It would be useful to add the possibility to determine the distance with the latitude and longitude collected by the gps.

I am using the NMEA TCP/IP stream of an "emid reach M+" module with this library and had an issue with the SENTENCE_LIMIT of 76. In my case it has to be at least 81 (suggsest 90 to be save).

Example:
$GNGGA,223832.20,ddmm.mmmmmmm,N,dddmm.mmmmmmm,W,1,07,1.2,298.903,M,-30.571,M,0.0,*61\r\n

I have received IndexError in micropyGPS.py in gpgsa(). I made a clunky temporary fix by catching IndexError in 2 places. I am running micropyGPS on a Tiny 2040, using MicroPython rp2-pico-20210202-v1.14.

Code from sentence_test.py from the repo pasted directly into PyBoard v1.1 REPL connected to a GT-U7 module:

from pyb import UART
from micropyGPS import MicropyGPS
uart = UART(3, 9600)
my_gps = MicropyGPS()
while True:
    if uart.any():
        stat = my_gps.update(chr(uart.readchar())) # Note the conversion to to chr, UART outputs ints normally
        if stat:
            print(stat)
            stat = None

Output:

GPRMC
GPGGA
GPGSA
GPGSV
GPGSV
GPGSV
GPGLL
GPRMC
GPGGA
GPGSA
GPGSV
GPGSV
GPGSV

It's hard to "see what isn't there", so one doesn't immediately realize there is a problem above, but note that lack of GPVTG detections. Of course, one might simply theorize there are no GPVTG sentences. But now look as this modified code which will expose the invisible missing data.

Code pasted into PyBoard v1.1 REPL connected to a GT-U7 module:

from pyb import UART
from micropyGPS import MicropyGPS
uart = UART(3, 9600)
gps = MicropyGPS()
while True:
	if uart.any():
		c = chr(uart.readchar())
		print(c, end='')
		stat = gps.update(c) # Note the conversion to chr, UART outputs ints normally
		if stat:
			print()
			print('    Sentence completed: ' + stat, end='')
			stat = None

And observe the output, which shows the GPVTG sentences on the wire but failing to trigger a end-of-sentence event:

$GPRMC,025631.00,V,,,,,,,080421,,,N*71
    Sentence completed: GPRMC
$GPVTG,,,,,,,,,N*30
$GPGGA,025631.00,,,,,0,04,25.50,,,,,,*63
    Sentence completed: GPGGA
$GPGSA,A,1,29,05,25,20,,,,,,,,,27.96,25.50,11.49*3E
    Sentence completed: GPGSA
$GPGSV,3,1,11,02,05,077,,05,47,061,22,13,14,101,24,15,15,137,*79
    Sentence completed: GPGSV
$GPGSV,3,2,11,16,16,321,,18,53,266,13,20,48,197,14,23,20,203,21*72
    Sentence completed: GPGSV
$GPGSV,3,3,11,25,22,190,16,26,33,299,08,29,79,111,21*48
    Sentence completed: GPGSV
$GPGLL,,,,,025631.00,V,N*49
    Sentence completed: GPGLL
$GPRMC,025632.00,V,,,,,,,080421,,,N*72
    Sentence completed: GPRMC
$GPVTG,,,,,,,,,N*30
$GPGGA,025632.00,,,,,0,04,25.44,,,,,,*65
    Sentence completed: GPGGA
$GPGSA,A,1,29,05,25,20,,,,,,,,,27.90,25.44,11.45*31
    Sentence completed: GPGSA
$GPGSV,3,1,11,02,05,077,,05,47,061,18,13,14,101,20,15,15,137,*74
    Sentence completed: GPGSV
$GPGSV,3,2,11,16,17,321,,18,53,266,12,20,48,197,12,23,20,203,19*7F
    Sentence completed: GPGSV
$GPGSV,3,3,11,25,22,190,12,26,33,299,08,29,79,111,18*46
    Sentence completed: GPGSV
$GPGLL,,,,,025632.00,V,N*4A
    Sentence completed: GPGLL
$GPRMC,025633.00,V,,,,,,,080421,,,N*73
    Sentence completed: GPRMC
$GPVTG,,,,,,,,,N*30
$GPGGA,025633.00,,,,,0,04,25.38,,,,,,*6F
    Sentence completed: GPGGA

BDS statement is not supported! But after all ,Thank you:-)

when there are 12 satellites in view the last sentence is not added to the satellite data. however 11 satellites works fine.

here is sample data and output
sentences
$GPGSV,3,1,12,13,65,002,50,02,61,098,47,39,60,352,,05,56,183,4970
$GPGSV,3,2,12,15,35,325,50,29,32,229,49,06,25,070,44,30,16,096,3870
$GPGSV,3,3,12,19,08,022,35,07,07,122,,12,06,316,49,25,03,278,36*7D
(my_gps.satellite_data)
{2: (61, 98, 47), 5: (56, 183, 49), 6: (25, 70, 44), 39: (60, 352, None), 13: (65, 2, 50), 15: (35, 325, 50), 29: (32, 229, 49), 30: (16, 96, 38)}

Hello,
I am currently using micro python on STM32Fdisc board, attempting to interface with GPS module. I have tried your code and also tested my code but i am unable to read data from GPS.
Do you have any idea about this problem, why it happen ?

