查找以'＃'开头的unix文件中的所有单词

Question

输入文件看起来像这样： -

mary #had a little #lamb its #fleece #was white as snow

输出应该看起来像

#had #lamb #fleece #was

Answer 1

使用sed

sed -r 's/(^| )[^#][^ ]+//g'

<强>测试

echo "mary #had a little #lamb its #fleece #was white as snow" | sed -r 's/(^| )[^#][^ ]+//g'

#had #lamb #fleece #was

使用grep

grep -o "#[^ ]*"

<强>测试

$ echo "mary #had a little #lamb its #fleece #was white as snow" | grep -o "#[^ ]*"
#had
#lamb
#fleece
#was

Answer 2

如果你不介意一个领先的空间，并且不想匹配中间可能包含#的单词，例如lit#tle，你也可以试试这个：

grep -o -P "(\s|^)#\w+"

示例：

echo "#mary #had a lit#tle #lamb its #fleece #was white as snow" | grep -o -P "(\s|^)#\w+"
#mary
 #had
 #lamb
 #fleece
 #was