Unix 中的脚本以从文件中删除 XML 标记和内容

Question

基本上需要从以 number.xml 命名的一组单独的 XML 文件中删除当事方实体（以及介于两者之间的所有内容）。我尝试了以下方法，但它并不能完全满足我的需求：

    cat test.xml | sed "s;<parties>;\do_opentag ;" | sed "s;</parties>;\do_closetag ;" |  awk 'BEGIN { doPrint = 1; } /do_opentag/ { doPrint = 0; print $0; } /do_closetag/ { doPrint = 1; } { if (doPrint) print $0; }' | grep -v 'do_opentag\|do_closetag'

<?xml version="1.0" encoding="UTF-8"?>
<patent-document xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" pid="58326519" doc-generation-date="2016-10-11">
  <bibliographic-data>
    <application-reference>
      <pan>46422</pan>
    </application-reference>
    <publication-reference>
      <publication-office>KR</publication-office>
      <patent-publication-date>
        <year>2016</year>
        <month>10</month>
        <day>11</day>
      </patent-publication-date>
    </publication-reference>
    <parties>
      <applicants>
        <applicant sequence="1">
          <name lang="EN"></name>
          <address>
            <location-of-work>KR</location-of-work>M
          </address>
        </applicant>
      </applicants>
    </parties>
  </bibliographic-data>
  <vendor>Any</vendor>
  <document-translation-date>2016-11-24</document-translation-date>M
  <invention-title lang="EN">Cell preservation container for liquid-based cell inspection</invention-title>
  <abstract lang="EN">The present invention relates to a liquid for discharging liquid containing cells and cell may be a sampling which is simply eminent generated in </abstract>
  <comment lang="EN"></comment>
</patent-document>

Answer 1

解析 XML 需要一个 XML 解析器。 xmlstarlet 使用起来相当简单。删除parties 节点：

xmlstarlet ed -P -d '//parties' file.xml

生产

<?xml version="1.0" encoding="UTF-8"?>
<patent-document xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" pid="58326519" doc-generation-date="2016-10-11">
  <bibliographic-data>
    <application-reference>
      <pan>46422</pan>
    </application-reference>
    <publication-reference>
      <publication-office>KR</publication-office>
      <patent-publication-date>
        <year>2016</year>
        <month>10</month>
        <day>11</day>
      </patent-publication-date>
    </publication-reference>

  </bibliographic-data>
  <vendor>Any</vendor>
  <document-translation-date>2016-11-24</document-translation-date>M
  <invention-title lang="EN">Cell preservation container for liquid-based cell inspection</invention-title>
  <abstract lang="EN">The present invention relates to a liquid for discharging liquid containing cells and cell may be a sampling which is simply eminent generated in </abstract>
  <comment lang="EN"/>
</patent-document>

Answer 2

sed -e '/<parties>/,/<\/parties>/d' test.xml

在 sed 中，您可以使用由逗号分隔的两个模式来将一个或多个 sed 命令应用于包括匹配模式在内的行范围以及匹配模式之间的行。这里我说的是d-- 删除这一行-- 从/<parties>/ 到/<\/parties>/。

这取决于您的 XML 格式。您需要保留的匹配行中没有其他内容。

如果您想就地编辑文件，请在 sed 中添加 -i 标志。