMBOL, PdfName.SYMBOL);
BuiltinFonts14.put(TIMES_ROMAN, PdfName.TIMES_ROMAN);
BuiltinFonts14.put(TIMES_BOLD, PdfName.TIMES_BOLD);
BuiltinFonts14.put(TIMES_BOLDITALIC, PdfName.TIMES_BOLDITALIC);
BuiltinFonts14.put(TIMES_ITALIC, PdfName.TIMES_ITALIC);
BuiltinFonts14.put(ZAPFDINGBATS, PdfName.ZAPFDINGBATS);
}
/** Generates the PDF stream with the Type1 and Truetype fonts returning
* a PdfStream.
*/
static class StreamFont extends PdfStream {
/** Generates the PDF stream with the Type1 and Truetype fonts returning
* a PdfStream.
* @param contents the content of the stream
* @param lengths an array of int that describes the several lengths of each part of the font
* @param compressionLevel the compression level of the Stream
* @throws DocumentException error in the stream compression
* @since 2.1.3 (replaces the constructor without param compressionLevel)
*/
public StreamFont(final byte contents[], final int lengths[], final int compressionLevel) throws DocumentException {
try {
this.bytes = contents;
put(PdfName.LENGTH, new PdfNumber(this.bytes.length));
for (int k = 0; k < lengths.length; ++k) {
put(new PdfName("Length" + (k + 1)), new PdfNumber(lengths[k]));
}
flateCompress(compressionLevel);
}
catch (final Exception e) {
throw new DocumentException(e);
}
}
/**
* Generates the PDF stream for a font.
* @param contents the content of a stream
* @param subType the subtype of the font.
* @param compressionLevel the compression level of the Stream
* @throws DocumentException error in the stream compression
* @since 2.1.3 (replaces the constructor without param compressionLevel)
*/
public StreamFont(final byte contents[], final String subType, final int compressionLevel) throws DocumentException {
try {
this.bytes = contents;
put(PdfName.LENGTH, new PdfNumber(this.bytes.length));
if (subType != null) {
put(PdfName.SUBTYPE, new PdfName(subType));
}
flateCompress(compressionLevel);
}
catch (final Exception e) {
throw new DocumentException(e);
}
}
}
/**
*Creates new BaseFont
*/
protected BaseFont() {
}
/**
* Creates a new font. This will always be the default Helvetica font (not embedded).
* This method is introduced because Helvetica is used in many examples.
* @return a BaseFont object (Helvetica, Winansi, not embedded)
* @throws IOException This shouldn't occur ever
* @throws DocumentException This shouldn't occur ever
* @since 2.1.1
*/
public static BaseFont createFont() throws DocumentException, IOException {
return createFont(BaseFont.HELVETICA, BaseFont.WINANSI, BaseFont.NOT_EMBEDDED);
}
/**
* Creates a new font. This font can be one of the 14 built in types,
* a Type1 font referred to by an AFM or PFM file, a TrueType font (simple or collection) or a CJK font from the
* Adobe Asian Font Pack. TrueType fonts and CJK fonts can have an optional style modifier
* appended to the name. These modifiers are: Bold, Italic and BoldItalic. An
* example would be "STSong-Light,Bold". Note that this modifiers do not work if
* the font is embedded. Fonts in TrueType collections are addressed by index such as "msgothic.ttc,1".
* This would get the second font (indexes start at 0), in this case "MS PGothic".
*
* The fonts are cached and if they already exist they are extracted from the cache,
* not parsed again.
*
* Besides the common encodings described by name, custom encodings
* can also be made. These encodings will only work for the single byte fonts
* Type1 and TrueType. The encoding string starts with a '#'
* followed by "simple" or "full". If "simple" there is a decimal for the first character position and then a list
* of hex values representing the Unicode codes that compose that encoding.
* The "simple" encoding is recommended for TrueType fonts
* as the "full" encoding risks not matching the character with the right glyph
* if not done with care.
* The "full" encoding is specially aimed at Type1 fonts where the glyphs have to be
* described by non standard names like the Tex math fonts. Each group of three elements
* compose a code position: the one byte code order in decimal or as 'x' (x cannot be the space), the name and the Unicode character
* used to access the glyph. The space must be assigned to character position 32 otherwise
* text justification will not work.
*
* Example for a "simple" encoding that includes the Unicode
* character space, A, B and ecyrillic:
*
* "# simple 32 0020 0041 0042 0454"
*
*
* Example for a "full" encoding for a Type1 Tex font:
*
* "# full 'A' nottriangeqlleft 0041 'B' dividemultiply 0042 32 space 0020"
*
*
* This method calls:
*
* createFont(name, encoding, embedded, true, null, null);
*
* @param name the name of the font or its location on file
* @param encoding the encoding to be applied to this font
* @param embedded true if the font is to be embedded in the PDF
* @return returns a new font. This font may come from the cache
* @throws DocumentException the font is invalid
* @throws IOException the font file could not be read
*/
public static BaseFont createFont(final String name, final String encoding, final boolean embedded) throws DocumentException, IOException {
return createFont(name, encoding, embedded, true, null, null, false);
}
/**
* Creates a new font. This font can be one of the 14 built in types,
* a Type1 font referred to by an AFM or PFM file, a TrueType font (simple or collection) or a CJK font from the
* Adobe Asian Font Pack. TrueType fonts and CJK fonts can have an optional style modifier
* appended to the name. These modifiers are: Bold, Italic and BoldItalic. An
* example would be "STSong-Light,Bold". Note that this modifiers do not work if
* the font is embedded. Fonts in TrueType collections are addressed by index such as "msgothic.ttc,1".
* This would get the second font (indexes start at 0), in this case "MS PGothic".
*
* The fonts are cached and if they already exist they are extracted from the cache,
* not parsed again.
*
* Besides the common encodings described by name, custom encodings
* can also be made. These encodings will only work for the single byte fonts
* Type1 and TrueType. The encoding string starts with a '#'
* followed by "simple" or "full". If "simple" there is a decimal for the first character position and then a list
* of hex values representing the Unicode codes that compose that encoding.
* The "simple" encoding is recommended for TrueType fonts
* as the "full" encoding risks not matching the character with the right glyph
* if not done with care.
* The "full" encoding is specially aimed at Type1 fonts where the glyphs have to be
* described by non standard names like the Tex math fonts. Each group of three elements
* compose a code position: the one byte code order in decimal or as 'x' (x cannot be the space), the name and the Unicode character
* used to access the glyph. The space must be assigned to character position 32 otherwise
* text justification will not work.
*
* Example for a "simple" encoding that includes the Unicode
* character space, A, B and ecyrillic:
*
* "# simple 32 0020 0041 0042 0454"
*
*
* Example for a "full" encoding for a Type1 Tex font:
*
* "# full 'A' nottriangeqlleft 0041 'B' dividemultiply 0042 32 space 0020"
*
*
* This method calls:
*
* createFont(name, encoding, embedded, true, null, null);
*
* @param name the name of the font or its location on file
* @param encoding the encoding to be applied to this font
* @param embedded true if the font is to be embedded in the PDF
* @param forceRead in some cases (TrueTypeFont, Type1Font), the full font file will be read and kept in memory if forceRead is true
* @return returns a new font. This font may come from the cache
* @throws DocumentException the font is invalid
* @throws IOException the font file could not be read
* @since 2.1.5
*/
public static BaseFont createFont(final String name, final String encoding, final boolean embedded, final boolean forceRead) throws DocumentException, IOException {
return createFont(name, encoding, embedded, true, null, null, forceRead);
}
/** Creates a new font. This font can be one of the 14 built in types,
* a Type1 font referred to by an AFM or PFM file, a TrueType font (simple or collection) or a CJK font from the
* Adobe Asian Font Pack. TrueType fonts and CJK fonts can have an optional style modifier
* appended to the name. These modifiers are: Bold, Italic and BoldItalic. An
* example would be "STSong-Light,Bold". Note that this modifiers do not work if
* the font is embedded. Fonts in TrueType collections are addressed by index such as "msgothic.ttc,1".
* This would get the second font (indexes start at 0), in this case "MS PGothic".
*
* The fonts may or may not be cached depending on the flag cached.
* If the byte arrays are present the font will be
* read from them instead of the name. A name is still required to identify
* the font type.
*
* Besides the common encodings described by name, custom encodings
* can also be made. These encodings will only work for the single byte fonts
* Type1 and TrueType. The encoding string starts with a '#'
* followed by "simple" or "full". If "simple" there is a decimal for the first character position and then a list
* of hex values representing the Unicode codes that compose that encoding.
* The "simple" encoding is recommended for TrueType fonts
* as the "full" encoding risks not matching the character with the right glyph
* if not done with care.
* The "full" encoding is specially aimed at Type1 fonts where the glyphs have to be
* described by non standard names like the Tex math fonts. Each group of three elements
* compose a code position: the one byte code order in decimal or as 'x' (x cannot be the space), the name and the Unicode character
* used to access the glyph. The space must be assigned to character position 32 otherwise
* text justification will not work.
*
* Example for a "simple" encoding that includes the Unicode
* character space, A, B and ecyrillic:
*
* "# simple 32 0020 0041 0042 0454"
*
*
* Example for a "full" encoding for a Type1 Tex font:
*
* "# full 'A' nottriangeqlleft 0041 'B' dividemultiply 0042 32 space 0020"
*
* @param name the name of the font or its location on file
* @param encoding the encoding to be applied to this font
* @param embedded true if the font is to be embedded in the PDF
* @param cached true if the font comes from the cache or is added to
* the cache if new, false if the font is always created new
* @param ttfAfm the true type font or the afm in a byte array
* @param pfb the pfb in a byte array
* @return returns a new font. This font may come from the cache but only if cached
* is true, otherwise it will always be created new
* @throws DocumentException the font is invalid
* @throws IOException the font file could not be read
* @since iText 0.80
*/
public static BaseFont createFont(final String name, final String encoding, final boolean embedded, final boolean cached, final byte ttfAfm[], final byte pfb[]) throws DocumentException, IOException {
return createFont(name, encoding, embedded, cached, ttfAfm, pfb, false);
}
/** Creates a new font. This font can be one of the 14 built in types,
* a Type1 font referred to by an AFM or PFM file, a TrueType font (simple or collection) or a CJK font from the
* Adobe Asian Font Pack. TrueType fonts and CJK fonts can have an optional style modifier
* appended to the name. These modifiers are: Bold, Italic and BoldItalic. An
* example would be "STSong-Light,Bold". Note that this modifiers do not work if
* the font is embedded. Fonts in TrueType collections are addressed by index such as "msgothic.ttc,1".
* This would get the second font (indexes start at 0), in this case "MS PGothic".
*
* The fonts may or may not be cached depending on the flag cached.
* If the byte arrays are present the font will be
* read from them instead of the name. A name is still required to identify
* the font type.
*
* Besides the common encodings described by name, custom encodings
* can also be made. These encodings will only work for the single byte fonts
* Type1 and TrueType. The encoding string starts with a '#'
* followed by "simple" or "full". If "simple" there is a decimal for the first character position and then a list
* of hex values representing the Unicode codes that compose that encoding.
* The "simple" encoding is recommended for TrueType fonts
* as the "full" encoding risks not matching the character with the right glyph
* if not done with care.
* The "full" encoding is specially aimed at Type1 fonts where the glyphs have to be
* described by non standard names like the Tex math fonts. Each group of three elements
* compose a code position: the one byte code order in decimal or as 'x' (x cannot be the space), the name and the Unicode character
* used to access the glyph. The space must be assigned to character position 32 otherwise
* text justification will not work.
*
* Example for a "simple" encoding that includes the Unicode
* character space, A, B and ecyrillic:
*
* "# simple 32 0020 0041 0042 0454"
*
*
* Example for a "full" encoding for a Type1 Tex font:
*
* "# full 'A' nottriangeqlleft 0041 'B' dividemultiply 0042 32 space 0020"
*
* @param name the name of the font or its location on file
* @param encoding the encoding to be applied to this font
* @param embedded true if the font is to be embedded in the PDF
* @param cached true if the font comes from the cache or is added to
* the cache if new, false if the font is always created new
* @param ttfAfm the true type font or the afm in a byte array
* @param pfb the pfb in a byte array
* @param noThrow if true will not throw an exception if the font is not recognized and will return null, if false will throw
* an exception if the font is not recognized. Note that even if true an exception may be thrown in some circumstances.
* This parameter is useful for FontFactory that may have to check many invalid font names before finding the right one
* @return returns a new font. This font may come from the cache but only if cached
* is true, otherwise it will always be created new
* @throws DocumentException the font is invalid
* @throws IOException the font file could not be read
* @since 2.0.3
*/
public static BaseFont createFont(final String name, final String encoding, final boolean embedded, final boolean cached, final byte ttfAfm[], final byte pfb[], final boolean noThrow) throws DocumentException, IOException {
return createFont(name, encoding, embedded, cached, ttfAfm, pfb, false, false);
}
/** Creates a new font. This font can be one of the 14 built in types,
* a Type1 font referred to by an AFM or PFM file, a TrueType font (simple or collection) or a CJK font from the
* Adobe Asian Font Pack. TrueType fonts and CJK fonts can have an optional style modifier
* appended to the name. These modifiers are: Bold, Italic and BoldItalic. An
* example would be "STSong-Light,Bold". Note that this modifiers do not work if
* the font is embedded. Fonts in TrueType collections are addressed by index such as "msgothic.ttc,1".
* This would get the second font (indexes start at 0), in this case "MS PGothic".
*
* The fonts may or may not be cached depending on the flag cached.
* If the byte arrays are present the font will be
* read from them instead of the name. A name is still required to identify
* the font type.
*
* Besides the common encodings described by name, custom encodings
* can also be made. These encodings will only work for the single byte fonts
* Type1 and TrueType. The encoding string starts with a '#'
* followed by "simple" or "full". If "simple" there is a decimal for the first character position and then a list
* of hex values representing the Unicode codes that compose that encoding.
* The "simple" encoding is recommended for TrueType fonts
* as the "full" encoding risks not matching the character with the right glyph
* if not done with care.
* The "full" encoding is specially aimed at Type1 fonts where the glyphs have to be
* described by non standard names like the Tex math fonts. Each group of three elements
* compose a code position: the one byte code order in decimal or as 'x' (x cannot be the space), the name and the Unicode character
* used to access the glyph. The space must be assigned to character position 32 otherwise
* text justification will not work.
*
* Example for a "simple" encoding that includes the Unicode
* character space, A, B and ecyrillic:
*
* "# simple 32 0020 0041 0042 0454"
*
*
* Example for a "full" encoding for a Type1 Tex font:
*
* "# full 'A' nottriangeqlleft 0041 'B' dividemultiply 0042 32 space 0020"
*
* @param name the name of the font or its location on file
* @param encoding the encoding to be applied to this font
* @param embedded true if the font is to be embedded in the PDF
* @param cached true if the font comes from the cache or is added to
* the cache if new, false if the font is always created new
* @param ttfAfm the true type font or the afm in a byte array
* @param pfb the pfb in a byte array
* @param noThrow if true will not throw an exception if the font is not recognized and will return null, if false will throw
* an exception if the font is not recognized. Note that even if true an exception may be thrown in some circumstances.
* This parameter is useful for FontFactory that may have to check many invalid font names before finding the right one
* @param forceRead in some cases (TrueTypeFont, Type1Font), the full font file will be read and kept in memory if forceRead is true
* @return returns a new font. This font may come from the cache but only if cached
* is true, otherwise it will always be created new
* @throws DocumentException the font is invalid
* @throws IOException the font file could not be read
* @since 2.1.5
*/
public static BaseFont createFont(final String name, String encoding, boolean embedded, final boolean cached, final byte ttfAfm[], final byte pfb[], final boolean noThrow, final boolean forceRead) throws DocumentException, IOException {
final String nameBase = getBaseName(name);
encoding = normalizeEncoding(encoding);
final boolean isBuiltinFonts14 = BuiltinFonts14.containsKey(name);
final boolean isCJKFont = isBuiltinFonts14 ? false : CJKFont.isCJKFont(nameBase, encoding);
if (isBuiltinFonts14 || isCJKFont) {
embedded = false;
} else if (encoding.equals(IDENTITY_H) || encoding.equals(IDENTITY_V)) {
embedded = true;
}
BaseFont fontFound = null;
BaseFont fontBuilt = null;
final String key = name + "\n" + encoding + "\n" + embedded;
if (cached) {
synchronized (fontCache) {
fontFound = (BaseFont)fontCache.get(key);
}
if (fontFound != null) {
return fontFound;
}
}
if (isBuiltinFonts14 || name.toLowerCase().endsWith(".afm") || name.toLowerCase().endsWith(".pfm")) {
fontBuilt = new Type1Font(name, encoding, embedded, ttfAfm, pfb, forceRead);
fontBuilt.fastWinansi = encoding.equals(CP1252);
}
else if (nameBase.toLowerCase().endsWith(".ttf") || nameBase.toLowerCase().endsWith(".otf") || nameBase.toLowerCase().indexOf(".ttc,") > 0) {
if (encoding.equals(IDENTITY_H) || encoding.equals(IDENTITY_V)) {
fontBuilt = new TrueTypeFontUnicode(name, encoding, embedded, ttfAfm, forceRead);
} else {
fontBuilt = new TrueTypeFont(name, encoding, embedded, ttfAfm, false, forceRead);
fontBuilt.fastWinansi = encoding.equals(CP1252);
}
}
else if (isCJKFont) {
fontBuilt = new CJKFont(name, encoding, embedded);
} else if (noThrow) {
return null;
} else {
throw new DocumentException("Font '" + name + "' with '" + encoding + "' is not recognized.");
}
if (cached) {
synchronized (fontCache) {
fontFound = (BaseFont)fontCache.get(key);
if (fontFound != null) {
return fontFound;
}
fontCache.put(key, fontBuilt);
}
}
return fontBuilt;
}
/**
* Creates a font based on an existing document font. The created font font may not
* behave as expected, depending on the encoding or subset.
