[gs-cvs] rev 8178 - in trunk/gs: doc lib src

leonardo at ghostscript.com leonardo at ghostscript.com
Sun Aug 5 06:55:36 PDT 2007


Author: leonardo
Date: 2007-08-05 06:55:26 -0700 (Sun, 05 Aug 2007)
New Revision: 8178

Added:
   trunk/gs/src/ConvertUTF.c
   trunk/gs/src/ConvertUTF.h
Modified:
   trunk/gs/doc/Ps2pdf.htm
   trunk/gs/lib/gs_agl.ps
   trunk/gs/lib/gs_pdfwr.ps
   trunk/gs/src/devs.mak
   trunk/gs/src/gdevpdfb.h
   trunk/gs/src/gdevpdfe.c
   trunk/gs/src/gdevpdfp.c
   trunk/gs/src/gdevpdfx.h
Log:
Fix (pdfwrite) : A wrong encoding of Metadata.

DETAILS :

Bug 689244 "Wrong encoding of title in metadata for PDF/A".

Thanks to mpsuzuki for preliminary analyzis.

1. See the documentation change.
2. gs_pdfwr.ps converts the specified name into a translation table.
   We believe that popolar encodings translate into single Unicode
   characters, so we don't implement Unicode compositions now.
   This may be a subject of further improvements. In this case
   the generated integer array should use a complex format
   for representing the translation table.
3. The translation table is passed to pdfwrite 
   as a new device parameter DSCEncodingToUnicode.
4. When it is specified, gdevpdfe.c performs the translation
   first into UTF-8, then into XML.
5. The new modules ConvertUTF.c, ConvertUTF.h are obtained from
   Unicode Inc.

EXPECTED DIFFERENCES :

None.


Modified: trunk/gs/doc/Ps2pdf.htm
===================================================================
--- trunk/gs/doc/Ps2pdf.htm	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/doc/Ps2pdf.htm	2007-08-05 13:55:26 UTC (rev 8178)
@@ -433,6 +433,21 @@
 and <b><tt>InstanceUUID</tt></b> are specified, <b><tt>DocumentTimeSeq</tt></b> is being ignored.
 </dl>
 
+<dl>
+<dt><b><tt>-sDSCEncoding=</tt></b><em>string</em>
+<dd>Defines a name of Postscript encoding in which
+DSC comments are encoded in the source document.
+If specified, the comments are converted
+from that encoding into Unicode UTF-8 when writing <b><tt>Metadata</b></tt>.
+If not specified, the comments are copied to <b><tt>Metadata</b></tt>
+with no conversion. Note that Adobe Distiller for 
+Windows uses the default locale's code page
+for this translation, so it's result may differ from Ghostscript.
+Adobe Acrobat appears to use <b><tt>PDFDocEncoding</b></tt> when
+displaying document's properties,
+so we recommend this value.
+</dl>
+
 <p>
 <b><tt>ps2pdf</tt></b> recognizes all of the Acrobat Distiller 5 parameters
 defined in the DistillerParameters document included in the 

Modified: trunk/gs/lib/gs_agl.ps
===================================================================
--- trunk/gs/lib/gs_agl.ps	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/lib/gs_agl.ps	2007-08-05 13:55:26 UTC (rev 8178)
@@ -23,7 +23,7 @@
 % The AdobeGlyphList dictionary maps glyph names to Unicode values.
 % Nearly all names map to a single value; however, a few names map to
 % multiple values.  See the above file, and also
-%	http://partners.adobe.com/asn/developer/typeforum/unicodegn.html
+%	http://www.adobe.com/devnet/opentype/archives/glyph.html
 % for more details.
 
 /AdobeGlyphList 1500 dict

Modified: trunk/gs/lib/gs_pdfwr.ps
===================================================================
--- trunk/gs/lib/gs_pdfwr.ps	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/lib/gs_pdfwr.ps	2007-08-05 13:55:26 UTC (rev 8178)
@@ -682,6 +682,7 @@
   /EmbedAllFonts { }
   /MaxSubsetPct { }
   /SubsetFonts { }
+  /DSCEncodingToUnicode { }
 .dicttomark readonly def
 /.distillerdevice
  { currentdevice .devicename dup /pdfwrite eq exch /ps2write eq or
@@ -754,6 +755,37 @@
   } ifelse
 } .bind odef
 
+% Create a table for translation DSC comments into Unicode UTF-16
+/.make_DSC_translation_table {        % <encoding> .make_DSC_translation_table <array>
+  dup type /stringtype eq {
+    cvn
+  } if
+  dup type /nametype eq {
+    /Encoding findresource
+  } if
+  dup length array exch                   % [out] [in]
+  0 1 2 index length 1 sub {              % [out] [in] i
+    dup 2 index exch get                       % [out] [in] i n
+    dup dup null eq exch /.notdef eq or {
+      pop -1
+    } {
+      dup //AdobeGlyphList exch .knownget {  % [out] [in] i n c
+      dup type dup /arraytype eq exch /packedarraytype eq or {
+        0 get % fixme: using the 1st char
+      } if
+      exch pop                               % [out] [in] i c
+      } {                                    % [out] [in] i n
+        (DSCEncoding defines a character that is not in AdobeGlyphList : ) 
+        exch .namestring concatstrings =
+        /.make_DSC_translation_table cvx /undefined signalerror 
+      } ifelse
+    } ifelse
+    3 index                                  % [out] [in] i c [out]
+    3 1 roll put                             % [out] [in]
+  } for
+  pop                                        % [out]
+} bind def 
+
 1000 % priority
 {
   % Note, this may not work if the initial device is not pdfwrite
@@ -792,6 +824,14 @@
         systemdict 2 index known { pop pop } { def } ifelse
       } forall
     } repeat
+    currentdevice .devicename /pdfwrite eq {
+      systemdict /PDFX known systemdict /PDFA known or {
+      /DSCEncoding where {
+          /DSCEncoding get .make_DSC_translation_table
+          /DSCEncodingToUnicode exch def 
+        } if
+      } if
+    } if
     currentdict end .setdistillerparams
     .distillerdevice //null false mark .putdeviceparams 
     dup type /booleantype eq not { cleartomark pop } if pop pop

