def return_data(self, command=None):
"""Read until exit status is returned.
Returns:
data: List of right-stripped strings containing output
of the command.
Raises:
AtCommandError: If an error is returned by the modem.
"""
data = list()
while 1:
# Read in one line of input.
input_line = self.readline().rstrip()
# Check for errors and raise exception with specific error code.
errors.check_for_errors(input_line)
if input_line == 'OK':
return data
# Append only related data (starting with "command" contents).
if command:
if input_line.startswith(command):
prefix_length = len(command)+2
data.append(input_line[prefix_length:])
else:
# Append only non-empty data.
if input_line:
data.append(input_line)
def send_at(self, cmd, suffix, prefixed=True):
"""Send serial text to the modem.
Arguments:
self -- serial port to send to,
text -- text value to send,
prefixed -- boolean determining weather to strip the AT
command prefix from each output line.
Returns:
List of strings.
"""
self.write('AT%s%s\r' % (cmd, suffix))
#print "D: - Sending " + ('AT%s%s\r' % (cmd, suffix))
# Read in the echoed text.
# Check for errors and raise exception with specific error code.
input_line = self.readline()
errors.check_for_errors(input_line)
# Return the result.
if prefixed:
# If the text being sent is an AT command, only relevant context
# answer (starting with '+command:' value) will be returned by
#return_data(). Otherwise any string will be returned.
return self.return_data(cmd)
else:
return self.return_data()
def return_data(self, command=None):
"""Read until exit status is returned.
Returns:
data: List of right-stripped strings containing output
of the command.
Raises:
AtCommandError: If an error is returned by the modem.
"""
data = list()
while 1:
# Read in one line of input.
input_line = self.readline().rstrip()
# Check for errors and raise exception with specific error code.
errors.check_for_errors(input_line)
if input_line == 'OK':
return data
# Append only related data (starting with "command" contents).
if command:
if input_line.startswith(command):
prefix_length = len(command) + 2
data.append(input_line[prefix_length:])
else:
# Append only non-empty data.
if input_line:
data.append(input_line)
def send_at(self, cmd, suffix, prefixed=True):
"""Send serial text to the modem.
Arguments:
self -- serial port to send to,
text -- text value to send,
prefixed -- boolean determining weather to strip the AT
command prefix from each output line.
Returns:
List of strings.
"""
self.write('AT%s%s\r' % (cmd, suffix))
#print "D: - Sending " + ('AT%s%s\r' % (cmd, suffix))
# Read in the echoed text.
# Check for errors and raise exception with specific error code.
input_line = self.readline()
errors.check_for_errors(input_line)
# Return the result.
if prefixed:
# If the text being sent is an AT command, only relevant context
# answer (starting with '+command:' value) will be returned by
#return_data(). Otherwise any string will be returned.
return self.return_data(cmd)
else:
return self.return_data()
def call_tesseract(input_filename, output_filename, language, pagesegmode):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
current_dir = os.getcwd()
error_stream = StringIO.StringIO()
try:
os.chdir(_working_dir)
args = [tesseract_exe_name, input_filename, output_filename]
if len(language) > 0:
args.append("-l")
args.append(language)
if len(str(pagesegmode)) > 0:
args.append("-psm")
args.append(str(pagesegmode))
try:
proc = subprocess.Popen(args)
except (TypeError, AttributeError):
proc = subprocess.Popen(args, shell=True)
retcode = proc.wait()
if retcode!=0:
error_text = error_stream.getvalue()
errors.check_for_errors(error_stream_text = error_text)
finally: # Guarantee that we return to the original directory
error_stream.close()
os.chdir(current_dir)
def call_tesseract(input_filename, output_filename, language, pagesegmode):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
current_dir = os.getcwd()
error_stream = StringIO.StringIO()
try:
os.chdir(_working_dir)
args = [tesseract_exe_name, input_filename, output_filename]
if len(language) > 0:
args.append("-l")
args.append(language)
if len(str(pagesegmode)) > 0:
args.append("-psm")
args.append(str(pagesegmode))
try:
proc = subprocess.Popen(args)
except (TypeError, AttributeError):
proc = subprocess.Popen(args, shell=True)
retcode = proc.wait()
if retcode != 0:
error_text = error_stream.getvalue()
errors.check_for_errors(error_stream_text=error_text)
finally: # Guarantee that we return to the original directory
error_stream.close()
os.chdir(current_dir)
def call_tesseract(input_filename, output_filename): """Calls external tesseract.exe on input file (restrictions on types), outputting output_filename+'txt'""" args = [tesseract_exe_name, input_filename, output_filename] proc = subprocess.Popen(args) retcode = proc.wait() if retcode!=0: errors.check_for_errors()
def call_tesseract(input_filename, output_filename): """Calls external tesseract.exe on input file (restrictions on types), outputting output_filename+'txt'""" args = [tesseract_exe_name, input_filename, output_filename] proc = subprocess.Popen(args) retcode = proc.wait() if retcode!=0: errors.check_for_errors()
def call_tesseract(input_filename, output_filename): """Calls external tesseract.exe on input file (restrictions on types), outputting output_filename+'txt'""" os.environ["TESSDATA_PREFIX"] = os.path.abspath( os.path.dirname(__file__) )+"/" args = [tesseract_exe_name, input_filename, output_filename, "nobatch", "my" ] proc = subprocess.Popen(args) retcode = proc.wait() if retcode!=0: errors.check_for_errors()
def call_tesseract(input_filename, output_filename): """Calls external tesseract.exe on input file (restrictions on types), outputting output_filename+'txt'""" devnull = open(os.devnull, 'w') #modified args = [tesseract_exe_name, input_filename, output_filename] proc = subprocess.Popen(args, stderr=devnull) #modified retcode = proc.wait() if retcode!=0: errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
devnull = open(os.devnull, 'w') #modified
args = [tesseract_exe_name, input_filename, output_filename]
