[转]POI : How to Create and Use User Defined Functions
本文转自:http://poi.apache.org/spreadsheet/user-defined-functions.html
How to Create and Use User Defined Functions
Description
This document describes the User Defined Functions within POI. User defined functions allow you to take code that is written in VBA and re-write in Java and use within POI. Consider the following example.
An Example
Suppose you are given a spreadsheet that can calculate the principal and interest payments for a mortgage. The user enters the principal loan amount, the interest rate and the term of the loan. The Excel spreadsheet does the rest.
When you actually look at the workbook you discover that rather than having the formula in a cell it has been written as VBA function. You review the function and determine that it could be written in Java:
If we write a small program to try to evaluate this cell, we'll fail. Consider this source code:
import java.io.File ;
import java.io.FileInputStream ;
import java.io.FileNotFoundException ;
import java.io.IOException ; import org.apache.poi.openxml4j.exceptions.InvalidFormatException ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ;
import org.apache.poi.ss.formula.udf.AggregatingUDFFinder ;
import org.apache.poi.ss.formula.udf.DefaultUDFFinder ;
import org.apache.poi.ss.formula.udf.UDFFinder ;
import org.apache.poi.ss.usermodel.Cell ;
import org.apache.poi.ss.usermodel.CellValue ;
import org.apache.poi.ss.usermodel.Row ;
import org.apache.poi.ss.usermodel.Sheet ;
import org.apache.poi.ss.usermodel.Workbook ;
import org.apache.poi.ss.usermodel.WorkbookFactory ;
import org.apache.poi.ss.util.CellReference ; public class Evaluator { public static void main( String[] args ) { System.out.println( "fileName: " + args[0] ) ;
System.out.println( "cell: " + args[1] ) ; File workbookFile = new File( args[0] ) ; try {
FileInputStream fis = new FileInputStream(workbookFile);
Workbook workbook = WorkbookFactory.create(fis); FormulaEvaluator evaluator = workbook.getCreationHelper().createFormulaEvaluator(); CellReference cr = new CellReference( args[1] ) ;
String sheetName = cr.getSheetName() ;
Sheet sheet = workbook.getSheet( sheetName ) ;
int rowIdx = cr.getRow() ;
int colIdx = cr.getCol() ;
Row row = sheet.getRow( rowIdx ) ;
Cell cell = row.getCell( colIdx ) ; CellValue value = evaluator.evaluate( cell ) ; System.out.println("returns value: " + value ) ; } catch( FileNotFoundException e ) {
e.printStackTrace();
} catch( InvalidFormatException e ) {
e.printStackTrace();
} catch( IOException e ) {
e.printStackTrace();
}
}
}
If you run this code, you're likely to get the following error:
Exception in thread "main" org.apache.poi.ss.formula.eval.NotImplementedException: Error evaluating cell Sheet1!B4
at org.apache.poi.ss.formula.WorkbookEvaluator.addExceptionInfo(WorkbookEvaluator.java:321)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateAny(WorkbookEvaluator.java:288)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluate(WorkbookEvaluator.java:221)
at org.apache.poi.hssf.usermodel.HSSFFormulaEvaluator.evaluateFormulaCellValue(HSSFFormulaEvaluator.java:320)
at org.apache.poi.hssf.usermodel.HSSFFormulaEvaluator.evaluate(HSSFFormulaEvaluator.java:182)
at poi.tests.Evaluator.main(Evaluator.java:61)
Caused by: org.apache.poi.ss.formula.eval.NotImplementedException: calculatePayment
at org.apache.poi.ss.formula.UserDefinedFunction.evaluate(UserDefinedFunction.java:59)
at org.apache.poi.ss.formula.OperationEvaluatorFactory.evaluate(OperationEvaluatorFactory.java:129)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateFormula(WorkbookEvaluator.java:456)
at org.apache.poi.ss.formula.WorkbookEvaluator.evaluateAny(WorkbookEvaluator.java:279)
... 4 more
How would we make it so POI can use this sheet?
Defining Your Function
To 'convert' this code to Java and make it available to POI you need to implement a FreeRefFunction instance. FreeRefFunction is an interface in the org.apache.poi.ss.formula.functions package. This interface defines one method, evaluate(ValueEval[] args, OperationEvaluationContext ec), which is how you will receive the argument values from POI.
The evaluate() method as defined above is where you will convert the ValueEval instances to the proper number types. The following code snippet shows you how to get your values:
public class CalculateMortgage implements FreeRefFunction { @Override
public ValueEval evaluate( ValueEval[] args, OperationEvaluationContext ec ) {
if (args.length != 3) {
return ErrorEval.VALUE_INVALID;
} double principal, rate, years, result;
try {
ValueEval v1 = OperandResolver.getSingleValue( args[0],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v2 = OperandResolver.getSingleValue( args[1],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v3 = OperandResolver.getSingleValue( args[2],
ec.getRowIndex(),
ec.getColumnIndex() ) ; principal = OperandResolver.coerceValueToDouble( v1 ) ;
rate = OperandResolver.coerceValueToDouble( v2 ) ;
years = OperandResolver.coerceValueToDouble( v3 ) ;
The first thing we do is check the number of arguments being passed since there is no sense in attempting to go further if you are missing critical information.
Next we declare our variables, in our case we need variables for:
- principal - the amount of the loan
- rate - the interest rate as a decimal
- years - the length of the loan in years
- result - the result of the calculation
Next, we use the OperandResolver to convert the ValueEval instances to doubles, though not directly. First we start by getting discreet values. Using the OperandResolver.getSingleValue() method we retrieve each of the values passed in by the cell in the spreadsheet. Next, we use the OperandResolver again to convert the ValueEval instances to doubles, in this case. This class has other methods of coercion for gettings Strings, ints and booleans. Now that we've got our primitive values we can move on to calculating the value.
As shown previously, we have the VBA source. We need to add code to our class to calculate the payment. To do this you could simply add it to the method we've already created but I've chosen to add it as its own method. Add the following method:
public double calculateMortgagePayment( double p, double r, double y ) { double i = r / 12 ;
double n = y * 12 ; double principalAndInterest =
p * (( i * Math.pow((1 + i),n ) ) / ( Math.pow((1 + i),n) - 1)) ; return principalAndInterest ;
}
The biggest change necessary is related to the exponents; Java doesn't have a notation for this so we had to add calls to Math.pow(). Now we need to add this call to our previous method:
result = calculateMortgagePayment( principal, rate, years ) ;
Having done that, the last things we need to do are to check to make sure we didn't get a bad result and, if not, we need to return the value. Add the following code to the class:
private void checkValue(double result) throws EvaluationException {
if (Double.isNaN(result) || Double.isInfinite(result)) {
throw new EvaluationException(ErrorEval.NUM_ERROR);
}
}
Then add a line of code to our evaluate method to call this new static method, complete our try/catch and return the value:
checkValue(result); } catch (EvaluationException e) {
e.printStackTrace() ;
return e.getErrorEval();
} return new NumberEval( result ) ;
So the whole class would be as follows:
import org.apache.poi.ss.formula.OperationEvaluationContext ;
import org.apache.poi.ss.formula.eval.ErrorEval ;
import org.apache.poi.ss.formula.eval.EvaluationException ;
import org.apache.poi.ss.formula.eval.NumberEval ;
import org.apache.poi.ss.formula.eval.OperandResolver ;
import org.apache.poi.ss.formula.eval.ValueEval ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ; /**
* A simple function to calculate principal and interest.
*
* @author Jon Svede
*
*/
public class CalculateMortgage implements FreeRefFunction { @Override
public ValueEval evaluate( ValueEval[] args, OperationEvaluationContext ec ) {
if (args.length != 3) {
return ErrorEval.VALUE_INVALID;
} double principal, rate, years, result;
try {
ValueEval v1 = OperandResolver.getSingleValue( args[0],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v2 = OperandResolver.getSingleValue( args[1],
ec.getRowIndex(),
ec.getColumnIndex() ) ;
ValueEval v3 = OperandResolver.getSingleValue( args[2],
ec.getRowIndex(),
ec.getColumnIndex() ) ; principal = OperandResolver.coerceValueToDouble( v1 ) ;
rate = OperandResolver.coerceValueToDouble( v2 ) ;
years = OperandResolver.coerceValueToDouble( v3 ) ; result = calculateMortgagePayment( principal, rate, years ) ; checkValue(result); } catch (EvaluationException e) {
e.printStackTrace() ;
return e.getErrorEval();
} return new NumberEval( result ) ;
} public double calculateMortgagePayment( double p, double r, double y ) {
double i = r / 12 ;
double n = y * 12 ; //M = P [ i(1 + i)n ] / [ (1 + i)n - 1]
double principalAndInterest =
p * (( i * Math.pow((1 + i),n ) ) / ( Math.pow((1 + i),n) - 1)) ; return principalAndInterest ;
} /**
* Excel does not support infinities and NaNs, rather, it gives a #NUM! error in these cases
*
* @throws EvaluationException (#NUM!) if <tt>result</tt> is <tt>NaN</> or <tt>Infinity</tt>
*/
static final void checkValue(double result) throws EvaluationException {
if (Double.isNaN(result) || Double.isInfinite(result)) {
throw new EvaluationException(ErrorEval.NUM_ERROR);
}
} }
Great! Now we need to go back to our original program that failed to evaluate our cell and add code that will allow it run our new Java code.
Registering Your Function
Now we need to register our function in the Workbook, so that the Formula Evaluator can resolve the name "calculatePayment" and map it to the actual implementation (CalculateMortgage). This is done using the UDFFinder object. The UDFFinder manages FreeRefFunctions which are our analogy for the VBA code. We need to create a UDFFinder. There are a few things we need to know in order to do this:
- The name of the function in the VBA code (in our case it is calculatePayment)
- The Class name of our FreeRefFunction
UDFFinder is actually an interface, so we need to use an actual implementation of this interface. Therefore we use the org.apache.poi.ss.formula.udf.DefaultUDFFinder class. If you refer to the Javadocs you'll see that this class expects to get two arrays, one containing the alias and the other containing an instance of the class that will represent that alias. In our case our alias will be calculatePayment and our class instance will be of the CalculateMortgage type. This class needs to be available at compile and runtime. Be sure to keep these arrays well organized because you'll run into problems if these arrays are of different sizes or the alias aren't in the same relative position in their respective arrays. Add the following code:
String[] functionNames = { "calculatePayment" } ;
FreeRefFunction[] functionImpls = { new CalculateMortgage() } ; UDFFinder udfs = new DefaultUDFFinder( functionNames, functionImpls ) ;
UDFFinder udfToolpack = new AggregatingUDFFinder( udfs ) ;
Now we have our UDFFinder instance and we've created the AggregatingUDFFinder instance. The last step is to pass this to our Workbook:
workbook.addToolPack(udfToolpack);
So now the whole class will look like this:
import java.io.File ;
import java.io.FileInputStream ;
import java.io.FileNotFoundException ;
import java.io.IOException ; import org.apache.poi.openxml4j.exceptions.InvalidFormatException ;
import org.apache.poi.ss.formula.functions.FreeRefFunction ;
import org.apache.poi.ss.formula.udf.AggregatingUDFFinder ;
import org.apache.poi.ss.formula.udf.DefaultUDFFinder ;
import org.apache.poi.ss.formula.udf.UDFFinder ;
import org.apache.poi.ss.usermodel.Cell ;
import org.apache.poi.ss.usermodel.CellValue ;
import org.apache.poi.ss.usermodel.Row ;
import org.apache.poi.ss.usermodel.Sheet ;
import org.apache.poi.ss.usermodel.Workbook ;
import org.apache.poi.ss.usermodel.WorkbookFactory ;
import org.apache.poi.ss.util.CellReference ; public class Evaluator { public static void main( String[] args ) { System.out.println( "fileName: " + args[0] ) ;
System.out.println( "cell: " + args[1] ) ; File workbookFile = new File( args[0] ) ; try {
FileInputStream fis = new FileInputStream(workbookFile);
Workbook workbook = WorkbookFactory.create(fis); String[] functionNames = { "calculatePayment" } ;
FreeRefFunction[] functionImpls = { new CalculateMortgage() } ; UDFFinder udfs = new DefaultUDFFinder( functionNames, functionImpls ) ;
UDFFinder udfToolpack = new AggregatingUDFFinder( udfs ) ; workbook.addToolPack(udfToolpack); FormulaEvaluator evaluator = workbook.getCreationHelper().createFormulaEvaluator(); CellReference cr = new CellReference( args[1] ) ;
String sheetName = cr.getSheetName() ;
Sheet sheet = workbook.getSheet( sheetName ) ;
int rowIdx = cr.getRow() ;
int colIdx = cr.getCol() ;
Row row = sheet.getRow( rowIdx ) ;
Cell cell = row.getCell( colIdx ) ; CellValue value = evaluator.evaluate( cell ) ; System.out.println("returns value: " + value ) ; } catch( FileNotFoundException e ) {
e.printStackTrace();
} catch( InvalidFormatException e ) {
e.printStackTrace();
} catch( IOException e ) {
e.printStackTrace();
}
}
}
Now that our evaluator is aware of the UDFFinder which in turn is aware of our FreeRefFunction, we're ready to re-run our example:
Evaluator mortgage-calculation.xls Sheet1!B4
which prints the following output in the console:
fileName: mortgage-calculation.xls
cell: Sheet1!B4
returns value: org.apache.poi.ss.usermodel.CellValue [790.7936267415464]
That is it! Now you can create Java code and register it, allowing your POI based appliction to run spreadsheets that previously were inaccessible.
This example can be found in the src/examples/src/org/apache/poi/ss/examples/formula folder in the source.
[转]POI : How to Create and Use User Defined Functions的更多相关文章
- Read / Write Excel file in Java using Apache POI
Read / Write Excel file in Java using Apache POI 2014-04-18 BY DINESH LEAVE A COMMENT About a year o ...
- [Hive - LanguageManual] Create/Drop/Alter -View、 Index 、 Function
Create/Drop/Alter View Create View Drop View Alter View Properties Alter View As Select Version info ...
- PostgreSQL 之 CREATE FUNCTION
官方文档 语法: CREATE [ OR REPLACE ] FUNCTION name ( [ [ argmode ] [ argname ] argtype [ { DEFAULT | = } d ...
- MySql批处理的小窍门:排行榜类数据生成
MySql批处理的小窍门:排行榜类数据生成 最近在做新版本的开发,其中涉及到排行榜的批量预生成,在此分享给大家. 关键点 名次的计算(不考虑用游标) 单榜单查询 对于排行榜这种类型的数据,当只查一个排 ...
- 超级顽固的流方式读取doc,docx乱码问题
因为工作中需要一个把doc或者docx的office文档内容,需要读取出来,并且也没展示功能.代码中第一考虑可能就是通过读取流方式,结果写了以后,各种乱码,百科的解决方案也是千奇百怪,第一点:可能是文 ...
- Phoenix综述(史上最全Phoenix中文文档)
个人主页:http://www.linbingdong.com 简书地址:http://www.jianshu.com/users/6cb45a00b49c/latest_articles 网上关于P ...
- 【repost】JavaScript Scoping and Hoisting
JavaScript Scoping and Hoisting Do you know what value will be alerted if the following is executed ...
- PHP7函数大全(4553个函数)
转载来自: http://www.infocool.net/kb/PHP/201607/168683.html a 函数 说明 abs 绝对值 acos 反余弦 acosh 反双曲余弦 addcsla ...
- 有关binlog的那点事(mysql5.7.13)
binlog作为mysql中最重要的日志之一,能实现异常恢复以及主从复制. 我们主要讨论的是主从复制中的binlog,这里将以mysql5.7.13的源码为主要依据来分析binlog. 在主从复制中, ...
随机推荐
- ubuntu16.04 跑Apollo Demo
1.安装docker(参考网址:https://docs.docker.com/install/linux/docker-ce/ubuntu/) Uninstall old versions Olde ...
- C++之自己实现的String类全部
一:回顾 (1)c++中的string类是在面试中和笔试中经常考的题目: 工程代码免费下载 string类的自行实现 (2)c++中的string类和fstream类合起来是处理外部数据的利器: (3 ...
- VC解析XML--使用CMarkup类解析XML
经过今天尝试MFC解析XML串,也算有了不少收获,总结一下. 我是使用的CMarkup类对XML进行操作. CMarkup好象都是先从一个xml文件里 ...
- poj 2069 Super Star —— 模拟退火
题目:http://poj.org/problem?id=2069 仍是随机地模拟退火,然而却WA了: 看看网上的题解,都是另一种做法——向距离最远的点靠近: 于是也改成那样,竟然真的A了...感觉这 ...
- 从MySQL获取数据
安装 PM> install-package newtonsoft.json PM> install-package mysql.data string connectionString ...
- Ubuntu 安装indicator-sysmonitor
之前就像安装一个软件用来查看Ubuntu的CPU, 内存, 网速情况, 终于让我碰到了--indicator-sysmonitor 仅需三条命令, 你值得拥有: sudo add-apt-reposi ...
- seafile文档
https://manual-cn.seafile.com/deploy_windows/download_and_setup_seafile_windows_server.html
- Subsets Forming Perfect Squares
题意: 给出n个数字,选出若干个数字,使得这些数字的乘积是一个完全平方数,问有多少种选法. 解法: 考虑异或方程组,$x_i$表示第i个数字是否选, 注意到只要保证结果中各个质因数都出现偶数次就可保证 ...
- vector 大小
vector定义以后就最好确定大小resize(),否则在vector析构时可能出现 "double free or corruption"这样的错误
- 在linux上部署tomcat服务
在linux上部署tomcat 1.安装JDK 2.下载tomcat http://tomcat.apache.org/download-70.cgi 3.上传到服务器,并解压 4.上传war包或者已 ...