* @param fontRef the reference to the document font
* @return the font
*/
public static BaseFont createFont(final PRIndirectReference fontRef) {
return new DocumentFont(fontRef);
}
/**
* Gets the name without the modifiers Bold, Italic or BoldItalic.
* @param name the full name of the font
* @return the name without the modifiers Bold, Italic or BoldItalic
*/
protected static String getBaseName(final String name) {
if (name.endsWith(",Bold")) {
return name.substring(0, name.length() - 5);
} else if (name.endsWith(",Italic")) {
return name.substring(0, name.length() - 7);
} else if (name.endsWith(",BoldItalic")) {
return name.substring(0, name.length() - 11);
} else {
return name;
}
}
/**
* Normalize the encoding names. "winansi" is changed to "Cp1252" and
* "macroman" is changed to "MacRoman".
* @param enc the encoding to be normalized
* @return the normalized encoding
*/
protected static String normalizeEncoding(final String enc) {
if (enc.equals("winansi") || enc.equals("")) {
return CP1252;
} else if (enc.equals("macroman")) {
return MACROMAN;
} else {
return enc;
}
}
/**
* Creates the widths and the differences arrays
*/
protected void createEncoding() {
if (this.encoding.startsWith("#")) {
this.specialMap = new IntHashtable();
final StringTokenizer tok = new StringTokenizer(this.encoding.substring(1), " ,\t\n\r\f");
if (tok.nextToken().equals("full")) {
while (tok.hasMoreTokens()) {
final String order = tok.nextToken();
final String name = tok.nextToken();
final char uni = (char)Integer.parseInt(tok.nextToken(), 16);
int orderK;
if (order.startsWith("'")) {
orderK = order.charAt(1);
} else {
orderK = Integer.parseInt(order);
}
orderK %= 256;
this.specialMap.put(uni, orderK);
this.differences[orderK] = name;
this.unicodeDifferences[orderK] = uni;
this.widths[orderK] = getRawWidth(uni, name);
this.charBBoxes[orderK] = getRawCharBBox(uni, name);
}
}
else {
int k = 0;
if (tok.hasMoreTokens()) {
k = Integer.parseInt(tok.nextToken());
}
while (tok.hasMoreTokens() && k < 256) {
final String hex = tok.nextToken();
final int uni = Integer.parseInt(hex, 16) % 0x10000;
final String name = GlyphList.unicodeToName(uni);
if (name != null) {
this.specialMap.put(uni, k);
this.differences[k] = name;
this.unicodeDifferences[k] = (char)uni;
this.widths[k] = getRawWidth(uni, name);
this.charBBoxes[k] = getRawCharBBox(uni, name);
++k;
}
}
}
for (int k = 0; k < 256; ++k) {
if (this.differences[k] == null) {
this.differences[k] = notdef;
}
}
}
else if (this.fontSpecific) {
for (int k = 0; k < 256; ++k) {
this.widths[k] = getRawWidth(k, null);
this.charBBoxes[k] = getRawCharBBox(k, null);
}
}
else {
String s;
String name;
char c;
final byte b[] = new byte[1];
for (int k = 0; k < 256; ++k) {
b[0] = (byte)k;
s = PdfEncodings.convertToString(b, this.encoding);
if (s.length() > 0) {
c = s.charAt(0);
}
else {
c = '?';
}
name = GlyphList.unicodeToName(c);
if (name == null) {
name = notdef;
}
this.differences[k] = name;
this.unicodeDifferences[k] = c;
this.widths[k] = getRawWidth(c, name);
this.charBBoxes[k] = getRawCharBBox(c, name);
}
}
}
/**
* Gets the width from the font according to the Unicode char c
* or the name. If the name is null it's a symbolic font.
* @param c the unicode char
* @param name the glyph name
* @return the width of the char
*/
abstract int getRawWidth(int c, String name);
/**
* Gets the kerning between two Unicode chars.
* @param char1 the first char
* @param char2 the second char
* @return the kerning to be applied in normalized 1000 units
*/
public abstract int getKerning(int char1, int char2);
/**
* Sets the kerning between two Unicode chars.
* @param char1 the first char
* @param char2 the second char
* @param kern the kerning to apply in normalized 1000 units
* @return true if the kerning was applied, false otherwise
*/
public abstract boolean setKerning(int char1, int char2, int kern);
/**
* Gets the width of a char in normalized 1000 units.
