转载

Knowage汉化Document不能保存中文字符问题修复

Knoage-6.x社区版（spagoBI升级版）在使用时发现Document不能保存中文字符。

TAG：Knowage汉化,SpagoBI汉化,Knowage中文版,Knowage中文乱码

场景现象

如果想报表名或描述保存为中文时会发生如下错误：

Knowage汉化Document不能保存中文字符问题修复

提示错误：

ERRORS:
The field Description must be an alphanumeric string (letters and numbers with other character such as parenthesis and -_;:!?,.//)

只能输入字母数字及某些特殊字符的组合。

查询源代码：

9016 = The field %0 must be an alphanumeric string (letters and numbers with other character such as parenthesis and -_;:!?,./)

ERROR_EXTENDED_ALFANUMERIC="9016";

一、重置验证

页面代码：

knowage/src/main/webapp/WEB-INF/conf/analiticalmodel/documentBrowser.xml

验证规则：

knowage/src/main/webapp/WEB-INF/conf/analiticalmodel/validation.xml

重置验证：

<FIELD name="name" label="#SBIDev.docConf.docDet.nameField">
	<VALIDATOR validatorName="MANDATORY"/>
	<!-- <VALIDATOR validatorName="EXTENDED_ALFANUMERIC"/> ranying -->
	<VALIDATOR validatorName="MAXLENGTH" arg0="200"/>
</FIELD>
<FIELD name="description" label="#SBIDev.docConf.docDet.descriptionField">
	<!-- <VALIDATOR validatorName="EXTENDED_ALFANUMERIC"/> ranying -->
	<VALIDATOR validatorName="MAXLENGTH" arg0="400"/>
</FIELD>

重置之后可以保存了但内容是乱码：

Knowage汉化Document不能保存中文字符问题修复

二、修复乱码

跟踪代码发现在验证之前就已经是乱码了，

请求链路：

1.it.eng.spago.dispatching.httpchannel.AdapterHTTP

DetailBIObjectModule
- Service
- handleMultipartForm(request, requestContext);
- modBIObject

问题就在handleMultipartForm方法，原因是使用common-fileupload解析字段内容但使用的默认编码格式,改为使用UTF-8格式读取字段内容。

private void handleMultipartForm(HttpServletRequest request, RequestContextIFace requestContext)
    	throws Exception{
    	SourceBean serviceRequest = requestContext.getServiceRequest();

    	// Create a factory for disk-based file items
    	FileItemFactory factory = new DiskFileItemFactory();

    	// This is done to make upload work in Unix solaris
    	//((DiskFileItemFactory)factory).setSizeThreshold(5242880);

    	// Create a new file upload handler
    	ServletFileUpload upload = new ServletFileUpload(factory);

    	//upload.setFileSizeMax(5242880);
    	//upload.setSizeMax(5242880);


        // Parse the request
        List fileItems = upload.parseRequest(request);
        Iterator iter = fileItems.iterator();
        while (iter.hasNext()) {
            FileItem item = (FileItem) iter.next();

            if (item.isFormField()) {
                String name = item.getFieldName();
                // Support all characters . by ranying
                String value = item.getString("utf-8");
                serviceRequest.setAttribute(name, value);
            } else {
                processFileField(item, requestContext);
            }
        }
    }

修改之后可以正常保存中文名及描述了，Label建议保持只支持英文数字格式。