def test_horizontally_spanning(self):
log.info("Testing Horizontally Spanned")
g = Grid()
a = Cell(startRow=1, startCol=0, rowSpan=3, colSpan=2, content="a")
b = Cell(startRow=3, startCol=2, rowSpan=1, colSpan=1, content="b")
c = Cell(startRow=1, startCol=2, rowSpan=2, colSpan=1, content="c")
d = Cell(startRow=1, startCol=3, rowSpan=3, colSpan=1, content="d")
g.addCell(a)
g.addCell(b)
g.addCell(c)
g.addCell(d)
print(g)
self.assertEqual(features.horizontallySpanned(a, c, None), 1)
self.assertEqual(features.horizontallySpanned(a, b, None), 1)
# self.assertEqual(features.horizontallySpanned(c,d, None), -1)
self.assertEqual(features.horizontallySpanned(b, c, None), 0)
def test_relative_horizontal_position(self):
Cell.switchToContentRepr()
log.info("Testing relative horizontal position")
g = Grid()
a = Cell(startRow=1, startCol=1, rowSpan=3, colSpan=3, content="a")
b = Cell(startRow=4, startCol=1, rowSpan=1, colSpan=3, content="b")
c = Cell(startRow=0, startCol=0, rowSpan=1, colSpan=2, content="c")
d = Cell(startRow=4, startCol=4, rowSpan=1, colSpan=1, content="d")
e = Cell(startRow=1, startCol=0, rowSpan=2, colSpan=1, content="e")
g.addCell(e)
g.addCell(a)
g.addCell(b)
g.addCell(c)
g.addCell(d)
print(g)
self.assertEqual(features.relativeHorizontalPosition(c, a, None), 0)
self.assertEqual(features.relativeHorizontalPosition(e, a, None), 1)
self.assertEqual(features.relativeHorizontalPosition(a, b, None), 0)
self.assertEqual(features.relativeHorizontalPosition(a, d, None), 1)
def test_below_empty_row(self):
Cell.switchToContentRepr()
log.info("Testing below empty row")
g = Grid()
a = Cell(startRow=1, startCol=1, rowSpan=3, colSpan=5, content="a")
b = Cell(startRow=4, startCol=1, rowSpan=1, colSpan=3, content="b")
c = Cell(startRow=0, startCol=0, rowSpan=1, colSpan=2, content="c")
d = Cell(startRow=4, startCol=4, rowSpan=1, colSpan=1, content="d")
e = Cell(startRow=1, startCol=0, rowSpan=2, colSpan=1, content="e")
g.addCell(e)
g.addCell(a)
g.addCell(b)
g.addCell(c)
g.addCell(d)
Cell.switchToContentRepr()
print(g)
self.assertEqual(features.belowEmptyRow(e, c, g), -1)
self.assertEqual(features.belowEmptyRow(c, e, g), 1)
self.assertEqual(features.belowEmptyRow(a, b, g), 0)
self.assertEqual(features.belowEmptyRow(c, b, g), 1)
def test_vertically_spanning(self):
Cell.switchToContentRepr()
log.info("Testing Vertically Spanned")
g = Grid()
a = Cell(startRow=1, startCol=1, rowSpan=3, colSpan=5, content="a")
b = Cell(startRow=4, startCol=1, rowSpan=1, colSpan=3, content="b")
c = Cell(startRow=0, startCol=0, rowSpan=1, colSpan=2, content="c")
d = Cell(startRow=4, startCol=4, rowSpan=1, colSpan=1, content="d")
g.addCell(a)
g.addCell(b)
g.addCell(c)
g.addCell(d)
print(g)
self.assertEqual(features.verticallySpanned(a, b, None), 1)
self.assertEqual(features.verticallySpanned(b, a, None), -1)
self.assertEqual(features.verticallySpanned(c, a, None), 0)
self.assertEqual(features.verticallySpanned(b, c, None), 0)
self.assertEqual(features.verticallySpanned(a, c, None), 0)
self.assertEqual(features.verticallySpanned(a, d, None), 1)
self.assertEqual(features.verticallySpanned(d, a, None), -1)
def test_spanned_by_same_cell(self):
Cell.switchToContentRepr()
log.info("Testing spanned by same cell")
g = Grid()
a = Cell(startRow=1, startCol=1, rowSpan=3, colSpan=5, content="a")
b = Cell(startRow=4, startCol=1, rowSpan=1, colSpan=3, content="b")
c = Cell(startRow=2, startCol=6, rowSpan=1, colSpan=1, content="c")
d = Cell(startRow=4, startCol=4, rowSpan=1, colSpan=1, content="d")
e = Cell(startRow=1, startCol=6, rowSpan=2, colSpan=1, content="e")
g.addCell(e)
g.addCell(a)
g.addCell(b)
g.addCell(c)
g.addCell(d)
Cell.switchToContentRepr()
print(g)
self.assertEqual(features.spannedBySameCell(b, d, g), 1)
self.assertEqual(features.spannedBySameCell(c, e, g), 1)
self.assertEqual(features.belowEmptyRow(a, b, g), 0)
self.assertEqual(features.belowEmptyRow(c, b, g), 0)
def parse(fileIndex, dataset = 'us', cell_char_limit = 5, readFunction = True, readStructure = True, display = False):
grid.log.disabled = True
fileName = FILE_DIRECTORIES[dataset]['xml'] + dataset + '-' + fileIndex + "-str.xml"
grid.log.info("Parsing " + fileName)
grid.Cell.CELL_CHAR_LIMIT = cell_char_limit
xmldoc = minidom.parse(fileName)
tableList = xmldoc.getElementsByTagName("table")
id_grid_map = {}
for table in tableList:
xmlGrid = grid.Grid()
region = table.getElementsByTagName('region')[0]
cellList = table.getElementsByTagName('cell')
for i, cellElem in enumerate(cellList):
startRow = int(cellElem.attributes.get('start-row', 0).value)
startCol = int(cellElem.attributes.get('start-col', 0).value)
rowSpan = cellElem.attributes.get('end-row', -1)
if rowSpan == -1:
rowSpan = 1
else:
rowSpan = int(rowSpan.value) - startRow + 1
colSpan = cellElem.attributes.get('end-col', -1)
if colSpan == -1:
colSpan = 1
else:
colSpan = int(colSpan.value) - startCol + 1
x1 = int(cellElem.getElementsByTagName("bounding-box")[0].attributes.get('x1').value)
y1 = int(cellElem.getElementsByTagName("bounding-box")[0].attributes.get('y1').value)
x2 = int(cellElem.getElementsByTagName("bounding-box")[0].attributes.get('x2').value)
y2 = int(cellElem.getElementsByTagName("bounding-box")[0].attributes.get('y2').value)
content = cellElem.getElementsByTagName("content")[0].firstChild.data.replace("\n", " ")
function = grid.Cell.DATA_CELL
xmlGrid.addCell(grid.Cell(startRow, startCol, rowSpan, colSpan, x1, x2, y1, y2, content, function, i))
id_grid_map[int(table.attributes.get("id").value)] = xmlGrid
if display:
grid.Cell.switchToContentRepr()
print(xmlGrid)
grid.Cell.switchToIDRepr()
print(xmlGrid)
if (readFunction):
functionFileName = FILE_DIRECTORIES[dataset]['function'] + dataset + '-' + fileIndex + "_function.txt"
#read functional tagging
log.info("READING FUNCTIONAL TAGGING FROM" + functionFileName)
functionalFilePtr = open(functionFileName)
numGrids = int(functionalFilePtr.readline())
for i in range(1, numGrids+1):
functionalFilePtr.readline()
stubCorner = functionalFilePtr.readline().strip().split()[1:]
stubCornerRow, stubCornerCol = int(stubCorner[0]), int(stubCorner[1])
stubCell = id_grid_map[i].getCell(stubCornerRow, stubCornerCol)
stubCell.function = grid.Cell.STUB_CELL
id_grid_map[i].stubCell = stubCell
id_grid_map[i].stubRow = stubCornerRow
id_grid_map[i].stubCol = stubCornerCol
columnHeaders = functionalFilePtr.readline().strip().split()[1:]
for column in columnHeaders:
column = int(column)
for j in range(stubCornerRow + 1, id_grid_map[i].numRows):
cell = id_grid_map[i].getCell(j, column)
if cell != grid.Cell.EMPTY_CELL and cell != stubCell:
cell.function = grid.Cell.COL_HEADER_CELL
rowHeaders = functionalFilePtr.readline().strip().split()[1:]
for row in rowHeaders:
row = int(row)
for k in range(stubCornerCol + 1, id_grid_map[i].numCols):
cell = id_grid_map[i].getCell(row, k)
if cell != grid.Cell.EMPTY_CELL and cell != stubCell:
cell.function = grid.Cell.ROW_HEADER_CELL
functionalFilePtr.readline()
functionalFilePtr.close()
if (readStructure):
structureFileName = FILE_DIRECTORIES[dataset]['structure'] + dataset + '-' + fileIndex + "_structure.txt"
#read structural tagging
log.info("READING STRUCTURAL TAGGING FROM" + structureFileName)
structuralFilePtr = open(structureFileName)
numGrids = int(structuralFilePtr.readline())
for i in range(1, numGrids+1):
line = structuralFilePtr.readline().strip()
while (line[0] != "*"):
parentID, childrenIDList = line[0], line[1:]
parent = id_grid_map[i].getCellByCellID(int(parentID))
for childID in childrenIDList:
child = id_grid_map[i].getCellByCellID(int(childID))
parent.addChild(child)
child.addParent(parent)
line = structuralFilePtr.readline().strip().split()
structuralFilePtr.close()
return id_grid_map