seraekim/TableParser.java

## TableParser.java
package org.srkim.test;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.select.Elements;

/**
 * Search both row and col header of multiple tables in a html request.
 * even can search spans.
 *
 * there are some functions to add that are currently not applicable. see the following:
 *
 * <blockquote>
 *   <ol>
 *     <li>thead, tbody and th</li>
 *     <li>vertical and horizontal scroll fixed row and col header resulting in one shown table having multiple tables</li>
 *   </ol>
 * </blockquote>
 * @author srkim
 *
 */
public class TableParser {

	static String[] rowHeaders = {"정정일자","1. 정정관련 공시서류","2. 정정관련 공시서류제출일","3. 정정사유","4. 정정사항"
			,"1. 일시","3. 의안 주요내용","4. 이사회결의일(결정일)","2. 장소","-사외이사 참석여부","-감사(사외이사가 아닌 감사위원) 참석여부"
			,"-주주총회 구분","※관련공시","사업목적 추가","사업목적 변경","사업목적 삭제", "날짜", "시간", "참석(명)", "불참(명)"};
	static String[] colHeaders = {"정정항목","정정전","정정후","성명","출생년월","임기","신규선임여부","주요경력(현직포함)"
			,"이사 등으로 재직 중인 다른 법인명(직위)","상근여부","구분","내용","이유","변경전","변경후","5. 기타 투자판단에 참고할 사항"};

	public static void main(String[] args) throws IOException {
//		String html = "<!doctype html><html><head><meta charset=\"UTF-8\">"
//				+ "<title>Untitled Document</title><style>table * { border: solid 1px red;}</style></head>"
//				+ "<body><table><tr><td rowspan=\"2\">l1</td><td colspan=\"2\">t1</td><td colspan=\"2\">t2</td></tr>"
//				+ "<tr><td>1</td><td>2</td><td>3</td><td>4</td></tr><tr><td>l2</td><td>6</td><td>7</td><td>8</td><td>9</td>"
//				+ "</tr></table></body></html>";

		String urlStr = "http://dart.fss.or.kr/report/viewer.do?rcpNo=20171121900601&dcmNo=5855132&eleId=0&offset=0&length=0&dtd=HTML";
		Document document = Jsoup.connect(urlStr).get();
		Elements tables = document.select("table");

		for(int i = 0; i < tables.size(); i++) {
			System.out.println("start table idx : " + i);
			Elements rows = tables.eq(i).select("tr");
			List<List<Object>> parsedSingleTable = splitSpans(rows);
			// 검색 해보기
			search(parsedSingleTable, "- 제2-8호 의안", false);
			search(parsedSingleTable, "임시주주총회 소집일시 및 안건 변경", true); // there is only row header.
			search(parsedSingleTable, "전) 주식회사 SBT 투자 사장", false); // there are both headers.
		}
	}

	/**
	 * split spans scanning rows
	 *
	 * @param rows
	 * @return parsed double list of split colspans and colspans
	 *
	 */
	public static List<List<Object>> splitSpans(Elements rows) {
		int rowCount = rows.size();
//		System.out.println("rowCount : " + rowCount);
		List<List<Object>> r = new ArrayList<List<Object>>();
		for (int i = 0; i < rowCount; i++) {
			r.add(new ArrayList<Object>());
		}

		for (int i = 0; i < rowCount; i++) {
			Elements cols = rows.get(i).select("td");
			int colCount = cols.size();

			for (int j = 0; j < colCount; j++) {
				String text = cols.get(j).text();

				String rSpanStr = cols.get(j).attr("rowspan");
				int rSpan = 1;
				if (rSpanStr.length() != 0) {
					rSpan = Integer.parseInt(rSpanStr);
				}

				String cSpanStr = cols.get(j).attr("colspan");
				int cSpan = 1;
				if (cSpanStr.length() != 0) {
					cSpan = Integer.parseInt(cSpanStr);
				}

				r.get(i).add(text);

				for (int k = 1; k < rSpan; k++) {
					r.get(i + k).add(text);
				}

				for (int k = 1; k < cSpan; k++) {
					r.get(i).add(text);
				}
			}
		}
		return r;
	}
	/**
	 * search row and col header of a keyword in tds
	 *
	 * @param r double list of rows and cols of a table
	 * @param keyword search word
	 * @param rowHeaderIdx designate row header index
	 * @param colHeaderIdx designate col header index
	 * @param isPerfectMatch perfect equals literally otherwise contains filter applied
	 */
	public static void search(List<List<Object>> r, String keyword, boolean isPerfectMatch) {
		// scan and print all elements of rows
		if(keyword == null || keyword.equals("")) {
			for (int i = 0; i < r.size(); i++) {
				for (int j = 0; j < r.get(0).size(); j++) {
					System.out.print(r.get(i).get(j) + "(" + i + "," + j + ") ");
				}
				System.out.println();
			}
		} else {
			for (int i = 0; i < r.size(); i++) {
				for (int j = 0; j < r.get(0).size(); j++) {
					String v = (String) r.get(i).get(j);
					if(isPerfectMatch) {
						if(v.equals(keyword)) {
							printSearchResult(r, i, j);
							break; // prevent from duplicate search by split colspans
						}
					} else {
						if(v.contains(keyword)) {
							printSearchResult(r, i, j);
							break; // prevent from duplicate search by split colspans
						}
					}
				}
			}
		}
	}
	/**
	 * print headers with their indexes
	 *
	 * @param r
	 * @param i
	 * @param j
	 * @param rowHeaderIdx
	 * @param colHeaderIdx
	 */
	public static void printSearchResult(List<List<Object>> r, int i, int j, Integer rowHeaderIdx, Integer colHeaderIdx) {
		String rowHeader = "";
		String colHeader = "";

		if(rowHeaderIdx != null) {
			rowHeader = "[rowhead " + r.get(i).get(rowHeaderIdx) + "]";
		}
		if(colHeaderIdx != null) {
			colHeader = "[colhead " + r.get(colHeaderIdx).get(j) + "]";
		}
		System.out.println("(" + i + "," + j + ") " + rowHeader + colHeader + " ====>  " + r.get(i).get(j));
	}

	/**
	 * print headers with predefined dics.
	 * @param r
	 * @param i
	 * @param j
	 * @param rowHeaderIdx
	 * @param colHeaderIdx
	 */
	public static void printSearchResult(List<List<Object>> r, int i, int j) {
		String rowHeader = "[]";
		String colHeader = "[]";

		// row scan
		for (String rowName : rowHeaders) {
			if (r.get(i).get(0).equals(rowName)) {
				rowHeader = "["+rowName+"]";
				break;
			}
		}
		/*
		 * col san.
		 * scan the past indexes of row and col with i,j as starting position.
		 */
		for (int k = i - 1; k >= 0; k--) {
			if (colHeader.equals("[]")) {
				for (String colName : colHeaders) {
					if (r.get(k).get(j).equals(colName)) {
						colHeader = "["+colName+"]";
						break;
					}
				}
			} else {
//				System.out.println(colHeader);
				break;
			}
		}

		System.out.println("(" + i + "," + j + ") " + rowHeader + colHeader + " ====>  " + r.get(i).get(j));
	}
	/*
	 * execution result....
	 *
	 * start table idx : 0
start table idx : 1
(6,1) [3. 의안 주요내용][정정전] ====>  <부의안건> 제1호 의안 : 정관 일부 변경의 건 - ...
(2,1) [3. 정정사유][] ====>  임시주주총회 소집일시 및 안건 변경
start table idx : 2
start table idx : 3
start table idx : 4
(3,4) [][주요경력(현직포함)] ====>  전) 주식회사 SBT 투자 사장 현) 한국감사협회 이사 현) GRC코리아 감사
start table idx : 5
start table idx : 6
start table idx : 7

	 */
}
	package org.srkim.test;

	import java.io.IOException;
	import java.util.ArrayList;
	import java.util.List;

	import org.jsoup.Jsoup;
	import org.jsoup.nodes.Document;
	import org.jsoup.select.Elements;

	/**
	* Search both row and col header of multiple tables in a html request.
	* even can search spans.
	*
	* there are some functions to add that are currently not applicable. see the following:
	*
	* <blockquote>
	* <ol>
	* <li>thead, tbody and th</li>
	* <li>vertical and horizontal scroll fixed row and col header resulting in one shown table having multiple tables</li>
	* </ol>
	* </blockquote>
	* @author srkim
	*
	*/
	public class TableParser {

	static String[] rowHeaders = {"정정일자","1. 정정관련 공시서류","2. 정정관련 공시서류제출일","3. 정정사유","4. 정정사항"
	,"1. 일시","3. 의안 주요내용","4. 이사회결의일(결정일)","2. 장소","-사외이사 참석여부","-감사(사외이사가 아닌 감사위원) 참석여부"
	,"-주주총회 구분","※관련공시","사업목적 추가","사업목적 변경","사업목적 삭제", "날짜", "시간", "참석(명)", "불참(명)"};
	static String[] colHeaders = {"정정항목","정정전","정정후","성명","출생년월","임기","신규선임여부","주요경력(현직포함)"
	,"이사 등으로 재직 중인 다른 법인명(직위)","상근여부","구분","내용","이유","변경전","변경후","5. 기타 투자판단에 참고할 사항"};

	public static void main(String[] args) throws IOException {
	// String html = "<!doctype html><html><head><meta charset=\"UTF-8\">"
	// + "<title>Untitled Document</title><style>table * { border: solid 1px red;}</style></head>"
	// + "<body><table><tr><td rowspan=\"2\">l1</td><td colspan=\"2\">t1</td><td colspan=\"2\">t2</td></tr>"
	// + "<tr><td>1</td><td>2</td><td>3</td><td>4</td></tr><tr><td>l2</td><td>6</td><td>7</td><td>8</td><td>9</td>"
	// + "</tr></table></body></html>";

	String urlStr = "http://dart.fss.or.kr/report/viewer.do?rcpNo=20171121900601&dcmNo=5855132&eleId=0&offset=0&length=0&dtd=HTML";
	Document document = Jsoup.connect(urlStr).get();
	Elements tables = document.select("table");

	for(int i = 0; i < tables.size(); i++) {
	System.out.println("start table idx : " + i);
	Elements rows = tables.eq(i).select("tr");
	List<List<Object>> parsedSingleTable = splitSpans(rows);
	// 검색 해보기
	search(parsedSingleTable, "- 제2-8호 의안", false);
	search(parsedSingleTable, "임시주주총회 소집일시 및 안건 변경", true); // there is only row header.
	search(parsedSingleTable, "전) 주식회사 SBT 투자 사장", false); // there are both headers.
	}
	}

	/**
	* split spans scanning rows
	*
	* @param rows
	* @return parsed double list of split colspans and colspans
	*
	*/
	public static List<List<Object>> splitSpans(Elements rows) {
	int rowCount = rows.size();
	// System.out.println("rowCount : " + rowCount);
	List<List<Object>> r = new ArrayList<List<Object>>();
	for (int i = 0; i < rowCount; i++) {
	r.add(new ArrayList<Object>());
	}

	for (int i = 0; i < rowCount; i++) {
	Elements cols = rows.get(i).select("td");
	int colCount = cols.size();

	for (int j = 0; j < colCount; j++) {
	String text = cols.get(j).text();

	String rSpanStr = cols.get(j).attr("rowspan");
	int rSpan = 1;
	if (rSpanStr.length() != 0) {
	rSpan = Integer.parseInt(rSpanStr);
	}

