Adding upstream version 1.14~rc1.upstream/1.14_rc1

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

28 files changed, 966 insertions, 790 deletions

diff --git a/AUTHORS b/AUTHORS
index 95949f7..04be19e 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -1,7 +1,7 @@
 Clzip was written by Antonio Diaz Diaz.
 
 The ideas embodied in clzip are due to (at least) the following people:
-Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the
+Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the
 definition of Markov chains), G.N.N. Martin (for the definition of range
 encoding), Igor Pavlov (for putting all the above together in LZMA), and
 Julian Seward (for bzip2's CLI).
diff --git a/ChangeLog b/ChangeLog
index ca2d657..e3b1859 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,15 @@
+2023-12-20  Antonio Diaz Diaz  <antonio@gnu.org>
+
+	* Version 1.14-rc1 released.
+	* New options '--empty-error' and '--marking-error'.
+	* main.c: Reformat file diagnostics as 'PROGRAM: FILE: MESSAGE'.
+	  (show_option_error): New function showing argument and option name.
+	  (open_outstream): Create missing intermediate directories.
+	* lzip.h: Rename verify_* to check_*.
+	* configure, Makefile.in: New variable 'MAKEINFO'.
+	* INSTALL: Document use of CFLAGS+='--std=c99 -D_XOPEN_SOURCE=500'.
+	* testsuite: New test files fox6.lz, fox6_mark.lz.
+
 2022-01-24  Antonio Diaz Diaz  <antonio@gnu.org>
 
 	* Version 1.13 released.
@@ -160,7 +172,7 @@
 	* Translated to C from the C++ source of lzip 1.10.
 
 
-Copyright (C) 2010-2022 Antonio Diaz Diaz.
+Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
 This file is a collection of facts, and thus it is not copyrightable,
 but just in case, you have unlimited permission to copy, distribute, and
diff --git a/INSTALL b/INSTALL
index 8513993..007b0bc 100644
--- a/INSTALL
+++ b/INSTALL
@@ -18,8 +18,8 @@ Procedure
 or
 	lzip -cd clzip[version].tar.lz | tar -xf -
 
-This creates the directory ./clzip[version] containing the source from
-the main archive.
+This creates the directory ./clzip[version] containing the source code
+extracted from the archive.
 
 2. Change to clzip directory and run configure.
    (Try 'configure --help' for usage instructions).
@@ -27,6 +27,10 @@ the main archive.
 	cd clzip[version]
 	./configure
 
+   If you choose a C standard, enable the POSIX features explicitly:
+
+	./configure CFLAGS+='--std=c99 -D_XOPEN_SOURCE=500'
+
    If you are compiling on MinGW, use:
 
 	./configure CFLAGS+='-D __USE_MINGW_ANSI_STDIO'
@@ -38,7 +42,8 @@ the main archive.
 4. Optionally, type 'make check' to run the tests that come with clzip.
 
 5. Type 'make install' to install the program and any data files and
-   documentation.
+   documentation. You need root privileges to install into a prefix owned
+   by root.
 
    Or type 'make install-compress', which additionally compresses the
    info manual and the man page after installation.
@@ -62,15 +67,15 @@ object files and executables to go and run the 'configure' script.
 'configure' automatically checks for the source code in '.', in '..', and
 in the directory that 'configure' is in.
 
-'configure' recognizes the option '--srcdir=DIR' to control where to
-look for the sources. Usually 'configure' can determine that directory
+'configure' recognizes the option '--srcdir=DIR' to control where to look
+for the source code. Usually 'configure' can determine that directory
 automatically.
 
 After running 'configure', you can run 'make' and 'make install' as
 explained above.
 
 
-Copyright (C) 2010-2022 Antonio Diaz Diaz.
+Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
 This file is free documentation: you have unlimited permission to copy,
 distribute, and modify it.
diff --git a/Makefile.in b/Makefile.in
index 5f77295..55e2bcb 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -29,6 +29,10 @@ main.o : main.c
 %.o : %.c
 	$(CC) $(CPPFLAGS) $(CFLAGS) -c -o $@ $<
 
+# prevent 'make' from trying to remake source files
+$(VPATH)/configure $(VPATH)/Makefile.in $(VPATH)/doc/$(pkgname).texi : ;
+%.h %.c : ;
+
 $(objs)        : Makefile
 carg_parser.o  : carg_parser.h
 decoder.o      : lzip.h decoder.h
@@ -39,13 +43,12 @@ list.o         : lzip.h lzip_index.h
 lzip_index.o   : lzip.h lzip_index.h
 main.o         : carg_parser.h lzip.h decoder.h encoder_base.h encoder.h fast_encoder.h
 
-
 doc : info man
 
 info : $(VPATH)/doc/$(pkgname).info
 
 $(VPATH)/doc/$(pkgname).info : $(VPATH)/doc/$(pkgname).texi
-	cd $(VPATH)/doc && makeinfo $(pkgname).texi
+	cd $(VPATH)/doc && $(MAKEINFO) $(pkgname).texi
 
 man : $(VPATH)/doc/$(progname).1
 
@@ -127,6 +130,8 @@ dist : doc
 	  $(DISTNAME)/testsuite/test.txt \
 	  $(DISTNAME)/testsuite/fox.lz \
 	  $(DISTNAME)/testsuite/fox_*.lz \
+	  $(DISTNAME)/testsuite/fox6.lz \
+	  $(DISTNAME)/testsuite/fox6_mark.lz \
 	  $(DISTNAME)/testsuite/test.txt.lz \
 	  $(DISTNAME)/testsuite/test_em.txt.lz
 	rm -f $(DISTNAME)
diff --git a/NEWS b/NEWS
index f3e9de1..27003a0 100644
--- a/NEWS
+++ b/NEWS
@@ -1,11 +1,21 @@
-Changes in version 1.13:
+Changes in version 1.14:
 
-Decompression time has been reduced by 5-12% depending on the file.
+The option '--empty-error', which forces exit status 2 if any empty member
+is found, has been added.
 
-In case of error in a numerical argument to a command line option, clzip
-now shows the name of the option and the range of valid values.
+The option '--marking-error', which forces exit status 2 if the first LZMA
+byte is non-zero in any member, has been added.
 
-Several descriptions have been improved in manual, '--help', and man page.
+File diagnostics have been reformatted as 'PROGRAM: FILE: MESSAGE'.
 
-The texinfo category of the manual has been changed from 'Data Compression'
-to 'Compression' to match that of gzip. (Reported by Alfred M. Szmidt).
+Diagnostics caused by invalid arguments to command-line options now show the
+argument and the name of the option.
+
+The option '-o, --output' now creates missing intermediate directories when
+writing to a file.
+
+The variable MAKEINFO has been added to configure and Makefile.in.
+
+It has been documented in INSTALL that when choosing a C standard, the POSIX
+features need to be enabled explicitly:
+  ./configure CFLAGS+='--std=c99 -D_XOPEN_SOURCE=500'
diff --git a/README b/README
index 17f7825..a5e55d0 100644
--- a/README
+++ b/README
@@ -7,14 +7,15 @@ C++ compiler.
 
 Lzip is a lossless data compressor with a user interface similar to the one
 of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov
-chain-Algorithm' (LZMA) stream format and provides a 3 factor integrity
-checking to maximize interoperability and optimize safety. Lzip can compress
-about as fast as gzip (lzip -0) or compress most files more than bzip2
-(lzip -9). Decompression speed is intermediate between gzip and bzip2.
-Lzip is better than gzip and bzip2 from a data recovery perspective. Lzip
-has been designed, written, and tested with great care to replace gzip and
-bzip2 as the standard general-purpose compressed format for unix-like
-systems.
+chain-Algorithm' (LZMA) stream format to maximize interoperability. The
+maximum dictionary size is 512 MiB so that any lzip file can be decompressed
+on 32-bit machines. Lzip provides accurate and robust 3-factor integrity
+checking. Lzip can compress about as fast as gzip (lzip -0) or compress most
+files more than bzip2 (lzip -9). Decompression speed is intermediate between
+gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery
+perspective. Lzip has been designed, written, and tested with great care to
+replace gzip and bzip2 as the standard general-purpose compressed format for
+Unix-like systems.
 
 For compressing/decompressing large files on multiprocessor machines plzip
 can be much faster than lzip at the cost of a slightly reduced compression
@@ -52,9 +53,9 @@ Clzip uses the same well-defined exit status values used by bzip2, which
 makes it safer than compressors returning ambiguous warning values (like
 gzip) when it is used as a back end for other programs like tar or zutils.
 
-Clzip will automatically use for each file the largest dictionary size that
-does not exceed neither the file size nor the limit given. Keep in mind that
-the decompression memory requirement is affected at compression time by the
+Clzip automatically uses for each file the largest dictionary size that does
+not exceed neither the file size nor the limit given. Keep in mind that the
+decompression memory requirement is affected at compression time by the
 choice of dictionary size limit.
 
 The amount of memory required for compression is about 1 or 2 times the
@@ -74,20 +75,20 @@ filename.tlz   becomes   filename.tar
 anyothername   becomes   anyothername.out
 
 (De)compressing a file is much like copying or moving it. Therefore clzip
-preserves the access and modification dates, permissions, and, when
-possible, ownership of the file just as 'cp -p' does. (If the user ID or
-the group ID can't be duplicated, the file permission bits S_ISUID and
-S_ISGID are cleared).
+preserves the access and modification dates, permissions, and, if you have
+appropriate privileges, ownership of the file just as 'cp -p' does. (If the
+user ID or the group ID can't be duplicated, the file permission bits
+S_ISUID and S_ISGID are cleared).
 
 Clzip is able to read from some types of non-regular files if either the
 option '-c' or the option '-o' is specified.
 
 If no file names are specified, clzip compresses (or decompresses) from
-standard input to standard output. Clzip will refuse to read compressed data
+standard input to standard output. Clzip refuses to read compressed data
 from a terminal or write compressed data to a terminal, as this would be
 entirely incomprehensible and might leave the terminal in an abnormal state.
 
-Clzip will correctly decompress a file which is the concatenation of two or
+Clzip correctly decompresses a file which is the concatenation of two or
 more compressed files. The result is the concatenation of the corresponding
 decompressed files. Integrity testing of concatenated compressed files is
 also supported.
@@ -114,13 +115,13 @@ Clzip currently implements two variants of the LZMA algorithm: fast
 (used by option '-0') and normal (used by all other compression levels).
 
 The high compression of LZMA comes from combining two basic, well-proven
-compression ideas: sliding dictionaries (LZ77/78) and markov models (the
-thing used by every compression algorithm that uses a range encoder or
-similar order-0 entropy coder as its last stage) with segregation of
-contexts according to what the bits are used for.
+compression ideas: sliding dictionaries (LZ77) and markov models (the thing
+used by every compression algorithm that uses a range encoder or similar
+order-0 entropy coder as its last stage) with segregation of contexts
+according to what the bits are used for.
 
 The ideas embodied in clzip are due to (at least) the following people:
-Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the
+Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the
 definition of Markov chains), G.N.N. Martin (for the definition of range
 encoding), Igor Pavlov (for putting all the above together in LZMA), and
 Julian Seward (for bzip2's CLI).
@@ -130,7 +131,7 @@ been compressed. Decompressed is used to refer to data which have undergone
 the process of decompression.
 
 
-Copyright (C) 2010-2022 Antonio Diaz Diaz.
+Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
 This file is free documentation: you have unlimited permission to copy,
 distribute, and modify it.
diff --git a/carg_parser.c b/carg_parser.c
index 181ba23..8882b97 100644
--- a/carg_parser.c
+++ b/carg_parser.c
@@ -1,5 +1,5 @@
-/* Arg_parser - POSIX/GNU command line argument parser. (C version)
-   Copyright (C) 2006-2022 Antonio Diaz Diaz.
+/* Arg_parser - POSIX/GNU command-line argument parser. (C version)
+   Copyright (C) 2006-2023 Antonio Diaz Diaz.
 
    This library is free software. Redistribution and use in source and
    binary forms, with or without modification, are permitted provided
diff --git a/carg_parser.h b/carg_parser.h
index 0c64861..ec5d3d0 100644
--- a/carg_parser.h
+++ b/carg_parser.h
@@ -1,5 +1,5 @@
-/* Arg_parser - POSIX/GNU command line argument parser. (C version)
-   Copyright (C) 2006-2022 Antonio Diaz Diaz.
+/* Arg_parser - POSIX/GNU command-line argument parser. (C version)
+   Copyright (C) 2006-2023 Antonio Diaz Diaz.
 
    This library is free software. Redistribution and use in source and
    binary forms, with or without modification, are permitted provided
diff --git a/configure b/configure
index 72901a1..7d332b6 100755
--- a/configure
+++ b/configure
@@ -1,12 +1,12 @@
 #! /bin/sh
 # configure script for Clzip - LZMA lossless data compressor
-# Copyright (C) 2010-2022 Antonio Diaz Diaz.
+# Copyright (C) 2010-2023 Antonio Diaz Diaz.
 #
 # This configure script is free software: you have unlimited permission
 # to copy, distribute, and modify it.
 
 pkgname=clzip
-pkgversion=1.13
+pkgversion=1.14-rc1
 progname=clzip
 srctrigger=doc/${pkgname}.texi
 
@@ -24,6 +24,7 @@ CC=gcc
 CPPFLAGS=
 CFLAGS='-Wall -W -O2'
 LDFLAGS=
+MAKEINFO=makeinfo
 
 # checking whether we are using GNU C.
 /bin/sh -c "${CC} --version" > /dev/null 2>&1 || { CC=cc ; CFLAGS=-O2 ; }
@@ -57,7 +58,7 @@ while [ $# != 0 ] ; do
 		echo "Options and variables: [defaults in brackets]"
 		echo "  -h, --help            display this help and exit"
 		echo "  -V, --version         output version information and exit"
-		echo "  --srcdir=DIR          find the sources in DIR [. or ..]"
+		echo "  --srcdir=DIR          find the source code in DIR [. or ..]"
 		echo "  --prefix=DIR          install into DIR [${prefix}]"
 		echo "  --exec-prefix=DIR     base directory for arch-dependent files [${exec_prefix}]"
 		echo "  --bindir=DIR          user executables directory [${bindir}]"
@@ -65,10 +66,11 @@ while [ $# != 0 ] ; do
 		echo "  --infodir=DIR         info files directory [${infodir}]"
 		echo "  --mandir=DIR          man pages directory [${mandir}]"
 		echo "  CC=COMPILER           C compiler to use [${CC}]"
-		echo "  CPPFLAGS=OPTIONS      command line options for the preprocessor [${CPPFLAGS}]"
-		echo "  CFLAGS=OPTIONS        command line options for the C compiler [${CFLAGS}]"
+		echo "  CPPFLAGS=OPTIONS      command-line options for the preprocessor [${CPPFLAGS}]"
+		echo "  CFLAGS=OPTIONS        command-line options for the C compiler [${CFLAGS}]"
 		echo "  CFLAGS+=OPTIONS       append options to the current value of CFLAGS"
-		echo "  LDFLAGS=OPTIONS       command line options for the linker [${LDFLAGS}]"
+		echo "  LDFLAGS=OPTIONS       command-line options for the linker [${LDFLAGS}]"
+		echo "  MAKEINFO=NAME         makeinfo program to use [${MAKEINFO}]"
 		echo
 		exit 0 ;;
 	--version | -V)
@@ -96,6 +98,7 @@ while [ $# != 0 ] ; do
 	CFLAGS=*)      CFLAGS=${optarg} ;;
 	CFLAGS+=*)     CFLAGS="${CFLAGS} ${optarg}" ;;
 	LDFLAGS=*)    LDFLAGS=${optarg} ;;
+	MAKEINFO=*)  MAKEINFO=${optarg} ;;
 
 	--*)
 		echo "configure: WARNING: unrecognized option: '${option}'" 1>&2 ;;
@@ -115,7 +118,7 @@ while [ $# != 0 ] ; do
 	fi
 done
 
-# Find the source files, if location was not specified.
+# Find the source code, if location was not specified.
 srcdirtext=
 if [ -z "${srcdir}" ] ; then
 	srcdirtext="or . or .." ; srcdir=.
@@ -127,7 +130,7 @@ if [ -z "${srcdir}" ] ; then
 fi
 
 if [ ! -r "${srcdir}/${srctrigger}" ] ; then
-	echo "configure: Can't find sources in ${srcdir} ${srcdirtext}" 1>&2
+	echo "configure: Can't find source code in ${srcdir} ${srcdirtext}" 1>&2
 	echo "configure: (At least ${srctrigger} is missing)." 1>&2
 	exit 1
 fi
@@ -147,7 +150,7 @@ if [ -z "${no_create}" ] ; then
 # This script is free software: you have unlimited permission
 # to copy, distribute, and modify it.
 
-exec /bin/sh $0 ${args} --no-create
+exec /bin/sh "$0" ${args} --no-create
 EOF
 	chmod +x config.status
 fi
@@ -164,10 +167,11 @@ echo "CC = ${CC}"
 echo "CPPFLAGS = ${CPPFLAGS}"
 echo "CFLAGS = ${CFLAGS}"
 echo "LDFLAGS = ${LDFLAGS}"
+echo "MAKEINFO = ${MAKEINFO}"
 rm -f Makefile
 cat > Makefile << EOF
 # Makefile for Clzip - LZMA lossless data compressor
-# Copyright (C) 2010-2022 Antonio Diaz Diaz.
+# Copyright (C) 2010-2023 Antonio Diaz Diaz.
 # This file was generated automatically by configure. Don't edit.
 #
 # This Makefile is free software: you have unlimited permission
@@ -187,6 +191,7 @@ CC = ${CC}
 CPPFLAGS = ${CPPFLAGS}
 CFLAGS = ${CFLAGS}
 LDFLAGS = ${LDFLAGS}
+MAKEINFO = ${MAKEINFO}
 EOF
 cat "${srcdir}/Makefile.in" >> Makefile
 
diff --git a/decoder.c b/decoder.c
index fb662c5..0217fc6 100644
--- a/decoder.c
+++ b/decoder.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -98,24 +98,21 @@ void LZd_flush_data( struct LZ_decoder * const d )
   }
 
 
-static bool LZd_verify_trailer( struct LZ_decoder * const d,
-                                struct Pretty_print * const pp )
+static int LZd_check_trailer( struct LZ_decoder * const d,
+                              struct Pretty_print * const pp,
+                              const bool ignore_empty )
   {
   Lzip_trailer trailer;
   int size = Rd_read_data( d->rdec, trailer, Lt_size );
-  const unsigned long long data_size = LZd_data_position( d );
-  const unsigned long long member_size = Rd_member_position( d->rdec );
   bool error = false;
 
   if( size < Lt_size )
     {
     error = true;
     if( verbosity >= 0 )
-      {
-      Pp_show_msg( pp, 0 );
-      fprintf( stderr, "Trailer truncated at trailer position %d;"
-                       " some checks may fail.\n", size );
-      }
+      { Pp_show_msg( pp, 0 );
+        fprintf( stderr, "Trailer truncated at trailer position %d;"
+                         " some checks may fail.\n", size ); }
     while( size < Lt_size ) trailer[size++] = 0;
     }
 
@@ -124,35 +121,32 @@ static bool LZd_verify_trailer( struct LZ_decoder * const d,
     {
     error = true;
     if( verbosity >= 0 )
-      {
-      Pp_show_msg( pp, 0 );
-      fprintf( stderr, "CRC mismatch; stored %08X, computed %08X\n",
-               td_crc, LZd_crc( d ) );
-      }
+      { Pp_show_msg( pp, 0 );
+        fprintf( stderr, "CRC mismatch; stored %08X, computed %08X\n",
+                 td_crc, LZd_crc( d ) ); }
     }
+  const unsigned long long data_size = LZd_data_position( d );
   const unsigned long long td_size = Lt_get_data_size( trailer );
   if( td_size != data_size )
     {
     error = true;
     if( verbosity >= 0 )
-      {
-      Pp_show_msg( pp, 0 );
-      fprintf( stderr, "Data size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n",
-               td_size, td_size, data_size, data_size );
-      }
+      { Pp_show_msg( pp, 0 );
+        fprintf( stderr, "Data size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n",
+                 td_size, td_size, data_size, data_size ); }
     }
+  const unsigned long long member_size = Rd_member_position( d->rdec );
   const unsigned long long tm_size = Lt_get_member_size( trailer );
   if( tm_size != member_size )
     {
     error = true;
     if( verbosity >= 0 )
-      {
-      Pp_show_msg( pp, 0 );
-      fprintf( stderr, "Member size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n",
-               tm_size, tm_size, member_size, member_size );
-      }
+      { Pp_show_msg( pp, 0 );
+        fprintf( stderr, "Member size mismatch; stored %llu (0x%llX), computed %llu (0x%llX)\n",
+                 tm_size, tm_size, member_size, member_size ); }
     }
-  if( error ) return false;
+  if( error ) return 3;
+  if( !ignore_empty && data_size == 0 ) return 5;
   if( verbosity >= 2 )
     {
     if( verbosity >= 4 ) show_header( d->dictionary_size );
@@ -167,13 +161,15 @@ static bool LZd_verify_trailer( struct LZ_decoder * const d,
     if( verbosity >= 3 )
       fprintf( stderr, "%9llu out, %8llu in. ", data_size, member_size );
     }
-  return true;
+  return 0;
   }
 
 
 /* Return value: 0 = OK, 1 = decoder error, 2 = unexpected EOF,
-                 3 = trailer error, 4 = unknown marker found. */
+                 3 = trailer error, 4 = unknown marker found,
+                 5 = empty member found, 6 = marked member found. */
 int LZd_decode_member( struct LZ_decoder * const d,
+                       const struct Cl_options * const cl_opts,
                        struct Pretty_print * const pp )
   {
   struct Range_decoder * const rdec = d->rdec;
@@ -208,7 +204,7 @@ int LZd_decode_member( struct LZ_decoder * const d,
   Lm_init( &match_len_model );
   Lm_init( &rep_len_model );
 
