Man Linux: Main Page and Category List


       fmerge - merge files


       fmerge [ option...  ] basefile fileA fileB

       fmerge -Help

       fmerge -VERSion


       The fmerge program is used to compare the changes between two different
       descendants of a base file, and creates an output file  which  contains
       both sets of changes.  This is useful when two users both take the same
       version of a file and make independent edits to it, and then later want
       to create a file which contains both sets of edits.  In such a use, the
       original file that both sets of edits is derived  from  is  called  the
       base  file.   The  two files containing the edits are called file A and
       file B.

       The command:
              fmerge basefile fileA fileB -o outputfile
       produces the output file which contains the edits contained in  file  A
       and  file  B,  based  on the base file .  If the -Output option was not
       used, or if no outputfile is specified, then the merged lines are typed
       to  the  standard output.  The order of specifying file A and file B is
       usually unimportant.

       The fmerge program can also be used to remove earlier edits made  to  a
       module.   To do this, make the version containing the edits you want to
       delete be the basefile.  Make the version previous to the edit you want
       deleted  be  file A.  Finally, make the most recent version of the file
       which contains the other edits (including the one you want deleted)  be
       file  B.   Then the result of merging will be the newest version of the
       module minus the changes made by the  edit  you  wanted  removed.   For
       example,  if  three  successive  versions of some module have the names
       edit10, edit11 and edit12, and you want the changes done by  edit11  to
       be  undone, but still want the changes done by edit12, then you use the
              fmerge edit11 edit10 edit12 -o outputfile

       While merging the two sets of edits, fmerge may discover conflicts.   A
       conflict  occurs when the same line of the base file is changed by both
       of the two sets of edits.  The change can be due  to  new  lines  being
       inserted,  lines  being  deleted,  or  both.  When conflicts occur, the
       output file contains conflict identification  lines,  which  are  lines
       containing the string ////.  These lines indicate the region where
       the two sets of edits are incompatible.  You must then edit the  output
       file  and  remove  these  lines,  and in addition correct the conflicts
       manually in order to produce the correct result.


       The following options are understood:

       -Conflicts [ conflictfile ]
               Since conflicts due to deletions are invisible  in  the  output
               file,  and  inserts  do  not  specify  which  of  the two edits
               inserted the lines, there is an alternative output format  from
               the  fmerge program.  This output format describes what happens
               to each line of the base file, so that conflicts are easier  to
               detect and fix.  The command:
                      fmerge basefile fileA fileB -c conflictfile
               produces  the  file  describing  the  results  of  the merge in
               detail.  If the -Conflicts  option  is  specified  without  any
               conflictfile  name, then the conflicts are send to the standard

               If there are conflicts,  and  the  -Conflicts  options  is  not
               specified, the fmerge program will exit with a status of 1.

               The conflict file contains lines which contain three characters
               and then some text.  The first three characters  describe  what
               is  happening to the base file at that point.  These characters
               are the following:

                      IA      This line was inserted by file A.

                      DA      This line was deleted by file A.

                      IB      This line was inserted by file B.

                      DB      This line was deleted by file B.

                              This line is unchanged.

                      X       This is a conflict identification line.

                      U       There are unspecified unchanged lines here.

               Each set of conflicts is flagged by three identification lines.
               The  first  line  indicates  the beginning of the conflict, and
               specifies the line numbers for the base file and two  divergent
               files.  The second conflict identification line separates lines
               changed by file A from lines changed  by  file  B.   The  third
               conflict identification indicates the end of the conflict.

               You  can edit this conflict file to remove the conflicts.  This
               involves  deleting  the  conflict  identification  lines,   and
               changing   the  conflicting  lines  as  necessary  to  fix  the
               conflict.  While doing this,  remember  to  leave  three  blank
               characters  at  the  front  of  any  new lines you insert while
               correcting the conflicts.  When you are done, there  should  be
               no  lines which begin with an ’X’ in the file.  All other lines
               can remain.  Then you can use the command:
                      fmerge conflictfile -o outputfile
               to create the new output file which has the desired data.  Once
               again,  if  no -Output option or outputfile is used, the output
               is send to the standard output.

       -Unchanged number
               Besides physical conflicts, there  can  be  logical  conflicts.
               These are changes made to different lines in the base file such
               that the program is no longer correct.  Such  conflicts  cannot
               be  detected  by  a  program,  and  so  these  must  be checked
               manually.  In order to make this process easier, the -Unchanged
               option  can  be used to reduce the size of the conflict file to
               only include regions near changed lines.  This file can then be
               examined  in order to detect possible logical conflicts.  As an
               example, the command:
                      fmerge basefile fileA fileB -c -u 3
               will send to the standard output all  changes  made  by  either
               sets of edits, with only three unchanged lines surrounding each

               When using  the  -Unchanged  option,  the  conflict  file  will
               contain  lines  starting  with  ’U’.  These represent unchanged
               lines, and the number following the letter  is  the  number  of
               unchanged lines.  The resulting conflict file cannot be read to
               produce an output file because of the missing lines.   If  this
               is attempted, an error will be generated.