Added: trunk/gs/src/ConvertUTF.c
===================================================================
--- trunk/gs/src/ConvertUTF.c	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/ConvertUTF.c	2007-08-05 13:55:26 UTC (rev 8178)
@@ -0,0 +1,542 @@
+/*
+ * Copyright 2001-2004 Unicode, Inc.
+ * 
+ * Disclaimer
+ * 
+ * This source code is provided as is by Unicode, Inc. No claims are
+ * made as to fitness for any particular purpose. No warranties of any
+ * kind are expressed or implied. The recipient agrees to determine
+ * applicability of information provided. If this file has been
+ * purchased on magnetic or optical media from Unicode, Inc., the
+ * sole remedy for any claim will be exchange of defective media
+ * within 90 days of receipt.
+ * 
+ * Limitations on Rights to Redistribute This Code
+ * 
+ * Unicode, Inc. hereby grants the right to freely use the information
+ * supplied in this file in the creation of products supporting the
+ * Unicode Standard, and to make copies of this file in any form
+ * for internal or external distribution as long as this notice
+ * remains attached.
+ */
+
+/* $Id$*/
+
+
+/* ---------------------------------------------------------------------
+
+    Conversions between UTF32, UTF-16, and UTF-8. Source code file.
+    Author: Mark E. Davis, 1994.
+    Rev History: Rick McGowan, fixes & updates May 2001.
+    Sept 2001: fixed const & error conditions per
+	mods suggested by S. Parent & A. Lillich.
+    June 2002: Tim Dodd added detection and handling of incomplete
+	source sequences, enhanced error detection, added casts
+	to eliminate compiler warnings.
+    July 2003: slight mods to back out aggressive FFFE detection.
+    Jan 2004: updated switches in from-UTF8 conversions.
+    Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions.
+
+    See the header file "ConvertUTF.h" for complete documentation.
+
+------------------------------------------------------------------------ */
+
+
+#include "ConvertUTF.h"
+#ifdef CVTUTF_DEBUG
+#include <stdio.h>
+#endif
+
+static const int halfShift  = 10; /* used for shifting by 10 bits */
+
+static const UTF32 halfBase = 0x0010000UL;
+static const UTF32 halfMask = 0x3FFUL;
+
+#define UNI_SUR_HIGH_START  (UTF32)0xD800
+#define UNI_SUR_HIGH_END    (UTF32)0xDBFF
+#define UNI_SUR_LOW_START   (UTF32)0xDC00
+#define UNI_SUR_LOW_END     (UTF32)0xDFFF
+#define false	   0
+#define true	    1
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF32toUTF16 (
+	const UTF32** sourceStart, const UTF32* sourceEnd, 
+	UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF32* source = *sourceStart;
+    UTF16* target = *targetStart;
+    while (source < sourceEnd) {
+	UTF32 ch;
+	if (target >= targetEnd) {
+	    result = targetExhausted; break;
+	}
+	ch = *source++;
+	if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
+	    /* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */
+	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+		if (flags == strictConversion) {
+		    --source; /* return to the illegal value itself */
+		    result = sourceIllegal;
+		    break;
+		} else {
+		    *target++ = UNI_REPLACEMENT_CHAR;
+		}
+	    } else {
+		*target++ = (UTF16)ch; /* normal case */
+	    }
+	} else if (ch > UNI_MAX_LEGAL_UTF32) {
+	    if (flags == strictConversion) {
+		result = sourceIllegal;
+	    } else {
+		*target++ = UNI_REPLACEMENT_CHAR;
+	    }
+	} else {
+	    /* target is a character in range 0xFFFF - 0x10FFFF. */
+	    if (target + 1 >= targetEnd) {
+		--source; /* Back up source pointer! */
+		result = targetExhausted; break;
+	    }
+	    ch -= halfBase;
+	    *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
+	    *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
+	}
+    }
+    *sourceStart = source;
+    *targetStart = target;
+    return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF16toUTF32 (
+	const UTF16** sourceStart, const UTF16* sourceEnd, 
+	UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF16* source = *sourceStart;
+    UTF32* target = *targetStart;
+    UTF32 ch, ch2;
+    while (source < sourceEnd) {
+	const UTF16* oldSource = source; /*  In case we have to back up because of target overflow. */
+	ch = *source++;
+	/* If we have a surrogate pair, convert to UTF32 first. */
+	if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
+	    /* If the 16 bits following the high surrogate are in the source buffer... */
+	    if (source < sourceEnd) {
+		ch2 = *source;
+		/* If it's a low surrogate, convert to UTF32. */
+		if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
+		    ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
+			+ (ch2 - UNI_SUR_LOW_START) + halfBase;
+		    ++source;
+		} else if (flags == strictConversion) { /* it's an unpaired high surrogate */
+		    --source; /* return to the illegal value itself */
+		    result = sourceIllegal;
+		    break;
+		}
+	    } else { /* We don't have the 16 bits following the high surrogate. */
+		--source; /* return to the high surrogate */
+		result = sourceExhausted;
+		break;
+	    }
+	} else if (flags == strictConversion) {
+	    /* UTF-16 surrogate values are illegal in UTF-32 */
+	    if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
+		--source; /* return to the illegal value itself */
+		result = sourceIllegal;
+		break;
+	    }
+	}
+	if (target >= targetEnd) {
+	    source = oldSource; /* Back up source pointer! */
+	    result = targetExhausted; break;
+	}
+	*target++ = ch;
+    }
+    *sourceStart = source;
+    *targetStart = target;
+#ifdef CVTUTF_DEBUG
+if (result == sourceIllegal) {
+    fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2);
+    fflush(stderr);
+}
+#endif
+    return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Index into the table below with the first byte of a UTF-8 sequence to
+ * get the number of trailing bytes that are supposed to follow it.
+ * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
+ * left as-is for anyone who may want to do such conversion, which was
+ * allowed in earlier algorithms.