-  Rd_load( rdec );
+  if( !Rd_load( rdec, cl_opts->ignore_marking ) ) return 6;
   while( !Rd_finished( rdec ) )
     {
     const int pos_state = LZd_data_position( d ) & pos_state_mask;
@@ -273,13 +269,9 @@ int LZd_decode_member( struct LZ_decoder * const d,
             Rd_normalize( rdec );
             LZd_flush_data( d );
             if( len == min_match_len )		/* End Of Stream marker */
-              {
-              if( LZd_verify_trailer( d, pp ) ) return 0; else return 3;
-              }
+              return LZd_check_trailer( d, pp, cl_opts->ignore_empty );
             if( len == min_match_len + 1 )	/* Sync Flush marker */
-              {
-              Rd_load( rdec ); continue;
-              }
+              { Rd_load( rdec, true ); continue; }
             if( verbosity >= 0 )
               {
               Pp_show_msg( pp, 0 );
diff --git a/decoder.h b/decoder.h
index 2823d46..3b52eee 100644
--- a/decoder.h
+++ b/decoder.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -79,13 +79,16 @@ static inline int Rd_read_data( struct Range_decoder * const rdec,
   return sz;
   }
 
-static inline void Rd_load( struct Range_decoder * const rdec )
+static inline bool Rd_load( struct Range_decoder * const rdec,
+                            const bool ignore_marking )
   {
   int i;
   rdec->code = 0;
-  for( i = 0; i < 5; ++i ) rdec->code = (rdec->code << 8) | Rd_get_byte( rdec );
   rdec->range = 0xFFFFFFFFU;
-  rdec->code &= rdec->range;	/* make sure that first byte is discarded */
+  /* check and discard first byte of the LZMA stream */
+  if( Rd_get_byte( rdec ) != 0 && !ignore_marking ) return false;
+  for( i = 0; i < 4; ++i ) rdec->code = (rdec->code << 8) | Rd_get_byte( rdec );
+  return true;
   }
 
 static inline void Rd_normalize( struct Range_decoder * const rdec )
@@ -360,4 +363,5 @@ LZd_data_position( const struct LZ_decoder * const d )
   { return d->partial_data_pos + d->pos; }
 
 int LZd_decode_member( struct LZ_decoder * const d,
+                       const struct Cl_options * const cl_opts,
                        struct Pretty_print * const pp );
diff --git a/doc/clzip.1 b/doc/clzip.1
index 3525cb4..f801503 100644
--- a/doc/clzip.1
+++ b/doc/clzip.1
@@ -1,5 +1,5 @@
-.\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.47.16.
-.TH CLZIP "1" "January 2022" "clzip 1.13" "User Commands"
+.\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.49.2.
+.TH CLZIP "1" "December 2023" "clzip 1.14-rc1" "User Commands"
 .SH NAME
 clzip \- reduces the size of files
 .SH SYNOPSIS
@@ -13,14 +13,15 @@ C++ compiler.
 .PP
 Lzip is a lossless data compressor with a user interface similar to the one
 of gzip or bzip2. Lzip uses a simplified form of the 'Lempel\-Ziv\-Markov
-chain\-Algorithm' (LZMA) stream format and provides a 3 factor integrity
-checking to maximize interoperability and optimize safety. Lzip can compress
-about as fast as gzip (lzip \fB\-0\fR) or compress most files more than bzip2
-(lzip \fB\-9\fR). Decompression speed is intermediate between gzip and bzip2.
-Lzip is better than gzip and bzip2 from a data recovery perspective. Lzip
-has been designed, written, and tested with great care to replace gzip and
-bzip2 as the standard general\-purpose compressed format for unix\-like
-systems.
+chain\-Algorithm' (LZMA) stream format to maximize interoperability. The
+maximum dictionary size is 512 MiB so that any lzip file can be decompressed
+on 32\-bit machines. Lzip provides accurate and robust 3\-factor integrity
+checking. Lzip can compress about as fast as gzip (lzip \fB\-0\fR) or compress most
+files more than bzip2 (lzip \fB\-9\fR). Decompression speed is intermediate between
+gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery
+perspective. Lzip has been designed, written, and tested with great care to
+replace gzip and bzip2 as the standard general\-purpose compressed format for
+Unix\-like systems.
 .SH OPTIONS
 .TP
 \fB\-h\fR, \fB\-\-help\fR
@@ -39,7 +40,7 @@ set member size limit in bytes
 write to standard output, keep input files
 .TP
 \fB\-d\fR, \fB\-\-decompress\fR
-decompress
+decompress, test compressed file integrity
 .TP
 \fB\-f\fR, \fB\-\-force\fR
 overwrite existing output files
@@ -83,6 +84,12 @@ alias for \fB\-0\fR
 \fB\-\-best\fR
 alias for \fB\-9\fR
 .TP
+\fB\-\-empty\-error\fR
+exit with error status if empty member in file
+.TP
+\fB\-\-marking\-error\fR
+exit with error status if 1st LZMA byte not 0
+.TP
 \fB\-\-loose\-trailing\fR
 allow trailing data seeming corrupt header
 .PP
@@ -90,24 +97,24 @@ If no file names are given, or if a file is '\-', clzip compresses or
 decompresses from standard input to standard output.
 Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,
 Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...
-Dictionary sizes 12 to 29 are interpreted as powers of two, meaning 2^12
-to 2^29 bytes.
+Dictionary sizes 12 to 29 are interpreted as powers of two, meaning 2^12 to
+2^29 bytes.
 .PP
-The bidimensional parameter space of LZMA can't be mapped to a linear
-scale optimal for all files. If your files are large, very repetitive,
-etc, you may need to use the options \fB\-\-dictionary\-size\fR and \fB\-\-match\-length\fR
-directly to achieve optimal performance.
+The bidimensional parameter space of LZMA can't be mapped to a linear scale
+optimal for all files. If your files are large, very repetitive, etc, you
+may need to use the options \fB\-\-dictionary\-size\fR and \fB\-\-match\-length\fR directly
+to achieve optimal performance.
 .PP
 To extract all the files from archive 'foo.tar.lz', use the commands
 \&'tar \fB\-xf\fR foo.tar.lz' or 'clzip \fB\-cd\fR foo.tar.lz | tar \fB\-xf\fR \-'.
 .PP
-Exit status: 0 for a normal exit, 1 for environmental problems (file
-not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or
-invalid input file, 3 for an internal consistency error (e.g., bug) which
-caused clzip to panic.
+Exit status: 0 for a normal exit, 1 for environmental problems
+(file not found, invalid command\-line options, I/O errors, etc), 2 to
+indicate a corrupt or invalid input file, 3 for an internal consistency
+error (e.g., bug) which caused clzip to panic.
 .PP
 The ideas embodied in clzip are due to (at least) the following people:
-Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the
+Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the
 definition of Markov chains), G.N.N. Martin (for the definition of range
 encoding), Igor Pavlov (for putting all the above together in LZMA), and
 Julian Seward (for bzip2's CLI).
@@ -116,7 +123,7 @@ Report bugs to lzip\-bug@nongnu.org
 .br
 Clzip home page: http://www.nongnu.org/lzip/clzip.html
 .SH COPYRIGHT
-Copyright \(co 2022 Antonio Diaz Diaz.
+Copyright \(co 2023 Antonio Diaz Diaz.
 License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html>
 .br
 This is free software: you are free to change and redistribute it.
diff --git a/doc/clzip.info b/doc/clzip.info
index c6dd8ea..8556191 100644
--- a/doc/clzip.info
+++ b/doc/clzip.info
@@ -11,13 +11,13 @@ File: clzip.info,  Node: Top,  Next: Introduction,  Up: (dir)
 Clzip Manual
 ************
 
-This manual is for Clzip (version 1.13, 24 January 2022).
+This manual is for Clzip (version 1.14-rc1, 20 December 2023).
 
 * Menu:
 
 * Introduction::           Purpose and features of clzip
 * Output::                 Meaning of clzip's output
-* Invoking clzip::         Command line interface
+* Invoking clzip::         Command-line interface
 * Quality assurance::      Design, development, and testing of lzip
 * Algorithm::              How clzip compresses the data
 * File format::            Detailed format of the compressed file
@@ -29,7 +29,7 @@ This manual is for Clzip (version 1.13, 24 January 2022).
 * Concept index::          Index of concepts
 
 
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This manual is free documentation: you have unlimited permission to copy,
 distribute, and modify it.
@@ -47,14 +47,15 @@ C++ compiler.
 
    Lzip is a lossless data compressor with a user interface similar to the
 one of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov
-chain-Algorithm' (LZMA) stream format and provides a 3 factor integrity
-checking to maximize interoperability and optimize safety. Lzip can compress
-about as fast as gzip (lzip -0) or compress most files more than bzip2
-(lzip -9). Decompression speed is intermediate between gzip and bzip2. Lzip
-is better than gzip and bzip2 from a data recovery perspective. Lzip has
-been designed, written, and tested with great care to replace gzip and
-bzip2 as the standard general-purpose compressed format for unix-like
-systems.
+chain-Algorithm' (LZMA) stream format to maximize interoperability. The
+maximum dictionary size is 512 MiB so that any lzip file can be decompressed
+on 32-bit machines. Lzip provides accurate and robust 3-factor integrity
+checking. Lzip can compress about as fast as gzip (lzip -0) or compress most
+files more than bzip2 (lzip -9). Decompression speed is intermediate between
+gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery
+perspective. Lzip has been designed, written, and tested with great care to
+replace gzip and bzip2 as the standard general-purpose compressed format for
+Unix-like systems.
 
    For compressing/decompressing large files on multiprocessor machines
 plzip can be much faster than lzip at the cost of a slightly reduced
@@ -92,22 +93,22 @@ byte near the beginning is a thing of the past.
 
    The member trailer stores the 32-bit CRC of the original data, the size
 of the original data, and the size of the member. These values, together
-with the "End Of Stream" marker, provide a 3 factor integrity checking
-which guarantees that the decompressed version of the data is identical to
-the original. This guards against corruption of the compressed data, and
-against undetected bugs in clzip (hopefully very unlikely). The chances of
-data corruption going undetected are microscopic. Be aware, though, that
-the check occurs upon decompression, so it can only tell you that something
-is wrong. It can't help you recover the original uncompressed data.
+with the "End Of Stream" marker, provide a 3-factor integrity checking which
+guarantees that the decompressed version of the data is identical to the
+original. This guards against corruption of the compressed data, and against
+undetected bugs in clzip (hopefully very unlikely). The chances of data
+corruption going undetected are microscopic. Be aware, though, that the
+check occurs upon decompression, so it can only tell you that something is
+wrong. It can't help you recover the original uncompressed data.
 
    Clzip uses the same well-defined exit status values used by bzip2, which
 makes it safer than compressors returning ambiguous warning values (like
 gzip) when it is used as a back end for other programs like tar or zutils.
 
-   Clzip will automatically use for each file the largest dictionary size
-that does not exceed neither the file size nor the limit given. Keep in
-mind that the decompression memory requirement is affected at compression
-time by the choice of dictionary size limit.
+   Clzip automatically uses for each file the largest dictionary size that
+does not exceed neither the file size nor the limit given. Keep in mind
+that the decompression memory requirement is affected at compression time
+by the choice of dictionary size limit.
 
    The amount of memory required for compression is about 1 or 2 times the
 dictionary size limit (1 if input file size is less than dictionary size
@@ -126,22 +127,22 @@ filename.tlz   becomes   filename.tar
 anyothername   becomes   anyothername.out
 
    (De)compressing a file is much like copying or moving it. Therefore clzip
-preserves the access and modification dates, permissions, and, when
-possible, ownership of the file just as 'cp -p' does. (If the user ID or
-the group ID can't be duplicated, the file permission bits S_ISUID and
-S_ISGID are cleared).
+preserves the access and modification dates, permissions, and, if you have
+appropriate privileges, ownership of the file just as 'cp -p' does. (If the
+user ID or the group ID can't be duplicated, the file permission bits
+S_ISUID and S_ISGID are cleared).
 
    Clzip is able to read from some types of non-regular files if either the
 option '-c' or the option '-o' is specified.
 
-   Clzip will refuse to read compressed data from a terminal or write
-compressed data to a terminal, as this would be entirely incomprehensible
-and might leave the terminal in an abnormal state.
+   Clzip refuses to read compressed data from a terminal or write compressed
+data to a terminal, as this would be entirely incomprehensible and might
+leave the terminal in an abnormal state.
 
-   Clzip will correctly decompress a file which is the concatenation of two
-or more compressed files. The result is the concatenation of the
-corresponding decompressed files. Integrity testing of concatenated
-compressed files is also supported.
+   Clzip correctly decompresses a file which is the concatenation of two or
+more compressed files. The result is the concatenation of the corresponding
+decompressed files. Integrity testing of concatenated compressed files is
+also supported.
 
    Clzip can produce multimember files, and lziprecover can safely recover
 the undamaged members in case of file damage. Clzip can also split the
@@ -213,7 +214,8 @@ The format for running clzip is:
 If no file names are specified, clzip compresses (or decompresses) from
 standard input to standard output. A hyphen '-' used as a FILE argument
 means standard input. It can be mixed with other FILES and is read just
-once, the first time it appears in the command line.
+once, the first time it appears in the command line. Remember to prepend
+'./' to any file name beginning with a hyphen, or use '--'.
 
    clzip supports the following options: *Note Argument syntax:
 (arg_parser)Argument syntax.
@@ -253,13 +255,14 @@ once, the first time it appears in the command line.
 
 '-d'
 '--decompress'
-     Decompress the files specified. If a file does not exist, can't be
-     opened, or the destination file already exists and '--force' has not
-     been specified, clzip continues decompressing the rest of the files
-     and exits with error status 1. If a file fails to decompress, or is a
-     terminal, clzip exits immediately with error status 2 without
-     decompressing the rest of the files. A terminal is considered an
-     uncompressed file, and therefore invalid.
+     Decompress the files specified. The integrity of the files specified is
+     checked. If a file does not exist, can't be opened, or the destination
+     file already exists and '--force' has not been specified, clzip
+     continues decompressing the rest of the files and exits with error
+     status 1. If a file fails to decompress, or is a terminal, clzip exits
+     immediately with error status 2 without decompressing the rest of the
+     files. A terminal is considered an uncompressed file, and therefore
+     invalid.
 
 '-f'
 '--force'
@@ -286,26 +289,27 @@ once, the first time it appears in the command line.
      printed.
 
      If any file is damaged, does not exist, can't be opened, or is not
-     regular, the final exit status will be > 0. '-lq' can be used to verify
+     regular, the final exit status is > 0. '-lq' can be used to check
      quickly (without decompressing) the structural integrity of the files
-     specified. (Use '--test' to verify the data integrity). '-alq'
-     additionally verifies that none of the files specified contain
-     trailing data.
+     specified. (Use '--test' to check the data integrity). '-alq'
+     additionally checks that none of the files specified contain trailing
+     data.
 
 '-m BYTES'
 '--match-length=BYTES'
      When compressing, set the match length limit in bytes. After a match
      this long is found, the search is finished. Valid values range from 5
-     to 273. Larger values usually give better compression ratios but longer
-     compression times.
+     to 273. Larger values usually give better compression ratios but
+     longer compression times.
 
 '-o FILE'
 '--output=FILE'
-     If '-c' has not been also specified, write the (de)compressed output to
-     FILE; keep input files unchanged. If compressing several files, each
-     file is compressed independently. (The output consists of a sequence of
-     independently compressed members). This option (or '-c') is needed when
-     reading from a named pipe (fifo) or from a device. '-o -' is
+     If '-c' has not been also specified, write the (de)compressed output
+     to FILE, automatically creating any missing parent directories; keep
+     input files unchanged. If compressing several files, each file is
+     compressed independently. (The output consists of a sequence of
+     independently compressed members). This option (or '-c') is needed
+     when reading from a named pipe (fifo) or from a device. '-o -' is
      equivalent to '-c'. '-o' has no effect when testing or listing.
 
      In order to keep backward compatibility with clzip versions prior to
@@ -326,14 +330,14 @@ once, the first time it appears in the command line.
 
 '-s BYTES'
 '--dictionary-size=BYTES'
-     When compressing, set the dictionary size limit in bytes. Clzip will
-     use for each file the largest dictionary size that does not exceed
-     neither the file size nor this limit. Valid values range from 4 KiB to
-     512 MiB. Values 12 to 29 are interpreted as powers of two, meaning
-     2^12 to 2^29 bytes. Dictionary sizes are quantized so that they can be
-     coded in just one byte (*note coded-dict-size::). If the size specified
-     does not match one of the valid sizes, it will be rounded upwards by
-     adding up to (BYTES / 8) to it.
+     When compressing, set the dictionary size limit in bytes. Clzip uses
+     for each file the largest dictionary size that does not exceed neither
+     the file size nor this limit. Valid values range from 4 KiB to 512 MiB.
+     Values 12 to 29 are interpreted as powers of two, meaning 2^12 to 2^29
+     bytes. Dictionary sizes are quantized so that they can be coded in
+     just one byte (*note coded-dict-size::). If the size specified does
+     not match one of the valid sizes, it is rounded upwards by adding up
+     to (BYTES / 8) to it.
 
      For maximum compression you should use a dictionary size limit as large
      as possible, but keep in mind that the decompression memory requirement
@@ -355,7 +359,7 @@ once, the first time it appears in the command line.
      really performs a trial decompression and throws away the result. Use
      it together with '-v' to see information about the files. If a file
      fails the test, does not exist, can't be opened, or is a terminal,
-     clzip continues checking the rest of the files. A final diagnostic is
+     clzip continues testing the rest of the files. A final diagnostic is
      shown at verbosity level 1 or higher if any file fails the test when
      testing multiple files.
 
@@ -403,6 +407,16 @@ once, the first time it appears in the command line.
 '--best'
      Aliases for GNU gzip compatibility.
 
+'--empty-error'
+     Exit with error status 2 if any empty member is found in the input
+     files.
+
+'--marking-error'
+     Exit with error status 2 if the first LZMA byte is non-zero in any
+     member of the input files. This may be caused by data corruption or by
+     deliberate insertion of tracking information in the file. Use
+     'lziprecover --clear-marking' to clear any such non-zero bytes.
+
 '--loose-trailing'
      When decompressing, testing, or listing, allow trailing data whose
      first bytes are so similar to the magic bytes of a lzip header that
@@ -411,26 +425,29 @@ once, the first time it appears in the command line.
      corrupt header.
 
 
-   Numbers given as arguments to options may be followed by a multiplier
-and an optional 'B' for "byte".
+   Numbers given as arguments to options may be expressed in decimal,
+hexadecimal, or octal (using the same syntax as integer constants in C++),
+and may be followed by a multiplier and an optional 'B' for "byte".
 
    Table of SI and binary prefixes (unit multipliers):
 
-Prefix   Value                     |   Prefix   Value
-k        kilobyte  (10^3 = 1000)   |   Ki       kibibyte (2^10 = 1024)
-M        megabyte  (10^6)          |   Mi       mebibyte (2^20)
-G        gigabyte  (10^9)          |   Gi       gibibyte (2^30)
-T        terabyte  (10^12)         |   Ti       tebibyte (2^40)
-P        petabyte  (10^15)         |   Pi       pebibyte (2^50)
-E        exabyte   (10^18)         |   Ei       exbibyte (2^60)
-Z        zettabyte (10^21)         |   Zi       zebibyte (2^70)
-Y        yottabyte (10^24)         |   Yi       yobibyte (2^80)
+Prefix   Value                      |   Prefix   Value
+k        kilobyte   (10^3 = 1000)   |   Ki       kibibyte  (2^10 = 1024)
+M        megabyte   (10^6)          |   Mi       mebibyte  (2^20)
+G        gigabyte   (10^9)          |   Gi       gibibyte  (2^30)
+T        terabyte   (10^12)         |   Ti       tebibyte  (2^40)
+P        petabyte   (10^15)         |   Pi       pebibyte  (2^50)
+E        exabyte    (10^18)         |   Ei       exbibyte  (2^60)
+Z        zettabyte  (10^21)         |   Zi       zebibyte  (2^70)
+Y        yottabyte  (10^24)         |   Yi       yobibyte  (2^80)
+R        ronnabyte  (10^27)         |   Ri       robibyte  (2^90)
+Q        quettabyte (10^30)         |   Qi       quebibyte (2^100)
 
 
    Exit status: 0 for a normal exit, 1 for environmental problems (file not
-found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or invalid
-input file, 3 for an internal consistency error (e.g., bug) which caused
-clzip to panic.
+found, invalid command-line options, I/O errors, etc), 2 to indicate a
+corrupt or invalid input file, 3 for an internal consistency error (e.g.,
+bug) which caused clzip to panic.
 
 
 File: clzip.info,  Node: Quality assurance,  Next: Algorithm,  Prev: Invoking clzip,  Up: Top
@@ -444,6 +461,11 @@ make it so complicated that there are no obvious deficiencies. The first
 method is far more difficult.
 -- C.A.R. Hoare
 
+   Lzip has been designed, written, and tested with great care to replace
+gzip and bzip2 as the standard general-purpose compressed format for
+Unix-like systems. This chapter describes the lessons learned from these
+previous formats, and their application to the design of lzip.
+
    Lzip is developed by volunteers who lack the resources required for
 extensive testing in all circumstances. It is up to you to test lzip before
 using it in mission-critical applications. However, a compressor like lzip
@@ -451,11 +473,6 @@ is not a toy, and maintaining it is not a hobby. Many people's data depend
 on it. Therefore the lzip file format has been reviewed carefully and is
 believed to be free from negligent design errors.
 
