def test_page_has_word(self):
page = Page('5')
expected_line = Line('bob has a train', 1, StubSpellChecker(()))
page.append(expected_line)
# assert simple match
line, line_info = page.find_word('has')
self.assertEquals(expected_line, line)
self.assertIsNone(line_info)
next_line = Line('he has no friends', 1, StubSpellChecker(()))
page.append(next_line)
# assert picks first line
line, line_info = page.find_word('has')
self.assertEquals(expected_line, line)
# assert can find in second line
line, line_info = page.find_word('friends')
self.assertEquals(next_line, line)
line_info_1 = LineInfo(1)
line_info_2 = LineInfo(2)
line_info_3 = LineInfo(3)
page.line_infos = [line_info_1, line_info_2, line_info_3]
# assert finds correct line info no header
line, line_info = page.find_word('friends')
self.assertEquals(line_info_2, line_info)
page.has_header = True
# assert finds correct line info with header
line, line_info = page.find_word('friends')
self.assertEquals(line_info_3, line_info)
# assert raises if not found
self.assertRaises(NoWordException, page.find_word, 'notinstrings')
def set_page_nbr(self, page_nbr):
self.page_nbr = page_nbr
self.frame.image = Image.open('images/pages/{}.pbm'.format(page_nbr)).convert('RGB')
self.frame.page_nbr = page_nbr
self.frame.line_infos = []
with open('working/page_info/{}.csv'.format(page_nbr), 'r') as f:
reader = csv.reader(f)
for row in reader:
line_info = LineInfo(int(row[3]))
line_info.height = int(row[1])
line_info.left_margin = int(row[2])
line_info.width = int(row[4])
self.frame.line_infos.append(line_info)
self.frame.show_panel()
def load(self, raw_file_dir):
""" Creates lines out of all the lines in a directory."""
line_infos = defaultdict(lambda:[])
if os.path.exists('working/page_info'):
for filename in os.listdir('working/page_info'):
if filename.endswith('.csv'):
print 'parsing', filename
with open('working/page_info/{}'.format(filename), 'rb') as f:
reader = csv.reader(f)
for row in reader:
line_info = LineInfo(int(row[3]))
line_info.height = int(row[1])
line_info.left_margin = int(row[2])
line_info.width = int(row[4])
line_infos[row[0]].append(line_info)
headers = set()
if os.path.exists('working/headers.txt'):
with open('working/headers.txt', 'rb') as f:
for l in f:
try:
headers.add(l.split('|')[0])
except:
pass
for fn in sorted(os.listdir(raw_file_dir), key=lambda x: int(os.path.splitext(x)[0])):
basename, ext = os.path.splitext(fn)
if int(basename) < self.start_page or ext != '.txt':
continue
if self.end_page > 0 and int(basename) > self.end_page:
break
with codecs.open('{}/{}'.format(raw_file_dir, fn), mode='r', encoding='utf-8') as f:
if self.verbose:
print 'Loading page {:>3}'.format(basename)
self.pages[basename] = Page(basename, fn in headers, line_infos[basename])
idx = 1
for l in f:
line = Line(l.strip(), idx, self.spell_checker)
self.pages[basename].append(line)
idx += 1
self.average_length = self.calculate_average_length()
self.average_lines_per_page = sum([len(lines) for lines in self.pages.values()])/len(self.pages)
self.page_numbers = sorted(self.pages.keys(), key=lambda x: int(x))
def OnAddLine(self, event):
new_line_y = self.current_height + (IMG_HEIGHT / 2)
line_info = LineInfo(new_line_y)
inserted = False
for idx, li in enumerate(self.line_infos):
if li.y > new_line_y:
self.line_infos.insert(idx, line_info)
inserted = True
break
if not inserted:
self.line_infos.append(line_info)
self.set_image()
