class TextItem(DocItem):
"""Prints some form of text"""
def __init__(self, text = '', style = 'body'):
DocItem.__init__(self)
self.text = text
self.style = style
def writePDF(self, pdf = None):
"""Write itself to a FPDF object.
Args:
pdf (FPDF): the FPDF object to write to.
"""
return self.getText()
def resizePDF(self, pdf, x = 0, y = 0):
"""Resize internal Rect according to current settings of pdf"""
width = pdf.get_string_width( self.getText() )
height = pdf.font_size_pt / pdf.k
self.rect = Rect( x, y, x + width, y + height )
def cellPDF(self, pdf, r = None):
if r:
x_shift = r.x0()
y_shift = r.y0()
else:
x_shift = 0.0
y_shift = 0.0
pdf.set_y( self.rect.y0() - y_shift )
pdf.set_x( self.rect.x0() - x_shift )
pdf.cell( self.rect.width(), self.rect.height(), self.getText() )
def refit(self):
"""Doesn't need to do anything as cellPDF uses self.rect to output the content"""
pass
class MathFrac(MultiItem):
"""Container for inline maths"""
def __init__(self, num = None, denom = None):
MultiItem.__init__(self)
self.style = 'math-var', 1
if num:
if not denom:
denom = MathNumber('1')
self.appendItem(num)
self.appendItem(denom)
def resizePDF(self, pdf, x = 0, y = 0):
if len(self.items) < 2 or not self.items[0] or not self.items[1]:
raise Exception('MathFrac must have two items.')
self.rect = Rect(x,y,x,y)
dx = pdf.get_string_width(' ') * self.style[1]
self.margins.set(dx, 0.0)
setFontPDF(pdf, self.style, self.styles)
lineHeight = pdf.font_size_pt / pdf.k
numerator = self.items[0]
if hasattr(numerator,'style'):
setFontPDF(pdf, numerator.style, self.styles)
numerator.resizePDF(pdf,x + dx, y - lineHeight * 0.5)
denominator = self.items[1]
if hasattr(denominator,'style'):
setFontPDF(pdf, denominator.style, self.styles)
denominator.resizePDF(pdf, x + dx, numerator.rect.y1())
if numerator.rect.width() > denominator.rect.width():
denominator.rect.alignXCenter(numerator.rect)
else:
numerator.rect.alignXCenter(denominator.rect)
self.rect.unite(numerator.rect)
self.rect.unite(denominator.rect)
self.rect.adjust(rect.Point(0,0),rect.Point(dx,0))
def cellPDF(self, pdf, r = None):
MultiItem.cellPDF(self, pdf, r)
y = self.items[0].rect.y1()
pdf.set_line_width(0.2)
if r:
x_shift = r.x0()
y_shift = r.y0()
else:
x_shift = 0.0
y_shift = 0.0
pdf.line(self.rect.x0() - x_shift, y - y_shift, self.rect.x1() - x_shift, y - y_shift)
def process_file(path, images):
annotation = ET.Element("annotation")
tree = ElementTree(annotation)
filename = ET.SubElement(annotation, "filename")
name = os.path.splitext(os.path.split(path)[1])[0]
filename.text = name + ".jpeg"
size = ET.SubElement(annotation, "size")
width = ET.SubElement(size, "width")
width.text = str(images[name].get("width"))
height = ET.SubElement(size, "height")
height.text = str(images[name].get("height"))
depth = ET.SubElement(size, "depth")
depth.text = str(3)
segmented = ET.SubElement(annotation, "segmented")
segmented.text = str(0)
marmot_tree = ET.parse(path)
marmot_root = marmot_tree.getroot()
tables = marmot_root.findall("*/Composites/*[@Label='TableBody']")
if not tables:
return None
for table in tables:
obj = ET.SubElement(annotation, "object")
obj_name = ET.SubElement(obj, "name")
obj_name.text = "table"
bndbox = ET.SubElement(obj, "bndbox")
hexs = table.get("BBox").split(" ")
bbox_array = [hex_to_double(x) for x in hexs]
re = Rect(bbox_array[0], bbox_array[3], bbox_array[2], bbox_array[1])
xmin = ET.SubElement(bndbox, "xmin")
xmin.text = str(re.x0())
ymin = ET.SubElement(bndbox, "ymin")
fymin = images[name].get("height") - re.y1()
ymin.text = str(fymin)
xmax = ET.SubElement(bndbox, "xmax")
xmax.text = str(re.x1())
fymax = images[name].get("height") - re.y0()
ymax = ET.SubElement(bndbox, "ymax")
ymax.text = str(fymax)
return tree
