错误记录了 java.net.URISyntaxException：使用 Apache POI 读取 Excel 文件时，索引处的方案名称中存在非法字符

Question

从英国政府读取外部 Excel (.xlsx) 文件时（请参阅 https://assets.publishing.service.gov.uk/media/5d8b3abded915d0373d3540f/File_1_-_IMD2019_Index_of_Multiple_Deprivation.xlsx），不会引发异常，但记录了一个错误：

09:13:35.150 [nREPL-session-b3e5d112-80a2-45f3-9e26-095934eeb06a] ERROR org.apache.poi.openxml4j.opc.PackageRelationshipCollection - Cannot convert All%20of%20the%20data%20files%20and%20supporting%20documents%20for%20the%20English%20Indices%20of%20Deprivation%202019%20are%20available%20from:%20www.gov.uk/government/statistics/english-indices-of-deprivation-2019 in a valid relationship URI-> dummy-URI used
java.net.URISyntaxException: Illegal character in scheme name at index 3: All%20of%20the%20data%20files%20and%20supporting%20documents%20for%20the%20English%20Indices%20of%20Deprivation%202019%20are%20available%20from:%20www.gov.uk/government/statistics/english-indices-of-deprivation-2019
    at java.base/java.net.URI$Parser.fail(URI.java:2976) ~[?:?]
    at java.base/java.net.URI$Parser.checkChars(URI.java:3147) ~[?:?]
    at java.base/java.net.URI$Parser.parse(URI.java:3173) ~[?:?]
    at java.base/java.net.URI.<init>(URI.java:623) ~[?:?]
    at org.apache.poi.openxml4j.opc.PackagingURIHelper.toURI(PackagingURIHelper.java:723) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.PackageRelationshipCollection.parseRelationshipsPart(PackageRelationshipCollection.java:358) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.PackageRelationshipCollection.<init>(PackageRelationshipCollection.java:160) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.PackageRelationshipCollection.<init>(PackageRelationshipCollection.java:130) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.PackagePart.loadRelationships(PackagePart.java:565) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.OPCPackage.getParts(OPCPackage.java:751) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.openxml4j.opc.OPCPackage.open(OPCPackage.java:322) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.xssf.usermodel.XSSFWorkbookFactory.create(XSSFWorkbookFactory.java:97) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.xssf.usermodel.XSSFWorkbookFactory.create(XSSFWorkbookFactory.java:36) ~[poi-ooxml-5.2.3.jar:5.2.3]
    at org.apache.poi.ss.usermodel.WorkbookFactory.lambda$create$2(WorkbookFactory.java:224) ~[poi-5.2.3.jar:5.2.3]
    at org.apache.poi.ss.usermodel.WorkbookFactory.wp(WorkbookFactory.java:329) [poi-5.2.3.jar:5.2.3]
    at org.apache.poi.ss.usermodel.WorkbookFactory.create(WorkbookFactory.java:224) [poi-5.2.3.jar:5.2.3]
    at org.apache.poi.ss.usermodel.WorkbookFactory.create(WorkbookFactory.java:185) [poi-5.2.3.jar:5.2.3]

我推测源文件中有错误，或者部分内容被错误分类为 URI，然后 java.net.URI 无法解析它。尽管记录了错误，但我可以毫无困难地继续使用 POI 读取文件内容。我可以通过更改日志记录级别来抑制错误，但是还有其他方法可以防止记录此错误吗？

这里有一个 log4j2 配置，可以抑制此错误的错误日志记录，但这对于在其他良好的 xlsx 文件中似乎不是异常情况的情况来说似乎是一个相当戏剧性的步骤。

<?xml version="1.0" encoding="UTF-8"?>
<Configuration status="WARN">
    <Appenders>
        <Console name="Console" target="SYSTEM_OUT">
            <PatternLayout pattern="%d{HH:mm:ss.SSS} [%t] %-5level %logger{36} - %msg%n"/>
        </Console>
    </Appenders>
    <Loggers>
        <Root level="error">
            <AppenderRef ref="Console"/>
        </Root>
        <!-- Turn off ERROR logging from the internals of POI when a URI cannot be parsed -->
        <Logger name="org.apache.poi.openxml4j.opc.PackageRelationshipCollection" level="off">
        </Logger>
    </Loggers>
</Configuration>

Answer 1

当我下载文件并在 Excel 中打开它时，第一个工作表“注释”上的整个文本框都有一个超链接集。但该链接已损坏，甚至 Excel 也无法跟踪它。因此 Apache POI 也无法跟随它。这就是错误的原因。

可以通过将鼠标悬停在文本框的空白部分来检查。出现手形光标，单击时 Excel 告知它无法跟随该链接。

由于全部都是开源的，因此可以查看源代码以遵循程序流程。

PackagingURIHelper.toURI：

...
    return new URI(value);
...

这里 java.net.URI 如果给定字符串

java.net.URISyntaxException

违反了 RFC 2396，则抛出

value

，如

java.net.URI

构造函数中描述的偏差所增强。

PackagingURIHelper.toURI

由try-catch块中的PackageRelationshipCollection.parseRelationshipsPart调用

...
        try {
            // when parsing of the given uri fails, we can either
            // ignore this relationship, which leads to IllegalStateException
            // later on, or use a dummy value and thus enable processing of the
            // package
            target = PackagingURIHelper.toURI(value);
        } catch (URISyntaxException e) {
            LOG.atError().withThrowable(e).log("Cannot convert {} in a valid relationship URI-> dummy-URI used", value);
        }
        addRelationship(target, targetMode, type, id);
...

这里的评论说明了为什么

URISyntaxException

被记录。否则，它会在读取

*.xlsx

时破坏整个解析关系过程。这比错误日志条目要糟糕得多。

所以这种行为是设计使然的。人们可能会认为，日志条目不应该处于错误级别，因为它并不重要。但这只能由 Apache POI 开发人员改变。因此，请向 Apache POI 提交错误报告或更改请求。