+ */
+static const char trailingBytesForUTF8[256] = {
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
+    1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
+    2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5
+};
+
+/*
+ * Magic values subtracted from a buffer value during UTF8 conversion.
+ * This table contains as many values as there might be trailing bytes
+ * in a UTF-8 sequence.
+ */
+static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL, 
+		     0x03C82080UL, 0xFA082080UL, 0x82082080UL };
+
+/*
+ * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed
+ * into the first byte, depending on how many bytes follow.  There are
+ * as many entries in this table as there are UTF-8 sequence types.
+ * (I.e., one byte sequence, two byte... etc.). Remember that sequencs
+ * for *legal* UTF-8 will be 4 or fewer bytes total.
+ */
+static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
+
+/* --------------------------------------------------------------------- */
+
+/* The interface converts a whole buffer to avoid function-call overhead.
+ * Constants have been gathered. Loops & conditionals have been removed as
+ * much as possible for efficiency, in favor of drop-through switches.
+ * (See "Note A" at the bottom of the file for equivalent code.)
+ * If your compiler supports it, the "isLegalUTF8" call can be turned
+ * into an inline function.
+ */
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF16toUTF8 (
+	const UTF16** sourceStart, const UTF16* sourceEnd, 
+	UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF16* source = *sourceStart;
+    UTF8* target = *targetStart;
+    while (source < sourceEnd) {
+	UTF32 ch;
+	unsigned short bytesToWrite = 0;
+	const UTF32 byteMask = 0xBF;
+	const UTF32 byteMark = 0x80; 
+	const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
+	ch = *source++;
+	/* If we have a surrogate pair, convert to UTF32 first. */
+	if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
+	    /* If the 16 bits following the high surrogate are in the source buffer... */
+	    if (source < sourceEnd) {
+		UTF32 ch2 = *source;
+		/* If it's a low surrogate, convert to UTF32. */
+		if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
+		    ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
+			+ (ch2 - UNI_SUR_LOW_START) + halfBase;
+		    ++source;
+		} else if (flags == strictConversion) { /* it's an unpaired high surrogate */
+		    --source; /* return to the illegal value itself */
+		    result = sourceIllegal;
+		    break;
+		}
+	    } else { /* We don't have the 16 bits following the high surrogate. */
+		--source; /* return to the high surrogate */
+		result = sourceExhausted;
+		break;
+	    }
+	} else if (flags == strictConversion) {
+	    /* UTF-16 surrogate values are illegal in UTF-32 */
+	    if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
+		--source; /* return to the illegal value itself */
+		result = sourceIllegal;
+		break;
+	    }
+	}
+	/* Figure out how many bytes the result will require */
+	if (ch < (UTF32)0x80) {	     bytesToWrite = 1;
+	} else if (ch < (UTF32)0x800) {     bytesToWrite = 2;
+	} else if (ch < (UTF32)0x10000) {   bytesToWrite = 3;
+	} else if (ch < (UTF32)0x110000) {  bytesToWrite = 4;
+	} else {			    bytesToWrite = 3;
+					    ch = UNI_REPLACEMENT_CHAR;
+	}
+
+	target += bytesToWrite;
+	if (target > targetEnd) {
+	    source = oldSource; /* Back up source pointer! */
+	    target -= bytesToWrite; result = targetExhausted; break;
+	}
+	switch (bytesToWrite) { /* note: everything falls through. */
+	    case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 1: *--target =  (UTF8)(ch | firstByteMark[bytesToWrite]);
+	}
+	target += bytesToWrite;
+    }
+    *sourceStart = source;
+    *targetStart = target;
+    return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Utility routine to tell whether a sequence of bytes is legal UTF-8.
+ * This must be called with the length pre-determined by the first byte.
+ * If not calling this from ConvertUTF8to*, then the length can be set by:
+ *  length = trailingBytesForUTF8[*source]+1;
+ * and the sequence is illegal right away if there aren't that many bytes
+ * available.
+ * If presented with a length > 4, this returns false.  The Unicode
+ * definition of UTF-8 goes up to 4-byte sequences.
+ */
+
+static Boolean isLegalUTF8(const UTF8 *source, int length) {
+    UTF8 a;
+    const UTF8 *srcptr = source+length;
+    switch (length) {
+    default: return false;
+	/* Everything else falls through when "true"... */
+    case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+    case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
+    case 2: if ((a = (*--srcptr)) > 0xBF) return false;
+
+	switch (*source) {
+	    /* no fall-through in this inner switch */
+	    case 0xE0: if (a < 0xA0) return false; break;
+	    case 0xED: if (a > 0x9F) return false; break;
+	    case 0xF0: if (a < 0x90) return false; break;
+	    case 0xF4: if (a > 0x8F) return false; break;
+	    default:   if (a < 0x80) return false;
+	}
+
+    case 1: if (*source >= 0x80 && *source < 0xC2) return false;
+    }
+    if (*source > 0xF4) return false;
+    return true;
+}
+
+/* --------------------------------------------------------------------- */
+
+/*
+ * Exported function to return whether a UTF-8 sequence is legal or not.
+ * This is not used here; it's just exported.
+ */
+Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) {
+    int length = trailingBytesForUTF8[*source]+1;
+    if (source+length > sourceEnd) {
+	return false;
+    }
+    return isLegalUTF8(source, length);
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF8toUTF16 (
+	const UTF8** sourceStart, const UTF8* sourceEnd, 
+	UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF8* source = *sourceStart;
+    UTF16* target = *targetStart;
+    while (source < sourceEnd) {
+	UTF32 ch = 0;
+	unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
+	if (source + extraBytesToRead >= sourceEnd) {
+	    result = sourceExhausted; break;
+	}
+	/* Do this check whether lenient or strict */
+	if (! isLegalUTF8(source, extraBytesToRead+1)) {
+	    result = sourceIllegal;
+	    break;
+	}
+	/*
+	 * The cases all fall through. See "Note A" below.
+	 */
+	switch (extraBytesToRead) {
+	    case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
+	    case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
+	    case 3: ch += *source++; ch <<= 6;
+	    case 2: ch += *source++; ch <<= 6;
+	    case 1: ch += *source++; ch <<= 6;
+	    case 0: ch += *source++;
+	}
+	ch -= offsetsFromUTF8[extraBytesToRead];
+
+	if (target >= targetEnd) {
+	    source -= (extraBytesToRead+1); /* Back up source pointer! */
+	    result = targetExhausted; break;
+	}
+	if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
+	    /* UTF-16 surrogate values are illegal in UTF-32 */
+	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+		if (flags == strictConversion) {
+		    source -= (extraBytesToRead+1); /* return to the illegal value itself */
+		    result = sourceIllegal;
+		    break;
+		} else {
+		    *target++ = UNI_REPLACEMENT_CHAR;
+		}
+	    } else {
+		*target++ = (UTF16)ch; /* normal case */
+	    }
+	} else if (ch > UNI_MAX_UTF16) {
+	    if (flags == strictConversion) {
+		result = sourceIllegal;
+		source -= (extraBytesToRead+1); /* return to the start */
+		break; /* Bail out; shouldn't continue */
+	    } else {
+		*target++ = UNI_REPLACEMENT_CHAR;
+	    }
+	} else {
+	    /* target is a character in range 0xFFFF - 0x10FFFF. */
+	    if (target + 1 >= targetEnd) {
+		source -= (extraBytesToRead+1); /* Back up source pointer! */
+		result = targetExhausted; break;
+	    }
+	    ch -= halfBase;
+	    *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
+	    *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
+	}
+    }
+    *sourceStart = source;
+    *targetStart = target;
+    return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF32toUTF8 (
+	const UTF32** sourceStart, const UTF32* sourceEnd, 
+	UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF32* source = *sourceStart;
+    UTF8* target = *targetStart;
+    while (source < sourceEnd) {
+	UTF32 ch;
+	unsigned short bytesToWrite = 0;
+	const UTF32 byteMask = 0xBF;
+	const UTF32 byteMark = 0x80; 
+	ch = *source++;
+	if (flags == strictConversion ) {
+	    /* UTF-16 surrogate values are illegal in UTF-32 */
+	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+		--source; /* return to the illegal value itself */
+		result = sourceIllegal;
+		break;
+	    }
+	}
+	/*
+	 * Figure out how many bytes the result will require. Turn any
+	 * illegally large UTF32 things (> Plane 17) into replacement chars.
+	 */
+	if (ch < (UTF32)0x80) {	     bytesToWrite = 1;
+	} else if (ch < (UTF32)0x800) {     bytesToWrite = 2;
+	} else if (ch < (UTF32)0x10000) {   bytesToWrite = 3;
+	} else if (ch <= UNI_MAX_LEGAL_UTF32) {  bytesToWrite = 4;
+	} else {			    bytesToWrite = 3;
+					    ch = UNI_REPLACEMENT_CHAR;
+					    result = sourceIllegal;
+	}
+	
+	target += bytesToWrite;
+	if (target > targetEnd) {
+	    --source; /* Back up source pointer! */
+	    target -= bytesToWrite; result = targetExhausted; break;
+	}
+	switch (bytesToWrite) { /* note: everything falls through. */
+	    case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
+	    case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]);
+	}
+	target += bytesToWrite;
+    }
+    *sourceStart = source;
+    *targetStart = target;
+    return result;
+}
+
+/* --------------------------------------------------------------------- */
+
+ConversionResult ConvertUTF8toUTF32 (
+	const UTF8** sourceStart, const UTF8* sourceEnd, 
+	UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
+    ConversionResult result = conversionOK;
+    const UTF8* source = *sourceStart;
+    UTF32* target = *targetStart;
+    while (source < sourceEnd) {
+	UTF32 ch = 0;
+	unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
+	if (source + extraBytesToRead >= sourceEnd) {
+	    result = sourceExhausted; break;
+	}
+	/* Do this check whether lenient or strict */
+	if (! isLegalUTF8(source, extraBytesToRead+1)) {
+	    result = sourceIllegal;
+	    break;
+	}
+	/*
+	 * The cases all fall through. See "Note A" below.
+	 */
+	switch (extraBytesToRead) {
+	    case 5: ch += *source++; ch <<= 6;
+	    case 4: ch += *source++; ch <<= 6;
+	    case 3: ch += *source++; ch <<= 6;
+	    case 2: ch += *source++; ch <<= 6;
+	    case 1: ch += *source++; ch <<= 6;
+	    case 0: ch += *source++;
+	}
+	ch -= offsetsFromUTF8[extraBytesToRead];
+
+	if (target >= targetEnd) {
+	    source -= (extraBytesToRead+1); /* Back up the source pointer! */
+	    result = targetExhausted; break;
+	}
+	if (ch <= UNI_MAX_LEGAL_UTF32) {
+	    /*
+	     * UTF-16 surrogate values are illegal in UTF-32, and anything
+	     * over Plane 17 (> 0x10FFFF) is illegal.
+	     */
+	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
+		if (flags == strictConversion) {
+		    source -= (extraBytesToRead+1); /* return to the illegal value itself */
+		    result = sourceIllegal;
+		    break;
+		} else {
+		    *target++ = UNI_REPLACEMENT_CHAR;
+		}
+	    } else {
+		*target++ = ch;
+	    }
+	} else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
+	    result = sourceIllegal;
+	    *target++ = UNI_REPLACEMENT_CHAR;
+	}
+    }
+    *sourceStart = source;
+    *targetStart = target;
+    return result;
+}
+
+/* ---------------------------------------------------------------------
+
+    Note A.
+    The fall-through switches in UTF-8 reading code save a
+    temp variable, some decrements & conditionals.  The switches
+    are equivalent to the following loop:
+	{
+	    int tmpBytesToRead = extraBytesToRead+1;
+	    do {
+		ch += *source++;
+		--tmpBytesToRead;
+		if (tmpBytesToRead) ch <<= 6;
+	    } while (tmpBytesToRead > 0);
+	}
+    In UTF-8 writing code, the switches on "bytesToWrite" are
+    similarly unrolled loops.
+
+   --------------------------------------------------------------------- */


