对于我的一个案例，我发现NodeJS url库有最简单的解决方案。当然是YMMV

$ urlencode(){ node -e "console.log(require('url').parse(process.argv.slice(1).join('+')).href)" "$@"; }

$ urlencode "https://example.com?my_database_has=these 'nasty' query strings in it"
https://example.com/?my_database_has=these%20%27nasty%27%20query%20strings%20in%20it

下面是纯BASH的答案。

更新:由于已经讨论了许多变化，我已经把这个放在https://github.com/sfinktah/bash/blob/master/rawurlencode.inc.sh上，任何人都可以发布一个PR。

注意:这个解决方案并不打算编码unicode或多字节字符——这完全超出了BASH的简单本机功能。它仅用于编码那些在POST或GET请求中会破坏参数传递的符号，例如。'&'， '='等等。

非常重要的提示:不要尝试用任何语言编写你自己的UNICODE转换函数。见答案结尾。

rawurlencode() {
  local string="${1}"
  local strlen=${#string}
  local encoded=""
  local pos c o

  for (( pos=0 ; pos<strlen ; pos++ )); do
     c=${string:$pos:1}
     case "$c" in
        [-_.~a-zA-Z0-9] ) o="${c}" ;;
        * )               printf -v o '%%%02x' "'$c"
     esac
     encoded+="${o}"
  done
  echo "${encoded}"    # You can either set a return variable (FASTER) 
  REPLY="${encoded}"   #+or echo the result (EASIER)... or both... :p
}

你可以在两种情况下使用它:

easier:  echo http://url/q?=$( rawurlencode "$args" )
faster:  rawurlencode "$args"; echo http://url/q?${REPLY}

(编辑)

下面是匹配的rawurldecode()函数，它非常棒。

# Returns a string in which the sequences with percent (%) signs followed by
# two hex digits have been replaced with literal characters.
rawurldecode() {

  # This is perhaps a risky gambit, but since all escape characters must be
  # encoded, we can replace %NN with \xNN and pass the lot to printf -b, which
  # will decode hex for us

  printf -v REPLY '%b' "${1//%/\\x}" # You can either set a return variable (FASTER)

  echo "${REPLY}"  #+or echo the result (EASIER)... or both... :p
}

有了匹配集，我们现在可以执行一些简单的测试:

$ diff rawurlencode.inc.sh \
        <( rawurldecode "$( rawurlencode "$( cat rawurlencode.inc.sh )" )" ) \
        && echo Matched

Output: Matched

如果你真的觉得你需要一个外部工具(好吧，它会快得多，可能会处理二进制文件之类的…)我在我的OpenWRT路由器上发现了这个…

replace_value=$(echo $replace_value | sed -f /usr/lib/ddns/url_escape.sed)

url_escape的地方。Sed是一个包含以下规则的文件:

# sed url escaping
s:%:%25:g
s: :%20:g
s:<:%3C:g
s:>:%3E:g
s:#:%23:g
s:{:%7B:g
s:}:%7D:g
s:|:%7C:g
s:\\:%5C:g
s:\^:%5E:g
s:~:%7E:g
s:\[:%5B:g
s:\]:%5D:g
s:`:%60:g
s:;:%3B:g
s:/:%2F:g
s:?:%3F:g
s^:^%3A^g
s:@:%40:g
s:=:%3D:g
s:&:%26:g
s:\$:%24:g
s:\!:%21:g
s:\*:%2A:g

虽然在BASH中编写这样一个能够处理UTF-8输入的脚本(可能使用xxd和一个非常长的规则集)并不是不可能的，但是有更快更可靠的方法。试图将UTF-8解码为UTF-32是一项非常重要的任务，要做到准确，尽管很容易做得不准确，以至于您认为它有效，直到有一天它不起作用。

甚至Unicode联盟在发现样本代码不再100%与实际标准兼容后，也删除了样本代码。

Unicode标准在不断发展，变得非常微妙。您可以拼凑的任何实现都不会具有适当的遵从性，如果您通过一些极端的努力来管理它，它也不会保持遵从性。

Ruby，为了完整性

value="$(ruby -r cgi -e 'puts CGI.escape(ARGV[0])' "$2")"

Python 3基于@sandro在2010年的好答案:

echo "Test & /me" | python -c "import urllib.parse;print (urllib.parse.quote(input()))"

测试% 20% 26% 20 /我

在bash脚本的第二行中使用Perl的URI::Escape模块和uri_escape函数:

...

value="$(perl -MURI::Escape -e 'print uri_escape($ARGV[0]);' "$2")"
...

