def get_plurals(img, cols):
lines = utils._get_lines(img, cols, 5, 6, "pairplurals",
psm_mode=6, apply_morphology=True)
text = ""
for line in lines:
text += line.text.replace("\n", "").replace(" ", "")
return text
def get_reference(img, cols):
lines = utils._get_lines(img, cols, 0, 1, "pairrefs",
psm_mode=6, apply_morphology=True)
text = ''
for line in lines:
text += line.text.upper().replace(" ", "").replace("\n", "")
return text
def get_default_operator(img, cols):
lines = utils._get_lines(img, cols, 4, 5, "pairdefop",
psm_mode=6, apply_morphology=True)
text = ""
for line in lines:
text += line.text.replace("\n", "").replace(" ", "")
return text
def get_hits(img, cols):
lines = utils._get_lines(img, cols, 1, 2, "digits",
psm_mode=6, apply_morphology=True)
text = ''
for line in lines:
text += line.text.replace("\n", " ").replace(" ", "")
return text
def get_query(img, cols):
lines = utils._get_lines(img, cols, 2, 3, psm_mode=3)
text = ''
for line in lines:
text += line.text.replace("\n", " ")
# Correct a common reading error.
text = text.replace("cc|s", "ccls")
return text
def get_date_time(img, cols):
lines = utils._get_lines(img, cols, -2, -1, "pairdate")
text = ''
for line in lines:
text += line.text.replace("\n", " ").replace(" ", "")
# Extract the date string.
m = re.match(r'\d{4}/\d{2}/\d{2}', text)
try:
date = m.group(0)
except:
date = ''
# Extract the time string.
timepos = text.find(":")
time = ''
if timepos > 0:
try:
time = text[timepos - 2:timepos + 3]
except:
time = ''
return date, time
def get_databases(img, cols):
lines = utils._get_lines(img, cols, 3, 4, "pairdbs")
text = ''
for line in lines:
text += line.text.replace("\n", "").replace(" ", "")
return text
