Question

目标：从csv文件

创建xml文件

我正在尝试从csv文件创建一个xml文件，xml包含页眉，页脚和正文。

我创建了3个不同的功能。

输入csv文件

name,value,amt,position,qnty
abc,2,3,4,5
cde,7,8,9,10
abc,12,13,14,15
cde,7,8,9,10

从csv文件

创建xml的代码

  #!/usr/bin/sh

  FILENAME=$1
   #create body of xml file
   createBody()
   {
    echo ${FILENAME}

    while read var
    do
       filename=`echo $var | sed 's/ /_/g'`
       echo "<abc:archivingGroup xc:value=\"$var\">" >> ${filename}_Fixings.xml
      #create files for each unique group from groups.txt file
       grep $var ${FILENAME} > $filename.txt
      #traverse each file and create a separate xml file for each group
       while IFS=\; read a b c d e
       do
           TEST=$(cat <<EOF
                  <abc:index xc:value="$b">
                   </abc:index>
                    )
            echo $TEST >>  ${filename}_Fixings.xml
       done < $filename.txt
       rm $filename.txt
       echo "</abc:archivingGroup>" >> ${filename}_Fixings.xml
     done < Groups.txt

    }

  #header is common is all files
   createHeader()
       {
        filename=$1
        Header=$(cat << EOF
            <abc:Main>
            <abc:sub>
            <abc:head>
       )

     echo $Header > ${filename}_Fixings.xml

   }

  #footer is common in all files
  createFooter()
   {
      filename=$1;
      Footer=$(cat <<EOF
              </abc:Main>
              </abc:sub>
              </abc:head>
        )
     echo $Footer >> ${filename}_Fixings.xml
   }


  createarchFile()
  {
   IFS=\|
   #creates the unique groups (abc,cde as per above file)
   awk -F";" '{print $1}' ${FILENAME} | sed 1d | sort | uniq > Groups.txt
    #add header to each file, each uniq group is treated as a separate file
    while read var
    do
      filename=`echo $var | sed 's/ /_/g'`
      createHeader $filename
    done < Groups.txt

    #create body for all the files , header for all the files is appended above
    createBody

    #add footer to each file,each uniq group is treated as a separate file
    while read var
    do
      filename=`echo $var | sed 's/ /_/g'`
      createFooter $filename
    done < Groups.txt
  }
  createarchFile

上面的代码通过将csv作为输入

来创建一个xml文件

它从文件中获取唯一名称并存储在Groups.txt文件中
为Groups.txt
遍历每个组文件并创建标题
遍历每个组文件并创建正文
遍历每个组文件并创建页脚

这是正确的方法还是有其他方法来创建xml？

Answer 1

使用printf填写值可能是一个更清晰的选择。

在外部循环上构建第一个标记为

first_tag='<abc:archivingGroup xc:value="%s"> %s </abc:archivingGroup>' var='some_filename' first_tag=$(printf "$first_tag" "$var" "%s")

first_tag将包含

<abc:archivingGroup xc:value="some_filename"> %s </abc:archivingGroup>

在内部循环上使用该格式化字符串

#format element including placeholder for next element (if any)
ele='<abc:index xc:value="%s" />
    %s'
ele=$(printf "$ele" "$b" "%s")
# add to first_tag 
first-tag=$(printf "$first_tag" "$ele" "%s")

结果

<abc:archivingGroup xc:value="some_file_name"> <abc:index xc:value="abc" /> %s </abc:archivingGroup>

只要有占位符，就可以添加更多元素。

使用shell脚本从csv文件创建xml

1 个答案: