There is currently nothing in pypi called fbida and the top google results point back to this repo. What is fbida?

There are few FIX-ME in tui code, let's finish it.

first off, this is awesome i figured i was going to end up having to put something like this together myself but it's already done!

on that note; has anyone had any luck sending audio to their cam? this utility appears to send the audio to the cam just fine. However it always sounds completely messed up when outputting through the cam. Initially i started with just curl with the same result until i stumbled upon this.

im assuming it's some type of audio encoding issue. I have tried sending in most of the available formats with different bit/sample rates with no luck.

so far this encoding gets me the closest but it's still FAR from correct

G.711A as content type
ffmpeg -i 4353.mp3 -c:a pcm_alaw -ac 1 -ar 16000 -sample_fmt s16 output.wav

Due static rule in spec, 'make rpm' is generating python2 for fedora or distro based on RHEL >= 8. Might be useful to users to still select python2.

Hits 1 from spec:
%if 0%{?fedora} || 0%{?rhel} >= 8
%global with_python3 1
%else
%global with_python2 1
%endif

Hint2:
We might create a AC_SUBS in configure.ac to determine the "%global python{NUMBER_VERSION} 1"

Hi,

Probably, I'm doing something wrong but in my case method audio_stream_capture never returns. I call it like this:

camera.audio_send_stream(httptype="singlepart", path_file=<path_to_file>, channel=1, encode="AAC")
<... other_logic ...>

My camera plays the sound successfully (length of the file is ~3 sec) but the method never returns and "<... other_logic ...>" below never executes.
I also tried to play with httptype=multipart but in this case, I hear no sound from the camera.

I also tried pure Amcrest REST API and basically have the same result. HTTP request takes too long (> 30 secs for 3 sec audio file), I hear sound but never get HTTP-response since it just hits a timeout.

So the question is, what is the proper way to send audio to a camera? :)

Protect password on credentials file introduced by #10

HomeAssistant is python3, TUI is python3 only, most of distros now support python3 from a long time.

Thanks for this package! Is it possible to set the resolution when requesting a camera snapshot?

I'm getting much lower resolution images than I do if I click the 'snapshot' icon in the web UI, so I know the cameras are capable of more.

Is there a way to convert binary data from snapshot method to numpy array?

As far as I see, snapshot method can return binary data of the image, which then can be written into a .jpeg file.

I have tried np.frombuffer method to obtain the image matrix directly, without writing it to a file first, but the result is inconsistent (matrix dimension is 1-D vector and varies from timestamp to timestamp).

Fails on:
2.400.AC01.15.R
2.400.AC02.15.R.20170731

See: #86

**Description of problem:**After upgrade to 0.59.1 only 1 of my Amcrest cameras output appear in both of my Amcrest cameras displays on the Dashboard.

I upgrade my RPI 3 running on Raspbian Stretch Desktop to HA version 0.60.0. Still having the same trouble with my Amcrest Cameras. I did see this in the Info tab if this helps!!
Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))': /cgi-bin/snapshot.cgi?channel=0
2:18 PM components/camera/amcrest.py (WARNING)

We should be able to run minimal auto tests before merging a patch. Initial tests could be if the patch pass against pep8 validation (flake8 tool ?).

Hi All,

Since 1.5.4, I got the following error:

Traceback (most recent call last):
File "", line 1, in
File "/usr/local/lib/python3.7/site-packages/amcrest/init.py", line 33, in init
timeout_protocol=timeout_protocol
TypeError: init() got an unexpected keyword argument 'verify'

Using this code:

from amcrest import AmcrestCamera
myCamera = AmcrestCamera(IP, 80, user, pass).camera

Works fine in 1.5.3.

New parameter?

Thanks,
Joe

Getting this issue while accessing the camera. Anybody else seen this?
Would appreciate any help!

camera = AmcrestCamera('192.168.0.1', 80, 'admin', 'passd').camera
Traceback (most recent call last):
File "", line 1, in
File "/home/anaconda3/lib/python3.7/site-packages/amcrest/init.py", line 33, in init
timeout_protocol=timeout_protocol
TypeError: init() got an unexpected keyword argument 'verify'

Enhancement request:
Add the telnet password prefix somewhere in the repository.
The telenet password is prefixed onto the user's password.

See about halfway down in the following thread:
http://www.cctvforum.com/viewtopic.php?f=6&t=52356&start=45

Nice project. Thanks!
Could you implement the Post Audio Stream API (ref section 4.3.3 in the API manual 2.10 2016-05-25)?

Use case:
When motion or sound are detected, play an audio file over the camera's speaker to act as a deterrent. Examples: dog barking, TV or radio playing, background conversations, alarm sound, etc.

In the "Usage" section of the README.rst the example for moving the camera down has two extra parens at the end of the line:
camera.ptz_control_command(action="start", code="Down", arg1=0, arg2=0, arg3=0)))

The option --config-backup --save /tmp/cfg

» amcrest-cli  --camera amcrestcam1 --config-backup --save /tmp/cfg                                                                                 ~/d/h/homeassistant
Traceback (most recent call last):
  File "/home/mdemello/.virtualenvs/home_assistant/bin/amcrest-cli", line 863, in <module>
    main()
  File "/home/mdemello/.virtualenvs/home_assistant/bin/amcrest-cli", line 849, in main
    camera.config_backup(args.save)
TypeError: 'str' object is not callable

I can see that there is a config_backup() function that calls /cgi-bin/config.backup?action=all - which, in turn, retrieves a JSON with lots of settings.

I could not find how to import it back. The only thing that resembles import in https://s3.amazonaws.com/amcrest-files/Amcrest+HTTP+API+3.2017.pdf is configManager.cgi?action=restore - however it only accepts the list of names (and not JSON), and when I tried to run it (configManager.cgi?action=restore&names[0]=MobilePushNotification) I saw zero effect.

Do you know how is it supposed to work?

I have both Amcrest IPC and Dahua IPC cameras, and the latter seem to be working great via the Home Assistant Amcrest component - including motion detection!

There currently isn't a Dahua component in HA, and I had been using the generic and ffmpeg camera components to support them. Not great, since no other camera features are supported except snapshot and streaming. Some users have gone to great lengths with Python scripts and MQTT to get motion alerts from Dahua cameras via HA.

Are you able to confirm if the Amcrest HTTP API is essentially the same as for Dahua IPC cameras ? Are you aware of any differences or issues? I am wanting to get my facts straight before publicly announcing on the HA forums that the existing Amcrest support is fully compatible with Dahua.

Thanks!

I'm using this library through the latest Home Assistant, I successfully added my Amcrest NVR with this:

# Amcrest NVR
amcrest:
  - host: NVR_IP
    username: <login>
    password: <password>
    sensors:
      - motion_detector
      - sdcard
      - ptz_preset

My Overview page in HA started showing images from the camera set to the first channel in the NVR.
Is it currently possible to specify the channel so that I can get the feed from other cameras in the NVR? Are there other implications of doing this?

In general I'm trying to avoid connecting directly to the Camera's IP, I'm thinking it's more efficient to connect to the NVR.

Amcrest Doorbell works great with this by the way. Can you add a doorbell sensor too?

We need a native gui interface for linux.

It looks like the new firmware changes the software version string too:

Traceback (most recent call last):
  File "./DPtest.py", line 8, in <module>
    print('SW info:', a.camera.software_information)
  File "/root/python-amcrest/src/amcrest/system.py", line 70, in software_information
    version, build_date = ret.content.decode('utf-8').split()
ValueError: need more than 1 value to unpack

If I print ret.content it's now:
ret.content: b'version=2.420.AC00.17.R,build:2017-03-22\r\n'

I'll add a check for this and PR, thanks for your work!

I recently just implemented a realtime event notifier - for VideoMotion events, but you could use any event, or multiple events, it's a list.

Is anyone interested in implementing it here?

This is the basic code (part of a camera class):

def stream_motion_data(self):
        text = ''
        chars = 0
        content_len = 0
        api_URL = 'http://%s:%s/cgi-bin/eventManager.cgi?action=attach&codes=[VideoMotion]' % (self.ip, self.port)

        r=requests.get(api_URL, auth=HTTPDigestAuth(self.user, self.password), stream=True)
            if r.status_code != requests.codes.ok:
                self.log.error('bad connection: %s' % r.status_code)
                r.raise_for_status()

