Shell 组合具有相同起始模式的多条线

我有一个包含以下内容的文本文件

-country
-the
-elections
+countries
+be
+a
+chance
-the
-means
-we
+need
+people’s
+choice
-democracy
+democracies
-elections
-there
+increases
+their

我想合并线，这是有相同的开始模式。对于上述文件，输出应为

-country -the -elections 
+countries +be +a +chance
-the -means -we
+need +people’s +choice
-democracy
+democracies
-elections -there
+increases +their

我试过了

sed '/^-/{N;s/\n/ /}' diff_1.txt

---

但它的合并行以-开头，这也是错误的，并不像预期的那样。

您可以使用此

awk

：

awk '{ch=substr($0,1,1)} p != ch{if (NR>1) print s; s=""}
{p=ch; s = (s != "" ? s " " $0 : $0)} END{print s}' file

这是另一个awk：

awk '{c=substr($0,1,1); printf (c==t ? OFS : ORS) $0; t=c}'

然而，这将在一切之前引入一条空行

您可以通过以下方式消除此问题：

awk '{c=substr($0,1,1); printf (c==t ? OFS : (NR==1?"":ORS)) $0; t=c}'

使用Perl

> cat amol.txt
-country
-the
-elections
+countries
+be
+a
+chance
-the
-means
-we
+need
+people’s
+choice
-democracy
+democracies
-elections
-there
+increases
+their
> perl -lne ' $c=substr($_,0,1) ;$tp=$tc;$tc.="$_"." "; if($.>1 and $p ne $c) { print "$tp";$tc=$_." ";} $p=$c; END { print "$tc" } ' amol.txt
-country -the -elections
+countries +be +a +chance
-the -means -we
+need +people’s +choice
-democracy
+democracies
-elections -there
+increases +their
>

甚至更短

> perl -lne ' $c=substr($_,0,1) ;$tp=$tc;$tc.="$_"." "; print "$tp" and $tc=$_." " if $.>1 and $p ne $c ; $p=$c; END { print "$tc" } ' amol.txt

---

但它缺少文件的最后一部分，即+增加+他们的

> perl -lne ' $c=substr($_,0,1) ;$tp=$tc;$tc.="$_"." "; print "$tp" and $tc=$_." " if $.>1 and $p ne $c ; $p=$c; END { print "$tc" } ' amol.txt