TypeError: list indices must be integers or slices, not str

[solarssk]

Cześć! Czy tylko u mnie przestała działać integracja z API od ZTMu? Czy ktoś też ma ten sam przypadek? Ostatni pomyślny odczyt był 31.10 o 11:29, od tamtego momentu już cisza.

configuration.yaml

sensor:
  - platform: ztm
    api_key: XXX
    entries: 3
    name: "ZTM"
    lines:
      - number: XXX
        stop_id: XXXX
        stop_number: "01"
      - number: XXX
        stop_id: XXXX
        stop_number: "05"
      - number: XXX
        stop_id: XXXX
        stop_number: "15" 

Logs:

Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 942, in async_update_ha_state
    await self.async_device_update()
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 1300, in async_device_update
    await self.async_update()
  File "/config/custom_components/ztm/sensor.py", line 153, in async_update
    self._timetable = self.map_results(res.get('response', [0]))
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/custom_components/ztm/sensor.py", line 226, in map_results
    return [parse_raw_timetable(row) for row in response['result']]
            ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/custom_components/ztm/sensor.py", line 232, in parse_raw_timetable
    for val in raw_result['values']:
               ~~~~~~~~~~^^^^^^^^^^
TypeError: list indices must be integers or slices, not str

I co jakiś czas też od wczoraj:

Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 942, in async_update_ha_state
    await self.async_device_update()
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 1300, in async_device_update
    await self.async_update()
  File "/config/custom_components/ztm/sensor.py", line 153, in async_update
    self._timetable = self.map_results(res.get('response', [0]))
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/config/custom_components/ztm/sensor.py", line 226, in map_results
    return [parse_raw_timetable(row) for row in response['result']]
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: 'NoneType' object is not iterable

[miedzeszyn]

Hi So, not only me :) I can confirm the same issue. Also, have no data for waste collection schedule, so could be that something in API provided by Warszawa has been changed.

[solarssk]

Yap... Maybe in Monday I will get more information from IT department in Warsaw municipality.

[MG-PL-MG]

Hi, any luck with checking the issue with WAW IT ? I can only confirm last info form api was 31.10 and no update since then.

I generated query and sent it from browser and got a reply so api seems to work. So maybe reply structure changed somehow thus not recognized by integration ?

the example in documentation is like this:

so value followed by key and started by "values" In the reply I get has this structure:

so key followed by value and no "values" at the beginning.

not sure if this is an issue.

[1Bogdan1]

[mgorsk1]

I have recently refactored the code and for me this issue is not present, can you check?

https://github.com/peetereczek/ztm/pull/24/files

[MG-PL-MG]

I'm newbie here so not sure how to validate things but I have copied all the files in to my HA and sensor is almost back to normal got up to date information but can't see the destination.

`sensor:

• platform: ztm api_key:***** entries: 4 lines:
  • number: 189 stop_id: 4011 stop_number: "06"`

[mgorsk1]

thanks for letting me know @MG-PL-MG - it seems that 'kierunek' field is now missing from the response of ztm api, sadly nothing I can do about it.

[1Bogdan1]

Niedawno dokonałem refaktoryzacji kodu i u mnie ten problem nie występuje. Czy możesz to sprawdzić?

https://github.com/peetereczek/ztm/pull/24/files

I confirm - after re -invoicing everything works correctly.

[m-wyz]

How did you built version from PR? In HA I see possibility to download only branches from master, not PR made by someone else. For now waiting for @peetereczek to merge :)