diff --git a/250909_DIW_2508C_TEST.xlsx b/250909_DIW_2508C_TEST.xlsx
new file mode 100644
index 0000000..7297518
Binary files /dev/null and b/250909_DIW_2508C_TEST.xlsx differ
diff --git a/DIW_2508C.json b/DIW_2508C.json
index 3df41c9..ca67d29 100644
--- a/DIW_2508C.json
+++ b/DIW_2508C.json
@@ -351,89 +351,79 @@
"item": "② 다단 2단"
},
"4": {
- "path": "//RECTANGLE//CHAR[text()='{searchValue}']/ancestor::RECTANGLE/SHAPEOBJECT/SIZE/@Width",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE/SHAPEOBJECT/SIZE/@Width",
"value": "60",
"points": 2,
- "category": "mmSize",
+ "category": "Rectangle.mmSize",
"item": "문구 (구강건강관리)/① 크기-너비 (60 mm)"
},
"5": {
- "path": "//RECTANGLE//CHAR[text()='{searchValue}']/ancestor::RECTANGLE/SHAPEOBJECT/SIZE/@Height",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE/SHAPEOBJECT/SIZE/@Height",
"value": "12",
"points": 2,
- "category": "mmSize",
+ "category": "Rectangle.mmSize",
"item": "문구 (구강건강관리)/② 크기-높이 (12 mm)"
},
"6": {
- "path": "//RECTANGLE[.//CHAR[text()='{searchValue}']]//LINESHAPE",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE//LINESHAPE",
"value": {
"Style": "DoubleSlim",
"Width": "283"
},
"points": 2,
- "category": "LineShape",
+ "category": "Rectangle.LineShape",
"item": "문구 (구강건강관리)/③ 테두리 : 이중 실선(1.00mm)",
"desc": "1mm = 283pt value['Width']에 pt값 입력"
},
"7": {
- "path": "//RECTANGLE[.//CHAR[text()='{searchValue}']]/@Ratio",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE/@Ratio",
"value": "20",
"points": 2,
- "category": "OneAnswer",
+ "category": "Rectangle.OneAnswer",
"item": "문구 (구강건강관리)/④ 글상자 모서리 (반원)",
"desc": "모서리 비율 반원:50 / 둥근모양:20"
},
"8": {
- "path": "//RECTANGLE[.//CHAR[text()='{searchValue}']]//WINDOWBRUSH/@FaceColor",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE//WINDOWBRUSH/@FaceColor",
"value": "187,140,209",
"points": 2,
- "category": "Color",
+ "category": "Rectangle.Color",
"item": "문구 (구강건강관리)/⑤ 채우기 : 색상(RGB:187,140,209)"
},
"9": {
- "path": "//RECTANGLE[.//CHAR[text()='{searchValue}']]/SHAPEOBJECT/POSITION/@TreatAsChar",
- "searchValue": "구강건강관리",
+ "path": "//RECTANGLE/SHAPEOBJECT/POSITION/@TreatAsChar",
"value": "true",
"points": 1,
- "category": "OneAnswer",
+ "category": "Rectangle.OneAnswer",
"item": "문구 (구강건강관리)/⑥ 글상자 위치 (글자처럼 취급)"
},
"10": {
- "path": "//PARASHAPE[@Id=//RECTANGLE//CHAR[text()='{searchValue}']/ancestor::P[last()]/@ParaShape]/@Align",
- "searchValue": "구강건강관리",
+ "path": "//PARASHAPE[@Id='{rectangle_parashape_id}']/@Align",
"value": "Center",
"points": 1,
- "category": "OneAnswer",
+ "category": "Rectangle.TextBoxAlign",
"item": "문구 (구강건강관리)/⑦ 글상자 정렬 (가운데 정렬)"
},
"11": {
"path": "//TEXT[CHAR[text()='{searchValue}']]/@CharShape",
- "searchValue": "구강건강관리",
"value": "맑은 고딕",
"points": 1,
- "category": "FontName",
+ "category": "Rectangle.FontName",