* @param char1 the unicode char to get the width of
* @return the width in normalized 1000 units
*/
public int getWidth(final int char1) {
if (this.fastWinansi) {
if (char1 < 128 || char1 >= 160 && char1 <= 255) {
return this.widths[char1];
} else {
return this.widths[PdfEncodings.winansi.get(char1)];
}
}
else {
int total = 0;
final byte mbytes[] = convertToBytes((char)char1);
for (final byte mbyte : mbytes) {
total += this.widths[0xff & mbyte];
}
return total;
}
}
/**
* Gets the width of a String in normalized 1000 units.
* @param text the String to get the width of
* @return the width in normalized 1000 units
*/
public int getWidth(final String text) {
int total = 0;
if (this.fastWinansi) {
final int len = text.length();
for (int k = 0; k < len; ++k) {
final char char1 = text.charAt(k);
if (char1 < 128 || char1 >= 160 && char1 <= 255) {
total += this.widths[char1];
} else {
total += this.widths[PdfEncodings.winansi.get(char1)];
}
}
return total;
}
else {
final byte mbytes[] = convertToBytes(text);
for (final byte mbyte : mbytes) {
total += this.widths[0xff & mbyte];
}
}
return total;
}
/**
* Gets the descent of a String in normalized 1000 units. The descent will always be
* less than or equal to zero even if all the characters have an higher descent.
* @param text the String to get the descent of
* @return the descent in normalized 1000 units
*/
public int getDescent(final String text) {
int min = 0;
final char chars[] = text.toCharArray();
for (final char c : chars) {
final int bbox[] = getCharBBox(c);
if (bbox != null && bbox[1] < min) {
min = bbox[1];
}
}
return min;
}
/**
* Gets the ascent of a String in normalized 1000 units. The ascent will always be
* greater than or equal to zero even if all the characters have a lower ascent.
* @param text the String to get the ascent of
* @return the ascent in normalized 1000 units
*/
public int getAscent(final String text) {
int max = 0;
final char chars[] = text.toCharArray();
for (final char c : chars) {
final int bbox[] = getCharBBox(c);
if (bbox != null && bbox[3] > max) {
max = bbox[3];
}
}
return max;
}
/**
* Gets the descent of a String in points. The descent will always be
* less than or equal to zero even if all the characters have an higher descent.
* @param text the String to get the descent of
* @param fontSize the size of the font
* @return the descent in points
*/
public float getDescentPoint(final String text, final float fontSize)
{
return getDescent(text) * 0.001f * fontSize;
}
/**
* Gets the ascent of a String in points. The ascent will always be
* greater than or equal to zero even if all the characters have a lower ascent.
* @param text the String to get the ascent of
* @param fontSize the size of the font
* @return the ascent in points
*/
public float getAscentPoint(final String text, final float fontSize)
{
return getAscent(text) * 0.001f * fontSize;
}
// ia>
/**
* Gets the width of a String in points taking kerning
* into account.
* @param text the String to get the width of
* @param fontSize the font size
* @return the width in points
*/
public float getWidthPointKerned(final String text, final float fontSize) {
final float size = getWidth(text) * 0.001f * fontSize;
if (!hasKernPairs()) {
return size;
}
final int len = text.length() - 1;
int kern = 0;
final char c[] = text.toCharArray();
for (int k = 0; k < len; ++k) {
kern += getKerning(c[k], c[k + 1]);
}
return size + kern * 0.001f * fontSize;
}
/**
* Gets the width of a String in points.
* @param text the String to get the width of
* @param fontSize the font size
* @return the width in points
*/
public float getWidthPoint(final String text, final float fontSize) {
return getWidth(text) * 0.001f * fontSize;
}
/**
* Gets the width of a char in points.
* @param char1 the char to get the width of
* @param fontSize the font size
* @return the width in points
*/
public float getWidthPoint(final int char1, final float fontSize) {
return getWidth(char1) * 0.001f * fontSize;
}
/**
* Converts a String to a byte array according
* to the font's encoding.
* @param text the String to be converted
* @return an array of byte representing the conversion according to the font's encoding
*/
byte[] convertToBytes(final String text) {
if (this.directTextToByte) {
return PdfEncodings.convertToBytes(text, null);
}
if (this.specialMap != null) {
final byte[] b = new byte[text.length()];
int ptr = 0;
final int length = text.length();
for (int k = 0; k < length; ++k) {
final char c = text.charAt(k);
if (this.specialMap.containsKey(c)) {
b[ptr++] = (byte)this.specialMap.get(c);
}
}
if (ptr < length) {
final byte[] b2 = new byte[ptr];
System.arraycopy(b, 0, b2, 0, ptr);
return b2;
} else {
return b;
}
}
return PdfEncodings.convertToBytes(text, this.encoding);
}
/**
* Converts a char to a byte array according
* to the font's encoding.
* @param char1 the char to be converted
* @return an array of byte representing the conversion according to the font's encoding
*/
byte[] convertToBytes(final int char1) {
if (this.directTextToByte) {
return PdfEncodings.convertToBytes((char)char1, null);
}
if (this.specialMap != null) {
if (this.specialMap.containsKey(char1)) {
return new byte[]{(byte)this.specialMap.get(char1)};
} else {
return new byte[0];
}
}
return PdfEncodings.convertToBytes((char)char1, this.encoding);
}
/** Outputs to the writer the font dictionaries and streams.
* @param writer the writer for this document
* @param ref the font indirect reference
* @param params several parameters that depend on the font type
* @throws IOException on error
* @throws DocumentException error in generating the object
*/
abstract void writeFont(PdfWriter writer, PdfIndirectReference ref, Object params[]) throws DocumentException, IOException;
/**
* Returns a PdfStream object with the full font program (if possible).
* This method will return null for some types of fonts (CJKFont, Type3Font)
* or if there is no font program available (standard Type 1 fonts).
* @return a PdfStream with the font program
* @since 2.1.3
* @throws IOException Exception of a I/O
* @throws DocumentException Exception of a document
*/
abstract PdfStream getFullFontStream() throws IOException, DocumentException;
/** Gets the encoding used to convert String into byte[].
* @return the encoding name
*/
public String getEncoding() {
return this.encoding;
}
/** Gets the font parameter identified by key. Valid values
* for key are ASCENT, AWT_ASCENT, CAPHEIGHT,
* DESCENT, AWT_DESCENT,
* ITALICANGLE, BBOXLLX, BBOXLLY, BBOXURX
* and BBOXURY.
* @param key the parameter to be extracted
* @param fontSize the font size in points
* @return the parameter in points
*/
public abstract float getFontDescriptor(int key, float fontSize);
/** Gets the font type. The font types can be: FONT_TYPE_T1,
* FONT_TYPE_TT, FONT_TYPE_CJK and FONT_TYPE_TTUNI.
* @return the font type
*/
public int getFontType() {
return this.fontType;
}
/** Gets the embedded flag.
* @return true if the font is embedded.
*/
public boolean isEmbedded() {
return this.embedded;
}
/** Gets the symbolic flag of the font.
* @return true if the font is symbolic
*/
public boolean isFontSpecific() {
return this.fontSpecific;
}
/** Creates a unique subset prefix to be added to the font name when the font is embedded and subset.
* @return the subset prefix
*/
public static String createSubsetPrefix() {
String s = "";
for (int k = 0; k < 6; ++k) {
s += (char)(Math.random() * 26 + 'A');
}
return s + "+";
}
/** Gets the Unicode character corresponding to the byte output to the pdf stream.
* @param index the byte index
* @return the Unicode character
*/
char getUnicodeDifferences(final int index) {
return this.unicodeDifferences[index];
}
/** Gets the postscript font name.
* @return the postscript font name
*/
public abstract String getPostscriptFontName();
/**
* Sets the font name that will appear in the pdf font dictionary.
* Use with care as it can easily make a font unreadable if not embedded.
* @param name the new font name
*/
public abstract void setPostscriptFontName(String name);
/** Gets the full name of the font. If it is a True Type font
* each array element will have {Platform ID, Platform Encoding ID,
* Language ID, font name}. The interpretation of this values can be
* found in the Open Type specification, chapter 2, in the 'name' table.
* For the other fonts the array has a single element with {"", "", "",
* font name}.
* @return the full name of the font
*/
public abstract String[][] getFullFontName();
/** Gets all the entries of the names-table. If it is a True Type font
* each array element will have {Name ID, Platform ID, Platform Encoding ID,
* Language ID, font name}. The interpretation of this values can be
* found in the Open Type specification, chapter 2, in the 'name' table.
* For the other fonts the array has a single element with {"4", "", "", "",
* font name}.
* @return the full name of the font
* @since 2.0.8
*/
public abstract String[][] getAllNameEntries();
/** Gets the full name of the font. If it is a True Type font
* each array element will have {Platform ID, Platform Encoding ID,
* Language ID, font name}. The interpretation of this values can be
* found in the Open Type specification, chapter 2, in the 'name' table.
* For the other fonts the array has a single element with {"", "", "",
* font name}.
* @param name the name of the font
* @param encoding the encoding of the font
* @param ttfAfm the true type font or the afm in a byte array
* @throws DocumentException on error
* @throws IOException on error
* @return the full name of the font
*/
public static String[][] getFullFontName(final String name, final String encoding, final byte ttfAfm[]) throws DocumentException, IOException {
final String nameBase = getBaseName(name);
BaseFont fontBuilt = null;
if (nameBase.toLowerCase().endsWith(".ttf") || nameBase.toLowerCase().endsWith(".otf") || nameBase.toLowerCase().indexOf(".ttc,") > 0) {
fontBuilt = new TrueTypeFont(name, CP1252, false, ttfAfm, true, false);
} else {
fontBuilt = createFont(name, encoding, false, false, ttfAfm, null);
}
return fontBuilt.getFullFontName();
}
/** Gets all the names from the font. Only the required tables are read.
* @param name the name of the font
* @param encoding the encoding of the font
* @param ttfAfm the true type font or the afm in a byte array
* @throws DocumentException on error
* @throws IOException on error
* @return an array of Object[] built with {getPostscriptFontName(), getFamilyFontName(), getFullFontName()}
*/
public static Object[] getAllFontNames(final String name, final String encoding, final byte ttfAfm[]) throws DocumentException, IOException {
final String nameBase = getBaseName(name);
BaseFont fontBuilt = null;
if (nameBase.toLowerCase().endsWith(".ttf") || nameBase.toLowerCase().endsWith(".otf") || nameBase.toLowerCase().indexOf(".ttc,") > 0) {
fontBuilt = new TrueTypeFont(name, CP1252, false, ttfAfm, true, false);
} else {
fontBuilt = createFont(name, encoding, false, false, ttfAfm, null);
}
return new Object[]{fontBuilt.getPostscriptFontName(), fontBuilt.getFamilyFontName(), fontBuilt.getFullFontName()};
}
/** Gets all the entries of the namestable from the font. Only the required tables are read.