-   Lzip has been designed, written, and tested with great care to replace
-gzip and bzip2 as the standard general-purpose compressed format for
-unix-like systems. This chapter describes the lessons learned from these
-previous formats, and their application to the design of lzip.
-
 
 4.1 Format design
 =================
@@ -537,10 +554,10 @@ extraction of the decompressed data.
      Using an optional CRC for the header is not only a bad idea, it is an
      error; it circumvents the Hamming distance (HD) of the CRC and may
      prevent the extraction of perfectly good data. For example, if the CRC
-     is used and the bit enabling it is reset by a bit flip, the header
-     will appear to be intact (in spite of being corrupt) while the
-     compressed blocks will appear to be totally unrecoverable (in spite of
-     being intact). Very misleading indeed.
+     is used and the bit enabling it is reset by a bit flip, then the
+     header seems to be intact (in spite of being corrupt) while the
+     compressed blocks seem to be totally unrecoverable (in spite of being
+     intact). Very misleading indeed.
 
 'Metadata'
      The gzip format stores some metadata, like the modification time of the
@@ -555,8 +572,8 @@ extraction of the decompressed data.
 '64-bit size field'
      Probably the most frequently reported shortcoming of the gzip format
      is that it only stores the least significant 32 bits of the
-     uncompressed size. The size of any file larger than 4 GiB gets
-     truncated.
+     uncompressed size. The size of any file larger or equal than 4 GiB
+     gets truncated.
 
      Bzip2 does not store the uncompressed size of the file.
 
@@ -580,8 +597,12 @@ extraction of the decompressed data.
 4.2 Quality of implementation
 =============================
 
+Our civilization depends critically on software; it had better be quality
+software.
+-- Bjarne Stroustrup
+
 'Accurate and robust error detection'
-     The lzip format provides 3 factor integrity checking, and the
+     The lzip format provides 3-factor integrity checking, and the
      decompressors report mismatches in each factor separately. This method
      detects most false positives for corruption. If just one byte in one
      factor fails but the other two factors match the data, it probably
@@ -590,15 +611,15 @@ extraction of the decompressed data.
      trailer.
 
 'Multiple implementations'
-     Just like the lzip format provides 3 factor protection against
+     Just like the lzip format provides 3-factor protection against
      undetected data corruption, the development methodology of the lzip
-     family of compressors provides 3 factor protection against undetected
+     family of compressors provides 3-factor protection against undetected
      programming errors.
 
      Three related but independent compressor implementations, lzip, clzip,
      and minilzip/lzlib, are developed concurrently. Every stable release
-     of any of them is tested to verify that it produces identical output
-     to the other two. This guarantees that all three implement the same
+     of any of them is tested to check that it produces identical output to
+     the other two. This guarantees that all three implement the same
      algorithm, and makes it unlikely that any of them may contain serious
      undiscovered errors. In fact, no errors have been discovered in lzip
      since 2009.
@@ -642,10 +663,10 @@ using the LZMA coding scheme.
 (used by option '-0') and normal (used by all other compression levels).
 
    The high compression of LZMA comes from combining two basic, well-proven
-compression ideas: sliding dictionaries (LZ77/78) and markov models (the
-thing used by every compression algorithm that uses a range encoder or
-similar order-0 entropy coder as its last stage) with segregation of
-contexts according to what the bits are used for.
+compression ideas: sliding dictionaries (LZ77) and markov models (the thing
+used by every compression algorithm that uses a range encoder or similar
+order-0 entropy coder as its last stage) with segregation of contexts
+according to what the bits are used for.
 
    Clzip is a two stage compressor. The first stage is a Lempel-Ziv coder,
 which reduces redundancy by translating chunks of data to their
@@ -690,7 +711,7 @@ intervals get longer with higher compression levels because dictionary size
 increases (and compression speed decreases) with compression level.
 
 The ideas embodied in clzip are due to (at least) the following people:
-Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the
+Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the
 definition of Markov chains), G.N.N. Martin (for the definition of range
 encoding), Igor Pavlov (for putting all the above together in LZMA), and
 Julian Seward (for bzip2's CLI).
@@ -721,7 +742,7 @@ when there is no longer anything to take away.
    represents a variable number of bytes.
 
 
-   A lzip file consists of a series of independent "members" (compressed
+   A lzip file consists of one or more independent "members" (compressed
 data sets). The members simply appear one after another in the file, with no
 additional information before, between, or after them. Each member can
 encode in compressed form up to 16 EiB - 1 byte of uncompressed data. The
@@ -765,10 +786,10 @@ size of a multimember file is unlimited.
 
 'Member size (8 bytes)'
      Total size of the member, including header and trailer. This field acts
-     as a distributed index, allows the verification of stream integrity,
-     and facilitates the safe recovery of undamaged members from
-     multimember files. Member size should be limited to 2 PiB to prevent
-     the data size field from overflowing.
+     as a distributed index, improves the checking of stream integrity, and
+     facilitates the safe recovery of undamaged members from multimember
+     files. Lzip limits the member size to 2 PiB to prevent the data size
+     field from overflowing.
 
 
 
@@ -788,12 +809,12 @@ in the code.
 
    Lzip finishes the LZMA stream with an "End Of Stream" (EOS) marker (the
 distance-length pair 0xFFFFFFFFU, 2), which in conjunction with the 'member
-size' field in the member trailer allows the verification of stream
-integrity. The EOS marker is the only marker allowed in lzip files. The
-LZMA stream in lzip files always has these two features (default properties
-and EOS marker) and is referred to in this document as LZMA-302eos. This
-simplified form of the LZMA stream format has been chosen to maximize
-interoperability and safety.
+size' field in the member trailer allows the checking of stream integrity.
+The EOS marker is the only LZMA marker allowed in lzip files. The LZMA
+stream in lzip files always has these two features (default properties and
+EOS marker) and is referred to in this document as LZMA-302eos. This
+simplified and marker-terminated form of the LZMA stream format has been
+chosen to maximize interoperability and safety.
 
    The second stage of LZMA is a range encoder that uses a different
 probability model for each type of symbol: distances, lengths, literal
@@ -811,9 +832,9 @@ a real decompressor seems the only appropriate reference to use.
 
    What follows is a description of the decoding algorithm for LZMA-302eos
 streams using as reference the source code of "lzd", an educational
-decompressor for lzip files which can be downloaded from the lzip download
-directory. Lzd is written in C++11 and its source code is included in
-appendix A. *Note Reference source code::.
+decompressor for lzip files, included in appendix A. *Note Reference source
+code::. Lzd is written in C++11 and can be downloaded from the lzip download
+directory.
 
 
 7.1 What is coded
@@ -878,10 +899,10 @@ the distance is >= 4, the remaining bits are encoded as follows.
 'direct_bits' is the amount of remaining bits (from 1 to 30) needed to form
 a complete distance, and is calculated as (slot >> 1) - 1. If a distance
 needs 6 or more direct_bits, the last 4 bits are encoded separately. The
-last piece (all the direct_bits for distances 4 to 127, or the last 4 bits
-for distances >= 128) is context-coded in reverse order (from LSB to MSB).
-For distances >= 128, the 'direct_bits - 4' part is encoded with fixed 0.5
-probability.
+last piece (all the direct_bits for distances 4 to 127 (slots 4 to 13), or
+the last 4 bits for distances >= 128 (slot >= 14)) is context-coded in
+reverse order (from LSB to MSB). For distances >= 128, the
+'direct_bits - 4' part is encoded with fixed 0.5 probability.
 
 Bit sequence                           Description
 ----------------------------------------------------------------------------
@@ -999,8 +1020,8 @@ range decoder. This is done by shifting 5 bytes in the initialization of
 'code' instead of 4. (See the 'Range_decoder' constructor in the source).
 
 
-7.4 Decoding and verifying the LZMA stream
-==========================================
+7.4 Decoding and checking the LZMA stream
+=========================================
 
 After decoding the member header and obtaining the dictionary size, the
 range decoder is initialized and then the LZMA decoder enters a loop (see
@@ -1010,9 +1031,9 @@ repeated matches, and literal bytes), until the "End Of Stream" marker is
 decoded.
 
    Once the "End Of Stream" marker has been decoded, the decompressor reads
-and decodes the member trailer, and verifies that the three integrity
-factors stored there (CRC, data size, and member size) match those computed
-from the data.
+and decodes the member trailer, and checks that the three integrity factors
+stored there (CRC, data size, and member size) match those computed from the
+data.
 
 
 File: clzip.info,  Node: Trailing data,  Next: Examples,  Prev: Stream format,  Up: Top
@@ -1027,12 +1048,13 @@ member. Such trailing data may be:
      example when writing to a tape. It is safe to append any amount of
      padding zero bytes to a lzip file.
 
-   * Useful data added by the user; a cryptographically secure hash, a
+   * Useful data added by the user; an "End Of File" string (to check that
+     the file has not been truncated), a cryptographically secure hash, a
      description of file contents, etc. It is safe to append any amount of
-     text to a lzip file as long as none of the first four bytes of the text
-     match the corresponding byte in the string "LZIP", and the text does
-     not contain any zero bytes (null characters). Nonzero bytes and zero
-     bytes can't be safely mixed in trailing data.
+     text to a lzip file as long as none of the first four bytes of the
+     text matches the corresponding byte in the string "LZIP", and the text
+     does not contain any zero bytes (null characters). Nonzero bytes and
+     zero bytes can't be safely mixed in trailing data.
 
    * Garbage added by some not totally successful copy operation.
 
@@ -1048,7 +1070,7 @@ member. Such trailing data may be:
      discriminate trailing data from a corrupt header has a Hamming
      distance (HD) of 3, and the 3 bit flips must happen in different magic
      bytes for the test to fail. In any case, the option '--trailing-error'
-     guarantees that any corrupt header will be detected.
+     guarantees that any corrupt header is detected.
 
    Trailing data are in no way part of the lzip file format, but tools
 reading lzip files are expected to behave as correctly and usefully as
@@ -1068,12 +1090,12 @@ File: clzip.info,  Node: Examples,  Next: Problems,  Prev: Trailing data,  Up: T
 WARNING! Even if clzip is bug-free, other causes may result in a corrupt
 compressed file (bugs in the system libraries, memory errors, etc).
 Therefore, if the data you are going to compress are important, give the
-option '--keep' to clzip and don't remove the original file until you
-verify the compressed file with a command like
-'clzip -cd file.lz | cmp file -'. Most RAM errors happening during
-compression can only be detected by comparing the compressed file with the
-original because the corruption happens before clzip compresses the RAM
-contents, resulting in a valid compressed file containing wrong data.
+option '--keep' to clzip and don't remove the original file until you check
+the compressed file with a command like 'clzip -cd file.lz | cmp file -'.
+Most RAM errors happening during compression can only be detected by
+comparing the compressed file with the original because the corruption
+happens before clzip compresses the RAM contents, resulting in a valid
+compressed file containing wrong data.
 
 
 Example 1: Extract all the files from archive 'foo.tar.lz'.
@@ -1101,7 +1123,7 @@ the operation is successful, 'file.lz' is removed.
      clzip -d file.lz
 
 
-Example 5: Verify the integrity of the compressed file 'file.lz' and show
+Example 5: Check the integrity of the compressed file 'file.lz' and show
 status.
 
      clzip -tv file.lz
@@ -1175,7 +1197,7 @@ Appendix A Reference source code
 ********************************
 
 /* Lzd - Educational decompressor for the lzip format
-   Copyright (C) 2013-2022 Antonio Diaz Diaz.
+   Copyright (C) 2013-2023 Antonio Diaz Diaz.
 
    This program is free software. Redistribution and use in source and
    binary forms, with or without modification, are permitted provided
@@ -1194,8 +1216,8 @@ Appendix A Reference source code
 */
 /*
    Exit status: 0 for a normal exit, 1 for environmental problems
-   (file not found, invalid flags, I/O errors, etc), 2 to indicate a
-   corrupt or invalid input file.
+   (file not found, invalid command-line options, I/O errors, etc), 2 to
+   indicate a corrupt or invalid input file.
 */
 
 #include <algorithm>
@@ -1306,10 +1328,11 @@ public:
 const CRC32 crc32;
 
 
-typedef uint8_t Lzip_header[6];		// 0-3 magic bytes
-					//   4 version
-					//   5 coded dictionary size
-typedef uint8_t Lzip_trailer[20];
+enum { header_size = 6, trailer_size = 20 };
+typedef uint8_t Lzip_header[header_size]; // 0-3 magic bytes
+					  //   4 version
+					  //   5 coded dictionary size
+typedef uint8_t Lzip_trailer[trailer_size];
 			//  0-3  CRC32 of the uncompressed data
 			//  4-11 size of the uncompressed data
 			// 12-19 member size including header and trailer
@@ -1321,9 +1344,11 @@ class Range_decoder
   uint32_t range;
 
 public:
-  Range_decoder() : member_pos( 6 ), code( 0 ), range( 0xFFFFFFFFU )
+  Range_decoder()
+    : member_pos( header_size ), code( 0 ), range( 0xFFFFFFFFU )
     {
-    for( int i = 0; i < 5; ++i ) code = ( code << 8 ) | get_byte();
+    get_byte();			// discard first byte of the LZMA stream
+    for( int i = 0; i < 4; ++i ) code = ( code << 8 ) | get_byte();
     }
 
   uint8_t get_byte() { ++member_pos; return std::getc( stdin ); }
@@ -1356,8 +1381,8 @@ public:
       }
     else
       {
-      range -= bound;
       code -= bound;
+      range -= bound;
       bm.probability -= bm.probability >> bit_model_move_bits;
       symbol = 1;
       }
@@ -1407,11 +1432,12 @@ public:
   unsigned decode_len( Len_model & lm, const int pos_state )
     {
     if( decode_bit( lm.choice1 ) == 0 )
-      return decode_tree( lm.bm_low[pos_state], len_low_bits );
+      return min_match_len +
+             decode_tree( lm.bm_low[pos_state], len_low_bits );
     if( decode_bit( lm.choice2 ) == 0 )
-      return len_low_symbols +
+      return min_match_len + len_low_symbols +
              decode_tree( lm.bm_mid[pos_state], len_mid_bits );
-    return len_low_symbols + len_mid_symbols +
+    return min_match_len + len_low_symbols + len_mid_symbols +
            decode_tree( lm.bm_high, len_high_bits );
     }
   };
@@ -1484,7 +1510,7 @@ void LZ_decoder::flush_data()
   }
 
 
-bool LZ_decoder::decode_member()	// Returns false if error
+bool LZ_decoder::decode_member()	// Return false if error
   {
   Bit_model bm_literal[1<<literal_context_bits][0x300];
   Bit_model bm_match[State::states][pos_states];
@@ -1546,12 +1572,12 @@ bool LZ_decoder::decode_member()	// Returns false if error
         rep0 = distance;
         }
       state.set_rep();
-      len = min_match_len + rdec.decode_len( rep_len_model, pos_state );
+      len = rdec.decode_len( rep_len_model, pos_state );
       }
     else					// match
       {
       rep3 = rep2; rep2 = rep1; rep1 = rep0;
-      len = min_match_len + rdec.decode_len( match_len_model, pos_state );
+      len = rdec.decode_len( match_len_model, pos_state );
       const int len_state = std::min( len - min_match_len, len_states - 1 );
       rep0 = rdec.decode_tree( bm_dis_slot[len_state], dis_slot_bits );
       if( rep0 >= start_dis_model )
@@ -1570,7 +1596,7 @@ bool LZ_decoder::decode_member()	// Returns false if error
           if( rep0 == 0xFFFFFFFFU )		// marker found
             {
             flush_data();
-            return ( len == min_match_len );	// End Of Stream marker
+            return len == min_match_len;	// End Of Stream marker
             }
           }
         }
@@ -1591,11 +1617,11 @@ int main( const int argc, const char * const argv[] )
     {
     std::printf(
       "Lzd %s - Educational decompressor for the lzip format.\n"
-      "Study the source to learn how a lzip decompressor works.\n"
+      "Study the source code to learn how a lzip decompressor works.\n"
       "See the lzip manual for an explanation of the code.\n"
       "\nUsage: %s [-d] < file.lz > file\n"
       "Lzd decompresses from standard input to standard output.\n"
-      "\nCopyright (C) 2022 Antonio Diaz Diaz.\n"
+      "\nCopyright (C) 2023 Antonio Diaz Diaz.\n"
       "License 2-clause BSD.\n"
       "This is free software: you are free to change and redistribute it.\n"
       "There is NO WARRANTY, to the extent permitted by law.\n"
@@ -1612,8 +1638,8 @@ int main( const int argc, const char * const argv[] )
 
   for( bool first_member = true; ; first_member = false )
     {
-    Lzip_header header;				// verify header
-    for( int i = 0; i < 6; ++i ) header[i] = std::getc( stdin );
+    Lzip_header header;				// check header
+    for( int i = 0; i < header_size; ++i ) header[i] = std::getc( stdin );
     if( std::feof( stdin ) || std::memcmp( header, "LZIP\x01", 5 ) != 0 )
       {
       if( first_member )
@@ -1631,8 +1657,8 @@ int main( const int argc, const char * const argv[] )
     if( !decoder.decode_member() )
       { std::fputs( "Data error\n", stderr ); return 2; }
 
-    Lzip_trailer trailer;			// verify trailer
-    for( int i = 0; i < 20; ++i ) trailer[i] = decoder.get_byte();
+    Lzip_trailer trailer;			// check trailer
+    for( int i = 0; i < trailer_size; ++i ) trailer[i] = decoder.get_byte();
     int retval = 0;
     unsigned crc = 0;
     for( int i = 3; i >= 0; --i ) crc = ( crc << 8 ) + trailer[i];
@@ -1688,22 +1714,22 @@ Concept index
 
 Tag Table:
 Node: Top205
-Node: Introduction1207
-Node: Output7226
-Node: Invoking clzip8829
-Ref: --trailing-error9627
-Node: Quality assurance18961
-Node: Algorithm27986
-Node: File format31397
-Ref: coded-dict-size32827
-Node: Stream format34062
-Ref: what-is-coded36459
-Node: Trailing data45387
-Node: Examples47650
-Ref: concat-example49102
-Node: Problems50332
-Node: Reference source code50868
-Node: Concept index65727
+Node: Introduction1212
+Node: Output7342
+Node: Invoking clzip8945
+Ref: --trailing-error9823
+Node: Quality assurance19929
+Node: Algorithm29060
+Node: File format32468
+Ref: coded-dict-size33898
+Node: Stream format35129
+Ref: what-is-coded37525
+Node: Trailing data46478
+Node: Examples48816
+Ref: concat-example50266
+Node: Problems51496
+Node: Reference source code52032
+Node: Concept index67094
 
 End Tag Table
 
diff --git a/doc/clzip.texi b/doc/clzip.texi
index c00599c..633526b 100644
--- a/doc/clzip.texi
+++ b/doc/clzip.texi
@@ -6,8 +6,8 @@
 @finalout
 @c %**end of header
 
-@set UPDATED 24 January 2022
-@set VERSION 1.13
+@set UPDATED 20 December 2023
+@set VERSION 1.14-rc1
 
 @dircategory Compression
 @direntry
@@ -38,7 +38,7 @@ This manual is for Clzip (version @value{VERSION}, @value{UPDATED}).
 @menu
 * Introduction::           Purpose and features of clzip
 * Output::                 Meaning of clzip's output
-* Invoking clzip::         Command line interface
+* Invoking clzip::         Command-line interface
 * Quality assurance::      Design, development, and testing of lzip
 * Algorithm::              How clzip compresses the data
 * File format::            Detailed format of the compressed file
@@ -51,7 +51,7 @@ This manual is for Clzip (version @value{VERSION}, @value{UPDATED}).
 @end menu
 
 @sp 1
-Copyright @copyright{} 2010-2022 Antonio Diaz Diaz.
+Copyright @copyright{} 2010-2023 Antonio Diaz Diaz.
 
 This manual is free documentation: you have unlimited permission to copy,
 distribute, and modify it.
@@ -71,14 +71,15 @@ C++ compiler.
 @uref{http://www.nongnu.org/lzip/lzip.html,,Lzip}
 is a lossless data compressor with a user interface similar to the one
 of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov
-chain-Algorithm' (LZMA) stream format and provides a 3 factor integrity
-checking to maximize interoperability and optimize safety. Lzip can compress
-about as fast as gzip @w{(lzip -0)} or compress most files more than bzip2
-@w{(lzip -9)}. Decompression speed is intermediate between gzip and bzip2.
-Lzip is better than gzip and bzip2 from a data recovery perspective. Lzip
-has been designed, written, and tested with great care to replace gzip and
-bzip2 as the standard general-purpose compressed format for unix-like
-systems.
+chain-Algorithm' (LZMA) stream format to maximize interoperability. The
+maximum dictionary size is 512 MiB so that any lzip file can be decompressed
+on 32-bit machines. Lzip provides accurate and robust 3-factor integrity
+checking. Lzip can compress about as fast as gzip @w{(lzip -0)} or compress most
+files more than bzip2 @w{(lzip -9)}. Decompression speed is intermediate between
+gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery
+perspective. Lzip has been designed, written, and tested with great care to
+replace gzip and bzip2 as the standard general-purpose compressed format for
+Unix-like systems.
 
 For compressing/decompressing large files on multiprocessor machines
 @uref{http://www.nongnu.org/lzip/manual/plzip_manual.html,,plzip} can be
@@ -128,30 +129,29 @@ the nearer it is from the beginning of the file. Therefore, with the help of
 lziprecover, losing an entire archive just because of a corrupt byte near
 the beginning is a thing of the past.
 
-The member trailer stores the 32-bit CRC of the original data, the size
-of the original data, and the size of the member. These values, together
-with the "End Of Stream" marker, provide a 3 factor integrity checking
-which guarantees that the decompressed version of the data is identical
-to the original. This guards against corruption of the compressed data,
-and against undetected bugs in clzip (hopefully very unlikely). The
-chances of data corruption going undetected are microscopic. Be aware,
-though, that the check occurs upon decompression, so it can only tell
-you that something is wrong. It can't help you recover the original
-uncompressed data.
+The member trailer stores the 32-bit CRC of the original data, the size of
+the original data, and the size of the member. These values, together with
+the "End Of Stream" marker, provide a 3-factor integrity checking which
+guarantees that the decompressed version of the data is identical to the
+original. This guards against corruption of the compressed data, and against
+undetected bugs in clzip (hopefully very unlikely). The chances of data
+corruption going undetected are microscopic. Be aware, though, that the
+check occurs upon decompression, so it can only tell you that something is
+wrong. It can't help you recover the original uncompressed data.
 
 Clzip uses the same well-defined exit status values used by bzip2, which
 makes it safer than compressors returning ambiguous warning values (like
 gzip) when it is used as a back end for other programs like tar or zutils.
 
-Clzip will automatically use for each file the largest dictionary size that
-does not exceed neither the file size nor the limit given. Keep in mind that
-the decompression memory requirement is affected at compression time by the
+Clzip automatically uses for each file the largest dictionary size that does
+not exceed neither the file size nor the limit given. Keep in mind that the
+decompression memory requirement is affected at compression time by the
 choice of dictionary size limit.
 
 The amount of memory required for compression is about 1 or 2 times the
 dictionary size limit (1 if input file size is less than dictionary size
 limit, else 2) plus 9 times the dictionary size really used. The option
-@samp{-0} is special and only requires about @w{1.5 MiB} at most. The
+@option{-0} is special and only requires about @w{1.5 MiB} at most. The
 amount of memory required for decompression is about @w{46 kB} larger
 than the dictionary size really used.
 
@@ -167,19 +167,19 @@ file from that of the compressed file as follows:
 @end multitable
 
 (De)compressing a file is much like copying or moving it. Therefore clzip
-preserves the access and modification dates, permissions, and, when
-possible, ownership of the file just as @w{@samp{cp -p}} does. (If the user ID or
-the group ID can't be duplicated, the file permission bits S_ISUID and
-S_ISGID are cleared).
+preserves the access and modification dates, permissions, and, if you have
+appropriate privileges, ownership of the file just as @w{@samp{cp -p}} does.
+(If the user ID or the group ID can't be duplicated, the file permission
+bits S_ISUID and S_ISGID are cleared).
 
 Clzip is able to read from some types of non-regular files if either the
-option @samp{-c} or the option @samp{-o} is specified.
+option @option{-c} or the option @option{-o} is specified.
 
-Clzip will refuse to read compressed data from a terminal or write compressed
+Clzip refuses to read compressed data from a terminal or write compressed
 data to a terminal, as this would be entirely incomprehensible and might
 leave the terminal in an abnormal state.
 
-Clzip will correctly decompress a file which is the concatenation of two or
+Clzip correctly decompresses a file which is the concatenation of two or
 more compressed files. The result is the concatenation of the corresponding
 decompressed files. Integrity testing of concatenated compressed files is
 also supported.
@@ -261,7 +261,8 @@ clzip [@var{options}] [@var{files}]
 If no file names are specified, clzip compresses (or decompresses) from
 standard input to standard output. A hyphen @samp{-} used as a @var{file}
 argument means standard input. It can be mixed with other @var{files} and is
-read just once, the first time it appears in the command line.
+read just once, the first time it appears in the command line. Remember to
+prepend @file{./} to any file name beginning with a hyphen, or use @samp{--}.
 
 clzip supports the following
 @uref{http://www.nongnu.org/arg-parser/manual/arg_parser_manual.html#Argument-syntax,,options}:
@@ -299,19 +300,20 @@ compression ratio, so use it only when needed. Valid values range from
 Compress or decompress to standard output; keep input files unchanged. If
 compressing several files, each file is compressed independently. (The
 output consists of a sequence of independently compressed members). This
-option (or @samp{-o}) is needed when reading from a named pipe (fifo) or
+option (or @option{-o}) is needed when reading from a named pipe (fifo) or
 from a device. Use it also to recover as much of the decompressed data as
-possible when decompressing a corrupt file. @samp{-c} overrides @samp{-o}
-and @samp{-S}. @samp{-c} has no effect when testing or listing.
+possible when decompressing a corrupt file. @option{-c} overrides @option{-o}
+and @option{-S}. @option{-c} has no effect when testing or listing.
 
 @item -d
 @itemx --decompress
-Decompress the files specified. If a file does not exist, can't be opened,
-or the destination file already exists and @samp{--force} has not been
-specified, clzip continues decompressing the rest of the files and exits with
-error status 1. If a file fails to decompress, or is a terminal, clzip exits
-immediately with error status 2 without decompressing the rest of the files.
-A terminal is considered an uncompressed file, and therefore invalid.
+Decompress the files specified. The integrity of the files specified is
+checked. If a file does not exist, can't be opened, or the destination file
+already exists and @option{--force} has not been specified, clzip continues
+decompressing the rest of the files and exits with error status 1. If a file
+fails to decompress, or is a terminal, clzip exits immediately with error
+status 2 without decompressing the rest of the files. A terminal is
+considered an uncompressed file, and therefore invalid.
 
 @item -f
 @itemx --force
@@ -331,38 +333,39 @@ Keep (don't delete) input files during compression or decompression.
 Print the uncompressed size, compressed size, and percentage saved of the
 files specified. Trailing data are ignored. The values produced are correct
 even for multimember files. If more than one file is given, a final line
-containing the cumulative sizes is printed. With @samp{-v}, the dictionary
+containing the cumulative sizes is printed. With @option{-v}, the dictionary
 size, the number of members in the file, and the amount of trailing data (if
-any) are also printed. With @samp{-vv}, the positions and sizes of each
+any) are also printed. With @option{-vv}, the positions and sizes of each
 member in multimember files are also printed.
 
 If any file is damaged, does not exist, can't be opened, or is not regular,
-the final exit status will be @w{> 0}. @samp{-lq} can be used to verify
-quickly (without decompressing) the structural integrity of the files
-specified. (Use @samp{--test} to verify the data integrity). @samp{-alq}
-additionally verifies that none of the files specified contain trailing data.
+the final exit status is @w{> 0}. @option{-lq} can be used to check quickly
+(without decompressing) the structural integrity of the files specified.
+(Use @option{--test} to check the data integrity). @option{-alq}
+additionally checks that none of the files specified contain trailing data.
 
 @item -m @var{bytes}
 @itemx --match-length=@var{bytes}
-When compressing, set the match length limit in bytes. After a match
-this long is found, the search is finished. Valid values range from 5 to
-273. Larger values usually give better compression ratios but longer
-compression times.
+When compressing, set the match length limit in bytes. After a match this
+long is found, the search is finished. Valid values range from 5 to 273.
+Larger values usually give better compression ratios but longer compression
+times.
 
 @item -o @var{file}
 @itemx --output=@var{file}
-If @samp{-c} has not been also specified, write the (de)compressed output to
-@var{file}; keep input files unchanged. If compressing several files, each
-file is compressed independently. (The output consists of a sequence of
-independently compressed members). This option (or @samp{-c}) is needed when
-reading from a named pipe (fifo) or from a device. @w{@samp{-o -}} is
-equivalent to @samp{-c}. @samp{-o} has no effect when testing or listing.
+If @option{-c} has not been also specified, write the (de)compressed output
+to @var{file}, automatically creating any missing parent directories; keep
+input files unchanged. If compressing several files, each file is compressed
+independently. (The output consists of a sequence of independently
+compressed members). This option (or @option{-c}) is needed when reading
+from a named pipe (fifo) or from a device. @w{@option{-o -}} is equivalent
+to @option{-c}. @option{-o} has no effect when testing or listing.
 
 In order to keep backward compatibility with clzip versions prior to 1.12,
 when compressing from standard input and no other file names are given, the
 extension @samp{.lz} is appended to @var{file} unless it already ends in
 @samp{.lz} or @samp{.tlz}. This feature will be removed in a future version
-of clzip. Meanwhile, redirection may be used instead of @samp{-o} to write
+of clzip. Meanwhile, redirection may be used instead of @option{-o} to write
 the compressed output to a file without the extension @samp{.lz} in its
 name: @w{@samp{clzip < file > foo}}.
 
@@ -377,14 +380,14 @@ Quiet operation. Suppress all messages.
 
 @item -s @var{bytes}
 @itemx --dictionary-size=@var{bytes}
-When compressing, set the dictionary size limit in bytes. Clzip will use
-for each file the largest dictionary size that does not exceed neither
-the file size nor this limit. Valid values range from @w{4 KiB} to
-@w{512 MiB}. Values 12 to 29 are interpreted as powers of two, meaning
-2^12 to 2^29 bytes. Dictionary sizes are quantized so that they can be
-coded in just one byte (@pxref{coded-dict-size}). If the size specified
-does not match one of the valid sizes, it will be rounded upwards by
-adding up to @w{(@var{bytes} / 8)} to it.
+When compressing, set the dictionary size limit in bytes. Clzip uses for
+each file the largest dictionary size that does not exceed neither the file
+size nor this limit. Valid values range from @w{4 KiB} to @w{512 MiB}.
+Values 12 to 29 are interpreted as powers of two, meaning 2^12 to 2^29
+bytes. Dictionary sizes are quantized so that they can be coded in just one
+byte (@pxref{coded-dict-size}). If the size specified does not match one of
+the valid sizes, it is rounded upwards by adding up to @w{(@var{bytes} / 8)}
+to it.
 
 For maximum compression you should use a dictionary size limit as large
 as possible, but keep in mind that the decompression memory requirement
@@ -392,7 +395,7 @@ is affected at compression time by the choice of dictionary size limit.
 
 @item -S @var{bytes}
 @itemx --volume-size=@var{bytes}
-When compressing, and @samp{-c} has not been also specified, split the
+When compressing, and @option{-c} has not been also specified, split the
 compressed output into several volume files with names
 @samp{original_name00001.lz}, @samp{original_name00002.lz}, etc, and set the
 volume size limit to @var{bytes}. Input files are kept unchanged. Each
@@ -404,11 +407,11 @@ from @w{100 kB} to @w{4 EiB}.
 @itemx --test
 Check integrity of the files specified, but don't decompress them. This
 really performs a trial decompression and throws away the result. Use it
-together with @samp{-v} to see information about the files. If a file
+together with @option{-v} to see information about the files. If a file
 fails the test, does not exist, can't be opened, or is a terminal, clzip
-continues checking the rest of the files. A final diagnostic is shown at
-verbosity level 1 or higher if any file fails the test when testing
-multiple files.
+continues testing the rest of the files. A final diagnostic is shown at
+verbosity level 1 or higher if any file fails the test when testing multiple
+files.
 
 @item -v
 @itemx --verbose
@@ -420,23 +423,23 @@ verbosity level, showing status, compression ratio, dictionary size,
 trailer contents (CRC, data size, member size), and up to 6 bytes of
 trailing data (if any) both in hexadecimal and as a string of printable
 ASCII characters.@*
-Two or more @samp{-v} options show the progress of (de)compression.
+Two or more @option{-v} options show the progress of (de)compression.
 
 @item -0 .. -9
 Compression level. Set the compression parameters (dictionary size and
 match length limit) as shown in the table below. The default compression
-level is @samp{-6}, equivalent to @w{@samp{-s8MiB -m36}}. Note that
-@samp{-9} can be much slower than @samp{-0}. These options have no
+level is @option{-6}, equivalent to @w{@option{-s8MiB -m36}}. Note that
+@option{-9} can be much slower than @option{-0}. These options have no
 effect when decompressing, testing, or listing.
 
-The bidimensional parameter space of LZMA can't be mapped to a linear
-scale optimal for all files. If your files are large, very repetitive,
-etc, you may need to use the options @samp{--dictionary-size} and
-@samp{--match-length} directly to achieve optimal performance.
+The bidimensional parameter space of LZMA can't be mapped to a linear scale
+optimal for all files. If your files are large, very repetitive, etc, you
+may need to use the options @option{--dictionary-size} and
+@option{--match-length} directly to achieve optimal performance.
 
-If several compression levels or @samp{-s} or @samp{-m} options are
-given, the last setting is used. For example @w{@samp{-9 -s64MiB}} is
-equivalent to @w{@samp{-s64MiB -m273}}
+If several compression levels or @option{-s} or @option{-m} options are
+given, the last setting is used. For example @w{@option{-9 -s64MiB}} is
+equivalent to @w{@option{-s64MiB -m273}}
 
 @multitable {Level} {Dictionary size (-s)} {Match length limit (-m)}
 @item Level @tab Dictionary size (-s) @tab Match length limit (-m)
@@ -456,6 +459,15 @@ equivalent to @w{@samp{-s64MiB -m273}}
 @itemx --best
 Aliases for GNU gzip compatibility.
 
+@item --empty-error
+Exit with error status 2 if any empty member is found in the input files.
+
+@item --marking-error
+Exit with error status 2 if the first LZMA byte is non-zero in any member of
+the input files. This may be caused by data corruption or by deliberate
+insertion of tracking information in the file. Use
+@w{@samp{lziprecover --clear-marking}} to clear any such non-zero bytes.
+
 @item --loose-trailing
 When decompressing, testing, or listing, allow trailing data whose first
 bytes are so similar to the magic bytes of a lzip header that they can
@@ -464,28 +476,31 @@ be confused with a corrupt header. Use this option if a file triggers a
 
 @end table
 
-Numbers given as arguments to options may be followed by a multiplier
-and an optional @samp{B} for "byte".
+Numbers given as arguments to options may be expressed in decimal,
+hexadecimal, or octal (using the same syntax as integer constants in C++),
+and may be followed by a multiplier and an optional @samp{B} for "byte".
 
 Table of SI and binary prefixes (unit multipliers):
 
-@multitable {Prefix} {kilobyte  (10^3 = 1000)} {|} {Prefix} {kibibyte (2^10 = 1024)}
+@multitable {Prefix} {kilobyte   (10^3 = 1000)} {|} {Prefix} {kibibyte  (2^10 = 1024)}
 @item Prefix @tab Value               @tab | @tab Prefix @tab Value
-@item k @tab kilobyte  (10^3 = 1000)  @tab | @tab Ki @tab kibibyte (2^10 = 1024)
-@item M @tab megabyte  (10^6)         @tab | @tab Mi @tab mebibyte (2^20)
-@item G @tab gigabyte  (10^9)         @tab | @tab Gi @tab gibibyte (2^30)
-@item T @tab terabyte  (10^12)        @tab | @tab Ti @tab tebibyte (2^40)
-@item P @tab petabyte  (10^15)        @tab | @tab Pi @tab pebibyte (2^50)
-@item E @tab exabyte   (10^18)        @tab | @tab Ei @tab exbibyte (2^60)
-@item Z @tab zettabyte (10^21)        @tab | @tab Zi @tab zebibyte (2^70)
-@item Y @tab yottabyte (10^24)        @tab | @tab Yi @tab yobibyte (2^80)
+@item k @tab kilobyte   (10^3 = 1000) @tab | @tab Ki @tab kibibyte  (2^10 = 1024)
+@item M @tab megabyte   (10^6)        @tab | @tab Mi @tab mebibyte  (2^20)
+@item G @tab gigabyte   (10^9)        @tab | @tab Gi @tab gibibyte  (2^30)
+@item T @tab terabyte   (10^12)       @tab | @tab Ti @tab tebibyte  (2^40)
+@item P @tab petabyte   (10^15)       @tab | @tab Pi @tab pebibyte  (2^50)
+@item E @tab exabyte    (10^18)       @tab | @tab Ei @tab exbibyte  (2^60)
+@item Z @tab zettabyte  (10^21)       @tab | @tab Zi @tab zebibyte  (2^70)
+@item Y @tab yottabyte  (10^24)       @tab | @tab Yi @tab yobibyte  (2^80)
+@item R @tab ronnabyte  (10^27)       @tab | @tab Ri @tab robibyte  (2^90)
+@item Q @tab quettabyte (10^30)       @tab | @tab Qi @tab quebibyte (2^100)
 @end multitable
 
 @sp 1
-Exit status: 0 for a normal exit, 1 for environmental problems (file not
-found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or invalid
-input file, 3 for an internal consistency error (e.g., bug) which caused
-clzip to panic.
+Exit status: 0 for a normal exit, 1 for environmental problems
+(file not found, invalid command-line options, I/O errors, etc), 2 to
+indicate a corrupt or invalid input file, 3 for an internal consistency
+error (e.g., bug) which caused clzip to panic.
 
 
 @node Quality assurance
@@ -498,6 +513,11 @@ make it so complicated that there are no obvious deficiencies. The first
 method is far more difficult.@*
 --- C.A.R. Hoare
 
+Lzip has been designed, written, and tested with great care to replace gzip
+and bzip2 as the standard general-purpose compressed format for Unix-like
+systems. This chapter describes the lessons learned from these previous
+formats, and their application to the design of lzip.
+
 Lzip is developed by volunteers who lack the resources required for
 extensive testing in all circumstances. It is up to you to test lzip before
 using it in mission-critical applications. However, a compressor like lzip
@@ -505,11 +525,6 @@ is not a toy, and maintaining it is not a hobby. Many people's data depend
 on it. Therefore the lzip file format has been reviewed carefully and is
 believed to be free from negligent design errors.
 
-Lzip has been designed, written, and tested with great care to replace gzip
-and bzip2 as the standard general-purpose compressed format for unix-like
-systems. This chapter describes the lessons learned from these previous
-formats, and their application to the design of lzip.
-
 @sp 1
 @section Format design
 
@@ -593,9 +608,9 @@ compressed blocks.
 Using an optional CRC for the header is not only a bad idea, it is an error;
 it circumvents the Hamming distance (HD) of the CRC and may prevent the
 extraction of perfectly good data. For example, if the CRC is used and the
-bit enabling it is reset by a bit flip, the header will appear to be intact
-(in spite of being corrupt) while the compressed blocks will appear to be
-totally unrecoverable (in spite of being intact). Very misleading indeed.
+bit enabling it is reset by a bit flip, then the header seems to be intact
+(in spite of being corrupt) while the compressed blocks seem to be totally
+unrecoverable (in spite of being intact). Very misleading indeed.
 
 @item Metadata
 
@@ -613,7 +628,7 @@ from identical input).
 
 Probably the most frequently reported shortcoming of the gzip format is that
 it only stores the least significant 32 bits of the uncompressed size. The
-size of any file larger than @w{4 GiB} gets truncated.
+size of any file larger or equal than @w{4 GiB} gets truncated.
 
 Bzip2 does not store the uncompressed size of the file.
 
@@ -636,10 +651,14 @@ and may limit the number of members or the total uncompressed size.
 
 @section Quality of implementation
 
+Our civilization depends critically on software; it had better be quality
+software.@*
+--- Bjarne Stroustrup
+
 @table @samp
 @item Accurate and robust error detection
 
-The lzip format provides 3 factor integrity checking, and the decompressors
+The lzip format provides 3-factor integrity checking, and the decompressors
 report mismatches in each factor separately. This method detects most false
 positives for corruption. If just one byte in one factor fails but the other
 two factors match the data, it probably means that the data are intact and
@@ -648,14 +667,14 @@ member size) in the member trailer.
 
 @item Multiple implementations
 
-Just like the lzip format provides 3 factor protection against undetected
+Just like the lzip format provides 3-factor protection against undetected
 data corruption, the development methodology of the lzip family of
-compressors provides 3 factor protection against undetected programming
+compressors provides 3-factor protection against undetected programming
 errors.
 
 Three related but independent compressor implementations, lzip, clzip, and
 minilzip/lzlib, are developed concurrently. Every stable release of any of
-them is tested to verify that it produces identical output to the other two.
+them is tested to check that it produces identical output to the other two.
 This guarantees that all three implement the same algorithm, and makes it
 unlikely that any of them may contain serious undiscovered errors. In fact,
 no errors have been discovered in lzip since 2009.
@@ -692,7 +711,7 @@ concrete algorithm; it is more like "any algorithm using the LZMA coding
 scheme". LZMA compression consists in describing the uncompressed data as a
 succession of coding sequences from the set shown in Section @samp{What is
 coded} (@pxref{what-is-coded}), and then encoding them using a range
-encoder. For example, the option @samp{-0} of clzip uses the scheme in almost
+encoder. For example, the option @option{-0} of clzip uses the scheme in almost
 the simplest way possible; issuing the longest match it can find, or a
 literal byte if it can't find a match. Inversely, a much more elaborated way
 of finding coding sequences of minimum size than the one currently used by
@@ -700,13 +719,13 @@ clzip could be developed, and the resulting sequence could also be coded
 using the LZMA coding scheme.
 
 Clzip currently implements two variants of the LZMA algorithm: fast
-(used by option @samp{-0}) and normal (used by all other compression levels).
+(used by option @option{-0}) and normal (used by all other compression levels).
 
 The high compression of LZMA comes from combining two basic, well-proven
-compression ideas: sliding dictionaries (LZ77/78) and markov models (the
-thing used by every compression algorithm that uses a range encoder or
-similar order-0 entropy coder as its last stage) with segregation of
-contexts according to what the bits are used for.
+compression ideas: sliding dictionaries (LZ77) and markov models (the thing
+used by every compression algorithm that uses a range encoder or similar
+order-0 entropy coder as its last stage) with segregation of contexts
+according to what the bits are used for.
 
 Clzip is a two stage compressor. The first stage is a Lempel-Ziv coder,
 which reduces redundancy by translating chunks of data to their
@@ -752,7 +771,7 @@ get longer with higher compression levels because dictionary size increases
 
 @noindent
 The ideas embodied in clzip are due to (at least) the following people:
-Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the
+Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the
 definition of Markov chains), G.N.N. Martin (for the definition of range
 encoding), Igor Pavlov (for putting all the above together in LZMA), and
 Julian Seward (for bzip2's CLI).
@@ -786,7 +805,7 @@ represents one byte; a box like this:
 represents a variable number of bytes.
 
 @sp 1
-A lzip file consists of a series of independent "members" (compressed data
+A lzip file consists of one or more independent "members" (compressed data
 sets). The members simply appear one after another in the file, with no
 additional information before, between, or after them. Each member can
 encode in compressed form up to @w{16 EiB - 1 byte} of uncompressed data.
@@ -832,10 +851,10 @@ Size of the original uncompressed data.
 