"item": "문구 (구강건강관리)/⑧ 글씨체 (맑은 고딕)"
},
"12": {
"path": "//CHARSHAPE[@Id=//RECTANGLE//TEXT[./CHAR[text()='{searchValue}']]/@CharShape]/@Height",
- "searchValue": "구강건강관리",
"value": "2300",
"points": 1,
- "category": "OneAnswer",
+ "category": "Rectangle.OneAnswer",
"item": "문구 (구강건강관리)/⑨ 글씨크기 (2300)",
"desc": "1pt당 100"
},
"13": {
"path": "//PARASHAPE[@Id=//RECTANGLE//P[.//CHAR[text()='{searchValue}']]/@ParaShape]/@Align",
- "searchValue": "구강건강관리",
"value": "Center",
"points": 1,
- "category": "OneAnswer",
+ "category": "Rectangle.OneAnswer",
"item": "문구 (구강건강관리)/⑩ 정렬 (가운데 정렬)"
},
"14": {
diff --git a/diwScoring2.py b/diwScoring2.py
index da7ff8d..a01c983 100644
--- a/diwScoring2.py
+++ b/diwScoring2.py
@@ -3,7 +3,7 @@ import difflib
import json
from pathlib import Path
import os
-from lxml import etree as ET
+from lxml import etree
import re
from difflib import SequenceMatcher
import pandas as pd
@@ -133,32 +133,6 @@ class XMLScorer:
# 하나의 XML 파일 채점
def _score_xml_file(self, xml_file, chart_xml):
- # def parse_pages_by_bookmark(root):
- # """
- # P/TEXT/BOOKMARK 구조를 가진 XML에서 페이지 구간별 요소를 파싱하여 반환
- # """
- # pages = {}
- # all_p_tags = root.xpath('//P')
-
- # current_page = None
- # page_start_index = None
-
- # for i, p in enumerate(all_p_tags):
- # # BOOKMARK가 존재하는지 확인 (어디에 있든 탐색)
- # bookmarks = p.xpath('.//BOOKMARK')
- # for bm in bookmarks:
- # name = bm.get('Name')
- # if name and name.endswith('_start'):
- # current_page = name.replace('_start', '')
- # page_start_index = i
- # elif name and name.endswith('_end') and current_page is not None:
- # page_end_index = i
- # page_content = all_p_tags[page_start_index:page_end_index + 1]
- # pages[current_page] = page_content
- # current_page = None
- # page_start_index = None
-
- # return pages
def parse_pages_by_bookmark(root):
"""
BOOKMARK(Name="Page_X_start" ~ "Page_X_end") 사이의
요소들을
@@ -201,7 +175,7 @@ class XMLScorer:
return full_text
try:
- tree = ET.parse(xml_file)
+ tree = etree.parse(xml_file)
root = tree.getroot()
# XML문서 페이지 파싱 전처리
@@ -216,9 +190,9 @@ class XMLScorer:
# 차트 XML 파일이 없는 경우 0점 채점을 위헤 빈 XML 생성
if chart_xml is None:
- chart_tree = ET.fromstring('')
+ chart_tree = etree.fromstring('')
else:
- chart_tree = ET.fromstring(chart_xml)
+ chart_tree = etree.fromstring(chart_xml)
# 결과값을 Dictionary로 저장
# 하나의 xml파일 = 수험생 한명의 답안지
@@ -275,6 +249,9 @@ class XMLScorer:
}
try:
+ # 머릿말과 관련된 문항에서 1페이지에 머릿말이 없는 경우의 처리
+ # [1-25, 26, 27] 문항 'DIAT' 머릿말 채점시 1페이지에 머릿말이 없으면
+ # 채점하지 않고 0점 처리
if "Header" in (category or ""):
def has_elements(ptags, xpath):
for p in ptags:
@@ -285,9 +262,27 @@ class XMLScorer:
page1_ptags = pages.get('Page_1', [])
header_xpath = ".//HEADER//P"
- has_page1_element = has_elements(page1_ptags, header_xpath)
+ has_page1_header = has_elements(page1_ptags, header_xpath)
- if not has_page1_element:
+ if not has_page1_header:
+ user_answer = None
+ self.evaluate_answer(scoring, user_answer, right_answer, points, method="equal")
+ continue
+
+ has_page2_rectangle = False
+ if "Rectangle" in (category or ""):
+ def has_elements(ptags, xpath):
+ for p in ptags:
+ element_list = p.xpath(xpath) if xpath else []
+ if element_list:
+ return True
+ return False
+
+ page2_ptags = pages.get('Page_2', [])
+ rectangle_xpath = ".//RECTANGLE"
+ has_page2_rectangle = has_elements(page2_ptags, rectangle_xpath)
+
+ if not has_page2_rectangle:
user_answer = None
self.evaluate_answer(scoring, user_answer, right_answer, points, method="equal")
continue
@@ -458,13 +453,48 @@ class XMLScorer:
self.evaluate_answer(scoring, user_answer, right_answer, points)
if scoring['points'] > 0:
break
-
+
+ # 글상자 정렬 [2-10] 문항
+ # 2페이지의 글상자의 ParaShape ID를 동적으로 찾아서 채점
+ elif "TextBoxAlign" in (category or ""):
+ if has_page2_rectangle:
+ # 2페이지 내에서만 검색
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ rectangle_parashape_id = search_root.xpath(".//RECTANGLE/ancestor::P[last()]/@ParaShape")
+ else:
+ # 전체 root에서 검색
+ rectangle_parashape_id = root.xpath(".//RECTANGLE/ancestor::P[last()]/@ParaShape")
+
+ # ParaShape ID가 있는 경우에만 xpath 치환 & 실행
+ if rectangle_parashape_id:
+ xpath = xpath.replace('{rectangle_parashape_id}', rectangle_parashape_id[0])
+ items = root.xpath(xpath)
+ else:
+ # RECTANGLE이 없으면 items는 빈 리스트
+ items = [None]
+
+ for item in items:
+ user_answer = item
+ self.evaluate_answer(scoring, user_answer, right_answer, points)
+ if scoring['points'] > 0:
+ break
+
# 정답이 하나인 경우
# elif (category or "") in ["OneAnswer", "ChartOneAnswer"]:
- elif "OneAnswer" in (category or ""):
- items = root.xpath(xpath) if xpath else []
- items2 = root.xpath(xpath2) if xpath2 else []
+ elif "OneAnswer" in (category or ""):
+ if has_page2_rectangle:
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ items = search_root.xpath(xpath) if xpath else []
+ items2 = search_root.xpath(xpath2) if xpath2 else []
+
+ else:
+ items = root.xpath(xpath) if xpath else []
+ items2 = root.xpath(xpath2) if xpath2 else []
# 차트 XML에서 정답을 찾는 경우
# 차트 종류가
@@ -496,8 +526,15 @@ class XMLScorer:
break
# [2-6] 테두리 이중실선 1.00mm
- elif (category or "") == "LineShape":
- line_shapes = root.xpath(xpath) if xpath else []
+ elif "LineShape" in (category or ""):
+ if has_page2_rectangle:
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ line_shapes = search_root.xpath(xpath) if xpath else []
+
+ else:
+ line_shapes = root.xpath(xpath) if xpath else []
user_answer = {
'Style': None,
@@ -516,8 +553,16 @@ class XMLScorer:
break
# 사용자 입력값이 mm단위인 경우
- elif (category or "") == "mmSize":
- items = root.xpath(xpath)
+ # elif (category or "") == "mmSize":
+ elif "mmSize" in (category or ""):
+ if has_page2_rectangle:
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ items = search_root.xpath(xpath)
+
+ else:
+ items = root.xpath(xpath)
# 오차범위 설정
# 한글 프로그램 내부에서 드물게 0mm이지만 1pt로 저장되는 경우가 있음
#
@@ -567,10 +612,18 @@ class XMLScorer:
self.evaluate_answer(scoring, user_answer, right_answer, points)
# 채점기준표 파일에 작성된 rgb값을 그대로 읽어와 HML파일 요소의 int형 rgb값과 비교
- elif (category or "") == "Color":
- items = root.xpath(xpath) if xpath else []
- items2 = root.xpath(xpath2) if xpath2 else []
-
+ elif "Color" in (category or ""):
+ if has_page2_rectangle:
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ items = search_root.xpath(xpath) if xpath else []
+ items2 = search_root.xpath(xpath2) if xpath2 else []
+
+ else:
+ items = root.xpath(xpath) if xpath else []
+ items2 = root.xpath(xpath2) if xpath2 else []
+
rgb_text = right_answer
# 정규식을 이용해 숫자만 리스트로 추출
@@ -605,8 +658,15 @@ class XMLScorer:
# 폰트명
elif "FontName" in (category or ""):
- charshape_list = root.xpath(xpath)
-
+ if has_page2_rectangle:
+ search_root = etree.Element("Page_2")
+ for p in page2_ptags:
+ search_root.append(p)
+ charshape_list = search_root.xpath(xpath)
+
+ else:
+ charshape_list = root.xpath(xpath)
+
# 문자속성이 없는 경우
if not charshape_list:
user_answer = ""
@@ -978,7 +1038,7 @@ class XMLScorer:
onePersonResult['total_score'] = self.total_score
return onePersonResult
- except ET.ParseError as e:
+ except etree.ParseError as e:
return {
'filename': os.path.basename(xml_file),
'error': f"XML 파싱 오류: {str(e)}",
@@ -986,7 +1046,7 @@ class XMLScorer:
}
def binary_to_chartxml(self, xml_path):
- tree = ET.parse(xml_path)
+ tree = etree.parse(xml_path)
root = tree.getroot()
binary_data = root.xpath('//BINDATA[@Id=//BINITEM[@Format="OLE"]/@BinData]/text()')
@@ -1038,8 +1098,8 @@ class XMLScorer:
# 2. 공백제거, 특정 형식 제거
# 3. 리스트를 문자열로 변환
- user_answer_root = ET.parse(user_answer_file).getroot()
- correct_answer_root = ET.parse(correct_answer_file).getroot()
+ user_answer_root = etree.parse(user_answer_file).getroot()
+ correct_answer_root = etree.parse(correct_answer_file).getroot()
# xpath로 바이너리 부분추출
user_input_text = user_answer_root.xpath('//CHAR//text()[not(ancestor::HEADER) and not(ancestor::TABLE)]')
@@ -1054,7 +1114,7 @@ class XMLScorer:
# 차트 XML에서 차트제목 추출
if chart_xml is not None:
- chart_xml_tree = ET.fromstring(chart_xml)
+ chart_xml_tree = etree.fromstring(chart_xml)
ns = {'c': 'http://schemas.openxmlformats.org/drawingml/2006/chart',
'a': 'http://schemas.openxmlformats.org/drawingml/2006/main'}
xpath_expr = '/c:chartSpace/c:chart/c:title/c:tx/c:rich/a:p/a:r/a:t'
@@ -1325,12 +1385,12 @@ def main():
exam_types = [
# 'A',
# 'B',
- # 'C',
- 'D',
+ 'C',
+ # 'D',
]
- test_mode = False
- # test_mode = True #/TEST 폴더 채점시
+ # test_mode = False
+ test_mode = True #/TEST 폴더 채점시
output_excel_paths = []
for exam_type in exam_types:
