redvox 2.9.12

Library for working with RedVox files.

### RedVox Python SDK

This repository contains code for reading and working with the RedVox API 900 data format.

See: https://bitbucket.org/redvoxhi/redvox-api900-python-reader/src/master/docs/v2.9.10/redvox-api900-docs.md for SDK documentation.


## Changelog

### 2.9.12 (2020-07-17)

• Add support for 16 kHz audio streams

### 2.9.11 (2020-07-07)

• Fix bug where reading data that spanned across several days would leave out data towards end of time window

### 2.9.10 (2020-06-25)

• Authentication responses not return a copy of the claims for convenience

### 2.9.9 (2020-06-25)

• Update dependencies
• Encapsulate HTTP logic for HTTP Cloud Client
• Add tests for Cloud Client
• Add configurable timeout
• Add custom errors and better error handling
• Fix bug where connection would not be closed on authentication error

### 2.9.8 (2020-06-24)

• Large metadata requests are now chunked by the client.
• Change refresh token interval from 1 minute to 10 minutes
• Re-use HTTP client with keep-alive for more efficient HTTP requests
• Add chunked response for timing metadata request

### 2.9.5 (2020-06-23)

• Make Cloud API refresh token interval configurable.
• Allow Cloud API client to be used within “with” blocks for automatic closing of resources

### 2.9.2 (2020-06-11)

• Update dependencies (now dataclasses will only be pulled in on Python 3.6)

### 2.9.1 (2020-06-05)

• Fix bug where timing metadata was not converted into its associated data class.

### 2.9.0 (2020-06-05)

• Add full fledged cloud based API client. This client seamlessly manages authentication tokens behind the scenes.
• Update CLI data request methods to make use of new cloud based client.

### 2.8.7 (2020-06-03)

• Integrate ability to access extracted metadata from RedVox packets utilizing the cloud data API.

### 2.8.6 (2020-05-12)

• Add small HTTP interface to upcoming RedVox cloud API
• Added new sub-command to CLI data_req_report which takes a report ID and will download the report data for authenticated users

### 2.8.5 (2020-05-11)

• The auth_token CLI field for the data_req CLI command has been renamed to secret_auth to better reflect the fact that it is a shared secrete.
• CLI for data req now makes the shared secret auth key optional dependent on the settings of the remote server.

### 2.8.4 (2020-05-07)

• Add –protocol option to redvox-cli when making data request. This allows the data client to optionally connect over HTTP (mainly only useful for local testing)

### 2.8.3 (2020-05-06)

• Add mach_time_zero to TimeSyncData class
• Add best_tri_msg_indices to TimeSyncData class to identify which tri-message exchange indicated the best latency and offset
• Add validation checks to ensure that there is no change in sample rate or mach_time_zero in the analyzed packets
• Add check for change in mach_time_zero when identifying gaps

### 2.8.2 (2020-04-27)

• Add workaround for accessing mach_time_zero in incorrectly constructed Android packets

### 2.8.1 (2020-04-02)

• TriMessageStats will now append empty arrays in a tuple to tri_message_coeffs when the time sync sensor is empty or doesn’t exist.
• Functions compute_barometric_height and compute _barometric_height_array take additional optional arguments: surface temperature in K, molar mass of air in kg/mol, acceleration of gravity in m/s2 and the universal gas constant in (kg * m2)/(K * mol * s2)

### 2.8.0 (2020-03-31)

• Added a migration module that allows users to slowly begin migrating API 900 data towards API 1000. A flag can be set either through the API or by setting an environment variable ENABLE_MIGRATIONS=”1”. When enabled, all getters for numeric types will return floating point values (the only numeric type in API 1000).

### 2.7.9 (2020-03-25)

• Added properties: server_acquisition_times, packet_duration and mach_time_zero to redvox/api900/timesync/api900_timesync.py to assist with analyzing time sync data

### 2.7.8 (2020-03-23)

• Added a property: tri_message_coeffs to redvox/api900/timesync/api900_timesync.py to allow access to the tri-message coefficients
• Allowed access to the function evaluate_latencies_and_offsets in redvox/api900/timesync/api900_timesync.py

### 2.7.7 (2020-03-19)

• Added unit tests for redvox/common/stats_helper.py
• The functions mean_of_means, variance_of_means and mean_of_variance in redvox/common/stats_helper.py will not fail when sum(counts) is 0.

### 2.7.6 (2020-03-19)

• Added validation function that removes duplicated timestamps to redvox/api900/timesync/tri_message_stats.py
• Added get_latency_mean, get_latency_std_dev, get_offset_mean and get_offset_std_dev functions to redvox/api900/timesync/api900_timesync.py
• Updated unit tests

### 2.7.5 (2020-03-13)

• Expose new package redvox/api900/qa
• Expose new gap detection module redvox/api900/qa/gap_detection.py
• Provide a public identify_time_gaps method

### 2.7.1 (2020-03-03)

• Add and define useful constants to redvox/common/constants.py
• Add cross-correlation functions to redvox/common/cross_stats.py
• Add date time utilities to redvox/common/date_time_utils.py
• Add rdvxz file statistics functions to redvox/common/file_statistics.py
• Add statistical helper functions to redvox/common/stats_helper.py
• Add new time synchronization package at redvox/api900/timesync
• Add a location analyzer to redvox/api900/location_analyzer.py
• Updated unit tests and documentation
• Updated dependencies to latest versions
• Added additional code QA (pylint, mypy, coverage)

### 2.6.1 (2020-01-28)

• Update bulk data download client to utilize a RedVox provided authentication token

### 2.6.0 (2019-12-11)

• Update the redvox-cli

  • The CLI is now installed when this SDK is installed and available to the user as `redvox-cli`
  • Added improved logging and error handling
  • Added ability to download bulk RedVox data sets

### 2.5.1 (2019-10-23)

• Add top level setter for mach_time_zero

  • WrappedRedvoxPacket.set_mach_time_zero(self, mach_time_zero: int) -> ‘WrappedRedvoxPacket’

### 2.5.0 (2019-10-17)

• Add top level getters and setters for accessing time synchronization metrics stored in RedVox Packet metadata.

  • WrappedRedvoxPacket.best_latency(self) -> typing.Optional[float]
  • WrappedRedvoxPacket.set_best_latency(self, best_latency: float) -> ‘WrappedRedvoxPacket’
  • WrappedRedvoxPacket.best_offset(self) -> typing.Optional[float]
  • WrappedRedvoxPacket.set_best_offset(self, best_offset: float) -> ‘WrappedRedvoxPacket’
  • WrappedRedvoxPacket.is_synch_corrected(self) -> bool
  • WrappedRedvoxPacket.set_is_synch_corrected(self, is_synch_corrected: bool) -> ‘WrappedRedvoxPacket’

• Add shortcut for adding metadata to a RedVox Packet (previously the entire metadata needed to be set at a time)

  • WrappedRedvoxPacket.add_metadata(self, key: str, value: str) -> ‘WrappedRedvoxPacket’

### 2.4.0 (2019-10-8)

• Add mach_time_zero accessor to WrappedRedvoxPackets.

  • WrappedRedvoxPacket.mach_time_zero(self) -> typing.Optional[int]

### 2.3.0 (2019-9-25)

• concat._identify_gaps now only checks for dropped data from sensors. It checks for timing continuity by ensuring that the gap between packets is no larger than a configurable amount for a given sample rate.
• concat._identify_sensor_changes was added to identify sensor changes such as change in sample rate, change in sensor name, change in data type, or missing sensor data

### 2.2.1 (2019-5-14)

• Added stat utils tests and updated function
• Edited documentation typos; 2.2.0 documentation still valid for 2.2.1

### 2.2.0 (2019-4-26)

• Add sensor timing correction
• Update documentation
• Fix more cyclic dependency issues

#### 2.1.1 (2019-4-24)

• Start and end timestamps are now optional when reading .rdvxz files from a range. For timestamps that are not supplied, the timestamps are parsed from the data file names to find the earliest and latest timestamp.
• Fixed a bug that created a cyclic dependency between the reader and concat modules.

### 2.0.0 (2019-4-12)

• Add ability concatenate multiple .rdvxz files together [(docs)](https://bitbucket.org/redvoxhi/redvox-api900-python-reader/src/master/docs/v2.0.0/redvox-api900-docs.md#markdown-header-concatenating-wrappedredvoxpackets)
• Add ability to identify gaps in continuous data when concatenating multiple files
• Read a range of .rdvxz files from a directory with a given time window and optional redvox ids to filter against [(docs)](https://bitbucket.org/redvoxhi/redvox-api900-python-reader/src/master/docs/v2.0.0/redvox-api900-docs.md#markdown-header-loading-redvox-api-900-files-from-a-range)
• Read a range of .rdvxz files from a structured directory with a given time window and optional redvox ids to filter against
• Deprecated several public API methods for setting and accessing sensor fields
• Add objects for summarizing RedVox data ranges [(docs)](https://bitbucket.org/redvoxhi/redvox-api900-python-reader/src/master/docs/v2.0.0/redvox-api900-docs.md#markdown-header-summarizing-wrappedredvoxpackets)
• Add ability to plot summary RedVox data ranges
• Refactor reader.py into several modules
• Update dependency versions

### 1.5.0 (2019-3-20)

• Add setters for all fields
• Add the ability to easily create sensor channels and RedVox packets
• Add CLI that * Converts .rdvxz files to .json files * Displays the contents of .rdvxz files
• Add ability to compare files and sensor channels
• Update documentation and API documentation
• Add more examples

### 1.4.1 (2019-2-15)

• Update required libraries
• Add ability to get original compressed buffer from WrappedRedvoxPacket
• Add utility functions for LZ4 compression

### v1.4.0 (2018-12-5)

• Added support for serializing to/from JSON
• Fixed bug where has_time_synchronization_channel() would return true even if the payload was empty