            for line in r.iter_lines():
                # filter out keep-alive new lines
                if line:
                    decoded_line = line.decode('utf-8')
                    if decoded_line.startswith('--'):  # start of new chunk
                        text = ''
                        chars = -1
                    else:
                        if 'Content' not in decoded_line:
                            text += '%s\n' % decoded_line
                            chars = len(text)
                            self.log.debug('current content length: %s' % chars)
                        elif 'Content-Length:' in decoded_line:
                            content_len = int(decoded_line.replace('Content-Length:','').strip())
                            self.log.debug('content length: %s' % content_len)
                        if chars == content_len:
                            self.log.debug('decoding lines')
                            self.decode_line(text)

def decode_line(self, lines):  
        self.log.info(lines)
        data = lines.split(';')
        try:
            regions = json.loads(data[-1].replace('data=',''))
            regions = regions.get('RegionName', [])
        except json.JSONDecodeError:
            regions = []
            
        self.log.debug(' '.join(data))
        if 'VideoMotion' in data[0]:
            if 'Start' in data[1]:
                self.log.info('Motion Detected, regions: %s' % (regions))

It's a bit rough, as I just hacked it together to trigger a light when motion is detected via the camera.

The output looks like this:

[I 2020-01-09 10:39:49,528] (MainThread) Motion Detected, regions: ['Region1', 'Region2']

you'd have to run it in a thread if you wanted to do anything else with the camera while it's running.

I don't want to submit a PR if no-one is interested.

I can see the channel 1 (low res mpeg) stream, control the camera movement, but cannot use the high-rest stream due to access problems.

camera1 = amcrest.AmcrestCamera('192.168.1.xxx', 80 'admin', 'pswd', ).camera

vs = VideoStream(camera1.realtime_stream(
    channel=0, typeno=None, path_file=None`)).start

frame = vs.read()
cv2.imshow("Frame", frame)

The result is
Trying again due to error 401 Client Error: Unauthorized for url: http://192.168.1.35:80/cgi-bin/realmonitor.cgi?action=getStream&channel=0&subtype=None

just bought one of these and am now sad to see the doorbell isnt supported (I know I should have done better research..)
any way this might be supported in the future ?

Does anyone know how to trigger a motion detection event via the API?

The problem I have is that the camera reacts to lights turning on as if they were motion. I have the camera recording 24/7 (ie scheduled), and when motion is detected, it records that motion flagged as motion (ie stupid file name with '[M]' in it in place of '[R]').

This works fine, but is constantly recording the hallway light turning on and off.
I also use motion (the linux program) to detect motion, and track the motion using pan and tilt. This allows me to fine tune the motion detection parameters, and filter out 'lightswitch' events (as motion calls them).
I can also trigger external events on motion detection.

My thought was to turn the camera motion detection sensitivity down (so it doesn't trigger on the light), and send a command via the API to trigger a motion event (like the external input trigger).

I can't figure how to do this via the API though. Is there a way to trigger external event or motion recording via the API?

Thanks.

Please implement the APIs on section 4.11.5-4.11.7 of the AMCREST_HTTP_API_SDK_V2.10.pdf. These media file APIs would enable SD card media file search and download, which would add some useful functionalities to the Amcrest IP camera. For example, one can write a simple automation script to download a motion-triggered recorded video on camera SD card, based on certain conditions that are practically useful, e.g. when there is no network traffic and/or within certain time frames. Currently, Amcrest Ip cams can either record motion-triggered video to sd or send off to an FTP server. While this limitation is practically good to prevent network traffic (when multiple ip cams trying to send files to an FTP server), it is a big disadvantage not having a way to store the recorded video in a safer place, other than the cam sd card.

Tested and appears to be working with Amcrest IP3M-941 ( https://smile.amazon.com/Amcrest-UltraHD-Security-3-Megapixel-IP3M-941W/dp/B01M0IHF9K ). Feel free to update the Matrix :)

Thanks for the hard work!!

Password with '%' character on ~/.config/amcrest.conf fails with the errro below:

Traceback (most recent call last):
  File "/home/mdemello/.virtualenvs/amcrestchime/bin/amcrest-cli", line 960, in <module>
    main()
  File "/home/mdemello/.virtualenvs/amcrestchime/bin/amcrest-cli", line 508, in main
    args.password = config.get(section, 'password')
  File "/usr/lib64/python3.5/configparser.py", line 797, in get
    d)
  File "/usr/lib64/python3.5/configparser.py", line 393, in before_get
    self._interpolate_some(parser, option, L, value, section, defaults, 1)
  File "/usr/lib64/python3.5/configparser.py", line 443, in _interpolate_some
    "found: %r" % (rest,))
configparser.InterpolationSyntaxError: '%' must be followed by '%' or '(', found: "%test%"

Create function to subscribe to event message bus to allow notifications real-time upon Motion Detection.

