def test_possible_pairs(self):
schedule = Schedule()
with open("res/test_pair_2.json", "r", encoding="utf-8") as file:
json_pair_2 = json.load(file)
with open("res/test_pair_3.json", "r", encoding="utf-8") as file:
json_pair_3 = json.load(file)
schedule.add_pair(StudentPair.from_json(json_pair_2))
schedule.add_pair(StudentPair.from_json(json_pair_3))
def test_impossible_pairs(self):
schedule = Schedule()
with open("res/test_pair_1.json", "r", encoding="utf-8") as file:
json_pair_1 = json.load(file)
with open("res/test_pair_2.json", "r", encoding="utf-8") as file:
json_pair_2 = json.load(file)
schedule.add_pair(StudentPair.from_json(json_pair_1))
self.assertRaises(AlongTwoPairsException, schedule.add_pair,
StudentPair.from_json(json_pair_2))
def test_impossible_intersect(self):
schedule = Schedule()
with open("res/test_pair_4.json", "r", encoding="utf-8") as file:
json_pair_4 = json.load(file)
with open("res/test_pair_5.json", "r", encoding="utf-8") as file:
json_pair_5 = json.load(file)
with open("res/test_pair_6.json", "r", encoding="utf-8") as file:
json_pair_6 = json.load(file)
schedule.add_pair(StudentPair.from_json(json_pair_4))
schedule.add_pair(StudentPair.from_json(json_pair_5))
self.assertRaises(AlongTwoPairsException, schedule.add_pair,
StudentPair.from_json(json_pair_6))
def import_from_pdf(process_id, manager: ImportManager) -> None:
tesseract = TesseractWrapper(tesseract_path=manager.tesseract_path)
while not manager.queue.empty():
try:
file_path = manager.queue.get(True, 1)
file = QFileInfo(file_path)
# convert from pdf to PIL image
img_pdf = pdf2image.convert_from_path(
file.absoluteFilePath(),
dpi=manager.dpi,
poppler_path=manager.poppler_path)
img_pdf = img_pdf[0].convert('RGB')
# convert to NumPy array
img = np.array(img_pdf)
img = img[:, :, ::-1].copy()
# set mask
grey = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
thresh = cv.threshold(grey, 127, 255, 0)[1]
if manager.flags["stop"]:
break
# found contours
contours = cv.findContours(thresh, cv.RETR_TREE,
cv.CHAIN_APPROX_SIMPLE)[0]
height, width = img.shape[:2]
# max and min area of rect
max_area = height * width / 20
min_area = max_area / 40
cells = []
time_cells = dict()
contours_number = len(contours)
number = 0
for k, contour in enumerate(contours, 1):
rect = cv.minAreaRect(contour)
area = int(rect[1][0] * rect[1][1])
if min_area < area < max_area:
if manager.flags["stop"]:
break
x, y, w, h = cv.boundingRect(contour)
crop_img = img[int(y):int(y + h), int(x):int(x + w)]
txt = tesseract.to_string(crop_img)
found = False
for i in range(8):
if TimePair.time_starts(
)[i] in txt and TimePair.time_ends()[i]:
time_cells[i] = (x, x + w)
found = True
break
if not found:
cells.append((x, x + w, " ".join(txt.split())))
# draw debug rect with number
if manager.debug_image:
box = cv.boxPoints(rect)
box = np.int0(box)
blue_color = (255, 0, 0)
center = (int(rect[0][0]), int(rect[0][1]))
cv.drawContours(img, [box], 0, blue_color, 2)
cv.putText(img, str(number),
(center[0] - 100, center[1] - 40),
cv.FONT_HERSHEY_SIMPLEX, 3, blue_color, 12)
number += 1
process = int(k / contours_number * 70)
manager.progress_value_list[process_id] = process
manager.progress_text_list[process_id] = "{} {}%".format(
file.baseName(), process)
if manager.debug_image:
cv.imwrite(file_path[0:-4] + "-debug.jpg", img)
if manager.flags["stop"]:
break
schedule = Schedule()
cells_number = len(cells)
for k, cell in enumerate(cells):
if manager.flags["stop"]:
break
start_x, end_x, text = cell
first_start_time, first_end_time = time_cells[0]
if not abs(end_x - first_start_time) < abs(start_x -
first_start_time):
text = "\n".join(re.findall(r".*?\]", text))
while True:
try:
pairs = parse_pair(manager, text)
break
except InvalidDatePair as ex:
text = confuse_loop(
process_id, manager,
ConfuseSituationException(
file.absoluteFilePath(),
text,
confuse=str(ex)))
except ConfuseSituationException as ex:
ex.filename = file.absoluteFilePath(
)[0:-4] + "-debug.jpg"
ex.cell = k
ex.context = text
if ex.maybe_answer == "":
ex.maybe_answer = text
text = confuse_loop(process_id, manager, ex)
if len(pairs) != 0:
diff_start = abs(start_x - first_start_time)
diff_end = abs(end_x - first_end_time)
start, end = 0, 0
for number, (start_time,
end_time) in time_cells.items():
diff = abs(start_x - start_time)
if diff < diff_start:
diff_start = diff
start = number
diff = abs(end_x - end_time)
if diff < diff_end:
diff_end = diff
end = number
for pair in pairs:
pair["time"].set_time(
TimePair.time_starts()[start],
TimePair.time_ends()[end])
schedule.add_pair(pair)
process = int(70 + k / cells_number * 30)
manager.progress_value_list[process_id] = process
manager.progress_text_list[process_id] = "{} {}%".format(
file.baseName(), process)
schedule.save(file.absoluteFilePath()[0:-4] + ".json")
print(file.absoluteFilePath()[0:-4] + ".json")
if manager.flags["stop"]:
break
if manager.weekly:
export_weeks_to_pdf(
schedule, file.baseName(), True,
file.absoluteFilePath()[0:-4] + "-weekly.pdf",
manager.font_name, manager.font_path, manager.encoding,
manager.start, manager.end, manager.color_a,
manager.color_b)
if manager.full:
export_full_to_pdf(schedule, file.baseName(),
file.absoluteFilePath()[0:-4] + "-full.pdf",
manager.font_name, manager.font_path,
manager.encoding)
except Exception as ex:
print("Exception, process:", process_id, "is:", ex)
traceback.print_exc()
manager.progress_value_list[process_id] = 100
manager.progress_text_list[process_id] = "Work complete"