Here is the code:

import pyb
import pyb,micropython
from pyb import LED, Pin, UART

rx = bytearray(250)
i = None
uart = UART(2,115200)

data = bytearray(255)
j = 0
#Capture Data in the Rx buffer
for i in range(255):
uart.readinto(rx)
print(rx)

x = uart.readchar()

print(x)

if i == 255:
break

I printed the buffer and the value is 0. So do you have any idea what can be the reason ? I have checked the GPS module its working using C/C++ and also the UART is working fine.

Hello,
I'm getting this error whilst trying to parse a string:

Traceback (most recent call last):
  File "gps.py", line 30, in <module>
    my_gps.update(x)
  File "/home/pi/MNT-GPS/micropyGPS.py", line 573, in update
    ascii_char = ord(new_char)
TypeError: ord() expected string of length 1, but int found

Here is my full code:

import serial
import time
import os
from micropyGPS import MicropyGPS
import pifacecad

my_gps = MicropyGPS()

cad = pifacecad.PiFaceCAD()
cad.lcd.backlight_on()
cad.lcd.blink_off() 

# Set up serial:
ser = serial.Serial(
    port='/dev/ttyUSB0',\
    baudrate=4800,\
    parity=serial.PARITY_NONE,\
    stopbits=serial.STOPBITS_ONE,\
    bytesize=serial.EIGHTBITS,\
        timeout=1)


# Main program loop:
while True:
    
    my_sentence = ser.readline()
    for x in my_sentence:
            my_gps.update(x)
    lat = my_gps.latitude
    lon = my_gps.longitude
    cad.lcd.clear()
    cad.lcd.write("%s\n%s" % (lat, lon)) 
    print("%s\n%s" % (lat, lon)) 
    time.sleep(2)

Thanks for the help :)

this line should read:

elif self.date[0] in (3, 23):

Hi. Can the library give me the latitude and longitude in decimal? I'd find this easier to work with. For example the statue of liberty would be at 40.689167, -74.044444

Unhandled exception in thread started by <bound_method>
Traceback (most recent call last):
File "/flash/lib/L76micropyGPS.py", line 60, in feedMicroGPS
File "/flash/lib/micropyGPS.py", line 615, in update
File "/flash/lib/micropyGPS.py", line 372, in gpgga
File "/flash/lib/micropyGPS.py", line 363, in gpgga
IndexError: list index out of range
I created the object from repl, it was working fine.
Then I left the computer, and this is what I saw.

This library works extremely well but there are potential optimisations for use on platforms with little RAM. My asynchronous GPS device driver implements these. The principal categories are as follows:

• Replace the string addition operator with the format method (since Python strings are immutable the string addition operator causes needless allocations). This also reduces code size.
• Use lists rather than tuples for bound variables which are updated, changing each element individually to avoid allocating a temporary tuple.
• Have common _fix and _set_timestamp methods to avoid code duplication.
• Replace string comparisons with integer comparisons and use the micropython.const function. This may be an optimisation too far for an existing library as it changes the API.

The first optimisation can substantially simplify the MTK_commands library. See this method for an example (implements update_sentences).

Clearly allocation can't be eliminated as floats are necessarily employed and satellite information is dynamic, but minimising it helps reduce RAM fragmentation.

I'm a getting an Out of Memory Error whenever I import micropyGPS on a NodeMcu v2 board with Micropython 1.9.1 (see https://www.amazon.com/HiLetgo-Version-NodeMCU-Internet-Development/dp/B010O1G1ES).

Based on your project's descrioption, you are using a Pyboard which has 192KiB RAM while the NodeMcu board has 128KiB RAM. Even with a simple import with no classes at all, I still get the Out of Memory Error. I am surprised the extra 64KiB makes a big difference in loading this library.

I am wondering if you were able to load your library on a NodeMcu board previously. I'm planning to strip out the library to the bare essentials (based on my needs), but before that I was thinking if you've done prior testing with a NodeMcu before. Or if you can provide a quick tip on which ones I can strip out quickly.

I will forego any error checking to save memory and assume happy path. All I want is read the coordinates from my Adafruit GPS device. Once that's read, send it to Raspberry for further processing. I am thinking of porting your implementation to Raspberry (if it doesn't exist yet).

I think by the time you read this, I would already start hacking your code tonight or tomorrow.

In this line (which occurs three times in the code) the hours value should be reduced modulo 24 to avoid illegal values.

The code produces decimal degrees that are incorrectly formatted.

This can be seen in the init function definition:

        Setup GPS Object Status Flags, Internal Data Registers, etc
            local_offset (int): Timzone Difference to UTC
            location_formatting (str): Style For Presenting Longitude/Latitude:
                                       Decimal Degree Minute (ddm) - 40° 26.767′ N
                                       Degrees Minutes Seconds (dms) - 40° 26′ 46″ N
                                       Decimal Degrees (dd) - 40.446° N

The correct format for "dd" is +/-90° for latitude, and +/-180° for longitude, with no compass heading suffix. The compass heading is encoded into the sign of the number. (See here)
So in the example from the code would resolve to +40.446 not 40.446 N.
In the southern hemisphere, it would resolve to -40.446, not 40.446 S.

Any chance of including a command to enable DGPS?

Cheers,

Kip

Our adafruit Ultimate GPS has an RTC w/ clock backup, and we noticed that when we were getting RMC strings w/no fix, time would be correct, but the date would be wrong.
else: # Clear Position Data if Sentence is 'Invalid' self._latitude = (0, 0.0, 'N') self._longitude = (0, 0.0, 'W') self.speed = (0.0, 0.0, 0.0) self.course = 0.0 self.date = (0, 0, 0) self.valid = False
Notice that self.date = 0, but timestamp is left alone. I realize this probably is the opposite of the previous issue's intent, but it seems to me in the case of an RTC enabled GPS, both date and time should come through if the GPS has them stored.

Unfortunately have no more information. Code was running for 3+ hours.

Traceback (most recent call last):
File "", line 195, in
File "", line 43, in getGPS
File "micropyGPS.py", line 621, in update
File "micropyGPS.py", line 456, in gpgsa
IndexError: list index out of range

I've updated my code, line 459, to add the exception IndexError as seen in issue #26 .

Hi!
I'm a Japanese.
I set local_offset to 9 (Tokyo JST).
And find date is yesterday.
Date is not maintained.
I make a correction.
This is my code

Hello, thanks for micropyGPS.
I am testing it right now. Unfortunately, there is a problem with parsing GPRMC sentance for my GPS module.
Example sentance:
$GPRMC,165904.00,A,xxxx.85300,N,yyyyy.85129,E,0.564,,011217,,,A*77
gps_segments (from PyCharm debugger):

gps_segments = {list} <class 'list'>: ['GPRMC', '165904.00', 'A', 'xxxx.85300', 'N', 'yyyyy.85129', 'E', '0.564', '', '011217', '', '', 'A', '77']
 00 = {str} 'GPRMC'
 01 = {str} '165904.00'
 02 = {str} 'A'
 03 = {str} 'xxxx.85300'
 04 = {str} 'N'
 05 = {str} 'yyyyy.85129'
 06 = {str} 'E'
 07 = {str} '0.564'
 08 = {str} ''
 09 = {str} '011217'
 10 = {str} ''
 11 = {str} ''
 12 = {str} 'A'
 13 = {str} '77'
 __len__ = {int} 14

Within function gprmc there is a code:

# Course
try:
    course = float(self.gps_segments[8])
except ValueError:
    return False

Since there is no Course data the parsing fails traying to do course = float(str)

Traceback (most recent call last):
  File "xxx", line 3, in Exec
    exec(exp, global_vars, local_vars)
  File "<input>", line 1, in <module>
ValueError: could not convert string to float: 

I just wanted to let you know.
But still its a great SW and the fix is simple.

When power is first applied to the MTK3339 chip it emits RMC messages with gps_segments[2] != 'A'. The 'A' value only appears once a satellite fix has been achieved.

Unfortunately those invalid messages can have invalid date/time contents. These can pass subsequent tests, yielding an incorrect timestamp. In my opinion the 'A' value should be checked at the start of parsing, and the sentence rejected if it is invalid.

The local_offset variable makes an adjustment to the 'hours' value in any timestamps that are recorded. However, many locations have a half hour offset (0.5) and this is not accounted for by also adjusting the 'minutes' value.

Additionally, from my reading of the code, I don't believe there is any adjustment being made to the date when local_offset results in a day shift forwards or backwards.

Hello

When using micropyGPS to parse NMEA messages in a file, I noticed that the my_gps.update() method always returns 'None', for RMC messages, even if the sentence has been parsed. The RMC messages are valid and are counted as "clean sentences" by the statistics parser.

I have tried a number of files with different receiver types, with consistent results.

Thanks for making a great tool. Other than this, I have had no issues and this was exactly what i was looking for in a NMEA parser.

Decimal Degrees precision is 0.00000 how to get 0.000000 (6 digits)?

Hi, i'm using this library on my wipy + pytrack shield.

it uses the quectel l76 gps + glonass receiver. here the spec: http://ecksteinimg.de/Datasheet/QuectelL76-LGNSS.pdf

i see that glonass specific sentences are: GLGSV, GNGLL, GNGSA, that are not implemented in your code. You think you will add those features? If I have some spare time i'd like to contribute.

from the spec file, i see that the only difference between:

[gp|gn] rmc
[gp|gl] gsv
[gp|gn] gll
[gp|gn] gsa

is only the "name". as the nmea sentence has the same format. so i think they can be added into
supported_sentences without new parsers.

however, as I understand from the notes in 2.1 chapter in the spec file, gp-gll uses only gps satellites and gn-gll uses gps + glonass satellites. so during parsing, gngll sentence should have some sort of priority on gpgll.

what do you think?

:D

This line should be

elif self.date[0] in (3, 23):

Whenever I try to import micropyGPS no further part of the main.py code executes. Prints put inside micropyGPS don't execute either. This fork of Mycropython doesn't have pyb. Can that be a problem?