Example:

curl -k -u 'admin:secret' \
"http://amcrestcam2.local.home/cgi-bin/eventManager.cgi?action=attach&codes=\[VideoMotion\]"
-- myboundary
Content-Type: text/plain
Content-Length:37
Code=VideoMotion;action=Start;index=0

-- myboundary
Content-Type: text/plain
Content-Length:36
Code=VideoMotion;action=Stop;index=0

-- myboundary
Content-Type: text/plain
Content-Length:37
Code=VideoMotion;action=Start;index=0

-- myboundary
Content-Type: text/plain
Content-Length:36
Code=VideoMotion;action=Stop;index=0

Create unittest coverage.

Hello, I'm loving this project so far which has greatly simplified the API calls for the amcrest camera.

I am currently testing with this camera [IP2M-841B]. And have run into one issue so far.

I'm using the pypip package to run the funcions.

For recording which works:
camera = AmcrestCamera(ip, 80, 'user', 'password').camera camera.audio_stream_capture( httptype="singlepart", channel=1, path_file="audio.wav")

But trying to "stream" the same file just plays back about 10 seconds of static:
camera = AmcrestCamera(ip, 80, 'user', 'password').camera camera.play_wav("singlepart", channel=1, path_file="audio.wav")

Any and all help would be appreciated.

Thanks!

amcrest-cli was created to support the tons of calls the python library provides.
However, as the project grows fast, now we have a bunch of calls and the cli was basically copy/past and now it's pretty big one single file. IMO, would be the best if we could split it in topics as we do in the library for easy coding.

As mentioned in the docs, the script to store the continuous stream is
camera.realtime_stream(path_file="/home/user/Desktop/myvideo")
CTRL-C to stop the continuous video flow or use a timer
I have a couple of questions -

1. How to use the timer for stopping the stream?
2. Where to put the print statement so that I can log whenever it is updating /home/user/Desktop/myvideo

if path_file:
_try:
I am putting a print statement here to see whenever it gets copied but it prints only once despite the file getting updated continuously
with open(path_file, 'wb') as out_file:
shutil.copyfileobj(ret.raw, out_file)
except HTTPError as error:
LOGGER.debug("%s Realtime stream capture to file failed due "
"to error: %s", self, repr(error))
raise CommError(error)

The documentation states there is MotionDetection but does this allow the ability to enable/disable motion detection?

The 6.9 GetEventIndexes section of the API documentation gives a way to know for which channel an event happened:

URL Syntax http://<ip>/cgi-bin/eventManager.cgi?action=getEventIndexes&code=<eventCode>

Comment | Get channels indexes that event of code eventCode happens. eventCode includes:VideoMotion: VideoLoss: VideoBlind: AlarmLocal:motion detection event video loss detection eventvideo blind detection event. alarm detection event.
-- | --
Response | channels[0]=0 channels[1]=2 channels[2]=3 ...(This response means event happened on channel 0, channel 2, and channel 3.)

Using this, we should be able to have a method telling if a motion detected event has occured for a given channel or not.

Currently, it seems that we only check that channels are present in the response, but we do not give access to the filtered information.

I'm working with a IP2M-841, just upgraded to new firmware V2.420.AC00.17.R 3/22/17

After upgrading, all my third party tools and custom code using python-amcrest broke on any request:

Traceback (most recent call last):
  File "/home/homeassistant/DPtest.py", line 7, in <module>
    print(c.camera.current_time)
  File "/home/homeassistant/.homeassistant/deps/amcrest/system.py", line 19, in current_time
    'global.cgi?action=getCurrentTime'
  File "/home/homeassistant/.homeassistant/deps/amcrest/http.py", line 102, in command
    resp.raise_for_status()
  File "/home/homeassistant/.homeassistant/deps/requests/models.py", line 909, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 401 Client Error: Unauthorized for url: http://xyz.xyz.xyz.xyz:80/cgi-bin/global.cgi?action=getCurrentTime

I verified by tcpdump/wireshark that we're sending basic auth in the request, and the camera responds 401 and asks for digest auth.

