一天一个shell命令 linux文本操作系列-diff命令详解

hello
  world v1_echo
  v1_diff
  v1_comm
 shell is easy!

hello world!
v2_echo
 v1_diff
 easy shell!

1,2c1,2
 < hello world
 &lt; v1_echo
 ---
 > hello world!
 &gt; v2_echo
 4c4
 < v1_comm
 ---
 > easy shell!

-v1_echo
 +hello world!
 +v2_echo
 v1_diff
 -v1_comm
 +easy shell!
 shell is easy!

 4c
 easy shell!
 .
 1,2c
 hello world!
 v2_echo
 .

Mandatory arguments to long options are mandatory for short options too.
 --normal                                        output a normal diff (the default)
 -q, --brief                                      report only when files differ
 -s, --report-identical-files               report when two files are the same
 -c, -C NUM, --context[=NUM]   output NUM (default 3) lines of copied context
 -u, -U NUM, --unified[=NUM]     output NUM (default 3) lines of unified context
 -e, --ed                                          output an ed script
 -n, --rcs                                         output an RCS format diff
 -y, --side-by-side                           output in two columns

FILES 的格式可以是‘FILE1 FILE2'、‘DIR1 DIR2'、‘DIR FILE...'或
‘FILE... DIR'。

 hello world               | hello world!
 v1_echo                    | v2_echo
 v1_diff                       v1_diff
 v1_comm                   | easy shell!
 shell is easy!               shell is easy!

用法：diff [选项]... FILES
Compare FILES line by line.

Mandatory arguments to long options are mandatory for short options too.
      --normal                  output a normal diff (the default)
  -q, --brief                   report only when files differ
  -s, --report-identical-files  report when two files are the same
  -c, -C NUM, --context[=NUM]   output NUM (default 3) lines of copied context
  -u, -U NUM, --unified[=NUM]   output NUM (default 3) lines of unified context
  -e, --ed                      output an ed script
  -n, --rcs                     output an RCS format diff
  -y, --side-by-side            output in two columns
  -W, --width=NUM               output at most NUM (default 130) print columns
      --left-column             output only the left column of common lines
      --suppress-common-lines   do not output common lines

  -p, --show-c-function         show which C function each change is in
  -F, --show-function-line=RE   show the most recent line matching RE
      --label LABEL             use LABEL instead of file name
                                  (can be repeated)

  -t, --expand-tabs             expand tabs to spaces in output
  -T, --initial-tab             make tabs line up by prepending a tab
      --tabsize=NUM             tab stops every NUM (default 8) print columns
      --suppress-blank-empty    suppress space or tab before empty output lines
  -l, --paginate                pass output through `pr' to paginate it

  -r, --recursive                 recursively compare any subdirectories found
  -N, --new-file                  treat absent files as empty
      --unidirectional-new-file   treat absent first files as empty
      --ignore-file-name-case     ignore case when comparing file names
      --no-ignore-file-name-case  consider case when comparing file names
  -x, --exclude=PAT               exclude files that match PAT
  -X, --exclude-from=FILE         exclude files that match any pattern in FILE
  -S, --starting-file=FILE        start with FILE when comparing directories
      --from-file=FILE1           compare FILE1 to all operands;
                                    FILE1 can be a directory
      --to-file=FILE2             compare all operands to FILE2;
                                    FILE2 can be a directory

  -i, --ignore-case               ignore case differences in file contents
  -E, --ignore-tab-expansion      ignore changes due to tab expansion
  -Z, --ignore-trailing-space     ignore white space at line end
  -b, --ignore-space-change       ignore changes in the amount of white space
  -w, --ignore-all-space          ignore all white space
  -B, --ignore-blank-lines        ignore changes whose lines are all blank
  -I, --ignore-matching-lines=RE  ignore changes whose lines all match RE

  -a, --text                      treat all files as text
      --strip-trailing-cr         strip trailing carriage return on input

  -D, --ifdef=NAME                output merged file with `#ifdef NAME' diffs
      --GTYPE-group-format=GFMT   format GTYPE input groups with GFMT
      --line-format=LFMT          format all input lines with LFMT
      --LTYPE-line-format=LFMT    format LTYPE input lines with LFMT
    These format options provide fine-grained control over the output
      of diff, generalizing -D/--ifdef.
    LTYPE 可以是‘old'、‘new'或‘unchanged'。GTYPE 可以是 LTYPE 的选择
    或是‘changed'。
    GFMT (only) may contain:
      %<  lines from FILE1
      %>  lines from FILE2
      %=  lines common to FILE1 and FILE2
      %[-][WIDTH][.[PREC]]{doxX}LETTER  printf-style spec for LETTER
        LETTERs are as follows for new group, lower case for old group:
          F  first line number
          L  last line number
          N  number of lines = L-F+1
          E  F-1
          M  L+1
      %(A=B?T:E)  if A equals B then T else E
    LFMT (only) may contain:
      %L  contents of line
      %l  contents of line, excluding any trailing newline
      %[-][WIDTH][.[PREC]]{doxX}n  printf-style spec for input line number
    Both GFMT and LFMT may contain:
      %%  %
      %c'C'  the single character C
      %c'\OOO'  the character with octal code OOO
      C    the character C (other characters represent themselves)

  -d, --minimal            try hard to find a smaller set of changes
      --horizon-lines=NUM  keep NUM lines of the common prefix and suffix
      --speed-large-files  assume large files and many scattered small changes

      --help               display this help and exit
  -v, --version            output version information and exit

  FILES 的格式可以是‘FILE1 FILE2'、‘DIR1 DIR2'、‘DIR FILE...'或
      ‘FILE... DIR'。
If --from-file or --to-file is given, there are no restrictions on FILE(s).
如果 FILE 是‘-'，则由标准输入读取内容。
如果输入相同，则退出状态为 0；1 表示输入不同；2 表示有错误产生。