 @item Member size (8 bytes)
 Total size of the member, including header and trailer. This field acts
-as a distributed index, allows the verification of stream integrity, and
+as a distributed index, improves the checking of stream integrity, and
 facilitates the safe recovery of undamaged members from multimember files.
-Member size should be limited to @w{2 PiB} to prevent the data size field
-from overflowing.
+Lzip limits the member size to @w{2 PiB} to prevent the data size field from
+overflowing.
 
 @end table
 
@@ -855,12 +874,12 @@ does not even appear in the code.
 
 Lzip finishes the LZMA stream with an "End Of Stream" (EOS) marker (the
 distance-length pair @w{0xFFFFFFFFU, 2}), which in conjunction with the
-@samp{member size} field in the member trailer allows the verification of
-stream integrity. The EOS marker is the only marker allowed in lzip files.
-The LZMA stream in lzip files always has these two features (default
-properties and EOS marker) and is referred to in this document as
-LZMA-302eos. This simplified form of the LZMA stream format has been chosen
-to maximize interoperability and safety.
+@samp{member size} field in the member trailer allows the checking of stream
+integrity. The EOS marker is the only LZMA marker allowed in lzip files. The
+LZMA stream in lzip files always has these two features (default properties
+and EOS marker) and is referred to in this document as LZMA-302eos. This
+simplified and marker-terminated form of the LZMA stream format has been
+chosen to maximize interoperability and safety.
 
 The second stage of LZMA is a range encoder that uses a different
 probability model for each type of symbol: distances, lengths, literal
@@ -878,9 +897,9 @@ code of a real decompressor seems the only appropriate reference to use.
 
 What follows is a description of the decoding algorithm for LZMA-302eos
 streams using as reference the source code of "lzd", an educational
-decompressor for lzip files which can be downloaded from the lzip download
-directory. Lzd is written in C++11 and its source code is included in
-appendix A. @xref{Reference source code}.
+decompressor for lzip files, included in appendix A. @xref{Reference source
+code}. Lzd is written in C++11 and can be downloaded from the lzip download
+directory.
 
 @sp 1
 @section What is coded
@@ -947,17 +966,17 @@ the distance is @w{>= 4}, the remaining bits are encoded as follows.
 @samp{direct_bits} is the amount of remaining bits (from 1 to 30) needed
 to form a complete distance, and is calculated as @w{(slot >> 1) - 1}.
 If a distance needs 6 or more direct_bits, the last 4 bits are encoded
-separately. The last piece (all the direct_bits for distances 4 to 127,
-or the last 4 bits for distances @w{>= 128}) is context-coded in reverse
-order (from LSB to MSB). For distances @w{>= 128}, the
-@w{@samp{direct_bits - 4}} part is encoded with fixed 0.5 probability.
+separately. The last piece (all the direct_bits for distances 4 to 127
+(slots 4 to 13), or the last 4 bits for distances @w{>= 128}
+@w{(slot >= 14)}) is context-coded in reverse order (from LSB to MSB). For
+distances @w{>= 128}, the @w{@samp{direct_bits - 4}} part is encoded with
+fixed 0.5 probability.
 
 @multitable @columnfractions .5 .5
 @headitem Bit sequence @tab Description
 @item slot @tab distances from 0 to 3
 @item slot + direct_bits @tab distances from 4 to 127
-@item slot + (direct_bits - 4) + 4 bits @tab distances from 128 to
-2^32 - 1
+@item slot + (direct_bits - 4) + 4 bits @tab distances from 128 to 2^32 - 1
 @end multitable
 
 @sp 1
@@ -1078,7 +1097,7 @@ range decoder. This is done by shifting 5 bytes in the initialization of
 the source).
 
 @sp 1
-@section Decoding and verifying the LZMA stream
+@section Decoding and checking the LZMA stream
 
 After decoding the member header and obtaining the dictionary size, the
 range decoder is initialized and then the LZMA decoder enters a loop
@@ -1088,7 +1107,7 @@ sequences (matches, repeated matches, and literal bytes), until the "End
 Of Stream" marker is decoded.
 
 Once the "End Of Stream" marker has been decoded, the decompressor reads and
-decodes the member trailer, and verifies that the three integrity factors
+decodes the member trailer, and checks that the three integrity factors
 stored there (CRC, data size, and member size) match those computed from the
 data.
 
@@ -1107,12 +1126,13 @@ example when writing to a tape. It is safe to append any amount of
 padding zero bytes to a lzip file.
 
 @item
-Useful data added by the user; a cryptographically secure hash, a
-description of file contents, etc. It is safe to append any amount of
-text to a lzip file as long as none of the first four bytes of the text
-match the corresponding byte in the string "LZIP", and the text does not
-contain any zero bytes (null characters). Nonzero bytes and zero bytes
-can't be safely mixed in trailing data.
+Useful data added by the user; an "End Of File" string (to check that the
+file has not been truncated), a cryptographically secure hash, a description
+of file contents, etc. It is safe to append any amount of text to a lzip
+file as long as none of the first four bytes of the text matches the
+corresponding byte in the string "LZIP", and the text does not contain any
+zero bytes (null characters). Nonzero bytes and zero bytes can't be safely
+mixed in trailing data.
 
 @item
 Garbage added by some not totally successful copy operation.
@@ -1130,8 +1150,8 @@ integrity information itself. Therefore it can be considered to be below
 the noise level. Additionally, the test used by clzip to discriminate
 trailing data from a corrupt header has a Hamming distance (HD) of 3,
 and the 3 bit flips must happen in different magic bytes for the test to
-fail. In any case, the option @samp{--trailing-error} guarantees that
-any corrupt header will be detected.
+fail. In any case, the option @option{--trailing-error} guarantees that
+any corrupt header is detected.
 @end itemize
 
 Trailing data are in no way part of the lzip file format, but tools
@@ -1141,7 +1161,7 @@ possible in the presence of trailing data.
 Trailing data can be safely ignored in most cases. In some cases, like
 that of user-added data, they are expected to be ignored. In those cases
 where a file containing trailing data must be rejected, the option
-@samp{--trailing-error} can be used. @xref{--trailing-error}.
+@option{--trailing-error} can be used. @xref{--trailing-error}.
 
 
 @node Examples
@@ -1151,8 +1171,8 @@ where a file containing trailing data must be rejected, the option
 WARNING! Even if clzip is bug-free, other causes may result in a corrupt
 compressed file (bugs in the system libraries, memory errors, etc).
 Therefore, if the data you are going to compress are important, give the
-option @samp{--keep} to clzip and don't remove the original file until you
-verify the compressed file with a command like
+option @option{--keep} to clzip and don't remove the original file until you
+check the compressed file with a command like
 @w{@samp{clzip -cd file.lz | cmp file -}}. Most RAM errors happening during
 compression can only be detected by comparing the compressed file with the
 original because the corruption happens before clzip compresses the RAM
@@ -1197,7 +1217,7 @@ clzip -d file.lz
 
 @sp 1
 @noindent
-Example 5: Verify the integrity of the compressed file @samp{file.lz} and
+Example 5: Check the integrity of the compressed file @samp{file.lz} and
 show status.
 
 @example
@@ -1295,7 +1315,7 @@ find by running @w{@samp{clzip --version}}.
 
 @verbatim
 /* Lzd - Educational decompressor for the lzip format
-   Copyright (C) 2013-2022 Antonio Diaz Diaz.
+   Copyright (C) 2013-2023 Antonio Diaz Diaz.
 
    This program is free software. Redistribution and use in source and
    binary forms, with or without modification, are permitted provided
@@ -1314,8 +1334,8 @@ find by running @w{@samp{clzip --version}}.
 */
 /*
    Exit status: 0 for a normal exit, 1 for environmental problems
-   (file not found, invalid flags, I/O errors, etc), 2 to indicate a
-   corrupt or invalid input file.
+   (file not found, invalid command-line options, I/O errors, etc), 2 to
+   indicate a corrupt or invalid input file.
 */
 
 #include <algorithm>
@@ -1426,10 +1446,11 @@ public:
 const CRC32 crc32;
 
 
-typedef uint8_t Lzip_header[6];		// 0-3 magic bytes
-					//   4 version
-					//   5 coded dictionary size
-typedef uint8_t Lzip_trailer[20];
+enum { header_size = 6, trailer_size = 20 };
+typedef uint8_t Lzip_header[header_size]; // 0-3 magic bytes
+					  //   4 version
+					  //   5 coded dictionary size
+typedef uint8_t Lzip_trailer[trailer_size];
 			//  0-3  CRC32 of the uncompressed data
 			//  4-11 size of the uncompressed data
 			// 12-19 member size including header and trailer
@@ -1441,9 +1462,11 @@ class Range_decoder
   uint32_t range;
 
 public:
-  Range_decoder() : member_pos( 6 ), code( 0 ), range( 0xFFFFFFFFU )
+  Range_decoder()
+    : member_pos( header_size ), code( 0 ), range( 0xFFFFFFFFU )
     {
-    for( int i = 0; i < 5; ++i ) code = ( code << 8 ) | get_byte();
+    get_byte();			// discard first byte of the LZMA stream
+    for( int i = 0; i < 4; ++i ) code = ( code << 8 ) | get_byte();
     }
 
   uint8_t get_byte() { ++member_pos; return std::getc( stdin ); }
@@ -1476,8 +1499,8 @@ public:
       }
     else
       {
-      range -= bound;
       code -= bound;
+      range -= bound;
       bm.probability -= bm.probability >> bit_model_move_bits;
       symbol = 1;
       }
@@ -1527,11 +1550,12 @@ public:
   unsigned decode_len( Len_model & lm, const int pos_state )
     {
     if( decode_bit( lm.choice1 ) == 0 )
-      return decode_tree( lm.bm_low[pos_state], len_low_bits );
+      return min_match_len +
+             decode_tree( lm.bm_low[pos_state], len_low_bits );
     if( decode_bit( lm.choice2 ) == 0 )
-      return len_low_symbols +
+      return min_match_len + len_low_symbols +
              decode_tree( lm.bm_mid[pos_state], len_mid_bits );
-    return len_low_symbols + len_mid_symbols +
+    return min_match_len + len_low_symbols + len_mid_symbols +
            decode_tree( lm.bm_high, len_high_bits );
     }
   };
@@ -1604,7 +1628,7 @@ void LZ_decoder::flush_data()
   }
 
 
-bool LZ_decoder::decode_member()	// Returns false if error
+bool LZ_decoder::decode_member()	// Return false if error
   {
   Bit_model bm_literal[1<<literal_context_bits][0x300];
   Bit_model bm_match[State::states][pos_states];
@@ -1666,12 +1690,12 @@ bool LZ_decoder::decode_member()	// Returns false if error
         rep0 = distance;
         }
       state.set_rep();
-      len = min_match_len + rdec.decode_len( rep_len_model, pos_state );
+      len = rdec.decode_len( rep_len_model, pos_state );
       }
     else					// match
       {
       rep3 = rep2; rep2 = rep1; rep1 = rep0;
-      len = min_match_len + rdec.decode_len( match_len_model, pos_state );
+      len = rdec.decode_len( match_len_model, pos_state );
       const int len_state = std::min( len - min_match_len, len_states - 1 );
       rep0 = rdec.decode_tree( bm_dis_slot[len_state], dis_slot_bits );
       if( rep0 >= start_dis_model )
@@ -1690,7 +1714,7 @@ bool LZ_decoder::decode_member()	// Returns false if error
           if( rep0 == 0xFFFFFFFFU )		// marker found
             {
             flush_data();
-            return ( len == min_match_len );	// End Of Stream marker
+            return len == min_match_len;	// End Of Stream marker
             }
           }
         }
@@ -1711,11 +1735,11 @@ int main( const int argc, const char * const argv[] )
     {
     std::printf(
       "Lzd %s - Educational decompressor for the lzip format.\n"
-      "Study the source to learn how a lzip decompressor works.\n"
+      "Study the source code to learn how a lzip decompressor works.\n"
       "See the lzip manual for an explanation of the code.\n"
       "\nUsage: %s [-d] < file.lz > file\n"
       "Lzd decompresses from standard input to standard output.\n"
-      "\nCopyright (C) 2022 Antonio Diaz Diaz.\n"
+      "\nCopyright (C) 2023 Antonio Diaz Diaz.\n"
       "License 2-clause BSD.\n"
       "This is free software: you are free to change and redistribute it.\n"
       "There is NO WARRANTY, to the extent permitted by law.\n"
@@ -1732,8 +1756,8 @@ int main( const int argc, const char * const argv[] )
 
   for( bool first_member = true; ; first_member = false )
     {
-    Lzip_header header;				// verify header
-    for( int i = 0; i < 6; ++i ) header[i] = std::getc( stdin );
+    Lzip_header header;				// check header
+    for( int i = 0; i < header_size; ++i ) header[i] = std::getc( stdin );
     if( std::feof( stdin ) || std::memcmp( header, "LZIP\x01", 5 ) != 0 )
       {
       if( first_member )
@@ -1751,8 +1775,8 @@ int main( const int argc, const char * const argv[] )
     if( !decoder.decode_member() )
       { std::fputs( "Data error\n", stderr ); return 2; }
 
-    Lzip_trailer trailer;			// verify trailer
-    for( int i = 0; i < 20; ++i ) trailer[i] = decoder.get_byte();
+    Lzip_trailer trailer;			// check trailer
+    for( int i = 0; i < trailer_size; ++i ) trailer[i] = decoder.get_byte();
     int retval = 0;
     unsigned crc = 0;
     for( int i = 3; i >= 0; --i ) crc = ( crc << 8 ) + trailer[i];
diff --git a/encoder.c b/encoder.c
index cf68614..a1fd214 100644
--- a/encoder.c
+++ b/encoder.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
diff --git a/encoder.h b/encoder.h
index 13ee448..f30a8b6 100644
--- a/encoder.h
+++ b/encoder.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
diff --git a/encoder_base.c b/encoder_base.c
index 91c4623..d0d53e2 100644
--- a/encoder_base.c
+++ b/encoder_base.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -173,9 +173,7 @@ void LZeb_full_flush( struct LZ_encoder_base * const eb, const State state )
   Lt_set_data_crc( trailer, LZeb_crc( eb ) );
   Lt_set_data_size( trailer, Mb_data_position( &eb->mb ) );
   Lt_set_member_size( trailer, Re_member_position( &eb->renc ) + Lt_size );
-  int i;
-  for( i = 0; i < Lt_size; ++i )
-    Re_put_byte( &eb->renc, trailer[i] );
+  int i; for( i = 0; i < Lt_size; ++i ) Re_put_byte( &eb->renc, trailer[i] );
   Re_flush_data( &eb->renc );
   }
 
diff --git a/encoder_base.h b/encoder_base.h
index f47ec48..49ed200 100644
--- a/encoder_base.h
+++ b/encoder_base.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -78,7 +78,7 @@ static inline int price1( const Bit_model probability )
   { return get_price( bit_model_total - probability ); }
 
 static inline int price_bit( const Bit_model bm, const bool bit )
-  { return ( bit ? price1( bm ) : price0( bm ) ); }
+  { return bit ? price1( bm ) : price0( bm ); }
 
 
 static inline int price_symbol3( const Bit_model bm[], int symbol )
@@ -263,7 +263,6 @@ static inline void Re_shift_low( struct Range_encoder * const renc )
 static inline void Re_reset( struct Range_encoder * const renc,
                              const unsigned dictionary_size )
   {
-  int i;
   renc->low = 0;
   renc->partial_member_pos = 0;
   renc->pos = 0;
@@ -271,8 +270,7 @@ static inline void Re_reset( struct Range_encoder * const renc,
   renc->ff_count = 0;
   renc->cache = 0;
   Lh_set_dictionary_size( renc->header, dictionary_size );
-  for( i = 0; i < Lh_size; ++i )
-    Re_put_byte( renc, renc->header[i] );
+  int i; for( i = 0; i < Lh_size; ++i ) Re_put_byte( renc, renc->header[i] );
   }
 
 static inline bool Re_init( struct Range_encoder * const renc,
diff --git a/fast_encoder.c b/fast_encoder.c
index f69b999..518a514 100644
--- a/fast_encoder.c
+++ b/fast_encoder.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
diff --git a/fast_encoder.h b/fast_encoder.h
index 64ea2cc..9e9d010 100644
--- a/fast_encoder.h
+++ b/fast_encoder.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
diff --git a/list.c b/list.c
index 390d83b..a1bb638 100644
--- a/list.c
+++ b/list.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -43,13 +43,14 @@ static void list_line( const unsigned long long uncomp_size,
 
 
 int list_files( const char * const filenames[], const int num_filenames,
-                const bool ignore_trailing, const bool loose_trailing )
+                const struct Cl_options * const cl_opts )
   {
   unsigned long long total_comp = 0, total_uncomp = 0;
   int files = 0, retval = 0;
   int i;
   bool first_post = true;
   bool stdin_used = false;
+
   for( i = 0; i < num_filenames; ++i )
     {
     const bool from_stdin = ( strcmp( filenames[i], "-" ) == 0 );
@@ -61,7 +62,7 @@ int list_files( const char * const filenames[], const int num_filenames,
     if( infd < 0 ) { set_retval( &retval, 1 ); continue; }
 
     struct Lzip_index lzip_index;
-    Li_init( &lzip_index, infd, ignore_trailing, loose_trailing );
+    Li_init( &lzip_index, infd, cl_opts );
     close( infd );
     if( lzip_index.retval != 0 )
       {
diff --git a/lzip.h b/lzip.h
index 9d4639e..fa8caf3 100644
--- a/lzip.h
+++ b/lzip.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -25,7 +25,6 @@
 typedef int State;
 
 enum { states = 12 };
-
 static inline bool St_is_char( const State st ) { return st < 7; }
 
 static inline State St_set_char( const State st )
@@ -33,17 +32,13 @@ static inline State St_set_char( const State st )
   static const State next[states] = { 0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 4, 5 };
   return next[st];
   }
-
 static inline State St_set_char_rep() { return 8; }
-
 static inline State St_set_match( const State st )
-  { return ( ( st < 7 ) ? 7 : 10 ); }
-
+  { return ( st < 7 ) ? 7 : 10; }
 static inline State St_set_rep( const State st )
-  { return ( ( st < 7 ) ? 8 : 11 ); }
-
+  { return ( st < 7 ) ? 8 : 11; }
 static inline State St_set_short_rep( const State st )
-  { return ( ( st < 7 ) ? 9 : 11 ); }
+  { return ( st < 7 ) ? 9 : 11; }
 
 
 enum {
@@ -62,7 +57,7 @@ enum {
   dis_slot_bits = 6,
   start_dis_model = 4,
   end_dis_model = 14,
-  modeled_distances = 1 << (end_dis_model / 2),		/* 128 */
+  modeled_distances = 1 << ( end_dis_model / 2 ),	/* 128 */
   dis_align_bits = 4,
   dis_align_size = 1 << dis_align_bits,
 
@@ -150,8 +145,8 @@ static inline void CRC32_update_buf( uint32_t * const crc,
 
 
 static inline bool isvalid_ds( const unsigned dictionary_size )
-  { return ( dictionary_size >= min_dictionary_size &&
-             dictionary_size <= max_dictionary_size ); }
+  { return dictionary_size >= min_dictionary_size &&
+           dictionary_size <= max_dictionary_size; }
 
 
 static inline int real_bits( unsigned value )
@@ -164,43 +159,43 @@ static inline int real_bits( unsigned value )
 
 static const uint8_t lzip_magic[4] = { 0x4C, 0x5A, 0x49, 0x50 }; /* "LZIP" */
 
-typedef uint8_t Lzip_header[6];		/* 0-3 magic bytes */
+enum { Lh_size = 6 };
+typedef uint8_t Lzip_header[Lh_size];	/* 0-3 magic bytes */
 					/*   4 version */
 					/*   5 coded dictionary size */
-enum { Lh_size = 6 };
 
 static inline void Lh_set_magic( Lzip_header data )
   { memcpy( data, lzip_magic, 4 ); data[4] = 1; }
 
-static inline bool Lh_verify_magic( const Lzip_header data )
-  { return ( memcmp( data, lzip_magic, 4 ) == 0 ); }
+static inline bool Lh_check_magic( const Lzip_header data )
+  { return memcmp( data, lzip_magic, 4 ) == 0; }
 
 /* detect (truncated) header */
-static inline bool Lh_verify_prefix( const Lzip_header data, const int sz )
+static inline bool Lh_check_prefix( const Lzip_header data, const int sz )
   {
   int i; for( i = 0; i < sz && i < 4; ++i )
     if( data[i] != lzip_magic[i] ) return false;
-  return ( sz > 0 );
+  return sz > 0;
   }
 
 /* detect corrupt header */
-static inline bool Lh_verify_corrupt( const Lzip_header data )
+static inline bool Lh_check_corrupt( const Lzip_header data )
   {
   int matches = 0;
   int i; for( i = 0; i < 4; ++i )
     if( data[i] == lzip_magic[i] ) ++matches;
-  return ( matches > 1 && matches < 4 );
+  return matches > 1 && matches < 4;
   }
 
 static inline uint8_t Lh_version( const Lzip_header data )
   { return data[4]; }
 
-static inline bool Lh_verify_version( const Lzip_header data )
-  { return ( data[4] == 1 ); }
+static inline bool Lh_check_version( const Lzip_header data )
+  { return data[4] == 1; }
 
 static inline unsigned Lh_get_dictionary_size( const Lzip_header data )
   {
-  unsigned sz = ( 1 << ( data[5] & 0x1F ) );
+  unsigned sz = 1 << ( data[5] & 0x1F );
   if( sz > min_dictionary_size )
     sz -= ( sz / 16 ) * ( ( data[5] >> 5 ) & 7 );
   return sz;
@@ -217,23 +212,23 @@ static inline bool Lh_set_dictionary_size( Lzip_header data, const unsigned sz )
     unsigned i;
     for( i = 7; i >= 1; --i )
       if( base_size - ( i * fraction ) >= sz )
-        { data[5] |= ( i << 5 ); break; }
+        { data[5] |= i << 5; break; }
     }
   return true;
   }
 