Property changes on: trunk/gs/src/ConvertUTF.c
___________________________________________________________________
Name: svn:keywords
   + Id
Name: svn:eol-style
   + native

Added: trunk/gs/src/ConvertUTF.h
===================================================================
--- trunk/gs/src/ConvertUTF.h	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/ConvertUTF.h	2007-08-05 13:55:26 UTC (rev 8178)
@@ -0,0 +1,152 @@
+/*
+ * Copyright 2001-2004 Unicode, Inc.
+ * 
+ * Disclaimer
+ * 
+ * This source code is provided as is by Unicode, Inc. No claims are
+ * made as to fitness for any particular purpose. No warranties of any
+ * kind are expressed or implied. The recipient agrees to determine
+ * applicability of information provided. If this file has been
+ * purchased on magnetic or optical media from Unicode, Inc., the
+ * sole remedy for any claim will be exchange of defective media
+ * within 90 days of receipt.
+ * 
+ * Limitations on Rights to Redistribute This Code
+ * 
+ * Unicode, Inc. hereby grants the right to freely use the information
+ * supplied in this file in the creation of products supporting the
+ * Unicode Standard, and to make copies of this file in any form
+ * for internal or external distribution as long as this notice
+ * remains attached.
+ */
+
+/* $Id$*/
+
+
+/* ---------------------------------------------------------------------
+
+    Conversions between UTF32, UTF-16, and UTF-8.  Header file.
+
+    Several funtions are included here, forming a complete set of
+    conversions between the three formats.  UTF-7 is not included
+    here, but is handled in a separate source file.
+
+    Each of these routines takes pointers to input buffers and output
+    buffers.  The input buffers are const.
+
+    Each routine converts the text between *sourceStart and sourceEnd,
+    putting the result into the buffer between *targetStart and
+    targetEnd. Note: the end pointers are *after* the last item: e.g. 
+    *(sourceEnd - 1) is the last item.
+
+    The return result indicates whether the conversion was successful,
+    and if not, whether the problem was in the source or target buffers.
+    (Only the first encountered problem is indicated.)
+
+    After the conversion, *sourceStart and *targetStart are both
+    updated to point to the end of last text successfully converted in
+    the respective buffers.
+
+    Input parameters:
+	sourceStart - pointer to a pointer to the source buffer.
+		The contents of this are modified on return so that
+		it points at the next thing to be converted.
+	targetStart - similarly, pointer to pointer to the target buffer.
+	sourceEnd, targetEnd - respectively pointers to the ends of the
+		two buffers, for overflow checking only.
+
+    These conversion functions take a ConversionFlags argument. When this
+    flag is set to strict, both irregular sequences and isolated surrogates
+    will cause an error.  When the flag is set to lenient, both irregular
+    sequences and isolated surrogates are converted.
+
+    Whether the flag is strict or lenient, all illegal sequences will cause
+    an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
+    or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
+    must check for illegal sequences.
+
+    When the flag is set to lenient, characters over 0x10FFFF are converted
+    to the replacement character; otherwise (when the flag is set to strict)
+    they constitute an error.
+
+    Output parameters:
+	The value "sourceIllegal" is returned from some routines if the input
+	sequence is malformed.  When "sourceIllegal" is returned, the source
+	value will point to the illegal value that caused the problem. E.g.,
+	in UTF-8 when a sequence is malformed, it points to the start of the
+	malformed sequence.  
+
+    Author: Mark E. Davis, 1994.
+    Rev History: Rick McGowan, fixes & updates May 2001.
+		 Fixes & updates, Sept 2001.
+
+------------------------------------------------------------------------ */
+
+/* ---------------------------------------------------------------------
+    The following 4 definitions are compiler-specific.
+    The C standard does not guarantee that wchar_t has at least
+    16 bits, so wchar_t is no less portable than unsigned short!
+    All should be unsigned values to avoid sign extension during
+    bit mask & shift operations.
+------------------------------------------------------------------------ */
+
+typedef unsigned long	UTF32;	/* at least 32 bits */
+typedef unsigned short	UTF16;	/* at least 16 bits */
+typedef unsigned char	UTF8;	/* typically 8 bits */
+typedef unsigned char	Boolean; /* 0 or 1 */
+
+/* Some fundamental constants */
+#define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
+#define UNI_MAX_BMP (UTF32)0x0000FFFF
+#define UNI_MAX_UTF16 (UTF32)0x0010FFFF
+#define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
+#define UNI_MAX_LEGAL_UTF32 (UTF32)0x0010FFFF
+
+typedef enum {
+	conversionOK, 		/* conversion successful */
+	sourceExhausted,	/* partial character in source, but hit end */
+	targetExhausted,	/* insuff. room in target for conversion */
+	sourceIllegal		/* source sequence is illegal/malformed */
+} ConversionResult;
+
+typedef enum {
+	strictConversion = 0,
+	lenientConversion
+} ConversionFlags;
+
+/* This is for C++ and does no harm in C */
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+ConversionResult ConvertUTF8toUTF16 (
+		const UTF8** sourceStart, const UTF8* sourceEnd, 
+		UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF16toUTF8 (
+		const UTF16** sourceStart, const UTF16* sourceEnd, 
+		UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
+		
+ConversionResult ConvertUTF8toUTF32 (
+		const UTF8** sourceStart, const UTF8* sourceEnd, 
+		UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF32toUTF8 (
+		const UTF32** sourceStart, const UTF32* sourceEnd, 
+		UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
+		
+ConversionResult ConvertUTF16toUTF32 (
+		const UTF16** sourceStart, const UTF16* sourceEnd, 
+		UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags);
+
+ConversionResult ConvertUTF32toUTF16 (
+		const UTF32** sourceStart, const UTF32* sourceEnd, 
+		UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
+
+Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd);
+
+#ifdef __cplusplus
+}
+#endif
+
+/* --------------------------------------------------------------------- */


