bash 如何从bash中的pom文件中删除版本号
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/29420216/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
How can I cut the version number out of a pom file in bash
提问by Alex Thomson
I have a pom file that has the following information:
我有一个 pom 文件，其中包含以下信息：
<modelVersion>4.0.0</modelVersion>
<groupId>com.site.camera</groupId>
<artifactId>proj3</artifactId>
<packaging>pom</packaging>
<version>2.6</version>
<name>Proj 3</name>
I am writing a bash script and want to be able to open this file (pom.xml) and cut out the version (eg. 2.6 in this case). I am doing this because as I update the version I want the rest of my script to update with it.
我正在编写一个 bash 脚本并希望能够打开这个文件 (pom.xml) 并删除版本（例如，在这种情况下为 2.6）。我这样做是因为当我更新版本时，我希望我的脚本的其余部分也随之更新。
Things that I have tried:
我尝试过的事情：
var=$(grep <version>2.6</version> pom.xml)
echo $var
Note that I have multiple version parameters in the pom file. And I need this specific version which is surrounded with this name, packaging, artifactId etc.
请注意，我在 pom 文件中有多个版本参数。我需要这个特定的版本，它被这个名称、包装、artifactId 等包围。
采纳答案by glenn Hymanman
The simplest way is to use GNU grep's perl-compatible regexes
最简单的方法是使用 GNU grep's perl-compatible regexes
grep -oP '<version>\K[^<]+' pom.xml
With sed, you'd write
sed -n 's/^[[:blank:]]*<version>\([^<]\+\).*//p' pom.xml
grep -oP '<version>\K[^<]+' pom.xml
使用 sed，你会写
sed -n 's/^[[:blank:]]*<version>\([^<]\+\).*//p' pom.xml
Thanks to David W.'s thoughtful commentary: the solution must use an XML parser. xmlstarletis one such tool:
感谢 David W. 深思熟虑的评论：解决方案必须使用 XML 解析器。xmlstarlet是这样一种工具：
ver=$( xmlstarlet sel -t -v /project/version pom.xml )
And just to be different, ruby comes with an XML parser in its standard library:
只是为了不同，ruby 在其标准库中附带了一个 XML 解析器：
ruby -r rexml/document -e 'puts REXML::Document.new(File.new(ARGV.shift)).elements["/project/version"].text' pom.xml
回答by that other guy
Some people object to parsing XML with regex. Here's how you can do it correctly and robustly with xmlstarlet:
有些人反对用 regex 解析 XML。以下是如何正确且稳健地执行此操作xmlstarlet：
$ cat pom.xml
<project>
<modelVersion>4.0.0</modelVersion>
<groupId>com.site.camera</groupId>
<artifactId>proj3</artifactId>
<packaging>pom</packaging>
<version>2.6</version>
<name>Proj 3</name>
</project>
$ xmlstarlet sel -t -v '/project/version' pom.xml
2.6
This works equally well after XML tools and editors have had their way with your document:
在 XML 工具和编辑器处理您的文档后，这同样有效：
$ cat pom.xml
<project>
<version><![CDATA[2]]>.6
</version>
</project>
$ xmlstarlet sel -t -v '/project/version' pom.xml
2.6
Edit: As a testament to Doing It Right(tm), this solution still works fine after it was pointed out that there will be a lot of different version tags.
编辑：作为对做对 (tm) 的证明，在指出会有很多不同的版本标签后，该解决方案仍然可以正常工作。
回答by Vytenis Bivainis
Using maven you do it this way:
使用 Maven 你这样做：
echo '${project.version}' | mvn help:evaluate | grep -v '^[[]'
回答by Chris Maes
try like this:
试试这样：
var=$(grep -Po '<version>\K[^<]*' pom.xml)
echo $var
output:
输出：
2.6
- -P : use perl regexp format
- -o : print matching part only
- -P : 使用 perl regexp 格式
- -o : 只打印匹配部分
回答by David C. Rankin
Or without perl compatible regular expressions using parameter expansion/substring extraction:
或者没有使用参数扩展/子字符串提取的perl 兼容正则表达式：
$ var=$(grep '<version>' pom.xml); ver=${var%<*}; ver=${ver#*>}; echo $ver
2.6
I would prefer the perl compatibile RE if you have that available.
如果你有可用的，我更喜欢 perl 兼容的 RE。
Note:As pointed out by David W. this works for the example given, but will return multipleversionstrings if the input file has more than one line containing <version>. The full pom.xmlfile referenced below in the comments contains multiple<version>strings. An additional loop will be required to process all versionlines and without a unique tag, it is not possible to determine which versionis the desired version. For example parsing the current full pom.xmlwith:
注：正如所指出的大卫W.这适用于给出的例子，但将返回多个version字符串如果输入文件具有包含多行<version>。pom.xml下面评论中引用的完整文件包含多个<version>字符串。将需要一个额外的循环来处理所有version行，如果没有唯一标签，则无法确定哪个version是所需的版本。例如解析当前完整pom.xml的：
$ for i in $(grep '<version>' pom.xml); do ver=${i%<*}; ver=${ver#*>}; echo "version: $ver"; done
Returns:
返回：
version: 33
version: 3.3.2
version: 4.11
version: 2.4
version: 3.2
version: 2.9.1
version: 2.5.2
version: 2.5.1
version: 2.4
version: 3.0.1
version: 2.4
version: 2.0
version: 1.7
回答by Tiwari
My Requirement was also similar. We are using SNAPSHOT at the end of our version so I did this to find out the Version number (Extension of David's answer)
我的要求也是类似的。我们在版本的末尾使用了 SNAPSHOT，所以我这样做是为了找出版本号（大卫的答案的扩展）
for i in $(grep '<version>' pom.xml); do
#echo "$i";
ver=${i%<*};
ver=${ver#*>};
if [[ $ver == *"SNAPSHOT"* ]]
then
echo "$ver";
fi
done
And It gives me an output; Which I wanted
它给了我一个输出；我想要的
1.09-SNAPSHOT
1.09-快照