* @param name the name of the font
* @param encoding the encoding of the font
* @param ttfAfm the true type font or the afm in a byte array
* @throws DocumentException on error
* @throws IOException on error
* @return an array of Object[] built with {getPostscriptFontName(), getFamilyFontName(), getFullFontName()}
* @since 2.0.8
*/
public static String[][] getAllNameEntries(final String name, final String encoding, final byte ttfAfm[]) throws DocumentException, IOException {
final String nameBase = getBaseName(name);
BaseFont fontBuilt = null;
if (nameBase.toLowerCase().endsWith(".ttf") || nameBase.toLowerCase().endsWith(".otf") || nameBase.toLowerCase().indexOf(".ttc,") > 0) {
fontBuilt = new TrueTypeFont(name, CP1252, false, ttfAfm, true, false);
} else {
fontBuilt = createFont(name, encoding, false, false, ttfAfm, null);
}
return fontBuilt.getAllNameEntries();
}
/** Gets the family name of the font. If it is a True Type font
* each array element will have {Platform ID, Platform Encoding ID,
* Language ID, font name}. The interpretation of this values can be
* found in the Open Type specification, chapter 2, in the 'name' table.
* For the other fonts the array has a single element with {"", "", "",
* font name}.
* @return the family name of the font
*/
public abstract String[][] getFamilyFontName();
/** Gets the code pages supported by the font. This has only meaning
* with True Type fonts.
* @return the code pages supported by the font
*/
public String[] getCodePagesSupported() {
return new String[0];
}
/** Enumerates the postscript font names present inside a
* True Type Collection.
* @param ttcFile the file name of the font
* @throws DocumentException on error
* @throws IOException on error
* @return the postscript font names
*/
public static String[] enumerateTTCNames(final String ttcFile) throws DocumentException, IOException {
return new EnumerateTTC(ttcFile).getNames();
}
/** Enumerates the postscript font names present inside a
* True Type Collection.
* @param ttcArray the font as a byte array
* @throws DocumentException on error
* @throws IOException on error
* @return the postscript font names
*/
public static String[] enumerateTTCNames(final byte ttcArray[]) throws DocumentException, IOException {
return new EnumerateTTC(ttcArray).getNames();
}
/** Gets the font width array.
* @return the font width array
*/
public int[] getWidths() {
return this.widths;
}
/** Gets the array with the names of the characters.
* @return the array with the names of the characters
*/
public String[] getDifferences() {
return this.differences;
}
/** Gets the array with the unicode characters.
* @return the array with the unicode characters
*/
public char[] getUnicodeDifferences() {
return this.unicodeDifferences;
}
/** Gets the state of the property.
* @return value of property forceWidthsOutput
*/
public boolean isForceWidthsOutput() {
return this.forceWidthsOutput;
}
/** Set to true to force the generation of the
* widths array.
* @param forceWidthsOutput true to force the generation of the
* widths array
*/
public void setForceWidthsOutput(final boolean forceWidthsOutput) {
this.forceWidthsOutput = forceWidthsOutput;
}
/** Gets the direct conversion of char to byte.
* @return value of property directTextToByte.
* @see #setDirectTextToByte(boolean directTextToByte)
*/
public boolean isDirectTextToByte() {
return this.directTextToByte;
}
/** Sets the conversion of char directly to byte
* by casting. This is a low level feature to put the bytes directly in
* the content stream without passing through String.getBytes().
* @param directTextToByte New value of property directTextToByte.
*/
public void setDirectTextToByte(final boolean directTextToByte) {
this.directTextToByte = directTextToByte;
}
/** Indicates if all the glyphs and widths for that particular
* encoding should be included in the document.
* @return false to include all the glyphs and widths.
*/
public boolean isSubset() {
return this.subset;
}
/** Indicates if all the glyphs and widths for that particular
* encoding should be included in the document. When set to true
* only the glyphs used will be included in the font. When set to false
* and {@link #addSubsetRange(int[])} was not called the full font will be included
* otherwise just the characters ranges will be included.
* @param subset new value of property subset
*/
public void setSubset(final boolean subset) {
this.subset = subset;
}
/** Gets the font resources.
* @param key the full name of the resource
* @return the InputStream to get the resource or
* null if not found
*/
public static InputStream getResourceStream(final String key) {
return getResourceStream(key, null);
}
/** Gets the font resources.
* @param key the full name of the resource
* @param loader the ClassLoader to load the resource or null to try the ones available
* @return the InputStream to get the resource or
* null if not found
*/
public static InputStream getResourceStream(String key, final ClassLoader loader) {
if (key.startsWith("/")) {
key = key.substring(1);
}
InputStream is = null;
if (loader != null) {
is = loader.getResourceAsStream(key);
if (is != null) {
return is;
}
}
// Try to use Context Class Loader to load the properties file.
try {
final ClassLoader contextClassLoader = Thread.currentThread().getContextClassLoader();
if (contextClassLoader != null) {
is = contextClassLoader.getResourceAsStream(key);
}
} catch (final Throwable e) {}
if (is == null) {
is = BaseFont.class.getResourceAsStream("/" + key);
}
if (is == null) {
is = ClassLoader.getSystemResourceAsStream(key);
}
return is;
}
/** Gets the Unicode equivalent to a CID.
* The (inexistent) CID <FF00> is translated as '\n'.
* It has only meaning with CJK fonts with Identity encoding.
* @param c the CID code
* @return the Unicode equivalent
*/
public int getUnicodeEquivalent(final int c) {
return c;
}
/** Gets the CID code given an Unicode.
* It has only meaning with CJK fonts.
* @param c the Unicode
* @return the CID equivalent
*/
public int getCidCode(final int c) {
return c;
}
/** Checks if the font has any kerning pairs.
* @return true if the font has any kerning pairs
*/
public abstract boolean hasKernPairs();
/**
* Checks if a character exists in this font.
* @param c the character to check
* @return true if the character has a glyph,
* false otherwise
*/
public boolean charExists(final int c) {
final byte b[] = convertToBytes(c);
return b.length > 0;
}
/**
* Sets the character advance.