Property changes on: trunk/gs/src/ConvertUTF.h
___________________________________________________________________
Name: svn:keywords
   + Id
Name: svn:eol-style
   + native

Modified: trunk/gs/src/devs.mak
===================================================================
--- trunk/gs/src/devs.mak	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/devs.mak	2007-08-05 13:55:26 UTC (rev 8178)
@@ -250,6 +250,7 @@
 gdevsvga_h=$(GLSRC)gdevsvga.h
 # Out of order
 gdevdljm_h=$(GLSRC)gdevdljm.h $(gdevpcl_h)
+ConvertUTF_h=$(GLSRC)ConvertUTF.h
 
 GDEVLDFJB2CC=$(CC_) $(I_)$(GLI_) $(II)$(LDF_JB2I_)$(_I) $(JB2CF_) $(GLF_)
 GDEVLWFJPXCC=$(CC_) $(I_)$(GLI_) $(II)$(LWF_JPXI_)$(_I) $(JPXCF_) $(GLF_)
@@ -807,7 +808,7 @@
 pdfwrite6_=$(GLOBJ)gdevpdfo.$(OBJ) $(GLOBJ)gdevpdfp.$(OBJ) $(GLOBJ)gdevpdft.$(OBJ)
 pdfwrite7_=$(GLOBJ)gdevpdfr.$(OBJ)
 pdfwrite8_=$(GLOBJ)gdevpdfu.$(OBJ) $(GLOBJ)gdevpdfv.$(OBJ)
-pdfwrite9_=
+pdfwrite9_= $(GLOBJ)ConvertUTF.$(OBJ)
 pdfwrite10_=$(GLOBJ)gsflip.$(OBJ)
 pdfwrite11_=$(GLOBJ)scantab.$(OBJ) $(GLOBJ)sfilter2.$(OBJ)
 pdfwrite_=$(pdfwrite1_) $(pdfwrite2_) $(pdfwrite3_) $(pdfwrite4_)\
@@ -880,8 +881,8 @@
 	$(GLCC) $(GLO_)gdevpdfd.$(OBJ) $(C_) $(GLSRC)gdevpdfd.c
 
 $(GLOBJ)gdevpdfe.$(OBJ) : $(GLSRC)gdevpdfe.c\
- $(gx_h) $(string__h) $(time__h) $(stream_h) $(gp_h) $(smd5_h) $(gscdefs_h)\
- $(gdevpdfx_h) $(gdevpdfg_h) $(gdevpdfo_h) $(gdevpdtf_h)
+ $(gx_h) $(gserrors_h) $(string__h) $(time__h) $(stream_h) $(gp_h) $(smd5_h) $(gscdefs_h)\
+ $(gdevpdfx_h) $(gdevpdfg_h) $(gdevpdfo_h) $(gdevpdtf_h) $(ConvertUTF_h)
 	$(GLCC) $(GLO_)gdevpdfe.$(OBJ) $(C_) $(GLSRC)gdevpdfe.c
 
 $(GLOBJ)gdevpdfg.$(OBJ) : $(GLSRC)gdevpdfg.c $(GXERR) $(math__h) $(string__h)\
@@ -955,6 +956,9 @@
  $(szlibx_h)
 	$(GLCC) $(GLO_)gdevpdfv.$(OBJ) $(C_) $(GLSRC)gdevpdfv.c
 
+$(GLOBJ)ConvertUTF.$(OBJ) : $(GLSRC)ConvertUTF.c $(ConvertUTF_h)
+	$(GLCC) $(GLO_)ConvertUTF.$(OBJ) $(C_) $(GLSRC)ConvertUTF.c
+
 ######## pdfwrite text
 
 # The text facilities for the PDF writer are so large and complex that

Modified: trunk/gs/src/gdevpdfb.h
===================================================================
--- trunk/gs/src/gdevpdfb.h	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/gdevpdfb.h	2007-08-05 13:55:26 UTC (rev 8178)
@@ -127,6 +127,7 @@
  max_long,                      /* MaxViewerMemorySize */
  256000,			/* MaxShadingBitmapSize */
  PDF_DEVICE_MaxInlineImageSize,	/* MaxInlineImageSize */
+ {0, 0},                        /* DSCEncodingToUnicode */
  {0, 0, 0},			/* OwnerPassword */
  {0, 0, 0},			/* UserPassword */
  0,				/* KeyLength */

Modified: trunk/gs/src/gdevpdfe.c
===================================================================
--- trunk/gs/src/gdevpdfe.c	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/gdevpdfe.c	2007-08-05 13:55:26 UTC (rev 8178)
@@ -14,6 +14,7 @@
 /* $Id$ */
 /* Metadata writer. */
 #include "gx.h"
+#include "gserrors.h"
 #include "string_.h"
 #include "time_.h"
 #include "stream.h"
@@ -24,7 +25,17 @@
 #include "gdevpdfg.h"
 #include "gdevpdfo.h"
 #include "gdevpdtf.h"
+#include "ConvertUTF.h"
 
+
+private void 
+copy_bytes(stream *s, const byte **data, int *data_length, int n)
+{
+    while (n-- && (*data_length)--) {
+	stream_putc(s, *((*data)++));
+    }
+}
+
 /* Write XML data */
 private void
 pdf_xml_data_write(stream *s, const byte *data, int data_length)
@@ -32,18 +43,34 @@
     int l = data_length;
     const byte *p = data;
 
-    for (; l--; p++) {
+    while (l > 0) {
 	switch (*p) {
-	    case '<' : stream_puts(s, "&lt;"); break;
-	    case '>' : stream_puts(s, "&gt;"); break;
-	    case '&' : stream_puts(s, "&amp;"); break;
-	    case '\'': stream_puts(s, "&apos;"); break;
-	    case '"' : stream_puts(s, "&quot;"); break;
+	    case '<' : stream_puts(s, "&lt;"); l--; p++; break;
+	    case '>' : stream_puts(s, "&gt;"); l--; p++; break;
+	    case '&' : stream_puts(s, "&amp;"); l--; p++; break;
+	    case '\'': stream_puts(s, "&apos;"); l--; p++; break;
+	    case '"' : stream_puts(s, "&quot;"); l--; p++; break;
 	    default:
-		if (*p < 32 || *p > 127)
+		if (*p < 32) {
+		    /* Not allowed in XML. */
 		    pprintd1(s, "&#%d;", *p);
-		else
+		    l--; p++;
+		} else if (*p >= 0x7F && *p <= 0x9f) {
+		    /* Control characters are discouraged in XML. */
+		    pprintd1(s, "&#%d;", *p);
+		} else if ((*p & 0xE0) == 0xC0) {
+		    /* A 2-byte UTF-8 sequence */
+		    copy_bytes(s, &p, &l, 2);
+		} else if ((*p & 0xF0) == 0xE0) {
+		    /* A 3-byte UTF-8 sequence */
+		    copy_bytes(s, &p, &l, 3);
+		} else if ((*p & 0xF0) == 0xF0) {
+		    /* A 4-byte UTF-8 sequence */
+		    copy_bytes(s, &p, &l, 4);
+		} else {
 		    stream_putc(s, *p);
+		    l--; p++;
+		}
 	}
     }
 }
@@ -253,7 +280,95 @@
     return l;
 }
 
-private void
+private inline byte
+decode_escape(const byte *data, int data_length, int *index)
+{
+    byte c;
+
+    (*index)++; /* skip '\' */
+    if (*index >= data_length)
+	return 0; /* Must_not_happen, because the string is PS encoded. */
+    c = data[*index];
+    switch (c) {
+	case 'n': (*index)++; return '\n';
+	case 'r': (*index)++; return '\r';
+	case 't': (*index)++; return '\t';
+	default:
+	    break;
+    }
+    if (c >= '0' && c <= '7') {
+	/* octal */
+	byte v = c - '0';
+
+	for (;;) {
+	    (*index)++;
+	    if (*index >= data_length)
+		return v;
+	    c = data[*index];
+	    if (c < '0' || c > '7')
+		break;
+	    v = v * 8 + (c - '0');
+	}
+	return v;	
+    }
+    return c; /* A wrong escapement sequence. */
+}
+
+private int
+pdf_xmp_write_translated(gx_device_pdf *pdev, stream *s, const byte *data, int data_length,
+			 void(*write)(stream *s, const byte *data, int data_length))
+{
+    if (pdev->DSCEncodingToUnicode.data == 0) {
+	write(s, data, data_length);
+	return 0;
+    } else {
+	UTF16 *buf0, *buf0b;
+	UTF8 *buf1, *buf1b;
+	int i, j = 0;
+
+	buf0 = (UTF16 *)gs_alloc_bytes(pdev->memory, data_length * sizeof(UTF16), 
+			"pdf_xmp_write_translated");
+	if (buf0 == NULL)
+	    return_error(gs_error_VMerror);
+	buf1 = (UTF8 *)gs_alloc_bytes(pdev->memory, data_length * 2, 
+			"pdf_xmp_write_translated");
+	if (buf1 == NULL)
+	    return_error(gs_error_VMerror);
+	buf0b = buf0;
+	buf1b = buf1;
+	for (i = 0; i < data_length; i++) {
+	    byte c = data[i];
+	    int v;
+
+	    if (c == '\\') 
+		c = decode_escape(data, data_length, &i);
+	    if (c > pdev->DSCEncodingToUnicode.size)
+		return_error(gs_error_rangecheck);
+
+	    v = pdev->DSCEncodingToUnicode.data[c];
+	    if (v == -1)
+		v = '?'; /* Arbitrary. */
+	    buf0[j] = v;
+	    j++;
+	}
+	switch (ConvertUTF16toUTF8(&buf0b, buf0 + j,
+			     &buf1b, buf1 + data_length * 2, strictConversion)) {
+	    case conversionOK:
+		write(s, buf1, buf1b - buf1);
+		break;
+	    case sourceExhausted:
+	    case targetExhausted:
+	    case sourceIllegal:
+	    default:
+		return_error(gs_error_rangecheck);
+	}
+	gs_free_object(pdev->memory, buf0, "pdf_xmp_write_translated");
+	gs_free_object(pdev->memory, buf1, "pdf_xmp_write_translated");
+	return 0;
+    }
+}
+
+private int
 pdf_xmp_write_docinfo_item(gx_device_pdf *pdev, stream *s, const char *key, const char *default_value,
 			   void(*write)(stream *s, const byte *data, int data_length))
 {
@@ -262,11 +377,15 @@
     if (v != NULL && (v->value_type == COS_VALUE_SCALAR || 
 			v->value_type == COS_VALUE_CONST)) {
 	if (v->contents.chars.size > 2 && v->contents.chars.data[0] == '(')
-	    write(s, v->contents.chars.data + 1, v->contents.chars.size - 2);
+	    return pdf_xmp_write_translated(pdev, s, v->contents.chars.data + 1, 
+			v->contents.chars.size - 2, write);
 	else
-	    write(s, v->contents.chars.data, v->contents.chars.size);
-    } else 
+	    return pdf_xmp_write_translated(pdev, s, v->contents.chars.data, 
+			v->contents.chars.size, write);
+    } else {
 	stream_puts(s, default_value);
+	return 0;
+    }
 }
 