               It  is  possible to use both -Output and -Conflicts in the same
               command.  Thus you can produce the output file which  you  hope
               is  correct,  and  also produce the conflict file which you can
               use to check for logical conflicts.

       -Verbose [ number ]
               This option can be specified with any other action, and outputs
               status  information  about the progress of the action.  This is
               useful for debugging of problems, or just  for  amusement  when
               the  system  is  slow  or  a large file is being processed.  It
               accepts a  numeric  argument  to  indicate  the  verbosity  for
               output.  The levels are as follows:

               0   No output at all (except for errors).

               1   Single‐line output describing action (default).

               2   Detailed status as action proceeds.

       -Failures number
               This option restricts the number of physical conflicts that are
               allowed before failing.  This is used if you are not interested
               in the results if there are too many conflicts.

               Give some help on how to use the fmerge program.

               Ignore all conflicts.

               The  option  may  be  used  to  suppress  conflicts  which make
               identical deletes, or identical inserts, or identical  changes.
               This  is often desirable when merging two source code branches.

               Show what version of fmerge is running.

       All options may be abbreviated; the abbreviation is documented  as  the
       upper  case  letters,  all  lower  case letters and underscores (_) are
       optional.  You must use consecutive sequences of optional letters.

       All options are case insensitive, you may type them in  upper  case  or
       lower case or a combination of both, case is not important.

       For example: the arguments "-help, "-HELP" and "-h" are all interpreted
       to mean the -Help option.  The argument "-hlp" will not be  understood,
       because consecutive optional characters were not supplied.

       Options  and  other  command line arguments may be mixed arbitrarily on
       the command line.

       The GNU long option names are understood.  Since all option  names  for
       fmerge  are  long,  this  means  ignoring  the  extra leading ’-’.  The
       "-option=value" convention is also understood.


       As a convenience, if a pathname begins with a period and a  environment
       variable  exists  with  that  name,  then  the value of the environment
       variable will be used as  the  actual  pathname.   For  example,  if  a
       environment  variable  of  .FOO has the value, then
       the command
              fmerge -o .FOO
       is actually equivilant to the command
              fmerge -o
       If you want to prevent the expansion of a pathname which begins with  a
       period, then you can use an alternate form for the pathname, as in:
              fmerge -o ./.FOO


       In  general,  fmerge  can  handle  all text files you throw at it, even
       international text with unusual encodings.  However, fmerge  is  unable
       to cope elegantly with files which contain the NUL character.

       The  fcomp(1)  program simply prints a warning, and continues, you need
       to know that it converts NUL  characters  into  an  0x80  value  before
       performing the comparison.

       The  fmerge(1) program also converts the NUL character to an 0x80 value
       before merging, after a warning, and any output file will contain  this
       value, rather than the original NUL character.

       The  fhist(1)  program,  however,  generates a fatal error if any input
       file contains NUL characters.  This is intended to protect your  source
       files  for  unintentional  corruption.   Use  -BINary  for  files which
       absolutely must contain NUL characters.


       The fmerge program will exit with a status of  1  on  any  error.   The
       fmerge  program  will  only  exit  with  a  status of 0 if there are no


       This program is based on the algorithm in
              An O(ND) Difference Algorithm  and  Its  Variations,  Eugene  W.
              Myers,  TR  85‐6, 10‐April‐1985, Department of Computer Science,
              The University of Arizona, Tuscon, Arizona 85721.
       See also:
              A File Comparison Program, Webb  Miller  and  Eugene  W.  Myers,
              Software  Practice  and  Experience, Volume 15, No. 11, November


       fmerge version 1.18.D001
       Copyright (C) 1991, 1992, 1993, 1994, 1995,  1996,  1997,  1998,  1999,
       2000, 2001, 2002, 2003, 2004, 2005, 2006, 2008, 2009 Peter Miller;

       This program is derived from a work
       Copyright (C) 1990 David I. Bell.

       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  the
       Free  Software Foundation; either version 3 of the License, or (at your
       option) any later version.

       This program is distributed in the hope that it  will  be  useful,  but
       WITHOUT   ANY   WARRANTY;   without   even   the  implied  warranty  of
       General Public License for more details.

       You should have received a copy of the GNU General Public License along
       with this program. If not, see <>.


       Peter Miller       Web:
       /\/\*           E‐Mail:

       David I. Bell      Web: