albert's blog

2014/12/25

Apache CollectionUtils 應用

Assume we had 2 list data

      //create a value object and add into list1
        UnpaymentDataVo vo1 = new UnpaymentDataVo();
        vo1.setDebtType("B");
        vo1.setBudgetCode(new BigDecimal(1));
        vo1.setDebitName("總預算");
        vo1.setDamt(new BigDecimal(5156340000000L));
        vo1.setCamt(new BigDecimal(3489209150690L));
        vo1.setIamt(new BigDecimal(69360745739L));
        vo1.setRamt(new BigDecimal(1597620103571L));

        List<UnpaymentDataVo> list1 = new ArrayList<UnpaymentDataVo>();
        list1.add(vo1);

 //create 2 value objects and add into list2
        UnpaymentDataVo vo2 = new UnpaymentDataVo();
        vo2.setDebtType("A");
        vo2.setBudgetCode(new BigDecimal(2));
        vo2.setDebitName("易淹水地區水患治理計劃第2期特別預算");
        vo2.setDamt(new BigDecimal(7500000000L));
        vo2.setCamt(new BigDecimal(0));
        vo2.setIamt(new BigDecimal(637500000L));
        vo2.setRamt(new BigDecimal(7500000000L));

        UnpaymentDataVo vo3 = new UnpaymentDataVo();
        vo3.setDebtType("A");
        vo3.setBudgetCode(new BigDecimal(1));
        vo3.setDebitName("振興經濟擴大公共建設特別預算");
        vo3.setDamt(new BigDecimal(204400000000L));
        vo3.setCamt(new BigDecimal(0));
        vo3.setIamt(new BigDecimal(14465750000L));
        vo3.setRamt(new BigDecimal(204400000000L));

        List<UnpaymentDataVo> list2 = new ArrayList<UnpaymentDataVo>();
        list2.add(vo2);
        list2.add(vo3);

 //print list1 and list2
        log.debug("list1=" + list1.toString());
        log.debug("list2=" + list2.toString());

The data structure looks like this:

1
2
3

- list1=[UnpaymentDataVo(debtType=B, budgetCode=1, debitName=總預算, damt=5156340000000, camt=3489209150690, iamt=69360745739, ramt=1597620103571)]
- list2=[UnpaymentDataVo(debtType=A, budgetCode=2, debitName=易淹水地區水患治理計劃第2期特別預算, damt=7500000000, camt=0, iamt=637500000, ramt=7500000000), 
         UnpaymentDataVo(debtType=A, budgetCode=1, debitName=振興經濟擴大公共建設特別預算, damt=204400000000, camt=0, iamt=14465750000, ramt=204400000000)]

If you would like to find out the data from list2 which condition is debt_type='A' and budget_code=2

        // find out the data from list2 which condition is debt_type='A' and budget_code=2
        List<UnpaymentDataVo> searchResult = new ArrayList<UnpaymentDataVo>();
        CollectionUtils.select(list2, new Predicate() {
            @Override
            public boolean evaluate(Object object) {
                UnpaymentDataVo vo = (UnpaymentDataVo) object;
                return "A".equals(vo.getDebtType())
                        && vo.getBudgetCode().compareTo(new BigDecimal(2)) == 0;
            }
        }, searchResult);

        log.debug("searchResult = " + searchResult.toString());

The search result looks like this:

1 2	- searchResult = [UnpaymentDataVo(debtType=A, budgetCode=2, debitName=易淹水地區水患治理計劃第2期特別預算, damt=7500000000, camt=0, iamt=637500000, ramt=7500000000), ]

If you would like to get a Collection containing the union of list1 and list2

        // Returns a Collection containing the union of the given Collections
        List<UnpaymentDataVo> unionResult = (List<UnpaymentDataVo>) CollectionUtils.union(list1,
                list2);
        log.debug("unionResult = " + unionResult.toString());

The union result looks like this:

1
2
3

- unionResult = [UnpaymentDataVo(debtType=A, budgetCode=2, debitName=易淹水地區水患治理計劃第2期特別預算, damt=7500000000, camt=0, iamt=637500000, ramt=7500000000), 
                 UnpaymentDataVo(debtType=A, budgetCode=1, debitName=振興經濟擴大公共建設特別預算, damt=204400000000, camt=0, iamt=14465750000, ramt=204400000000), 
   UnpaymentDataVo(debtType=B, budgetCode=1, debitName=總預算, damt=5156340000000, camt=3489209150690, iamt=69360745739, ramt=1597620103571)]

If you would like to return a collection which contains unionResult - list2

        // Returns a new Collection containing a - b.
        List<UnpaymentDataVo> subtractResult = (List<UnpaymentDataVo>) CollectionUtils.subtract(
                unionResult, list2);
        log.debug("subtractResult = " + subtractResult.toString());

The subtract result looks like this:

1	- subtractResult = [UnpaymentDataVo(debtType=B, budgetCode=1, debitName=總預算, damt=5156340000000, camt=3489209150690, iamt=69360745739, ramt=1597620103571)]

Reference
[1] https://commons.apache.org/proper/commons-collections/javadocs/api-3.2.1/org/apache/commons/collections/CollectionUtils.html#subtract(java.util.Collection, java.util.Collection)

[Oracle] Use to_char to format date column

Question

I have two column, ISSUE_DATE & DUE_DATE, with date data type.

But I only new YYYYMM not YYYYMMDD in report, how to format them?

Answer
We can use TO_CHAR to fulfill this requirement.

The to_char syntax is:

TO_CHAR( value, [ format_mask ], [ nls_language ] )

value can either be a number or date that will be converted to a string.
format_mask is optional. This is the format that will be used to convert value to a string.
nls_language is optional. This is the nls language used to convert value to a string.

Example.

SELECT ID, 
        DUE_DATE, 
        TO_CHAR( DUE_DATE, 'YYYYMM') FORMATED_DUE_DATE,
        ISSUE_DATE,
        TO_CHAR( ISSUE_DATE, 'YYYYMM') FORMATED_ISSUE_DATE
FROM ISSUE_MAIN

Reference
[1] http://www.techonthenet.com/oracle/functions/to_char.php

2014/12/22

市場乘數 (平均本益比)

無論是漫步華爾街或是彼得學林的學以致富的書籍中，作者都有提到不要買價格超過其實際價值的股票，強調用本益比來挑選股票的重要性。

如果想知道目前本隻本益比是偏高或偏低，可以看過去的歷史紀錄來判斷。
以中華電信為例，目前中華電信(2412)的本益比是18.05

我們可以去查一下過去五年來的最高平均、最低平均以及平均本益比的值

根據下圖的資訊，可以得知99~103年的歷年最高、最低本益比資訊，故：

平均最高本益比 = (19 + 20 + 18 + 18 + 16) / 5 = 18.2
平均最低本益比 = (17 + 17 + 14 + 14 + 13) / 5 = 15
平均本益比 = 16.6

由上述資訊可以看出目前中華電信本益比為18.05，屬於接近偏高

然而以上是用單支股票的方式來看，在學以致富此本書中有提到，我們也可以一次挑選多家公司，把他們的股價加起來，然後除以它們的盈餘，就可以得出一個平均本益比。在華爾街，人們用此方式計算道瓊工業指數、S&P 500指數的本益比，得出所謂的市場乘數(Market Multiple)，也就是平均本益比，其代表著人們願意以多高的價格買下多少盈餘的公司。

在台灣的話，我們可以參考台灣50的成分股，以下是50檔台灣50的本日股價與EPS相關資訊：

代號	公司名稱	股價	EPS
3474	華亞科	47.60	7.10
4938	和碩	72.50	4.94
3481	群創	14.95	0.97
2330	台積電	138.50	8.82
2303	聯電	14.45	0.66
2882	國泰金	46.15	3.63
2357	華碩	335.00	27.68
1303	南亞	61.60	4.43
2883	開發金	9.83	0.68
1301	台塑	68.00	3.79
2002	中鋼	26.10	1.22
2311	日月光	39.10	2.55
2317	鴻海	88.00	8.21
1402	遠東新	30.05	2.23
2892	第一金	18.45	1.44
2880	華南金	17.40	1.30
2801	彰銀	18.00	1.36
1216	統一	48.60	2.00
1101	台泥	42.45	2.89
1102	亞泥	37.85	2.76
2382	廣達	78.50	4.88
2308	台達電	186.00	8.31
1326	台化	64.50	3.39
2886	兆豐金	24.10	2.28
2891	中信金	20.40	2.66
2325	矽品	48.20	3.50
2105	正新	71.60	5.13
2395	研華	216.00	7.44
2408	南科	73.90	10.30
2412	中華電	92.60	5.12
2409	友達	15.10	1.30
2207	和泰車	463.00	16.63
2301	光寶科	36.00	3.24
9904	寶成	38.10	2.94
2912	統一超	245.00	8.76
2354	鴻準	87.60	5.21
2474	可成	242.00	19.40
3045	台灣大	99.90	5.78
2454	聯發科	455.00	29.64
2881	富邦金	49.45	5.81
2887	台新金	12.90	1.62
4904	遠傳	69.90	3.71
2885	元大金	15.20	1.53
3008	大立光	2,325.00	113.67
2498	宏達電	136.50	1.62
2884	玉山金	19.55	1.50
2890	永豐金	12.80	1.53
6505	台塑化	66.20	3.02
5880	合庫金	18.45	1.26
2227	裕日車	327.00	20.60

上述50檔台灣50的股價total為 6785.03，盈餘的total為 390.44，把股價 (6785.03) 除以盈餘 (390.44) ，可以得到這50檔的平均本益比為17.38

Total 股價	Total 盈餘	市場乘數 (市場平均本益比)
6785.03	390.44	17.38

我們就可以拿17.38這個數字做為參考，人們願意購買的本益比的股票約為17.38，代表著人們願意以多高的價格買下多少盈餘的公司股票，若過高的話就要考慮一下了，這個或許也可以作為一個股票買賣的一個參考

Reference
[1] http://www.twse.com.tw/ch/trading/indices/twco/tai50i.php
[2] http://www.cnyes.com/twstock/profile/2412.htm
[3] http://jsjustweb.jihsun.com.tw/z/zc/zca/zca_2412.djhtm

How to Determine if the Date Format is Correct or Not in Java

Problem
We hope the date parameter should be yyyyMMdd format and only accept Western calendar system.
If user provide a date parameter with Minguo calendar, it should return error to user.

Solution
This DateUtils class provides an isCorrectDateFormatWithWesternYear method, it will

return false if date format is incorrect
return true if date format is correct

We define the expected format in Line 24.
Set lenient to false with strict parsing, and call parse method to check the parse result.

package gov.nta.dbm.utils;

import java.text.ParseException;
import java.text.SimpleDateFormat;

import lombok.extern.slf4j.Slf4j;

/**
 * The Class DateUtils.
 */
@Slf4j
public class DateUtils {

    /**
     * Checks if is correct date format with western year.
     * 
     * @param dateStr
     *            the date str
     * @return the boolean
     */
    public static Boolean isCorrectDateFormatWithWesternYear(String dateStr) {
        Boolean isCollect = Boolean.FALSE;
        // instantiate SimpleDateFormat object with specific date format
        SimpleDateFormat dateFormat = new SimpleDateFormat("yyyyMMdd");
        try {
            // Specify whether or not date/time parsing is to be lenient. With lenient parsing, the
            // parser may use heuristics to interpret inputs that do not precisely match this
            // object's format. With strict parsing, inputs must match this object's format
            dateFormat.setLenient(false);
            // call parse method, return true if parse successfully, return false if fail to parse.
            dateFormat.parse(dateStr);
            isCollect = Boolean.TRUE;
        } catch (ParseException e) {
            isCollect = Boolean.FALSE;
            e.printStackTrace();
        }
        return isCollect;
    }

    /**
     * The main method.
     * 
     * @param args
     *            the arguments
     */
    public static void main(String[] args) {
        log.debug("test1=" + DateUtils.isCorrectDateFormatWithWesternYear("20141231"));
        log.debug("test1=" + DateUtils.isCorrectDateFormatWithWesternYear("1031222"));
    }

}

In main method, we input "20141231" and "1031222" to do test, and here is the test result:

15:34:00.446 [main] DEBUG gov.nta.dbm.utils.DateUtils - test1=true
java.text.ParseException: Unparseable date: "1031222"
 at java.text.DateFormat.parse(DateFormat.java:357)
 at gov.nta.dbm.utils.DateUtils.isCorrectDateFormatWithWesternYear(DateUtils.java:31)
 at gov.nta.dbm.utils.DateUtils.main(DateUtils.java:48)
15:34:00.453 [main] DEBUG gov.nta.dbm.utils.DateUtils - test1=false

Reference
[1] http://docs.oracle.com/javase/7/docs/api/java/text/DateFormat.html

2014/12/19

How to delete a branch in EGit?

Question

Assume I created a branch, DBM_CommonService_getPaymentData, to implement a new function.

After I finished this function and commit & push to GitLab, and GitLab project owner merged this branch into master.

If I had switch to master and would like to delete DBM_CommonService_getPaymentData branch, how to delete it?

Answer

Step1. right click => Team => Advanced => Delete branch

Step2. Selected the branch name you want to delete => Click "OK"

See..as you right click => Team => Switch to, you won't see DBM_CommonService_getPaymentData branch anymore.

Reference

[1] http://www.saros-project.org/node/142

2014/12/17

How to Run Stored Procedure in Oracle SQL Developer

Problem
Assume I created a stored procedure which named PROC_FMS406R_TAB5_RPT1_STEP1

create or replace PROCEDURE PROC_FMS406R_TAB5_RPT1_STEP1 
(
  INPUT_YYY_MM IN VARCHAR2 
, INPUT_LAST_YEAR IN VARCHAR2 
, INPUT_USER_ID IN VARCHAR2 
) AS 
BEGIN
  
  --1.5.2 計算『歲入估測分析表(含特別預算)』上一年度的一月到12月的實際數
 -- ignore execution detail

  
  
 --1.5.3 計算『歲入估測分析表(含特別預算)』上一年度的合計數
 -- ignore execution detail
  
END PROC_FMS406R_TAB5_RPT1_STEP1;

I tried to execute PROC_FMS406R_TAB5_RPT1_STEP1 in SQL worksheet

It will fail to execute and show this error message:

命令的第 2 行開始發生錯誤 -
PROC_FMS406R_TAB5_RPT1_STEP1(:INPUT_YYY_MM, :INPUT_LAST_YEAR, :INPUT_USER_ID)
錯誤報告 -
不明的命令

How to Fix This Problem
Add BEGIN before the stored procedure and add END; after the stored procedure.

1
2
3

BEGIN
PROC_FMS406R_TAB5_RPT1_STEP1(:INPUT_YYY_MM, :INPUT_LAST_YEAR, :INPUT_USER_ID);
END;

Reference
[1] http://dba.stackexchange.com/questions/57163/how-to-run-procedure-in-oracle-sql-developer

Total Pageviews