proc = subprocess.Popen(args, stderr=devnull) #modified
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
args.append("-psm") #added grzegorz pietrusza
args.append(str(6)) #added grzegorz pietrusza
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode!=0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
args.append("-psm") #added grzegorz pietrusza
args.append(str(6)) #added grzegorz pietrusza
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
path = os.path.join(os.getcwd(), 'pyspider\\pytesser', tesseract_exe_name)
args = [path, input_filename, output_filename]
print args
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, language=""):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
if len(language) > 0:
args.append('-l')
args.append(language)
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
if _platform_=='linux' or _platform_=='linux2':
args = ['/usr/bin/wine', tesseract_exe_name, input_filename, output_filename]
else:
args = [tesseract_exe_name, input_filename, output_filename]
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
dest = open(os.devnull, 'w') # Jabba
if DEBUG:
dest = STDOUT
args = [tesseract_exe_name, input_filename, output_filename]
proc = Popen(args, stdout=dest, stderr=dest) # Jabba
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, language=""):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
if len(language) > 0:
args.append('-l')
args.append(language)
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode!=0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
dest = open(os.devnull, 'w') # Jabba
if DEBUG:
dest = STDOUT
args = [tesseract_exe_name, input_filename, output_filename]
proc = Popen(args, stdout=dest, stderr=dest) # Jabba
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
os.environ["TESSDATA_PREFIX"] = os.path.abspath(
os.path.dirname(__file__)) + "/"
args = [
tesseract_exe_name, input_filename, output_filename, "nobatch", "my"
]
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
try:
proc = subprocess.Popen(args, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
except:
print "\n[!] Please install tesseract-ocr first!"
print "[!] Run 'sudo apt-get install tesseract-ocr'"
exit()
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
args = [tesseract_exe_name, input_filename, output_filename]
args.append("-l eng2+eng3+eng4+eng5+eng6+eng7+eng8")
args.append("-psm")
args.append(str(10))
args.append("letters")
proc = subprocess.Popen(args,
stderr=subprocess.PIPE,
stdout=subprocess.PIPE)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, bool_digits=False):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
# args = [tesseract_exe_name, input_filename, output_filename]
if bool_digits:
# args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l eng -psm 7 nobatch eng_digits" # price
args = tesseract_exe_name + " " + input_filename + " " + output_filename + " -l test_digits -psm 7 nobatch" # price
else:
args = tesseract_exe_name + " " + input_filename + " " + output_filename + " -l eng -psm 7 nobatch eng_characters" # English letters
# args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l test_eng -psm 7 nobatch" # English letters
# print args
proc = subprocess.Popen(args, shell=True)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, bool_digits=False):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
# args = [tesseract_exe_name, input_filename, output_filename]
if bool_digits:
# args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l eng -psm 7 nobatch eng_digits" # price
args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l test_digits -psm 7 nobatch" # price
else:
args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l eng -psm 7 nobatch eng_characters" # English letters
# args = tesseract_exe_name+" "+input_filename+" "+output_filename+" -l test_eng -psm 7 nobatch" # English letters
# print args
proc = subprocess.Popen(args, shell=True)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, charset=None):
"""Calls external tesseract on input file (restrictions on types),
outputting output_filename + '.txt'
Optionally, you can provide a character set to limit the OCR characters
the ones specified"""
devnull = open(os.devnull, 'w')
args = [settings.TESSERACT_PATH, input_filename, output_filename]
if charset:
util.charset_to_scratch(charset, scratch_charset)
args.append(scratch_charset)
proc = subprocess.Popen(args, stderr=devnull)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, charset=None):
"""Calls external tesseract on input file (restrictions on types),
outputting output_filename + '.txt'
Optionally, you can provide a character set to limit the OCR characters
the ones specified"""
devnull = open(os.devnull, 'w')
args = [settings.TESSERACT_PATH, input_filename, output_filename]
if charset:
util.charset_to_scratch(charset, scratch_charset)
args.append(scratch_charset)
proc = subprocess.Popen(args, stderr=devnull)
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename, psm, oem):
"""Calls external tesseract.exe on input file (restrictions on types),
outputting output_filename+'txt'"""
# https://github.com/tesseract-ocr/tesseract/wiki/Command-Line-Usage
cmdline = tesseract_exe_name + ' -l eng --psm ' + str(psm) + ' --oem ' + str(oem) + \
' ' + input_filename + ' ' + output_filename
args = cmdline.split()
# hide output (echo off)
FNULL = open(os.devnull, 'w')
proc = subprocess.Popen(args, stdout=FNULL, stderr=subprocess.STDOUT, cwd=os.getcwd())
retcode = proc.wait()
if retcode != 0:
errors.check_for_errors()
def make_clear():
args = ['./make_clear']
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode!=0:
errors.check_for_errors()
def call_tesseract(input_filename, output_filename): """Calls external tesseract.exe on input file (restrictions on types), outputting output_filename+'txt'""" args = [tesseract_exe_name, input_filename, output_filename] if retcode!=0: errors.check_for_errors()
def call_tesseract(input_filename, output_filename):
args = [tesseract_exe_name, input_filename, output_filename]
proc = subprocess.Popen(args)
retcode = proc.wait()
if retcode!=0:
errors.check_for_errors()