-static inline bool Lh_verify( const Lzip_header data )
+static inline bool Lh_check( const Lzip_header data )
   {
-  return Lh_verify_magic( data ) && Lh_verify_version( data ) &&
+  return Lh_check_magic( data ) && Lh_check_version( data ) &&
          isvalid_ds( Lh_get_dictionary_size( data ) );
   }
 
 
-typedef uint8_t Lzip_trailer[20];
+enum { Lt_size = 20 };
+typedef uint8_t Lzip_trailer[Lt_size];
 			/*  0-3  CRC32 of the uncompressed data */
 			/*  4-11 size of the uncompressed data */
 			/* 12-19 member size including header and trailer */
-enum { Lt_size = 20 };
 
 static inline unsigned Lt_get_data_crc( const Lzip_trailer data )
   {
@@ -266,7 +261,7 @@ static inline void Lt_set_member_size( Lzip_trailer data, unsigned long long sz
   { int i; for( i = 12; i <= 19; ++i ) { data[i] = (uint8_t)sz; sz >>= 8; } }
 
 /* check internal consistency */
-static inline bool Lt_verify_consistency( const Lzip_trailer data )
+static inline bool Lt_check_consistency( const Lzip_trailer data )
   {
   const unsigned crc = Lt_get_data_crc( data );
   const unsigned long long dsize = Lt_get_data_size( data );
@@ -281,12 +276,27 @@ static inline bool Lt_verify_consistency( const Lzip_trailer data )
   }
 
 
+struct Cl_options		/* command-line options */
+  {
+  bool ignore_empty;
+  bool ignore_marking;
+  bool ignore_trailing;
+  bool loose_trailing;
+  };
+
+static inline void Cl_options_init( struct Cl_options * cl_opts )
+  { cl_opts->ignore_empty = true; cl_opts->ignore_marking = true;
+    cl_opts->ignore_trailing = true; cl_opts->loose_trailing = false; }
+
+
 static inline void set_retval( int * retval, const int new_val )
   { if( *retval < new_val ) *retval = new_val; }
 
 static const char * const bad_magic_msg = "Bad magic number (file not in lzip format).";
 static const char * const bad_dict_msg = "Invalid dictionary size in member header.";
 static const char * const corrupt_mm_msg = "Corrupt header in multimember file.";
+static const char * const empty_msg = "Empty member not allowed.";
+static const char * const marking_msg = "Marking data not allowed.";
 static const char * const trailing_msg = "Trailing data not allowed.";
 static const char * const mem_msg = "Not enough memory.";
 
@@ -296,7 +306,7 @@ int writeblock( const int fd, const uint8_t * const buf, const int size );
 
 /* defined in list.c */
 int list_files( const char * const filenames[], const int num_filenames,
-                const bool ignore_trailing, const bool loose_trailing );
+                const struct Cl_options * const cl_opts );
 
 /* defined in main.c */
 struct stat;
diff --git a/lzip_index.c b/lzip_index.c
index 41d0e13..6f64205 100644
--- a/lzip_index.c
+++ b/lzip_index.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -88,17 +88,17 @@ static void Li_reverse_member_vector( struct Lzip_index * const li )
   }
 
 
-static bool Li_check_header_error( struct Lzip_index * const li,
-                                   const Lzip_header header )
+static bool Li_check_header( struct Lzip_index * const li,
+                             const Lzip_header header )
   {
-  if( !Lh_verify_magic( header ) )
-    { add_error( li, bad_magic_msg ); li->retval = 2; return true; }
-  if( !Lh_verify_version( header ) )
+  if( !Lh_check_magic( header ) )
+    { add_error( li, bad_magic_msg ); li->retval = 2; return false; }
+  if( !Lh_check_version( header ) )
     { add_error( li, bad_version( Lh_version( header ) ) ); li->retval = 2;
-      return true; }
+      return false; }
   if( !isvalid_ds( Lh_get_dictionary_size( header ) ) )
-    { add_error( li, bad_dict_msg ); li->retval = 2; return true; }
-  return false;
+    { add_error( li, bad_dict_msg ); li->retval = 2; return false; }
+  return true;
   }
 
 static void Li_set_errno_error( struct Lzip_index * const li,
@@ -119,10 +119,13 @@ static void Li_set_num_error( struct Lzip_index * const li,
 
 
 static bool Li_read_header( struct Lzip_index * const li, const int fd,
-                            Lzip_header header, const long long pos )
+            Lzip_header header, const long long pos, const bool ignore_marking )
   {
   if( seek_read( fd, header, Lh_size, pos ) != Lh_size )
     { Li_set_errno_error( li, "Error reading member header: " ); return false; }
+  uint8_t byte;
+  if( !ignore_marking && readblock( fd, &byte, 1 ) == 1 && byte != 0 )
+    { add_error( li, marking_msg ); li->retval = 2; return false; }
   return true;
   }
 
@@ -130,8 +133,7 @@ static bool Li_read_header( struct Lzip_index * const li, const int fd,
 /* If successful, push last member and set pos to member header. */
 static bool Li_skip_trailing_data( struct Lzip_index * const li, const int fd,
                                    unsigned long long * const pos,
-                                   const bool ignore_trailing,
-                                   const bool loose_trailing )
+                                   const struct Cl_options * const cl_opts )
   {
   if( *pos < min_member_size ) return false;
   enum { block_size = 16384,
@@ -157,31 +159,34 @@ static bool Li_skip_trailing_data( struct Lzip_index * const li, const int fd,
         const unsigned long long member_size = Lt_get_member_size( *trailer );
         if( member_size == 0 )			/* skip trailing zeros */
           { while( i > Lt_size && buffer[i-9] == 0 ) --i; continue; }
-        if( member_size > ipos + i || !Lt_verify_consistency( *trailer ) )
+        if( member_size > ipos + i || !Lt_check_consistency( *trailer ) )
           continue;
         Lzip_header header;
-        if( !Li_read_header( li, fd, header, ipos + i - member_size ) )
-          return false;
-        if( !Lh_verify( header ) ) continue;
+        if( !Li_read_header( li, fd, header, ipos + i - member_size,
+                             cl_opts->ignore_marking ) ) return false;
+        if( !Lh_check( header ) ) continue;
         const Lzip_header * header2 = (const Lzip_header *)( buffer + i );
         const bool full_h2 = bsize - i >= Lh_size;
-        if( Lh_verify_prefix( *header2, bsize - i ) )	/* last member */
+        if( Lh_check_prefix( *header2, bsize - i ) )	/* last member */
           {
           if( !full_h2 ) add_error( li, "Last member in input file is truncated." );
-          else if( !Li_check_header_error( li, *header2 ) )
+          else if( Li_check_header( li, *header2 ) )
             add_error( li, "Last member in input file is truncated or corrupt." );
           li->retval = 2; return false;
           }
-        if( !loose_trailing && full_h2 && Lh_verify_corrupt( *header2 ) )
+        if( !cl_opts->loose_trailing && full_h2 && Lh_check_corrupt( *header2 ) )
           { add_error( li, corrupt_mm_msg ); li->retval = 2; return false; }
-        if( !ignore_trailing )
+        if( !cl_opts->ignore_trailing )
           { add_error( li, trailing_msg ); li->retval = 2; return false; }
-        *pos = ipos + i - member_size;
+        const unsigned long long data_size = Lt_get_data_size( *trailer );
+        if( !cl_opts->ignore_empty && data_size == 0 )
+          { add_error( li, empty_msg ); li->retval = 2; return false; }
+        *pos = ipos + i - member_size;			/* good member */
         const unsigned dictionary_size = Lh_get_dictionary_size( header );
         if( li->dictionary_size < dictionary_size )
           li->dictionary_size = dictionary_size;
-        return push_back_member( li, 0, Lt_get_data_size( *trailer ), *pos,
-                                 member_size, dictionary_size );
+        return push_back_member( li, 0, data_size, *pos, member_size,
+                                 dictionary_size );
         }
     if( ipos == 0 )
       { Li_set_num_error( li, "Bad trailer at pos ", *pos - Lt_size );
@@ -196,7 +201,7 @@ static bool Li_skip_trailing_data( struct Lzip_index * const li, const int fd,
 
 
 bool Li_init( struct Lzip_index * const li, const int infd,
-              const bool ignore_trailing, const bool loose_trailing )
+              const struct Cl_options * const cl_opts )
   {
   li->member_vector = 0;
   li->error = 0;
@@ -215,8 +220,8 @@ bool Li_init( struct Lzip_index * const li, const int infd,
       li->retval = 2; return false; }
 
   Lzip_header header;
-  if( !Li_read_header( li, infd, header, 0 ) ) return false;
-  if( Li_check_header_error( li, header ) ) return false;
+  if( !Li_read_header( li, infd, header, 0, cl_opts->ignore_marking ) ||
+      !Li_check_header( li, header ) ) return false;
 
   unsigned long long pos = li->insize;	/* always points to a header or to EOF */
   while( pos >= min_member_size )
@@ -225,32 +230,33 @@ bool Li_init( struct Lzip_index * const li, const int infd,
     if( seek_read( infd, trailer, Lt_size, pos - Lt_size ) != Lt_size )
       { Li_set_errno_error( li, "Error reading member trailer: " ); break; }
     const unsigned long long member_size = Lt_get_member_size( trailer );
-    if( member_size > pos || !Lt_verify_consistency( trailer ) )
+    if( member_size > pos || !Lt_check_consistency( trailer ) )
       {							/* bad trailer */
       if( li->members <= 0 )
-        { if( Li_skip_trailing_data( li, infd, &pos, ignore_trailing,
-              loose_trailing ) ) continue; else return false; }
-      Li_set_num_error( li, "Bad trailer at pos ", pos - Lt_size );
-      break;
+        { if( Li_skip_trailing_data( li, infd, &pos, cl_opts ) ) continue;
+          return false; }
+      Li_set_num_error( li, "Bad trailer at pos ", pos - Lt_size ); break;
       }
-    if( !Li_read_header( li, infd, header, pos - member_size ) ) break;
-    if( !Lh_verify( header ) )				/* bad header */
+    if( !Li_read_header( li, infd, header, pos - member_size,
+                         cl_opts->ignore_marking ) ) break;
+    if( !Lh_check( header ) )				/* bad header */
       {
       if( li->members <= 0 )
-        { if( Li_skip_trailing_data( li, infd, &pos, ignore_trailing,
-              loose_trailing ) ) continue; else return false; }
-      Li_set_num_error( li, "Bad header at pos ", pos - member_size );
-      break;
+        { if( Li_skip_trailing_data( li, infd, &pos, cl_opts ) ) continue;
+          return false; }
+      Li_set_num_error( li, "Bad header at pos ", pos - member_size ); break;
       }
-    pos -= member_size;
+    const unsigned long long data_size = Lt_get_data_size( trailer );
+    if( !cl_opts->ignore_empty && data_size == 0 )
+      { add_error( li, empty_msg ); li->retval = 2; break; }
+    pos -= member_size;					/* good member */
     const unsigned dictionary_size = Lh_get_dictionary_size( header );
     if( li->dictionary_size < dictionary_size )
       li->dictionary_size = dictionary_size;
-    if( !push_back_member( li, 0, Lt_get_data_size( trailer ), pos,
-                           member_size, dictionary_size ) )
-      return false;
+    if( !push_back_member( li, 0, data_size, pos, member_size,
+                           dictionary_size ) ) return false;
     }
-  if( pos != 0 || li->members <= 0 )
+  if( pos != 0 || li->members <= 0 || li->retval != 0 )
     {
     Li_free_member_vector( li );
     if( li->retval == 0 )
diff --git a/lzip_index.h b/lzip_index.h
index b76560e..5e2f77f 100644
--- a/lzip_index.h
+++ b/lzip_index.h
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -16,13 +16,13 @@
 */
 
 #ifndef INT64_MAX
-#define INT64_MAX  0x7FFFFFFFFFFFFFFFLL
+#define INT64_MAX 0x7FFFFFFFFFFFFFFFLL
 #endif
 
 
 struct Block
   {
-  long long pos, size;			/* pos + size <= INT64_MAX */
+  long long pos, size;		/* pos >= 0, size >= 0, pos + size <= INT64_MAX */
   };
 
 static inline void init_block( struct Block * const b,
@@ -40,10 +40,10 @@ struct Member
   };
 
 static inline void init_member( struct Member * const m,
-                                const long long dp, const long long ds,
-                                const long long mp, const long long ms,
+                                const long long dpos, const long long dsize,
+                                const long long mpos, const long long msize,
                                 const unsigned dict_size )
-  { init_block( &m->dblock, dp, ds ); init_block( &m->mblock, mp, ms );
+  { init_block( &m->dblock, dpos, dsize ); init_block( &m->mblock, mpos, msize );
     m->dictionary_size = dict_size; }
 
 struct Lzip_index
@@ -58,7 +58,7 @@ struct Lzip_index
   };
 
 bool Li_init( struct Lzip_index * const li, const int infd,
-              const bool ignore_trailing, const bool loose_trailing );
+              const struct Cl_options * const cl_opts );
 
 void Li_free( struct Lzip_index * const li );
 
diff --git a/main.c b/main.c
index 1205f48..e5a8dd5 100644
--- a/main.c
+++ b/main.c
@@ -1,5 +1,5 @@
 /* Clzip - LZMA lossless data compressor
-   Copyright (C) 2010-2022 Antonio Diaz Diaz.
+   Copyright (C) 2010-2023 Antonio Diaz Diaz.
 
    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
@@ -16,9 +16,9 @@
 */
 /*
    Exit status: 0 for a normal exit, 1 for environmental problems
-   (file not found, invalid flags, I/O errors, etc), 2 to indicate a
-   corrupt or invalid input file, 3 for an internal consistency error
-   (e.g., bug) which caused clzip to panic.
+   (file not found, invalid command-line options, I/O errors, etc), 2 to
+   indicate a corrupt or invalid input file, 3 for an internal consistency
+   error (e.g., bug) which caused clzip to panic.
 */
 
 #define _FILE_OFFSET_BITS 64
@@ -26,10 +26,10 @@
 #include <ctype.h>
 #include <errno.h>
 #include <fcntl.h>
-#include <limits.h>
+#include <limits.h>		/* SSIZE_MAX */
 #include <signal.h>
 #include <stdbool.h>
-#include <stdint.h>
+#include <stdint.h>		/* SIZE_MAX */
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -80,7 +80,7 @@
 int verbosity = 0;
 
 static const char * const program_name = "clzip";
-static const char * const program_year = "2022";
+static const char * const program_year = "2023";
 static const char * invocation_name = "clzip";		/* default value */
 
 static const struct { const char * from; const char * to; } known_extensions[] = {
@@ -111,14 +111,15 @@ static void show_help( void )
           "C++ compiler.\n"
           "\nLzip is a lossless data compressor with a user interface similar to the one\n"
           "of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov\n"
-          "chain-Algorithm' (LZMA) stream format and provides a 3 factor integrity\n"
-          "checking to maximize interoperability and optimize safety. Lzip can compress\n"
-          "about as fast as gzip (lzip -0) or compress most files more than bzip2\n"
-          "(lzip -9). Decompression speed is intermediate between gzip and bzip2.\n"
-          "Lzip is better than gzip and bzip2 from a data recovery perspective. Lzip\n"
-          "has been designed, written, and tested with great care to replace gzip and\n"
-          "bzip2 as the standard general-purpose compressed format for unix-like\n"
-          "systems.\n"
+          "chain-Algorithm' (LZMA) stream format to maximize interoperability. The\n"
+          "maximum dictionary size is 512 MiB so that any lzip file can be decompressed\n"
+          "on 32-bit machines. Lzip provides accurate and robust 3-factor integrity\n"
+          "checking. Lzip can compress about as fast as gzip (lzip -0) or compress most\n"
+          "files more than bzip2 (lzip -9). Decompression speed is intermediate between\n"
+          "gzip and bzip2. Lzip is better than gzip and bzip2 from a data recovery\n"
+          "perspective. Lzip has been designed, written, and tested with great care to\n"
+          "replace gzip and bzip2 as the standard general-purpose compressed format for\n"
+          "Unix-like systems.\n"
           "\nUsage: %s [options] [files]\n", invocation_name );
   printf( "\nOptions:\n"
           "  -h, --help                     display this help and exit\n"
@@ -126,7 +127,7 @@ static void show_help( void )
           "  -a, --trailing-error           exit with error status if trailing data\n"
           "  -b, --member-size=<bytes>      set member size limit in bytes\n"
           "  -c, --stdout                   write to standard output, keep input files\n"
-          "  -d, --decompress               decompress\n"
+          "  -d, --decompress               decompress, test compressed file integrity\n"
           "  -f, --force                    overwrite existing output files\n"
           "  -F, --recompress               force re-compression of compressed files\n"
           "  -k, --keep                     keep (don't delete) input files\n"
@@ -141,25 +142,27 @@ static void show_help( void )
           "  -0 .. -9                       set compression level [default 6]\n"
           "      --fast                     alias for -0\n"
           "      --best                     alias for -9\n"
+          "      --empty-error              exit with error status if empty member in file\n"
+          "      --marking-error            exit with error status if 1st LZMA byte not 0\n"
           "      --loose-trailing           allow trailing data seeming corrupt header\n"
           "\nIf no file names are given, or if a file is '-', clzip compresses or\n"
           "decompresses from standard input to standard output.\n"
           "Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,\n"
           "Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...\n"
-          "Dictionary sizes 12 to 29 are interpreted as powers of two, meaning 2^12\n"
-          "to 2^29 bytes.\n"
-          "\nThe bidimensional parameter space of LZMA can't be mapped to a linear\n"
-          "scale optimal for all files. If your files are large, very repetitive,\n"
-          "etc, you may need to use the options --dictionary-size and --match-length\n"
-          "directly to achieve optimal performance.\n"
+          "Dictionary sizes 12 to 29 are interpreted as powers of two, meaning 2^12 to\n"
+          "2^29 bytes.\n"
+          "\nThe bidimensional parameter space of LZMA can't be mapped to a linear scale\n"
+          "optimal for all files. If your files are large, very repetitive, etc, you\n"
+          "may need to use the options --dictionary-size and --match-length directly\n"
+          "to achieve optimal performance.\n"
           "\nTo extract all the files from archive 'foo.tar.lz', use the commands\n"
           "'tar -xf foo.tar.lz' or 'clzip -cd foo.tar.lz | tar -xf -'.\n"
-          "\nExit status: 0 for a normal exit, 1 for environmental problems (file\n"
-          "not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or\n"
-          "invalid input file, 3 for an internal consistency error (e.g., bug) which\n"
-          "caused clzip to panic.\n"
+          "\nExit status: 0 for a normal exit, 1 for environmental problems\n"
+          "(file not found, invalid command-line options, I/O errors, etc), 2 to\n"
+          "indicate a corrupt or invalid input file, 3 for an internal consistency\n"
+          "error (e.g., bug) which caused clzip to panic.\n"
           "\nThe ideas embodied in clzip are due to (at least) the following people:\n"
-          "Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the\n"
+          "Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrei Markov (for the\n"
           "definition of Markov chains), G.N.N. Martin (for the definition of range\n"
           "encoding), Igor Pavlov (for putting all the above together in LZMA), and\n"
           "Julian Seward (for bzip2's CLI).\n"
@@ -264,16 +267,15 @@ const char * bad_version( const unsigned version )
 
 const char * format_ds( const unsigned dictionary_size )
   {
-  enum { bufsize = 16, factor = 1024 };
+  enum { bufsize = 16, factor = 1024, n = 3 };
   static char buf[bufsize];
-  const char * const prefix[8] =
-    { "Ki", "Mi", "Gi", "Ti", "Pi", "Ei", "Zi", "Yi" };
+  const char * const prefix[n] = { "Ki", "Mi", "Gi" };
   const char * p = "";
   const char * np = "  ";
   unsigned num = dictionary_size;
   bool exact = ( num % factor == 0 );
 
-  int i; for( i = 0; i < 8 && ( num > 9999 || ( exact && num >= factor ) ); ++i )
+  int i; for( i = 0; i < n && ( num > 9999 || ( exact && num >= factor ) ); ++i )
     { num /= factor; if( num % factor != 0 ) exact = false;
       p = prefix[i]; np = ""; }
   snprintf( buf, bufsize, "%s%4u %sB", np, num, p );
@@ -287,12 +289,12 @@ void show_header( const unsigned dictionary_size )
   }
 
 
-/* separate large numbers >= 100_000 in groups of 3 digits using '_' */
+/* separate numbers of 5 or more digits in groups of 3 digits using '_' */
 static const char * format_num3( unsigned long long num )
   {
-  const char * const si_prefix = "kMGTPEZY";
-  const char * const binary_prefix = "KMGTPEZY";
-  enum { buffers = 8, bufsize = 4 * sizeof (long long) };
+  enum { buffers = 8, bufsize = 4 * sizeof num, n = 10 };
+  const char * const si_prefix = "kMGTPEZYRQ";
+  const char * const binary_prefix = "KMGTPEZYRQ";
   static char buffer[buffers][bufsize];	/* circle of static buffers for printf */
   static int current = 0;
   int i;
@@ -302,15 +304,15 @@ static const char * format_num3( unsigned long long num )
   if( num > 1024 )
     {
     char prefix = 0;			/* try binary first, then si */
-    for( i = 0; i < 8 && num >= 1024 && num % 1024 == 0; ++i )
+    for( i = 0; i < n && num != 0 && num % 1024 == 0; ++i )
       { num /= 1024; prefix = binary_prefix[i]; }
     if( prefix ) *(--p) = 'i';
     else
-      for( i = 0; i < 8 && num >= 1000 && num % 1000 == 0; ++i )
+      for( i = 0; i < n && num != 0 && num % 1000 == 0; ++i )
         { num /= 1000; prefix = si_prefix[i]; }
     if( prefix ) *(--p) = prefix;
     }
-  const bool split = num >= 100000;
+  const bool split = num >= 10000;
 
   for( i = 0; ; )
     {
@@ -321,6 +323,16 @@ static const char * format_num3( unsigned long long num )
   }
 