编辑:修复引用问题，正如Chris Johnsen在评论中建议的那样。谢谢!

awk版本的直接链接:http://www.shelldorado.com/scripts/cmds/urlencode 我用了很多年了，效果很好

:
##########################################################################
# Title      :  urlencode - encode URL data
# Author     :  Heiner Steven (heiner.steven@odn.de)
# Date       :  2000-03-15
# Requires   :  awk
# Categories :  File Conversion, WWW, CGI
# SCCS-Id.   :  @(#) urlencode  1.4 06/10/29
##########################################################################
# Description
#   Encode data according to
#       RFC 1738: "Uniform Resource Locators (URL)" and
#       RFC 1866: "Hypertext Markup Language - 2.0" (HTML)
#
#   This encoding is used i.e. for the MIME type
#   "application/x-www-form-urlencoded"
#
# Notes
#    o  The default behaviour is not to encode the line endings. This
#   may not be what was intended, because the result will be
#   multiple lines of output (which cannot be used in an URL or a
#   HTTP "POST" request). If the desired output should be one
#   line, use the "-l" option.
#
#    o  The "-l" option assumes, that the end-of-line is denoted by
#   the character LF (ASCII 10). This is not true for Windows or
#   Mac systems, where the end of a line is denoted by the two
#   characters CR LF (ASCII 13 10).
#   We use this for symmetry; data processed in the following way:
#       cat | urlencode -l | urldecode -l
#   should (and will) result in the original data
#
#    o  Large lines (or binary files) will break many AWK
#       implementations. If you get the message
#       awk: record `...' too long
#        record number xxx
#   consider using GNU AWK (gawk).
#
#    o  urlencode will always terminate it's output with an EOL
#       character
#
# Thanks to Stefan Brozinski for pointing out a bug related to non-standard
# locales.
#
# See also
#   urldecode
##########################################################################

PN=`basename "$0"`          # Program name
VER='1.4'

: ${AWK=awk}

Usage () {
    echo >&2 "$PN - encode URL data, $VER
usage: $PN [-l] [file ...]
    -l:  encode line endings (result will be one line of output)

The default is to encode each input line on its own."
    exit 1
}

Msg () {
    for MsgLine
    do echo "$PN: $MsgLine" >&2
    done
}

Fatal () { Msg "$@"; exit 1; }

set -- `getopt hl "$@" 2>/dev/null` || Usage
[ $# -lt 1 ] && Usage           # "getopt" detected an error

EncodeEOL=no
while [ $# -gt 0 ]
do
    case "$1" in
        -l) EncodeEOL=yes;;
    --) shift; break;;
    -h) Usage;;
    -*) Usage;;
    *)  break;;         # First file name
    esac
    shift
done

LANG=C  export LANG
$AWK '
    BEGIN {
    # We assume an awk implementation that is just plain dumb.
    # We will convert an character to its ASCII value with the
    # table ord[], and produce two-digit hexadecimal output
    # without the printf("%02X") feature.

    EOL = "%0A"     # "end of line" string (encoded)
    split ("1 2 3 4 5 6 7 8 9 A B C D E F", hextab, " ")
    hextab [0] = 0
    for ( i=1; i<=255; ++i ) ord [ sprintf ("%c", i) "" ] = i + 0
    if ("'"$EncodeEOL"'" == "yes") EncodeEOL = 1; else EncodeEOL = 0
    }
    {
    encoded = ""
    for ( i=1; i<=length ($0); ++i ) {
        c = substr ($0, i, 1)
        if ( c ~ /[a-zA-Z0-9.-]/ ) {
        encoded = encoded c     # safe character
        } else if ( c == " " ) {
        encoded = encoded "+"   # special handling
        } else {
        # unsafe character, encode it as a two-digit hex-number
        lo = ord [c] % 16
        hi = int (ord [c] / 16);
        encoded = encoded "%" hextab [hi] hextab [lo]
        }
    }
    if ( EncodeEOL ) {
        printf ("%s", encoded EOL)
    } else {
        print encoded
    }
    }
    END {
        #if ( EncodeEOL ) print ""
    }
' "$@"