 private uint64_t
@@ -398,8 +517,10 @@
 	    pdf_xml_attribute_name(s, "xmlns:pdf");
 	    pdf_xml_attribute_value(s, "http://ns.adobe.com/pdf/1.3/");
 	    pdf_xml_attribute_name(s, "pdf:Producer");
-	    pdf_xmp_write_docinfo_item(pdev, s,  "/Producer", "UnknownProduicer",
+	    code = pdf_xmp_write_docinfo_item(pdev, s,  "/Producer", "UnknownProduicer",
 			pdf_xml_attribute_value_data);
+	    if (code < 0)
+		return code;
 	    pdf_xml_tag_end_empty(s);
 	    pdf_xml_newline(s);
 
@@ -416,8 +537,10 @@
 	    {
 		pdf_xml_tag_open_beg(s, "xap:CreatorTool");
 		pdf_xml_tag_end(s);
-		pdf_xmp_write_docinfo_item(pdev, s,  "/Creator", "UnknownApplication",
+		code = pdf_xmp_write_docinfo_item(pdev, s,  "/Creator", "UnknownApplication",
 			pdf_xml_data_write);
+		if (code < 0)
+		    return code;
 		pdf_xml_tag_close(s, "xap:CreatorTool");
 	    }
 	    pdf_xml_tag_close(s, "rdf:Description");
@@ -451,8 +574,10 @@
 			pdf_xml_attribute_value(s, "x-default");
 			pdf_xml_tag_end(s);
 			{
-			   pdf_xmp_write_docinfo_item(pdev, s,  "/Title", "Untitled", 
+			   code = pdf_xmp_write_docinfo_item(pdev, s,  "/Title", "Untitled", 
 				    pdf_xml_data_write);
+			    if (code < 0)
+				return code;
 			}
 			pdf_xml_tag_close(s, "rdf:li");
 		    }
@@ -470,8 +595,10 @@
 			{
 			    pdf_xml_tag_open(s, "rdf:li");
 			    {
-    				pdf_xmp_write_docinfo_item(pdev, s,  "/Author", "Unknown", 
+    				code = pdf_xmp_write_docinfo_item(pdev, s,  "/Author", "Unknown", 
 					    pdf_xml_data_write);
+				if (code < 0)
+				    return code;
 			    }
 			    pdf_xml_tag_close(s, "rdf:li");
 			}

Modified: trunk/gs/src/gdevpdfp.c
===================================================================
--- trunk/gs/src/gdevpdfp.c	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/gdevpdfp.c	2007-08-05 13:55:26 UTC (rev 8178)
@@ -73,6 +73,7 @@
     pi("CompressFonts", gs_param_type_bool, CompressFonts),
     pi("PrintStatistics", gs_param_type_bool, PrintStatistics),
     pi("MaxInlineImageSize", gs_param_type_long, MaxInlineImageSize),
+    pi("DSCEncodingToUnicode", gs_param_type_int_array, DSCEncodingToUnicode),
 
 	/* PDF Encryption */
     pi("OwnerPassword", gs_param_type_string, OwnerPassword),

Modified: trunk/gs/src/gdevpdfx.h
===================================================================
--- trunk/gs/src/gdevpdfx.h	2007-08-05 03:08:49 UTC (rev 8177)
+++ trunk/gs/src/gdevpdfx.h	2007-08-05 13:55:26 UTC (rev 8178)
@@ -450,6 +450,7 @@
 			      a bitmap representation of a shading.
 			      (Bigger shadings to be downsampled). */
     long MaxInlineImageSize;
+    gs_param_int_array DSCEncodingToUnicode;
     /* Encryption parameters */
     gs_param_string OwnerPassword;
     gs_param_string UserPassword;
@@ -673,8 +674,8 @@
  m(28,sbstack) m(29,substream_Resources) m(30,font3)\
  m(31,accumulating_substream_resource) \
  m(32,pres_soft_mask_dict) m(33,PDFXTrimBoxToMediaBoxOffset.data)\
- m(34,PDFXBleedBoxToTrimBoxOffset.data)
-#define gx_device_pdf_num_ptrs 35
+ m(34,PDFXBleedBoxToTrimBoxOffset.data) m(35, DSCEncodingToUnicode.data)
+#define gx_device_pdf_num_ptrs 36
 #define gx_device_pdf_do_param_strings(m)\
     m(0, OPDFReadProcsetPath) m(1, OwnerPassword) m(2, UserPassword) m(3, NoEncrypt)\
     m(4, DocumentUUID) m(5, InstanceUUID)



More information about the gs-cvs mailing list