 
+void show_option_error( const char * const arg, const char * const msg,
+                        const char * const option_name )
+  {
+  if( verbosity >= 0 )
+    fprintf( stderr, "%s: '%s': %s option '%s'.\n",
+             program_name, arg, msg, option_name );
+  }
+
+
+/* Recognized formats: <num>k, <num>Ki, <num>[MGTPEZYRQ][i] */
 static unsigned long long getnum( const char * const arg,
                                   const char * const option_name,
                                   const unsigned long long llimit,
@@ -330,12 +342,8 @@ static unsigned long long getnum( const char * const arg,
   errno = 0;
   unsigned long long result = strtoull( arg, &tail, 0 );
   if( tail == arg )
-    {
-    if( verbosity >= 0 )
-      fprintf( stderr, "%s: Bad or missing numerical argument in "
-               "option '%s'.\n", program_name, option_name );
-    exit( 1 );
-    }
+    { show_option_error( arg, "Bad or missing numerical argument in",
+                         option_name ); exit( 1 ); }
 
   if( !errno && tail[0] )
     {
@@ -344,6 +352,8 @@ static unsigned long long getnum( const char * const arg,
     int i;
     switch( tail[0] )
       {
+      case 'Q': exponent = 10; break;
+      case 'R': exponent = 9; break;
       case 'Y': exponent = 8; break;
       case 'Z': exponent = 7; break;
       case 'E': exponent = 6; break;
@@ -355,12 +365,8 @@ static unsigned long long getnum( const char * const arg,
       case 'k': if( factor == 1000 ) exponent = 1; break;
       }
     if( exponent <= 0 )
-      {
-      if( verbosity >= 0 )
-        fprintf( stderr, "%s: Bad multiplier in numerical argument of "
-                 "option '%s'.\n", program_name, option_name );
-      exit( 1 );
-      }
+      { show_option_error( arg, "Bad multiplier in numerical argument of",
+                           option_name ); exit( 1 ); }
     for( i = 0; i < exponent; ++i )
       {
       if( ulimit / factor >= result ) result *= factor;
@@ -371,8 +377,8 @@ static unsigned long long getnum( const char * const arg,
   if( errno )
     {
     if( verbosity >= 0 )
-      fprintf( stderr, "%s: Numerical argument out of limits [%s,%s] "
-               "in option '%s'.\n", program_name, format_num3( llimit ),
+      fprintf( stderr, "%s: '%s': Value out of limits [%s,%s] in "
+               "option '%s'.\n", program_name, arg, format_num3( llimit ),
                format_num3( ulimit ), option_name );
     exit( 1 );
     }
@@ -453,7 +459,7 @@ static void set_d_outname( const char * const name, const int eindex )
   strcpy( output_filename, name );
   strcat( output_filename, ".out" );
   if( verbosity >= 1 )
-    fprintf( stderr, "%s: Can't guess original name for '%s' -- using '%s'\n",
+    fprintf( stderr, "%s: %s: Can't guess original name -- using '%s'\n",
              program_name, name, output_filename );
   }
 
@@ -474,9 +480,9 @@ int open_instream( const char * const name, struct stat * const in_statsp,
     if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || one_to_one ) ) )
       {
       if( verbosity >= 0 )
-        fprintf( stderr, "%s: Input file '%s' is not a regular file%s.\n",
+        fprintf( stderr, "%s: %s: Input file is not a regular file%s.\n",
                  program_name, name, ( can_read && one_to_one ) ?
-                 ",\n       and neither '-c' nor '-o' were specified" : "" );
+                 ",\n  and neither '-c' nor '-o' were specified" : "" );
       close( infd );
       infd = -1;
       }
@@ -492,7 +498,7 @@ static int open_instream2( const char * const name, struct stat * const in_stats
   if( program_mode == m_compress && !recompress && eindex >= 0 )
     {
     if( verbosity >= 0 )
-      fprintf( stderr, "%s: Input file '%s' already has '%s' suffix.\n",
+      fprintf( stderr, "%s: %s: Input file already has '%s' suffix.\n",
                program_name, name, known_extensions[eindex].from );
     return -1;
     }
@@ -500,6 +506,33 @@ static int open_instream2( const char * const name, struct stat * const in_stats
   }
 
 
+static bool make_dirs( const char * const name )
+  {
+  int i = strlen( name );
+  while( i > 0 && name[i-1] != '/' ) --i;	/* remove last component */
+  while( i > 0 && name[i-1] == '/' ) --i;	/* remove slash(es) */
+  const int dirsize = i;	/* size of dirname without trailing slash(es) */
+
+  for( i = 0; i < dirsize; )	/* if dirsize == 0, dirname is '/' or empty */
+    {
+    while( i < dirsize && name[i] == '/' ) ++i;
+    const int first = i;
+    while( i < dirsize && name[i] != '/' ) ++i;
+    if( first < i )
+      {
+      char partial[i+1]; memcpy( partial, name, i ); partial[i] = 0;
+      const mode_t mode = S_IRWXU | S_IRGRP | S_IXGRP | S_IROTH | S_IXOTH;
+      struct stat st;
+      if( stat( partial, &st ) == 0 )
+        { if( !S_ISDIR( st.st_mode ) ) { errno = ENOTDIR; return false; } }
+      else if( mkdir( partial, mode ) != 0 && errno != EEXIST )
+        return false;		/* if EEXIST, another process created the dir */
+      }
+    }
+  return true;
+  }
+
+
 static bool open_outstream( const bool force, const bool protect )
   {
   const mode_t usr_rw = S_IRUSR | S_IWUSR;
@@ -508,18 +541,21 @@ static bool open_outstream( const bool force, const bool protect )
   int flags = O_CREAT | O_WRONLY | O_BINARY;
   if( force ) flags |= O_TRUNC; else flags |= O_EXCL;
 
-  outfd = open( output_filename, flags, outfd_mode );
-  if( outfd >= 0 ) delete_output_on_interrupt = true;
-  else if( verbosity >= 0 )
-    {
+  outfd = -1;
+  const int len = strlen( output_filename );
+  if( len > 0 && output_filename[len-1] == '/' ) errno = EISDIR;
+  else {
+    if( !protect && !make_dirs( output_filename ) )
+      { show_file_error( output_filename,
+          "Error creating intermediate directory", errno ); return false; }
+    outfd = open( output_filename, flags, outfd_mode );
+    if( outfd >= 0 ) { delete_output_on_interrupt = true; return true; }
     if( errno == EEXIST )
-      fprintf( stderr, "%s: Output file '%s' already exists, skipping.\n",
-               program_name, output_filename );
-    else
-      fprintf( stderr, "%s: Can't create output file '%s': %s\n",
-               program_name, output_filename, strerror( errno ) );
+      { show_file_error( output_filename,
+          "Output file already exists, skipping.", 0 ); return false; }
     }
-  return ( outfd >= 0 );
+  show_file_error( output_filename, "Can't create output file", errno );
+  return false;
   }
 
 
@@ -537,12 +573,10 @@ void cleanup_and_fail( const int retval )
   if( delete_output_on_interrupt )
     {
     delete_output_on_interrupt = false;
-    if( verbosity >= 0 )
-      fprintf( stderr, "%s: Deleting output file '%s', if it exists.\n",
-               program_name, output_filename );
+    show_file_error( output_filename, "Deleting output file, if it exists.", 0 );
     if( outfd >= 0 ) { close( outfd ); outfd = -1; }
     if( remove( output_filename ) != 0 && errno != ENOENT )
-      show_error( "WARNING: deletion of output file (apparently) failed.", 0, false );
+      show_error( "warning: deletion of output file failed", errno, false );
     }
   exit( retval );
   }
@@ -596,10 +630,8 @@ static void close_and_set_permissions( const struct stat * const in_statsp )
         warning = true;
     }
   if( close( outfd ) != 0 )
-    {
-    show_error( "Error closing output file", errno, false );
-    cleanup_and_fail( 1 );
-    }
+    { show_file_error( output_filename, "Error closing output file", errno );
+      cleanup_and_fail( 1 ); }
   outfd = -1;
   delete_output_on_interrupt = false;
   if( in_statsp )
@@ -610,7 +642,8 @@ static void close_and_set_permissions( const struct stat * const in_statsp )
     if( utime( output_filename, &t ) != 0 ) warning = true;
     }
   if( warning && verbosity >= 1 )
-    show_error( "Can't change output file attributes.", 0, false );
+    show_file_error( output_filename,
+                     "warning: can't change output file attributes", errno );
   }
 
 
@@ -724,7 +757,7 @@ static int compress( const unsigned long long cfile_size,
   }
 
 
-static unsigned char xdigit( const unsigned value )
+static unsigned char xdigit( const unsigned value )	/* hex digit for 'value' */
   {
   if( value <= 9 ) return '0' + value;
   if( value <= 15 ) return 'A' + value - 10;
@@ -756,13 +789,13 @@ static bool show_trailing_data( const uint8_t * const data, const int size,
     Pp_show_msg( pp, buf );
     if( ignore_trailing == 0 ) show_file_error( pp->name, trailing_msg, 0 );
     }
-  return ( ignore_trailing > 0 );
+  return ignore_trailing > 0;
   }
 
 
 static int decompress( const unsigned long long cfile_size, const int infd,
-                struct Pretty_print * const pp, const bool ignore_trailing,
-                const bool loose_trailing, const bool testing )
+                       const struct Cl_options * const cl_opts,
+                       struct Pretty_print * const pp, const bool testing )
   {
   unsigned long long partial_file_pos = 0;
   struct Range_decoder rdec;
@@ -781,28 +814,25 @@ static int decompress( const unsigned long long cfile_size, const int infd,
       if( first_member )
         { show_file_error( pp->name, "File ends unexpectedly at member header.", 0 );
           retval = 2; }
-      else if( Lh_verify_prefix( header, size ) )
+      else if( Lh_check_prefix( header, size ) )
         { Pp_show_msg( pp, "Truncated header in multimember file." );
-          show_trailing_data( header, size, pp, true, -1 );
-          retval = 2; }
-      else if( size > 0 && !show_trailing_data( header, size, pp,
-                                                true, ignore_trailing ) )
-        retval = 2;
+          show_trailing_data( header, size, pp, true, -1 ); retval = 2; }
+      else if( size > 0 && !show_trailing_data( header, size, pp, true,
+                                 cl_opts->ignore_trailing ) ) retval = 2;
       break;
       }
-    if( !Lh_verify_magic( header ) )
+    if( !Lh_check_magic( header ) )
       {
       if( first_member )
         { show_file_error( pp->name, bad_magic_msg, 0 ); retval = 2; }
-      else if( !loose_trailing && Lh_verify_corrupt( header ) )
+      else if( !cl_opts->loose_trailing && Lh_check_corrupt( header ) )
         { Pp_show_msg( pp, corrupt_mm_msg );
-          show_trailing_data( header, size, pp, false, -1 );
-          retval = 2; }
-      else if( !show_trailing_data( header, size, pp, false, ignore_trailing ) )
-        retval = 2;
+          show_trailing_data( header, size, pp, false, -1 ); retval = 2; }
+      else if( !show_trailing_data( header, size, pp, false,
+                                    cl_opts->ignore_trailing ) ) retval = 2;
       break;
       }
-    if( !Lh_verify_version( header ) )
+    if( !Lh_check_version( header ) )
       { Pp_show_msg( pp, bad_version( Lh_version( header ) ) );
         retval = 2; break; }
     const unsigned dictionary_size = Lh_get_dictionary_size( header );
@@ -816,7 +846,7 @@ static int decompress( const unsigned long long cfile_size, const int infd,
     if( !LZd_init( &decoder, &rdec, dictionary_size, outfd ) )
       { Pp_show_msg( pp, mem_msg ); retval = 1; break; }
     show_dprogress( cfile_size, partial_file_pos, &rdec, pp );	/* init */
-    const int result = LZd_decode_member( &decoder, pp );
+    const int result = LZd_decode_member( &decoder, cl_opts, pp );
     partial_file_pos += Rd_member_position( &rdec );
     LZd_free( &decoder );
     if( result != 0 )
@@ -828,6 +858,8 @@ static int decompress( const unsigned long long cfile_size, const int infd,
                  "File ends unexpectedly" : "Decoder error",
                  partial_file_pos );
         }
+      else if( result == 5 ) Pp_show_msg( pp, empty_msg );
+      else if( result == 6 ) Pp_show_msg( pp, marking_msg );
       retval = 2; break;
       }
     if( verbosity >= 2 )
@@ -933,8 +965,8 @@ void show_dprogress( const unsigned long long cfile_size,
 
 int main( const int argc, const char * const argv[] )
   {
-  /* Mapping from gzip/bzip2 style 1..9 compression modes
-     to the corresponding LZMA compression modes. */
+  /* Mapping from gzip/bzip2 style 0..9 compression levels to the
+     corresponding LZMA compression parameters. */
   const struct Lzma_options option_mapping[] =
     {
     { 1 << 16,  16 },		/* -0 */
@@ -955,48 +987,50 @@ int main( const int argc, const char * const argv[] )
   const char * default_output_filename = "";
   enum Mode program_mode = m_compress;
   int i;
+  struct Cl_options cl_opts;		/* command-line options */
+  Cl_options_init( &cl_opts );
   bool force = false;
-  bool ignore_trailing = true;
   bool keep_input_files = false;
-  bool loose_trailing = false;
   bool recompress = false;
   bool to_stdout = false;
   bool zero = false;
   if( argc > 0 ) invocation_name = argv[0];
 
-  enum { opt_lt = 256 };
+  enum { opt_eer = 256, opt_lt, opt_mer };
   const struct ap_Option options[] =
     {
-    { '0', "fast",              ap_no  },
-    { '1', 0,                   ap_no  },
-    { '2', 0,                   ap_no  },
-    { '3', 0,                   ap_no  },
-    { '4', 0,                   ap_no  },
-    { '5', 0,                   ap_no  },
-    { '6', 0,                   ap_no  },
-    { '7', 0,                   ap_no  },
-    { '8', 0,                   ap_no  },
-    { '9', "best",              ap_no  },
-    { 'a', "trailing-error",    ap_no  },
-    { 'b', "member-size",       ap_yes },
-    { 'c', "stdout",            ap_no  },
-    { 'd', "decompress",        ap_no  },
-    { 'f', "force",             ap_no  },
-    { 'F', "recompress",        ap_no  },
-    { 'h', "help",              ap_no  },
-    { 'k', "keep",              ap_no  },
-    { 'l', "list",              ap_no  },
-    { 'm', "match-length",      ap_yes },
-    { 'n', "threads",           ap_yes },
-    { 'o', "output",            ap_yes },
-    { 'q', "quiet",             ap_no  },
-    { 's', "dictionary-size",   ap_yes },
-    { 'S', "volume-size",       ap_yes },
-    { 't', "test",              ap_no  },
-    { 'v', "verbose",           ap_no  },
-    { 'V', "version",           ap_no  },
-    { opt_lt, "loose-trailing", ap_no  },
-    {  0, 0,                    ap_no  } };
+    { '0', "fast",               ap_no  },
+    { '1', 0,                    ap_no  },
+    { '2', 0,                    ap_no  },
+    { '3', 0,                    ap_no  },
+    { '4', 0,                    ap_no  },
+    { '5', 0,                    ap_no  },
+    { '6', 0,                    ap_no  },
+    { '7', 0,                    ap_no  },
+    { '8', 0,                    ap_no  },
+    { '9', "best",               ap_no  },
+    { 'a', "trailing-error",     ap_no  },
+    { 'b', "member-size",        ap_yes },
+    { 'c', "stdout",             ap_no  },
+    { 'd', "decompress",         ap_no  },
+    { 'f', "force",              ap_no  },
+    { 'F', "recompress",         ap_no  },
+    { 'h', "help",               ap_no  },
+    { 'k', "keep",               ap_no  },
+    { 'l', "list",               ap_no  },
+    { 'm', "match-length",       ap_yes },
+    { 'n', "threads",            ap_yes },
+    { 'o', "output",             ap_yes },
+    { 'q', "quiet",              ap_no  },
+    { 's', "dictionary-size",    ap_yes },
+    { 'S', "volume-size",        ap_yes },
+    { 't', "test",               ap_no  },
+    { 'v', "verbose",            ap_no  },
+    { 'V', "version",            ap_no  },
+    { opt_eer, "empty-error",    ap_no  },
+    { opt_lt,  "loose-trailing", ap_no  },
+    { opt_mer, "marking-error",  ap_no  },
+    {  0, 0,                     ap_no  } };
 
   CRC32_init();
 
@@ -1020,7 +1054,7 @@ int main( const int argc, const char * const argv[] )
       case '5': case '6': case '7': case '8': case '9':
                 zero = ( code == '0' );
                 encoder_options = option_mapping[code-'0']; break;
-      case 'a': ignore_trailing = false; break;
+      case 'a': cl_opts.ignore_trailing = false; break;
       case 'b': member_size = getnum( arg, pn, 100000, max_member_size ); break;
       case 'c': to_stdout = true; break;
       case 'd': set_mode( &program_mode, m_decompress ); break;
@@ -1042,8 +1076,10 @@ int main( const int argc, const char * const argv[] )
       case 't': set_mode( &program_mode, m_test ); break;
       case 'v': if( verbosity < 4 ) ++verbosity; break;
       case 'V': show_version(); return 0;
-      case opt_lt: loose_trailing = true; break;
-      default : internal_error( "uncaught option." );
+      case opt_eer: cl_opts.ignore_empty = false; break;
+      case opt_lt:  cl_opts.loose_trailing = true; break;
+      case opt_mer: cl_opts.ignore_marking = false; break;
+      default: internal_error( "uncaught option." );
       }
     } /* end process options */
 
@@ -1065,7 +1101,7 @@ int main( const int argc, const char * const argv[] )
     }
 
   if( program_mode == m_list )
-    return list_files( filenames, num_filenames, ignore_trailing, loose_trailing );
+    return list_files( filenames, num_filenames, &cl_opts );
 
   if( program_mode == m_compress )
     {
@@ -1119,7 +1155,7 @@ int main( const int argc, const char * const argv[] )
                              eindex, one_to_one, recompress );
       if( infd < 0 ) { set_retval( &retval, 1 ); continue; }
       if( !check_tty_in( pp.name, infd, program_mode, &retval ) ) continue;
-      if( one_to_one )			/* open outfd after verifying infd */
+      if( one_to_one )			/* open outfd after checking infd */
         {
         if( program_mode == m_compress )
           set_c_outname( input_filename, true, true, volume_size > 0 );
@@ -1132,7 +1168,7 @@ int main( const int argc, const char * const argv[] )
     if( one_to_one && !check_tty_out( program_mode ) )
       { set_retval( &retval, 1 ); return retval; }	/* don't delete a tty */
 
-    if( to_file && outfd < 0 )		/* open outfd after verifying infd */
+    if( to_file && outfd < 0 )		/* open outfd after checking infd */
       {
       if( program_mode == m_compress ) set_c_outname( default_output_filename,
                                        filenames_given, false, volume_size > 0 );
@@ -1154,8 +1190,7 @@ int main( const int argc, const char * const argv[] )
       tmp = compress( cfile_size, member_size, volume_size, infd,
                       &encoder_options, &pp, in_statsp, zero );
     else
-      tmp = decompress( cfile_size, infd, &pp, ignore_trailing,
-                        loose_trailing, program_mode == m_test );
+      tmp = decompress( cfile_size, infd, &cl_opts, &pp, program_mode == m_test );
     if( close( infd ) != 0 )
       { show_file_error( pp.name, "Error closing input file", errno );
         set_retval( &tmp, 1 ); }
diff --git a/testsuite/check.sh b/testsuite/check.sh
index 0b80dd1..f880280 100755
--- a/testsuite/check.sh
+++ b/testsuite/check.sh
@@ -1,6 +1,6 @@
 #! /bin/sh
 # check script for Clzip - LZMA lossless data compressor
-# Copyright (C) 2010-2022 Antonio Diaz Diaz.
+# Copyright (C) 2010-2023 Antonio Diaz Diaz.
 #
 # This script is free software: you have unlimited permission
 # to copy, distribute, and modify it.
@@ -32,6 +32,8 @@ cat "${testdir}"/test.txt > in || framework_failure
 in_lz="${testdir}"/test.txt.lz
 in_em="${testdir}"/test_em.txt.lz
 fox_lz="${testdir}"/fox.lz
+fox6_lz="${testdir}"/fox6.lz
+f6mk_lz="${testdir}"/fox6_mark.lz
 fail=0
 test_failed() { fail=1 ; printf " $1" ; [ -z "$2" ] || printf "($2)" ; }
 
@@ -70,12 +72,13 @@ done
 [ ! -e out.lz ] || test_failed $LINENO
 "${LZIP}" -qf -S100k -o out in in
 [ $? = 1 ] || test_failed $LINENO
+{ [ ! -e out ] && [ ! -e out.lz ] ; } || test_failed $LINENO
 # these are for code coverage
 "${LZIP}" -lt "${in_lz}" 2> /dev/null
 [ $? = 1 ] || test_failed $LINENO
-"${LZIP}" -cdl "${in_lz}" > out 2> /dev/null
+"${LZIP}" -cdl "${in_lz}" 2> /dev/null
 [ $? = 1 ] || test_failed $LINENO
-"${LZIP}" -cdt "${in_lz}" > out 2> /dev/null
+"${LZIP}" -cdt "${in_lz}" 2> /dev/null
 [ $? = 1 ] || test_failed $LINENO
 "${LZIP}" -t -- nx_file.lz 2> /dev/null
 [ $? = 1 ] || test_failed $LINENO
@@ -100,39 +103,43 @@ done
 printf "LZIP\001-.............................." | "${LZIP}" -t 2> /dev/null
 printf "LZIP\002-.............................." | "${LZIP}" -t 2> /dev/null
 printf "LZIP\001+.............................." | "${LZIP}" -t 2> /dev/null
-rm -f out || framework_failure
 
 printf "\ntesting decompression..."
 
