From 1e49e903b786f4bd8a8ae5b254faad4e477cf6ac Mon Sep 17 00:00:00 2001 From: Cornelis Hoeflake Date: Fri, 31 Oct 2014 10:48:14 +0100 Subject: [PATCH] separates font substitute logic from ExternalFonts PDFBOX-2467 --- .../apache/pdfbox/pdmodel/font/ExternalFonts.java | 214 ++++----------------- .../pdmodel/font/SubstituteFontProvider.java | 201 +++++++++++++++++++ 2 files changed, 238 insertions(+), 177 deletions(-) create mode 100644 pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/SubstituteFontProvider.java diff --git a/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/ExternalFonts.java b/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/ExternalFonts.java index b04c1ca..8b0f49d 100644 --- a/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/ExternalFonts.java +++ b/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/ExternalFonts.java @@ -19,9 +19,7 @@ package org.apache.pdfbox.pdmodel.font; import java.io.IOException; import java.io.InputStream; import java.net.URL; -import java.util.ArrayList; import java.util.Arrays; -import java.util.Collections; import java.util.HashMap; import java.util.List; import java.util.Map; @@ -33,8 +31,8 @@ import org.apache.fontbox.cff.CFFFont; import org.apache.fontbox.cff.CFFParser; import org.apache.fontbox.cff.CFFType1Font; import org.apache.fontbox.ttf.TTFParser; -import org.apache.fontbox.ttf.Type1Equivalent; import org.apache.fontbox.ttf.TrueTypeFont; +import org.apache.fontbox.ttf.Type1Equivalent; import org.apache.fontbox.type1.Type1Font; import org.apache.pdfbox.io.IOUtils; @@ -59,8 +57,15 @@ public final class ExternalFonts /** fallback fonts, used as as a last resort */ private static final TrueTypeFont ttfFallbackFont; private static final CFFCIDFont cidFallbackFont; + private final static Map> registrySubstitutes = new HashMap>(); static { + // extra substitute mechanism for CJK CIDFonts when all we know is the ROS + registrySubstitutes.put("$Adobe-CNS1", Arrays.asList("AdobeMingStd-Light")); + registrySubstitutes.put("$Adobe-Japan1", Arrays.asList("KozMinPr6N-Regular")); + registrySubstitutes.put("$Adobe-Korea1", Arrays.asList("AdobeGothicStd-Bold")); + registrySubstitutes.put("$Adobe-GB1", Arrays.asList("AdobeHeitiStd-Regular")); + try { // ttf @@ -97,7 +102,19 @@ public final class ExternalFonts */ public static void setProvider(FontProvider fontProvider) { - ExternalFonts.fontProvider = fontProvider; + setProvider(fontProvider, true); + } + + /** + * Sets the font service provider. + */ + public static void setProvider(FontProvider fontProvider, boolean withSubstitutes) + { + if (withSubstitutes && !(fontProvider instanceof SubstituteFontProvider)) { + ExternalFonts.fontProvider = new SubstituteFontProvider(fontProvider); + } else { + ExternalFonts.fontProvider = fontProvider; + } } /** @@ -112,117 +129,6 @@ public final class ExternalFonts return fontProvider; } - /** Map of PostScript name substitutes, in priority order. */ - private final static Map> substitutes = new HashMap>(); - static - { - // substitutes for standard 14 fonts - substitutes.put("Courier", - Arrays.asList("CourierNew", "CourierNewPSMT", "LiberationMono", "NimbusMonL-Regu")); - substitutes.put("Courier-Bold", - Arrays.asList("CourierNewPS-BoldMT", "CourierNew-Bold", "LiberationMono-Bold", - "NimbusMonL-Bold")); - substitutes.put("Courier-Oblique", - Arrays.asList("CourierNewPS-ItalicMT","CourierNew-Italic", - "LiberationMono-Italic", "NimbusMonL-ReguObli")); - substitutes.put("Courier-BoldOblique", - Arrays.asList("CourierNewPS-BoldItalicMT","CourierNew-BoldItalic", - "LiberationMono-BoldItalic", "NimbusMonL-BoldObli")); - substitutes.put("Helvetica", - Arrays.asList("ArialMT", "Arial", "LiberationSans", "NimbusSanL-Regu")); - substitutes.put("Helvetica-Bold", - Arrays.asList("Arial-BoldMT", "Arial-Bold", "LiberationSans-Bold", - "NimbusSanL-Bold")); - substitutes.put("Helvetica-Oblique", - Arrays.asList("Arial-ItalicMT", "Arial-Italic", "Helvetica-Italic", - "LiberationSans-Italic", "NimbusSanL-ReguItal")); - substitutes.put("Helvetica-BoldOblique", - Arrays.asList("Arial-BoldItalicMT", "Helvetica-BoldItalic", - "LiberationSans-BoldItalic", "NimbusSanL-BoldItal")); - substitutes.put("Times-Roman", - Arrays.asList("TimesNewRomanPSMT", "TimesNewRoman", "TimesNewRomanPS", - "LiberationSerif", "NimbusRomNo9L-Regu")); - substitutes.put("Times-Bold", - Arrays.asList("TimesNewRomanPS-BoldMT", "TimesNewRomanPS-Bold", - "TimesNewRoman-Bold", "LiberationSerif-Bold", - "NimbusRomNo9L-Medi")); - substitutes.put("Times-Italic", - Arrays.asList("TimesNewRomanPS-ItalicMT", "TimesNewRomanPS-Italic", - "TimesNewRoman-Italic", "LiberationSerif-Italic", - "NimbusRomNo9L-ReguItal")); - substitutes.put("Times-BoldItalic", - Arrays.asList("TimesNewRomanPS-BoldItalicMT", "TimesNewRomanPS-BoldItalic", - "TimesNewRoman-BoldItalic", "LiberationSerif-BoldItalic", - "NimbusRomNo9L-MediItal")); - substitutes.put("Symbol", Arrays.asList("SymbolMT", "StandardSymL")); - substitutes.put("ZapfDingbats", Arrays.asList("ZapfDingbatsITC", "Dingbats")); - - // extra substitute mechanism for CJK CIDFonts when all we know is the ROS - substitutes.put("$Adobe-CNS1", Arrays.asList("AdobeMingStd-Light")); - substitutes.put("$Adobe-Japan1", Arrays.asList("KozMinPr6N-Regular")); - substitutes.put("$Adobe-Korea1", Arrays.asList("AdobeGothicStd-Bold")); - substitutes.put("$Adobe-GB1", Arrays.asList("AdobeHeitiStd-Regular")); - - // Acrobat also uses alternative names for Standard 14 fonts, which we map to those above - // these include names such as "Arial" and "TimesNewRoman" - for (String baseName : Standard14Fonts.getNames()) - { - if (!substitutes.containsKey(baseName)) - { - String mappedName = Standard14Fonts.getMappedFontName(baseName); - substitutes.put(baseName, copySubstitutes(mappedName)); - } - } - } - - /** - * Copies a list of font substitutes, adding the original font at the start of the list. - */ - private static List copySubstitutes(String postScriptName) - { - List copy = new ArrayList(substitutes.get(postScriptName)); - copy.add(0, postScriptName); - return copy; - } - - /** - * Adds a top-priority substitute for the given font. - * - * @param match PostScript name of the font to match - * @param replace PostScript name of the font to use as a replacement - */ - public static void addSubstitute(String match, String replace) - { - if (!substitutes.containsKey(match)) - { - substitutes.put(match, new ArrayList()); - } - substitutes.get(match).add(replace); - } - - /** - * Returns the substitutes for a given font. - */ - private static List getSubstitutes(String postScriptName) - { - List subs = substitutes.get(postScriptName.replaceAll(" ", "")); - if (subs != null) - { - return subs; - } - else - { - return Collections.emptyList(); - } - } - - /** - * Windows name (ArialNarrow,Bold) to PostScript name (ArialNarrow-Bold) - */ - private static String windowsToPs(String windowsName) - { - return windowsName.replaceAll(",", "-"); - } /** * Finds a CFF CID-Keyed font with the given PostScript name, or a suitable substitute, or null. @@ -237,16 +143,19 @@ public final class ExternalFonts // todo: this is a fairly primitive mechanism and could be improved if (registryOrdering != null) { - for (String substituteName : getSubstitutes("$" + registryOrdering)) - { - CFFFont cff = getProvider().getCFFFont(substituteName); - if (cff != null) - { - if (cff instanceof CFFCIDFont) - { - return (CFFCIDFont)cff; - } - } + List list = registrySubstitutes.get("$" + registryOrdering); + if (list != null) { + for (String substituteName : list) + { + CFFFont cff = getProvider().getCFFFont(substituteName); + if (cff != null) + { + if (cff instanceof CFFCIDFont) + { + return (CFFCIDFont)cff; + } + } + } } } return cidFallbackFont; @@ -376,23 +285,7 @@ public final class ExternalFonts */ public static TrueTypeFont getTrueTypeFont(String postScriptName) { - // first ask the font provider for the