From c99f320d5c0fd98fe88d9cea5407eb7ad9d50e8a Mon Sep 17 00:00:00 2001

From: Paul Eggert <eggert@cs.ucla.edu>

Date: Mon, 4 Apr 2022 23:52:49 -0700

Subject: zgrep: port to POSIX sed

MIME-Version: 1.0

Content-Type: text/plain; charset=UTF-8

Content-Transfer-Encoding: 8bit

* zgrep.in (res): When escaping the file name do not rely on GNU

sed’s extension to POSIX with respect to s/.../\n/.  Instead, use

features that should also work with AIX and/or Solaris sed.  This is

simpler anyway, and would have prevented the recently-fixed bug.

---

 zgrep.in | 15 ++++-----------

 1 file changed, 4 insertions(+), 11 deletions(-)

diff --git a/zgrep.in b/zgrep.in

index bdf7da2..6a16dd1 100644

--- a/zgrep.in

+++ b/zgrep.in

@@ -220,18 +220,11 @@ do

       case $i in

       (*'

 '* | *'&'* | *'\'* | *'|'*)

-        i=$(printf '%s\n' "$i" |

-            sed '

-              :start

-              $!{

-                N

-                b start

-              }

-              s/[&\|]/\\&/g

-              s/\n/\\n/g

-            ');;

+          icolon=$(printf '%s\n' "$i:" |

+                     sed -e 's/[&\|]/\\&/g' -e '$!s/$/\\/');;

+      (*) icolon="$i:";;

       esac

-      sed_script="s|^|$i:|"

+      sed_script="s|^|$icolon|"

       # Fail if grep or sed fails.

       r=$(

-- 

cgit v1.1