* @param c the character
* @param advance the character advance normalized to 1000 units
* @return true if the advance was set,
* false otherwise
*/
public boolean setCharAdvance(final int c, final int advance) {
final byte b[] = convertToBytes(c);
if (b.length == 0) {
return false;
}
this.widths[0xff & b[0]] = advance;
return true;
}
private static void addFont(final PRIndirectReference fontRef, final IntHashtable hits, final ArrayList fonts) {
final PdfObject obj = PdfReader.getPdfObject(fontRef);
if (obj == null || !obj.isDictionary()) {
return;
}
final PdfDictionary font = (PdfDictionary)obj;
final PdfName subtype = font.getAsName(PdfName.SUBTYPE);
if (!PdfName.TYPE1.equals(subtype) && !PdfName.TRUETYPE.equals(subtype)) {
return;
}
final PdfName name = font.getAsName(PdfName.BASEFONT);
fonts.add(new Object[]{PdfName.decodeName(name.toString()), fontRef});
hits.put(fontRef.getNumber(), 1);
}
private static void recourseFonts(final PdfDictionary page, final IntHashtable hits, final ArrayList fonts, int level) {
++level;
if (level > 50) {
return;
}
final PdfDictionary resources = page.getAsDict(PdfName.RESOURCES);
if (resources == null) {
return;
}
final PdfDictionary font = resources.getAsDict(PdfName.FONT);
if (font != null) {
for (final Object element : font.getKeys()) {
final PdfObject ft = font.get((PdfName)element);
if (ft == null || !ft.isIndirect()) {
continue;
}
final int hit = ((PRIndirectReference)ft).getNumber();
if (hits.containsKey(hit)) {
continue;
}
addFont((PRIndirectReference)ft, hits, fonts);
}
}
final PdfDictionary xobj = resources.getAsDict(PdfName.XOBJECT);
if (xobj != null) {
for (final Object element : xobj.getKeys()) {
recourseFonts(xobj.getAsDict((PdfName)element), hits, fonts, level);
}
}
}
/**
* Gets a list of all document fonts. Each element of the ArrayList
* contains a Object[]{String,PRIndirectReference} with the font name
* and the indirect reference to it.
* @param reader the document where the fonts are to be listed from
* @return the list of fonts and references
*/
public static ArrayList getDocumentFonts(final PdfReader reader) {
final IntHashtable hits = new IntHashtable();
final ArrayList fonts = new ArrayList();
final int npages = reader.getNumberOfPages();
for (int k = 1; k <= npages; ++k) {
recourseFonts(reader.getPageN(k), hits, fonts, 1);
}
return fonts;
}
/**
* Gets a list of the document fonts in a particular page. Each element of the ArrayList
* contains a Object[]{String,PRIndirectReference} with the font name
* and the indirect reference to it.
* @param reader the document where the fonts are to be listed from
* @param page the page to list the fonts from
* @return the list of fonts and references
*/
public static ArrayList getDocumentFonts(final PdfReader reader, final int page) {
final IntHashtable hits = new IntHashtable();
final ArrayList fonts = new ArrayList();
recourseFonts(reader.getPageN(page), hits, fonts, 1);
return fonts;
}
/**
* Gets the smallest box enclosing the character contours. It will return
* null if the font has not the information or the character has no
* contours, as in the case of the space, for example. Characters with no contours may
* also return [0,0,0,0].
* @param c the character to get the contour bounding box from
* @return an array of four floats with the bounding box in the format [llx,lly,urx,ury] or
* null
*/
public int[] getCharBBox(final int c) {
final byte b[] = convertToBytes(c);
if (b.length == 0) {
return null;
} else {
return this.charBBoxes[b[0] & 0xff];
}
}
protected abstract int[] getRawCharBBox(int c, String name);
/**
* iText expects Arabic Diactrics (tashkeel) to have zero advance but some fonts,
* most notably those that come with Windows, like times.ttf, have non-zero
* advance for those characters. This method makes those character to have zero
* width advance and work correctly in the iText Arabic shaping and reordering
* context.
*/
public void correctArabicAdvance() {
for (char c = '\u064b'; c <= '\u0658'; ++c) {
setCharAdvance(c, 0);
}
setCharAdvance('\u0670', 0);
for (char c = '\u06d6'; c <= '\u06dc'; ++c) {
setCharAdvance(c, 0);
}
for (char c = '\u06df'; c <= '\u06e4'; ++c) {
setCharAdvance(c, 0);
}
for (char c = '\u06e7'; c <= '\u06e8'; ++c) {
setCharAdvance(c, 0);
}
for (char c = '\u06ea'; c <= '\u06ed'; ++c) {
setCharAdvance(c, 0);
}
}
/**
* Adds a character range when subsetting. The range is an int array
* where the first element is the start range inclusive and the second element is the
* end range inclusive. Several ranges are allowed in the same array.
* @param range the character range
*/
public void addSubsetRange(final int[] range) {
if (this.subsetRanges == null) {
this.subsetRanges = new ArrayList();
}
this.subsetRanges.add(range);
}
/**
* Returns the compression level used for the font streams.
* @return the compression level (0 = best speed, 9 = best compression, -1 is default)
* @since 2.1.3
*/
public int getCompressionLevel() {
return this.compressionLevel;
}
/**
* Sets the compression level to be used for the font streams.
* @param compressionLevel a value between 0 (best speed) and 9 (best compression)
* @since 2.1.3
*/
public void setCompressionLevel(final int compressionLevel) {
if (compressionLevel < PdfStream.NO_COMPRESSION || compressionLevel > PdfStream.BEST_COMPRESSION) {
this.compressionLevel = PdfStream.DEFAULT_COMPRESSION;
} else {
this.compressionLevel = compressionLevel;
}
}
}