font - TrueTypeFont ttf = getProvider().getTrueTypeFont(postScriptName); - if (ttf == null) - { - // then try substitutes - for (String substituteName : getSubstitutes(postScriptName)) - { - ttf = getProvider().getTrueTypeFont(substituteName); - if (ttf != null) - { - return ttf; - } - } - // then Windows name - ttf = getProvider().getTrueTypeFont(windowsToPs(postScriptName)); - } - return ttf; + return getProvider().getTrueTypeFont(postScriptName); } /** @@ -402,23 +295,7 @@ public final class ExternalFonts */ public static Type1Font getType1Font(String postScriptName) { - // first ask the font provider for the font - Type1Font t1 = getProvider().getType1Font(postScriptName); - if (t1 == null) - { - // then try substitutes - for (String substituteName : getSubstitutes(postScriptName)) - { - t1 = getProvider().getType1Font(substituteName); - if (t1 != null) - { - return t1; - } - } - // then Windows name - t1 = getProvider().getType1Font(windowsToPs(postScriptName)); - } - return t1; + return getProvider().getType1Font(postScriptName); } /** @@ -458,24 +335,7 @@ public final class ExternalFonts */ private static CFFFont getCFFFont(String postScriptName) { - // first ask the font provider for the font - CFFFont cff = getProvider().getCFFFont(postScriptName); - if (cff == null) - { - // then try substitutes - for (String substituteName : getSubstitutes(postScriptName)) - { - cff = getProvider().getCFFFont(substituteName); - if (cff != null) - { - return cff; - } - } - - // then Windows name - cff = getProvider().getCFFFont(windowsToPs(postScriptName)); - } - return cff; + return getProvider().getCFFFont(postScriptName); } /** diff --git a/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/SubstituteFontProvider.java b/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/SubstituteFontProvider.java new file mode 100644 index 0000000..7034da5 --- /dev/null +++ b/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/SubstituteFontProvider.java @@ -0,0 +1,201 @@ +package org.apache.pdfbox.pdmodel.font; + +import java.util.ArrayList; +import java.util.Arrays; +import java.util.Collections; +import java.util.HashMap; +import java.util.List; +import java.util.Map; + +import org.apache.fontbox.cff.CFFFont; +import org.apache.fontbox.ttf.TrueTypeFont; +import org.apache.fontbox.type1.Type1Font; + +public class SubstituteFontProvider implements FontProvider { + private FontProvider delegate; + + /** Map of PostScript name substitutes, in priority order. */ + private final static Map> substitutes = new HashMap>(); + static + { + // substitutes for standard 14 fonts + substitutes.put("Courier", + Arrays.asList("CourierNew", "CourierNewPSMT", "LiberationMono", "NimbusMonL-Regu")); + substitutes.put("Courier-Bold", + Arrays.asList("CourierNewPS-BoldMT", "CourierNew-Bold", "LiberationMono-Bold", + "NimbusMonL-Bold")); + substitutes.put("Courier-Oblique", + Arrays.asList("CourierNewPS-ItalicMT","CourierNew-Italic", + "LiberationMono-Italic", "NimbusMonL-ReguObli")); + substitutes.put("Courier-BoldOblique", + Arrays.asList("CourierNewPS-BoldItalicMT","CourierNew-BoldItalic", + "LiberationMono-BoldItalic", "NimbusMonL-BoldObli")); + substitutes.put("Helvetica", + Arrays.asList("ArialMT", "Arial", "LiberationSans", "NimbusSanL-Regu")); + substitutes.put("Helvetica-Bold", + Arrays.asList("Arial-BoldMT", "Arial-Bold", "LiberationSans-Bold", + "NimbusSanL-Bold")); + substitutes.put("Helvetica-Oblique", + Arrays.asList("Arial-ItalicMT", "Arial-Italic", "Helvetica-Italic", + "LiberationSans-Italic", "NimbusSanL-ReguItal")); + substitutes.put("Helvetica-BoldOblique", + Arrays.asList("Arial-BoldItalicMT", "Helvetica-BoldItalic", + "LiberationSans-BoldItalic", "NimbusSanL-BoldItal")); + substitutes.put("Times-Roman", + Arrays.asList("TimesNewRomanPSMT", "TimesNewRoman", "TimesNewRomanPS", + "LiberationSerif", "NimbusRomNo9L-Regu")); + substitutes.put("Times-Bold", + Arrays.asList("TimesNewRomanPS-BoldMT", "TimesNewRomanPS-Bold", + "TimesNewRoman-Bold", "LiberationSerif-Bold", + "NimbusRomNo9L-Medi")); + substitutes.put("Times-Italic", + Arrays.asList("TimesNewRomanPS-ItalicMT", "TimesNewRomanPS-Italic", + "TimesNewRoman-Italic", "LiberationSerif-Italic", + "NimbusRomNo9L-ReguItal")); + substitutes.put("Times-BoldItalic", + Arrays.asList("TimesNewRomanPS-BoldItalicMT", "TimesNewRomanPS-BoldItalic", + "TimesNewRoman-BoldItalic", "LiberationSerif-BoldItalic", + "NimbusRomNo9L-MediItal")); + substitutes.put("Symbol", Arrays.asList("SymbolMT", "StandardSymL")); + substitutes.put("ZapfDingbats", Arrays.asList("ZapfDingbatsITC", "Dingbats")); + + // Acrobat also uses alternative names for Standard 14 fonts, which we map to those above + // these include names such as "Arial" and "TimesNewRoman" + for (String baseName : Standard14Fonts.getNames()) + { + if (!substitutes.containsKey(baseName)) + { + String mappedName = Standard14Fonts.getMappedFontName(baseName); + substitutes.put(baseName, copySubstitutes(mappedName)); + } + } + } + + /** + * Copies a list of font substitutes, adding the original font at the start of the list. + */ + private static List copySubstitutes(String postScriptName) + { + List copy = new ArrayList(substitutes.get(postScriptName)); + copy.add(0, postScriptName); + return copy; + } + + /** + * Adds a top-priority substitute for the given font. + * + * @param match PostScript name of the font to match + * @param replace PostScript name of the font to use as a replacement + */ + public static void addSubstitute(String match, String replace) + { + if (!substitutes.containsKey(match)) + { + substitutes.put(match, new ArrayList()); + } + substitutes.get(match).add(replace); + } + + /** + * Returns the substitutes for a given font. + */ + protected static List getSubstitutes(String postScriptName) + { + List subs = substitutes.get(postScriptName.replaceAll(" ", "")); + if (subs != null) + { + return subs; + } + else + { + return Collections.emptyList(); + } + } + + /** + * Windows name (ArialNarrow,Bold) to PostScript name (ArialNarrow-Bold) + */ + private static String windowsToPs(String windowsName) + { + return windowsName.replaceAll(",", "-"); + } + + public SubstituteFontProvider(FontProvider delegate) { + this.delegate = delegate; + } + + + @Override + public TrueTypeFont getTrueTypeFont(String postScriptName) { + // first ask the font provider for the font + TrueTypeFont ttf = delegate.getTrueTypeFont(postScriptName); + if (ttf == null) + { + // then try substitutes + for (String substituteName : getSubstitutes(postScriptName)) + { + ttf = delegate.getTrueTypeFont(substituteName); + if (ttf != null) + { + return ttf; + } + } + // then Windows name + ttf = delegate.getTrueTypeFont(windowsToPs(postScriptName)); + } + return ttf; + } + + @Override + public CFFFont getCFFFont(String postScriptName) { + // first ask the font provider for the font + CFFFont cff = delegate.getCFFFont(postScriptName); + if (cff == null) + { + // then try substitutes + for (String substituteName : getSubstitutes(postScriptName)) + { + cff = delegate.getCFFFont(substituteName); + if (cff != null) + { + return cff; + } + } + + // then Windows name + cff = delegate.getCFFFont(windowsToPs(postScriptName)); + } + return cff; + } + + @Override + public Type1Font getType1Font(String postScriptName) { + // first ask the font provider for the font + Type1Font t1 = delegate.getType1Font(postScriptName); + if (t1 == null) + { + // then try substitutes + for (String substituteName : getSubstitutes(postScriptName)) + { + t1 = delegate.getType1Font(substituteName); + if (t1 != null) + { + return t1; + } + } + // then Windows name + t1 = delegate.getType1Font(windowsToPs(postScriptName)); + } + return t1; + } + + @Override + public String toDebugString() { + String message = delegate.toDebugString(); + if (message != null) { + return "Substitude font provider with delegate: " + message; + } + return null; + } + +} -- 1.8.1.msysgit.1