	String cSpanStr = cols.get(j).attr("colspan");
	int cSpan = 1;
	if (cSpanStr.length() != 0) {
	cSpan = Integer.parseInt(cSpanStr);
	}

	r.get(i).add(text);

	for (int k = 1; k < rSpan; k++) {
	r.get(i + k).add(text);
	}

	for (int k = 1; k < cSpan; k++) {
	r.get(i).add(text);
	}
	}
	}
	return r;
	}
	/**
	* search row and col header of a keyword in tds
	*
	* @param r double list of rows and cols of a table
	* @param keyword search word
	* @param rowHeaderIdx designate row header index
	* @param colHeaderIdx designate col header index
	* @param isPerfectMatch perfect equals literally otherwise contains filter applied
	*/
	public static void search(List<List<Object>> r, String keyword, boolean isPerfectMatch) {
	// scan and print all elements of rows
	if(keyword == null \|\| keyword.equals("")) {
	for (int i = 0; i < r.size(); i++) {
	for (int j = 0; j < r.get(0).size(); j++) {
	System.out.print(r.get(i).get(j) + "(" + i + "," + j + ") ");
	}
	System.out.println();
	}
	} else {
	for (int i = 0; i < r.size(); i++) {
	for (int j = 0; j < r.get(0).size(); j++) {
	String v = (String) r.get(i).get(j);
	if(isPerfectMatch) {
	if(v.equals(keyword)) {
	printSearchResult(r, i, j);
	break; // prevent from duplicate search by split colspans
	}
	} else {
	if(v.contains(keyword)) {
	printSearchResult(r, i, j);
	break; // prevent from duplicate search by split colspans
	}
	}
	}
	}
	}
	}
	/**
	* print headers with their indexes
	*
	* @param r
	* @param i
	* @param j
	* @param rowHeaderIdx
	* @param colHeaderIdx
	*/
	public static void printSearchResult(List<List<Object>> r, int i, int j, Integer rowHeaderIdx, Integer colHeaderIdx) {
	String rowHeader = "";
	String colHeader = "";

	if(rowHeaderIdx != null) {
	rowHeader = "[rowhead " + r.get(i).get(rowHeaderIdx) + "]";
	}
	if(colHeaderIdx != null) {
	colHeader = "[colhead " + r.get(colHeaderIdx).get(j) + "]";
	}
	System.out.println("(" + i + "," + j + ") " + rowHeader + colHeader + " ====> " + r.get(i).get(j));
	}

	/**
	* print headers with predefined dics.
	* @param r
	* @param i
	* @param j
	* @param rowHeaderIdx
	* @param colHeaderIdx
	*/
	public static void printSearchResult(List<List<Object>> r, int i, int j) {
	String rowHeader = "[]";
	String colHeader = "[]";

	// row scan
	for (String rowName : rowHeaders) {
	if (r.get(i).get(0).equals(rowName)) {
	rowHeader = "["+rowName+"]";
	break;
	}
	}
	/*
	* col san.
	* scan the past indexes of row and col with i,j as starting position.
	*/
	for (int k = i - 1; k >= 0; k--) {
	if (colHeader.equals("[]")) {
	for (String colName : colHeaders) {
	if (r.get(k).get(j).equals(colName)) {
	colHeader = "["+colName+"]";
	break;
	}
	}
	} else {
	// System.out.println(colHeader);
	break;
	}
	}

	System.out.println("(" + i + "," + j + ") " + rowHeader + colHeader + " ====> " + r.get(i).get(j));
	}
	/*
	* execution result....
	*
	* start table idx : 0
	start table idx : 1
	(6,1) [3. 의안 주요내용][정정전] ====> <부의안건> 제1호 의안 : 정관 일부 변경의 건 - ...
	(2,1) [3. 정정사유][] ====> 임시주주총회 소집일시 및 안건 변경
	start table idx : 2
	start table idx : 3
	start table idx : 4
	(3,4) [][주요경력(현직포함)] ====> 전) 주식회사 SBT 투자 사장 현) 한국감사협회 이사 현) GRC코리아 감사
	start table idx : 5
	start table idx : 6
	start table idx : 7

	*/
	}