An alert application for observing supernovas.
Project description
SNEWS Publishing Tool
Docs: | https://snews-publishing-tools.readthedocs.io/en/latest/ |
This packages provides users with a Python API and CLI to publish observation messages to SNEWS
How to Install
Note: Make sure your hop credentials are set up !!
Follow the instructions here if needed
First you need to clone this repo. In your terminal run the following:
git clone https://github.com/SNEWS2/SNEWS_Publishing_Tools.git
Once cloned, install the package using pip (make sure you're in the cloned dir)
pip install .
Message Schemas
Note: Not the final schemas !!
Coincidence Tier
_id
detector_name (user input)
sent_time
machine_time (user input)
neutrino_time (user input)
p_val (user input)
Significance Tier
_id
detector_name (user input)
sent_time
machine_time (user input)
neutrino_time (user input)
p_value(s) (user input)
Time Series Tier
_id
detector_name (user input)
sent_time
machine_time (user input)
neutrino_time (user input)
timing_series (user input)
False Obs
_id
detector_name (user input)
false_id (user input, optional)
which_tier (user input)
N_retract_latest (user input)
retraction_reason (user input, optional)
sent_time
How to Publish
Before we get started, right now the publishing method will send your message to the test kafka server.
First you need to import your Publisher:
# Import the constructor for SNEWS Tiers and Publisher class
from SNEWS_PT.snews_pub import SNEWSTiersPublisher
Passing Message Parameters as Arguments.
To send a message you need initialize the Publisher
, construct your message by initializing SNEWSTiers
and
passing your parameters of choice. The backend will parse your arguments, check their data types and determine which
tiers you can send a message to (see Publishing Protocols). If you pass multiple parameters (see code bellow) the
sender will send a message all the appropriate tiers.
SNEWSTiersPublisher(detector_name='KamLAND', neutrino_time='22/02/28 4:31:08:565',
timing_series=['22/02/28 4:31:08:565', '22/02/28 4:31:08:765', '22/02/28 4:31:09:001'],
p_val=0.000007, machine_time='22/02/28 4:31:08:565',
).send_to_snews()
This instance has parameters for CoincidenceTier and TimingTier, thus it will send a message to both. The output should look like this: !
Passing Message Parameters from JSON File.
You can also pass your input from a json file, and make modifications on the spot. Let's first create an observation
object this time before sending it to snews;
observation = SNEWSTiersPublisher.from_json('my_input_asjson.json',
detector_name='XENONnT',
comment="This is submitted from a json file")
Here, we read the content from the 'my_input_asjson.json'
file, and overwrite detector_name
and also add a comment field (which will be accepted as a meta data). Notice we still haven't sent it to snews yet. You can display, and modify the parsed messages after you create the object instance. Depending on the fields you provided SNEWSTierPublisher
will decide where to submit your data (see above). You can see these tier(s) and the individual message contents. See, observation.tiernames
to get names of the tiers that your input message belongs, and observation.messages
to display their content, and modify if desired.
Once you are done, you can just send that observation
object to snews.
observation.send_to_snews()
See this examples notebook
for more tutorial scripts
Publishing Protocols
Coincidence Tier
p_value
andneutrino_time
need to be passed.p_val
must be afloat
.neutrino_time
must be astring
, format:'%y/%m/%d %H:%M:%S'
Significance Tier
p_values
needs to be passed.p_values
must be alist (float)
.
Timing Tier
p_value
andtiming_series
need to be passed.p_val
must be afloat
.timing_series
must be alist (string)
, format:'%y/%m/%d %H:%M:%S'
Retraction
n_retract_latest
andwhich
need to be passed.n_retract_latest
must be aint (and >0 )
. You can also pass it as a'ALL'
.which_tier
must be awhich_tier
, format:'%y/%m/%d %H:%M:%S'
Pre-SN Timing Tier
is_pre_sn
andtiming_series
need to be passed.is_pre_sn
must be abool
.timing_series
must be alist (string)
, format:'%y/%m/%d %H:%M:%S'
Notice that your message can contain fields that corresponds to several tiers e.g. if you have p_value
, neutrino_time
, and p_values
we submit two separate messages to Coincidence and Significance tiers by selecting the relevant fields from your input.
How to Subscribe
In two lines, one can subscribe to the alert topic specified in the default configuration.
This starts a stream, and waits for alert messages to be received.
from SNEWS_PT.snews_sub import Subscriber
Subscriber().subscribe()
Should there be an alert message, this will be both displayed on the screen and saved into your local machine
as SNEWS_MSGs/<today fmt="%y_%m_%d">/subscribed_messages.json
and if there are multiple messages in the same day e.g.
hype-mode is on and for the same supernova you kept receiving alerts with every coincidence message, these will be
appended in this file with the sent time as the first key. An example (partly missing) can be
found here
Command Line Interface (CLI)
There also exists tools for command line interactions. These are explained in detail here
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for SNEWS_PT-0.0.1.dev360-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 8a0e824a97ed189939698096a41ca6f46e200431c4801782846749081bff6b2e |
|
MD5 | c1aa39a53450a4b7c79638ddf470b940 |
|
BLAKE2b-256 | 8a83c5a4a4e4acb5a36be548aeae87a439a3e0108a38ead52ceb4dd09d97ff37 |