diff --git a/zzz.xbook b/zzz.xbook
index 0b62787..ccd3958 100644
--- a/zzz.xbook
+++ b/zzz.xbook
@@ -1 +1 @@
-[{"kind":2,"language":"xpath","value":"//a:t[text()='클라우드 보안투자']/ancestor::a:r//a:ea/@typeface"},{"kind":2,"language":"xpath","value":"boolean(//FONTFACE[@Lang='Hangul']/FONT[@Id=//CHARSHAPE/FONTID/@Hangul]/@Name='바탕' and //CHARSHAPE/@Height='1000' and //PARASHAPE/PARAMARGIN/@LineSpacing='160' and //PARASHAPE/@Align='Justify')"},{"kind":2,"language":"xpath","value":"//FONTFACE[@Lang='Hangul']/FONT[@Id=//CHARSHAPE/FONTID/@Hangul]/@Name='바탕'"},{"kind":2,"language":"xpath","value":"//RECTANGLE//CHAR[text()='구강건강관거리']/ancestor::RECTANGLE/SHAPEOBJECT/SIZE/@Width"},{"kind":2,"language":"xpath","value":"//CHAR[contains(text(),'예방')][contains(text(),'豫防')]"},{"kind":2,"language":"xpath","value":"//TEXT[CHAR[text()='DIAT']]"},{"kind":2,"language":"xpath","value":"//HEADER//P"},{"kind":2,"language":"xpath","value":"//P[.//FIELDBEGIN[@Type='Hyperlink'] and .//CHAR[contains(., 'http')]]"},{"kind":2,"language":"xpath","value":"//PICTURE[./IMAGE[@BinItem=//BINITEM[@Format='JPG']/@BinData]]/SHAPEOBJECT/POSITION[not(@TreatAsChar='true')]/@HorzOffset"},{"kind":2,"language":"xpath","value":"//CHAR[contains(string(.), '※')]/descendant-or-self::text()"},{"kind":2,"language":"xpath","value":"//P[@ParaShape=\"17\"]/TEXT[@CharShape='7']//CHAR[string(.)]"},{"kind":2,"language":"xpath","value":"//CHAR[contains(string(.), '기타')]/text()"}]
\ No newline at end of file
+[{"kind":2,"language":"xpath","value":"//a:t[text()='클라우드 보안투자']/ancestor::a:r//a:ea/@typeface"},{"kind":2,"language":"xpath","value":"boolean(//FONTFACE[@Lang='Hangul']/FONT[@Id=//CHARSHAPE/FONTID/@Hangul]/@Name='바탕' and //CHARSHAPE/@Height='1000' and //PARASHAPE/PARAMARGIN/@LineSpacing='160' and //PARASHAPE/@Align='Justify')"},{"kind":2,"language":"xpath","value":"//FONTFACE[@Lang='Hangul']/FONT[@Id=//CHARSHAPE/FONTID/@Hangul]/@Name='바탕'"},{"kind":2,"language":"xpath","value":"//PARASHAPE[@Id=//RECTANGLE/ancestor::P[last()]/@ParaShape]/@Align"},{"kind":2,"language":"xpath","value":"//RECTANGLE//LINESHAPE"},{"kind":2,"language":"xpath","value":"//RECTANGLE/SHAPEOBJECT/SIZE/@Width"},{"kind":2,"language":"xpath","value":"//HEADER//P"},{"kind":2,"language":"xpath","value":"//P[.//FIELDBEGIN[@Type='Hyperlink'] and .//CHAR[contains(., 'http')]]"},{"kind":2,"language":"xpath","value":"//PICTURE[./IMAGE[@BinItem=//BINITEM[@Format='JPG']/@BinData]]/SHAPEOBJECT/POSITION[not(@TreatAsChar='true')]/@HorzOffset"},{"kind":2,"language":"xpath","value":"//CHAR[contains(string(.), '※')]/descendant-or-self::text()"},{"kind":2,"language":"xpath","value":"//P[@ParaShape=\"17\"]/TEXT[@CharShape='7']//CHAR[string(.)]"},{"kind":2,"language":"xpath","value":"//CHAR[contains(string(.), '기타')]/text()"}]
\ No newline at end of file