使用svn命令统计两次上线之间的代码行数，以及不同作者的代码行数

原理

其原理如下：

1，取得svn的两个版本之间的diff数据

target=https://svn.xxx.com/erp/trunk/cms/cmsnew

start_version=113121

end_version=113128



步骤一，使用svn diff命令取得SVN两个版本diff文件

svn diff -r$start_version:$end_version$target --username lizejun --password xxx >svndiff.log



步骤二，然后将文件取得 开头为+的行，处理后去掉空行，计算行数

grep "^+" svndiff.log | grep -v"^+++" | sed 's/^.//'| sed s/[[:space:]]//g |sed '/^$/d'|wc -l



2，取得某个RD在两次上线版本之间增加的代码行数



步骤一，使用svn log命令取得$rdname提交代码的svn版本号

svn log -q $target -r$start_version:HEAD --username lizejun --password xxx |awk '{print $1,$3}'|grep$rdname |awk '{print \$1}'



步骤二，将取得的版本号与上一个版本做操作1中的工作，统计代码的增加情况

for rev in ${revs};do

rev=${rev:1}

last_rev=$((rev-1))



${SVN_CMD} diff -r${last_rev}:${rev} $target --username $USERNAME --password$PASSWD > $TEMPFILE

count=`grep "^+" $TEMPFILE | grep -v "^+++" |sed 's/^.//'| sed s/[[:space:]]//g |sed '/^$/d'|wc -l`

TOTAL=$((TOTAL+count))

rm -rf $TEMPFILE

done

代码脚本

//向我们公司伟大的作者致敬，为防止挖墙脚，去掉作者

#! /bin/bash
# 计算有效变更代码量的脚本
# 包括注释，但不包括新增的空行
version() {
    OS=`uname -o`
    echo "svn_counter ($OS) 0.0.1"
}

usage() {
    echo "usage: svn_counter [-t SVN_REPOSITORY_URL] [-s START_REVISION]"
    echo "                      [-e END_REVISION] [-u USER_NAME]"
    echo "                      [-p PASSWD]"
    echo "       svn_counter [-v|-h]"
    echo 
    echo "        -t,                 目标SVN库地址"
    echo "        -s,                 起始修订号"
    echo "        -e,                 结束修订号"
    echo "        -a,                 提交作者"
    echo "        -u,                 svn帐号"
    echo "        -p,                 svn密码"
    echo "        -h,                 帮助"
    echo "        -v,                 版本信息"
}

if [ $# -lt 1 ]; then
    usage
    exit 1 
fi

while getopts "t:s:e:a:u:p:vh" opt; do
    case $opt in
        t) target=$OPTARG;;
        s) start_revision=$OPTARG;;
        e) end_revision=$OPTARG;;
        a) author=$OPTARG;;
        u) user=$OPTARG;;
        p) passwd=$OPTARG;;
        v) version; exit 1;;
        h) usage; exit 1;;
    esac
done

if [ -z $target ]; then
    echo "请输入目标SVN库地址!"
    exit 1
fi

if [ -z $start_revision ]; then
    echo "请输入起始修订号!"
    exit 1
fi

#SVN_CMD='/home/work/local/svn/bin/svn'
SVN_CMD='svn'

TEMPFILE=temp_diff.log
USERNAME=${user}
PASSWD=${passwd}
TOTAL=0

if [ -z $author ]; then
    if [ -z $end_revision ]; then
        ${SVN_CMD} diff -r$start_revision $target --username $USERNAME --password $PASSWD > $TEMPFILE
    else
        ${SVN_CMD} diff -r$start_revision:$end_revision $target --username $USERNAME --password $PASSWD > $TEMPFILE
    fi

    #去掉含空格的空行
    TOTAL=`grep "^+" $TEMPFILE | grep -v "^+++" | sed 's/^.//'| sed s/[[:space:]]//g |sed '/^$/d'|wc -l`
    #没有去掉有的空行
    #TOTAL=`grep "^+" $TEMPFILE|grep -v "^+++"|sed 's/^.//'|sed '/^$/d'|wc -l`
    
    rm -fr $TEMPFILE
    echo "$TOTAL"
else
    if [ -z $end_revision ]; then   
        revs=`${SVN_CMD} log -q $target  -r $start_revision:HEAD --username $USERNAME --password $PASSWD |awk '{print \$1,\$3}'| grep  ${author}| awk '{print \$1}'`
    else 
        revs=`${SVN_CMD} log -q $target  -r $start_revision:$end_revision --username $USERNAME --password $PASSWD |awk '{print \$1,\$3}'| grep  ${author}| awk '{print \$1}'`
    fi 
    for rev in ${revs};do
        rev=${rev:1}
        last_rev=$((rev-1))

        ${SVN_CMD} diff -r${last_rev}:${rev} $target --username $USERNAME --password $PASSWD  > $TEMPFILE
        count=`grep "^+" $TEMPFILE | grep -v "^+++" | sed 's/^.//'| sed s/[[:space:]]//g |sed '/^$/d'|wc -l`
        TOTAL=$((TOTAL+count))
        rm -rf $TEMPFILE
    done    
    echo "$TOTAL" 
fi