Seems others are seeing it too:
https://amcrest.com/forum/technical-discussion-f3/cgi-sdk-no-longer-functioning-on-17r--t2401.html

I fixed it by changing to digest auth in http.py:
self._token = requests.auth.HTTPDigestAuth(self._user, self._password)

Everything is working now, happy to submit a PR

Remove 'http{s}://' and '/' from possible URLs when initiating the Amcrest object.

#from
cam = AmcrestCamera('http://192.168.1.10/', 80, 'admin', 'secret')
#to
cam = AmcrestCamera(' 192.168.1.10', 80, 'admin', 'secret')

Using Hassio component Amcrest IP Camera with "camera platform amcrest" & "sensor platform amcrest" for my "IP8M-2496EB for 1 month now.

Camera is displayed perfectly after setting to (resolution) low. I use the RSTP method and I'm curious if that could be the cause of my issue. (The sensor and switch errors below mention the HTTP transport protocol. Are they not to be used with RSTP

"Sensor" and "Switch" through a time out errors ever 2 seconds. Logs fill fast. But the associatied entities appear to be functioning correctly. (ie.. they yeld "state / data" that changes accurately.)

See details below.
` Log Details (WARNING)
Fri Dec 21 2018 04:35:51 GMT-0500 (Eastern Standard Time)

Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ReadTimeoutError("HTTPConnectionPool(host='192.168..', port=**): Read timed out. (read timeout=3.0)",)': /cgi-bin/magicBox.cgi?action=getSoftwareVersion `

I will adjust the scan itnerval. But I suspect this time out issue is related to setting in amcrest.py component used by hassio. Any help would be most appreciated.

I also get this error..

`Log Details (WARNING)
Fri Dec 21 2018 04:30:21 GMT-0500 (Eastern Standard Time)

Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'ConnectTimeoutError(<urllib3.connection.HTTPConnection object at 0x6f448930>, 'Connection to 192.168.. timed out. (connect timeout=3.0)')': /cgi-bin/configManager.cgi?action=getConfig&name=MotionDetect `

Obviously I redacted my internal IP of the camera...

Right now, it looks like scan_devices() might be the only function where you can pass in a custom timeout. In other cases, such as snapshot(), there's no way to override the default timeout of 3 seconds.

On a Dahua 2015 firmware, no upgrade possible (last firmware)

This URL gets a bad response (400)
http://192.168.1.41:80/cgi-bin/eventManager.cgi?action=attach&codes=%5BAll%5D

This URL works
http://192.168.1.41:80/cgi-bin/eventManager.cgi?action=attach&codes=[All]

A quick fix was detailed here (with Python Amcrest 1.7.0) in HA Reported bug
home-assistant/core#33875 (comment)

Which seems to be working for me.

Getting the following error when I try to run anything, even just getting cam info:

AttributeError: module 'socket' has no attribute 'TCP_KEEPIDLE'

Process finished with exit code 1

Create docs based on sphinx

I'm a bit perplexed as to why this isn't working, but I can't seem to run the eval line to install this. I'm running FreeBSD on this machine.

I am sendnig some audio which plays fine on the camera, however I am then unable to send subsequent streams or query the API for information such as SD card capacity. This is the documentation I followed.

        this_camera = get_camera(call);
        file=call.data.get(ATTR_FILE);
        _LOGGER.info("Playing {} on {}".format(file,call.data[ATTR_ENTITY_ID]));
        # this_camera.play_wav(httptype='singlepart', path_file=file);
        result = this_camera.audio_send_stream(httptype='singlepart', channel=1, path_file=file, encode='MPEG2');
        _LOGGER.info(dir(result));
        return result == 'OK';

The audio_send_stream method does not return, which explains the symptoms.
Is there anything I am missing in my code? Do I have to close some stream manually?
Tried singlepart and multipart and had the same problem with the commented out attempt at play_wav.
The only way to send more audio is by rebooting the camera.

I've got a couple of Amcrest IP2M-841W camera's setup at the house. I made a kludge smoke detector for one that feeds into the camera relay, using a DC-DC step up converter to get the 9v required by the sensor.

It works OK and the app on my phone pushes me a notification if the relay is triggered. But it's not really enough to wake me up or alert the whole house etc. It would be great if I could have those relays exposed to homeassisstant. I'm not a software developer and don't have a sense for what it would take to get them working. Just wanted to post here and see if there's any interest.