 for i in "${in_lz}" "${in_em}" ; do
 	"${LZIP}" -lq "$i" || test_failed $LINENO "$i"
 	"${LZIP}" -t "$i" || test_failed $LINENO "$i"
-	"${LZIP}" -d "$i" -o copy || test_failed $LINENO "$i"
-	cmp in copy || test_failed $LINENO "$i"
-	"${LZIP}" -cd "$i" > copy || test_failed $LINENO "$i"
-	cmp in copy || test_failed $LINENO "$i"
-	"${LZIP}" -d "$i" -o - > copy || test_failed $LINENO "$i"
-	cmp in copy || test_failed $LINENO "$i"
-	"${LZIP}" -d < "$i" > copy || test_failed $LINENO "$i"
-	cmp in copy || test_failed $LINENO "$i"
-	rm -f copy || framework_failure
+	"${LZIP}" -d "$i" -o out || test_failed $LINENO "$i"
+	cmp in out || test_failed $LINENO "$i"
+	"${LZIP}" -cd "$i" > out || test_failed $LINENO "$i"
+	cmp in out || test_failed $LINENO "$i"
+	"${LZIP}" -d "$i" -o - > out || test_failed $LINENO "$i"
+	cmp in out || test_failed $LINENO "$i"
+	"${LZIP}" -d < "$i" > out || test_failed $LINENO "$i"
+	cmp in out || test_failed $LINENO "$i"
+	rm -f out || framework_failure
 done
 
-lines=$("${LZIP}" -tvv "${in_em}" 2>&1 | wc -l) || test_failed $LINENO
+lines=`"${LZIP}" -tvv "${in_em}" 2>&1 | wc -l` || test_failed $LINENO
 [ "${lines}" -eq 8 ] || test_failed $LINENO "${lines}"
+"${LZIP}" -tq "${in_em}" --empty-error
+[ $? = 2 ] || test_failed $LINENO
 
-lines=$("${LZIP}" -lvv "${in_em}" | wc -l) || test_failed $LINENO
+lines=`"${LZIP}" -lvv "${in_em}" | wc -l` || test_failed $LINENO
 [ "${lines}" -eq 11 ] || test_failed $LINENO "${lines}"
+"${LZIP}" -lq "${in_em}" --empty-error
+[ $? = 2 ] || test_failed $LINENO
 
-"${LZIP}" -cd "${fox_lz}" > fox || test_failed $LINENO
-cat "${in_lz}" > copy.lz || framework_failure
-"${LZIP}" -dk copy.lz || test_failed $LINENO
-cmp in copy || test_failed $LINENO
-cat fox > copy || framework_failure
 cat "${in_lz}" > out.lz || framework_failure
+"${LZIP}" -dk out.lz || test_failed $LINENO
+cmp in out || test_failed $LINENO
 rm -f out || framework_failure
+"${LZIP}" -cd "${fox_lz}" > fox || test_failed $LINENO
+cat fox > copy || framework_failure
+cat "${in_lz}" > copy.lz || framework_failure
 "${LZIP}" -d copy.lz out.lz 2> /dev/null	# skip copy, decompress out
 [ $? = 1 ] || test_failed $LINENO
+[ ! -e out.lz ] || test_failed $LINENO
 cmp fox copy || test_failed $LINENO
 cmp in out || test_failed $LINENO
 "${LZIP}" -df copy.lz || test_failed $LINENO
@@ -140,15 +147,15 @@ cmp in out || test_failed $LINENO
 cmp in copy || test_failed $LINENO
 rm -f copy out || framework_failure
 
-cat "${in_lz}" > copy.lz || framework_failure
-"${LZIP}" -d -S100k copy.lz || test_failed $LINENO	# ignore -S
-[ ! -e copy.lz ] || test_failed $LINENO
-cmp in copy || test_failed $LINENO
+cat "${in_lz}" > out.lz || framework_failure
+"${LZIP}" -d -S100k out.lz || test_failed $LINENO	# ignore -S
+[ ! -e out.lz ] || test_failed $LINENO
+cmp in out || test_failed $LINENO
 
-printf "to be overwritten" > copy || framework_failure
-"${LZIP}" -df -o copy < "${in_lz}" || test_failed $LINENO
-cmp in copy || test_failed $LINENO
-rm -f out copy || framework_failure
+printf "to be overwritten" > out || framework_failure
+"${LZIP}" -df -o out < "${in_lz}" || test_failed $LINENO
+cmp in out || test_failed $LINENO
+rm -f out || framework_failure
 "${LZIP}" -d -o ./- "${in_lz}" || test_failed $LINENO
 cmp in ./- || test_failed $LINENO
 rm -f ./- || framework_failure
@@ -157,11 +164,11 @@ cmp in ./- || test_failed $LINENO
 rm -f ./- || framework_failure
 
 cat "${in_lz}" > anyothername || framework_failure
-"${LZIP}" -dv - anyothername - < "${in_lz}" > copy 2> /dev/null ||
+"${LZIP}" -dv - anyothername - < "${in_lz}" > out 2> /dev/null ||
 	test_failed $LINENO
-cmp in copy || test_failed $LINENO
+cmp in out || test_failed $LINENO
 cmp in anyothername.out || test_failed $LINENO
-rm -f copy anyothername.out || framework_failure
+rm -f out anyothername.out || framework_failure
 
 "${LZIP}" -lq in "${in_lz}"
 [ $? = 2 ] || test_failed $LINENO
@@ -171,62 +178,82 @@ rm -f copy anyothername.out || framework_failure
 [ $? = 2 ] || test_failed $LINENO
 "${LZIP}" -tq nx_file.lz "${in_lz}"
 [ $? = 1 ] || test_failed $LINENO
-"${LZIP}" -cdq in "${in_lz}" > copy
+"${LZIP}" -cdq in "${in_lz}" > out
 [ $? = 2 ] || test_failed $LINENO
-cat copy in | cmp in - || test_failed $LINENO		# copy must be empty
-"${LZIP}" -cdq nx_file.lz "${in_lz}" > copy
+cat out in | cmp in - || test_failed $LINENO		# out must be empty
+"${LZIP}" -cdq nx_file.lz "${in_lz}" > out	# skip nx_file, decompress in
 [ $? = 1 ] || test_failed $LINENO
-cmp in copy || test_failed $LINENO
-rm -f copy || framework_failure
-cat "${in_lz}" > copy.lz || framework_failure
+cmp in out || test_failed $LINENO
+rm -f out || framework_failure
+cat "${in_lz}" > out.lz || framework_failure
 for i in 1 2 3 4 5 6 7 ; do
-	printf "g" >> copy.lz || framework_failure
-	"${LZIP}" -alvv copy.lz "${in_lz}" > /dev/null 2>&1
+	printf "g" >> out.lz || framework_failure
+	"${LZIP}" -alvv out.lz "${in_lz}" > /dev/null 2>&1
 	[ $? = 2 ] || test_failed $LINENO $i
-	"${LZIP}" -atvvvv copy.lz "${in_lz}" 2> /dev/null
+	"${LZIP}" -atvvvv out.lz "${in_lz}" 2> /dev/null
 	[ $? = 2 ] || test_failed $LINENO $i
 done
-"${LZIP}" -dq in copy.lz
+"${LZIP}" -dq in out.lz
 [ $? = 2 ] || test_failed $LINENO
-[ -e copy.lz ] || test_failed $LINENO
-[ ! -e copy ] || test_failed $LINENO
+[ -e out.lz ] || test_failed $LINENO
+[ ! -e out ] || test_failed $LINENO
 [ ! -e in.out ] || test_failed $LINENO
-"${LZIP}" -dq nx_file.lz copy.lz
+"${LZIP}" -dq nx_file.lz out.lz
 [ $? = 1 ] || test_failed $LINENO
-[ ! -e copy.lz ] || test_failed $LINENO
+[ ! -e out.lz ] || test_failed $LINENO
 [ ! -e nx_file ] || test_failed $LINENO
-cmp in copy || test_failed $LINENO
+cmp in out || test_failed $LINENO
+rm -f out || framework_failure
 
 cat in in > in2 || framework_failure
 "${LZIP}" -lq "${in_lz}" "${in_lz}" || test_failed $LINENO
 "${LZIP}" -t "${in_lz}" "${in_lz}" || test_failed $LINENO
-"${LZIP}" -cd "${in_lz}" "${in_lz}" -o out > copy2 || test_failed $LINENO
+"${LZIP}" -cd "${in_lz}" "${in_lz}" -o out > out2 || test_failed $LINENO
 [ ! -e out ] || test_failed $LINENO			# override -o
-cmp in2 copy2 || test_failed $LINENO
-rm -f copy2 || framework_failure
-"${LZIP}" -d "${in_lz}" "${in_lz}" -o copy2 || test_failed $LINENO
-cmp in2 copy2 || test_failed $LINENO
-rm -f copy2 || framework_failure
-
-cat "${in_lz}" "${in_lz}" > copy2.lz || framework_failure
-printf "\ngarbage" >> copy2.lz || framework_failure
-"${LZIP}" -tvvvv copy2.lz 2> /dev/null || test_failed $LINENO
-"${LZIP}" -alq copy2.lz
+cmp in2 out2 || test_failed $LINENO
+rm -f out2 || framework_failure
+"${LZIP}" -d "${in_lz}" "${in_lz}" -o out2 || test_failed $LINENO
+cmp in2 out2 || test_failed $LINENO
+rm -f out2 || framework_failure
+
+cat "${in_lz}" "${in_lz}" > out2.lz || framework_failure
+printf "\ngarbage" >> out2.lz || framework_failure
+"${LZIP}" -tvvvv out2.lz 2> /dev/null || test_failed $LINENO
+"${LZIP}" -alq out2.lz
+[ $? = 2 ] || test_failed $LINENO
+"${LZIP}" -atq out2.lz
+[ $? = 2 ] || test_failed $LINENO
+"${LZIP}" -atq < out2.lz
 [ $? = 2 ] || test_failed $LINENO
-"${LZIP}" -atq copy2.lz
+"${LZIP}" -adkq out2.lz
 [ $? = 2 ] || test_failed $LINENO
-"${LZIP}" -atq < copy2.lz
+[ ! -e out2 ] || test_failed $LINENO
+"${LZIP}" -adkq -o out2 < out2.lz
 [ $? = 2 ] || test_failed $LINENO
-"${LZIP}" -adkq copy2.lz
+[ ! -e out2 ] || test_failed $LINENO
+printf "to be overwritten" > out2 || framework_failure
+"${LZIP}" -df out2.lz || test_failed $LINENO
+cmp in2 out2 || test_failed $LINENO
+rm -f out2 || framework_failure
+
+"${LZIP}" -cd "${fox6_lz}" > out || test_failed $LINENO
+"${LZIP}" -cd "${f6mk_lz}" > copy || test_failed $LINENO
+cmp copy out || test_failed $LINENO
+rm -f copy out || framework_failure
+"${LZIP}" -lq "${f6mk_lz}" --marking-error
 [ $? = 2 ] || test_failed $LINENO
-[ ! -e copy2 ] || test_failed $LINENO
-"${LZIP}" -adkq -o copy2 < copy2.lz
+"${LZIP}" -tq "${f6mk_lz}" --marking-error
 [ $? = 2 ] || test_failed $LINENO
-[ ! -e copy2 ] || test_failed $LINENO
-printf "to be overwritten" > copy2 || framework_failure
-"${LZIP}" -df copy2.lz || test_failed $LINENO
-cmp in2 copy2 || test_failed $LINENO
-rm -f copy2 || framework_failure
+
+"${LZIP}" -d "${fox_lz}" -o a/b/c/fox || test_failed $LINENO
+cmp fox a/b/c/fox || test_failed $LINENO
+rm -rf a || framework_failure
+"${LZIP}" -d -o a/b/c/fox < "${fox_lz}" || test_failed $LINENO
+cmp fox a/b/c/fox || test_failed $LINENO
+rm -rf a || framework_failure
+"${LZIP}" -dq "${fox_lz}" -o a/b/c/
+[ $? = 1 ] || test_failed $LINENO
+[ ! -e a ] || test_failed $LINENO
 
 printf "\ntesting   compression..."
 
@@ -234,14 +261,16 @@ printf "\ntesting   compression..."
 [ ! -e out3.lz ] || test_failed $LINENO			# override -o and -S
 "${LZIP}" -0f in in --output=copy2.lz || test_failed $LINENO
 "${LZIP}" -d copy2.lz -o out2 || test_failed $LINENO
+[ -e copy2.lz ] || test_failed $LINENO
 cmp in2 out2 || test_failed $LINENO
 rm -f in2 out2 copy2.lz || framework_failure
 
-"${LZIP}" -cf "${in_lz}" > out 2> /dev/null	# /dev/null is a tty on OS/2
+"${LZIP}" -cf "${in_lz}" > lzlz 2> /dev/null	# /dev/null is a tty on OS/2
 [ $? = 1 ] || test_failed $LINENO
-"${LZIP}" -Fvvm36 -o - "${in_lz}" > out 2> /dev/null || test_failed $LINENO
-"${LZIP}" -cd out | "${LZIP}" -d > copy || test_failed $LINENO
-cmp in copy || test_failed $LINENO
+"${LZIP}" -Fvvm36 -o - "${in_lz}" > lzlz 2> /dev/null || test_failed $LINENO
+"${LZIP}" -cd lzlz | "${LZIP}" -d > out || test_failed $LINENO
+cmp in out || test_failed $LINENO
+rm -f lzlz out || framework_failure
 
 "${LZIP}" -0 -o ./- in || test_failed $LINENO
 "${LZIP}" -cd ./- | cmp in - || test_failed $LINENO
@@ -253,10 +282,10 @@ rm -f ./-.lz || framework_failure
 
 for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do
 	"${LZIP}" -k -$i in || test_failed $LINENO $i
-	mv -f in.lz copy.lz || test_failed $LINENO $i
-	printf "garbage" >> copy.lz || framework_failure
-	"${LZIP}" -df copy.lz || test_failed $LINENO $i
-	cmp in copy || test_failed $LINENO $i
+	mv in.lz out.lz || test_failed $LINENO $i
+	printf "garbage" >> out.lz || framework_failure
+	"${LZIP}" -df out.lz || test_failed $LINENO $i
+	cmp in out || test_failed $LINENO $i
 
 	"${LZIP}" -$i in -c > out || test_failed $LINENO $i
 	"${LZIP}" -$i in -o o_out || test_failed $LINENO $i	# don't add .lz
@@ -278,7 +307,7 @@ for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do
 	"${LZIP}" -df -o copy < out.lz || test_failed $LINENO $i
 	cmp in copy || test_failed $LINENO $i
 done
-rm -f out out.lz || framework_failure
+rm -f copy out.lz || framework_failure
 
 cat in in in in in in in in > in8 || framework_failure
 "${LZIP}" -1s12 -S100k in8 || test_failed $LINENO
@@ -303,11 +332,12 @@ rm -f in8 || framework_failure
 "${LZIP}" -cd out00001.lz out00002.lz | cmp in8.lz - || test_failed $LINENO
 [ ! -e out00003.lz ] || test_failed $LINENO
 rm -f out00001.lz || framework_failure
-"${LZIP}" -1 -S100k -o out < in8.lz || test_failed $LINENO
-"${LZIP}" -t out00001.lz out00002.lz || test_failed $LINENO
-"${LZIP}" -cd out00001.lz out00002.lz | cmp in8.lz - || test_failed $LINENO
-[ ! -e out00003.lz ] || test_failed $LINENO
-rm -f out00001.lz out00002.lz || framework_failure
+"${LZIP}" -1 -S100k -o a/b/c/out < in8.lz || test_failed $LINENO
+"${LZIP}" -t a/b/c/out00001.lz a/b/c/out00002.lz || test_failed $LINENO
+"${LZIP}" -cd a/b/c/out00001.lz a/b/c/out00002.lz | cmp in8.lz - ||
+	test_failed $LINENO
+[ ! -e a/b/c/out00003.lz ] || test_failed $LINENO
+rm -rf a || framework_failure
 "${LZIP}" -0 -F -S100k in8.lz || test_failed $LINENO
 "${LZIP}" -t in8.lz00001.lz in8.lz00002.lz || test_failed $LINENO
 "${LZIP}" -cd in8.lz00001.lz in8.lz00002.lz | cmp in8.lz - || test_failed $LINENO
@@ -318,16 +348,23 @@ rm -f in8.lz00001.lz in8.lz00002.lz || framework_failure
 "${LZIP}" -cd in8.lz.lz | cmp in8.lz - || test_failed $LINENO
 rm -f in8.lz in8.lz.lz || framework_failure
 
+"${LZIP}" fox -o a/b/c/fox.lz || test_failed $LINENO
+cmp "${fox_lz}" a/b/c/fox.lz || test_failed $LINENO
+rm -rf a || framework_failure
+"${LZIP}" -o a/b/c/fox.lz < fox || test_failed $LINENO
+cmp "${fox_lz}" a/b/c/fox.lz || test_failed $LINENO
+rm -rf a || framework_failure
+
 printf "\ntesting bad input..."
 
 headers='LZIp LZiP LZip LzIP LzIp LziP lZIP lZIp lZiP lzIP'
 body='\001\014\000\203\377\373\377\377\300\000\000\000\000\000\000\000\000\000\000\000\000\000\000\000$\000\000\000\000\000\000\000'
-cat "${in_lz}" > int.lz
-printf "LZIP${body}" >> int.lz
+cat "${in_lz}" > int.lz || framework_failure
+printf "LZIP${body}" >> int.lz || framework_failure
 if "${LZIP}" -tq int.lz ; then
 	for header in ${headers} ; do
-		printf "${header}${body}" > int.lz	# first member
-		"${LZIP}" -lq int.lz
+		printf "${header}${body}" > int.lz || framework_failure
+		"${LZIP}" -lq int.lz			# first member
 		[ $? = 2 ] || test_failed $LINENO ${header}
 		"${LZIP}" -tq int.lz
 		[ $? = 2 ] || test_failed $LINENO ${header}
@@ -343,9 +380,9 @@ if "${LZIP}" -tq int.lz ; then
 		[ $? = 2 ] || test_failed $LINENO ${header}
 		"${LZIP}" -cdq --loose-trailing int.lz > /dev/null
 		[ $? = 2 ] || test_failed $LINENO ${header}
-		cat "${in_lz}" > int.lz
-		printf "${header}${body}" >> int.lz	# trailing data
-		"${LZIP}" -lq int.lz
+		cat "${in_lz}" > int.lz || framework_failure
+		printf "${header}${body}" >> int.lz || framework_failure
+		"${LZIP}" -lq int.lz			# trailing data
 		[ $? = 2 ] || test_failed $LINENO ${header}
 		"${LZIP}" -tq int.lz
 		[ $? = 2 ] || test_failed $LINENO ${header}
@@ -400,15 +437,15 @@ if dd if=in3.lz of=trunc.lz bs=14752 count=1 2> /dev/null &&
 		[ $? = 2 ] || test_failed $LINENO $i
 		"${LZIP}" -tq < trunc.lz
 		[ $? = 2 ] || test_failed $LINENO $i
-		"${LZIP}" -cdq trunc.lz > out
+		"${LZIP}" -cdq trunc.lz > /dev/null
 		[ $? = 2 ] || test_failed $LINENO $i
-		"${LZIP}" -dq < trunc.lz > out
+		"${LZIP}" -dq < trunc.lz > /dev/null
 		[ $? = 2 ] || test_failed $LINENO $i
 	done
 else
 	printf "\nwarning: skipping truncation test: 'dd' does not work on your system."
 fi
-rm -f in2.lz in3.lz trunc.lz out || framework_failure
+rm -f in2.lz in3.lz trunc.lz || framework_failure
 
 cat "${in_lz}" > ingin.lz || framework_failure
 printf "g" >> ingin.lz || framework_failure
@@ -419,17 +456,17 @@ cat "${in_lz}" >> ingin.lz || framework_failure
 [ $? = 2 ] || test_failed $LINENO
 "${LZIP}" -atq < ingin.lz
 [ $? = 2 ] || test_failed $LINENO
-"${LZIP}" -acdq ingin.lz > out
+"${LZIP}" -acdq ingin.lz > /dev/null
 [ $? = 2 ] || test_failed $LINENO
-"${LZIP}" -adq < ingin.lz > out
+"${LZIP}" -adq < ingin.lz > /dev/null
 [ $? = 2 ] || test_failed $LINENO
 "${LZIP}" -t ingin.lz || test_failed $LINENO
 "${LZIP}" -t < ingin.lz || test_failed $LINENO
-"${LZIP}" -cd ingin.lz > copy || test_failed $LINENO
-cmp in copy || test_failed $LINENO
-"${LZIP}" -d < ingin.lz > copy || test_failed $LINENO
-cmp in copy || test_failed $LINENO
-rm -f copy ingin.lz out || framework_failure
+"${LZIP}" -cd ingin.lz > out || test_failed $LINENO
+cmp in out || test_failed $LINENO
+"${LZIP}" -d < ingin.lz > out || test_failed $LINENO
+cmp in out || test_failed $LINENO
+rm -f out ingin.lz || framework_failure
 
 echo
 if [ ${fail} = 0 ] ; then
diff --git a/testsuite/fox6.lz b/testsuite/fox6.lz
new file mode 100644
index 0000000..8401b99
--- /dev/null
+++ b/testsuite/fox6.lz
diff --git a/testsuite/fox6_mark.lz b/testsuite/fox6_mark.lz
new file mode 100644
index 0000000..32b2ac0
--- /dev/null
+++ b/testsuite/fox6_mark.lz