How to only capute sub-matched character by grep

File content:

<a href="ceph-0.80.9-82.1.x86_64.rpm"><img src="/icons/rpm.gif" alt="[   ]" width="16" height="16" /></a> <a href="ceph-0.80.9-82.1.x86_64.rpm">ceph-0.80.9-82.1.x86_64.rpm</a>                                     11-Jun-2015 16:37  9.9M   <a href="ceph-0.80.9-82.1.x86_64.rpm.mirrorlist">Details</a>
<a href="ceph-common-0.80.9-82.1.x86_64.rpm"><img src="/icons/rpm.gif" alt="[   ]" width="16" height="16" /></a> <a href="ceph-common-0.80.9-82.1.x86_64.rpm">ceph-common-0.80.9-82.1.x86_64.rpm</a>                              11-Jun-2015 16:37  3.9M   <a href="ceph-common-0.80.9-82.1.x86_64.rpm.mirrorlist">Details</a>
<a href="ceph-common-debuginfo-0.80.9-82.1.x86_64.rpm"><img src="/icons/rpm.gif" alt="[   ]" width="16" height="16" /></a> <a href="ceph-common-debuginfo-0.80.9-82.1.x86_64.rpm">ceph-common-debuginfo-0.80.9-82.1.x86_64.rpm</a>                    11-Jun-2015 16:37   49M   <a href="ceph-common-debuginfo-0.80.9-82.1.x86_64.rpm.mirrorlist">Details</a>

......

 

Command:

grep -P -o ">\K[^ ]+?x86_64.rpm" <file name>

-P : perl format regular expression

-o : only matched content

\K : throw away content of before \K

 

Result:

xen-doc-html-4.4.2_06-3.1.x86_64.rpm
xen-kmp-default-4.4.2_06_k3.12.28_4-3.1.x86_64.rpm
xen-kmp-default-debuginfo-4.4.2_06_k3.12.28_4-3.1.x86_64.rpm
xen-libs-4.4.2_06-3.1.x86_64.rpm
xen-libs-debuginfo-4.4.2_06-3.1.x86_64.rpm
xen-tools-4.4.2_06-3.1.x86_64.rpm
xen-tools-debuginfo-4.4.2_06-3.1.x86_64.rpm
xen-tools-domU-4.4.2_06-3.1.x86_64.rpm