LUNA PLATFORM v.5.42.0#
Summary#
-
Now the Image Store service can store any files as objects (for example, a video file).
Previously, only the following types of objects were available:
json
,text
,zip
,pdf
.You can upload files using the "create objects" request. The bytes of the file must be specified in the request body, and the
Content-Type
header must contain the MIME type of the file (for example,video/mp4
). The response to the "get object" request contains the header Content-Disposition. This header contains the file name of the attachment object (for example,video_1.mp4
). The file name is generated based on theobject_id
and the MIME type.If the MIME type of the file could not be determined, then the file extension will be set as `.bin'.
Now the
Accept
header is ignored for the "get object" request. Previously, the service returned the error12023, Content type is unacceptable
with the status code 406, if the object content type did not match the MIME type of the file. -
Now, when executing any request with correct authorization, information about the corresponding account is displayed in the logs of the API service.
This functionality enables you to determine who exactly executed a particular request. This may be required for information security and system administrators.
If the request was executed with BasicAuth or LunaAccountIdAuth type authorization, the following message will be displayed in the logs:
Request invoked by user (account_id: '270531af-e52e-4538-9181-628d9900a0db')
If the request was executed with BearerAuth type authorization, the following message will be displayed in the logs:
Request invoked by user (account_id: '270531af-e52e-4538-9181-628d9900a0db' token_id: 'd57e16f5-e243-47d2-aa85-8b200c12d86f')
If the request was executed without authorization, the following message will be displayed in the logs:
Request invoked by user (account_id: null)
The logs of the Accounts service additionally display information about the creation of tokens by specific users:
User with account_id: '270531af-e52e-4538-9181-628d9900a0db' create token: 'd57e16f5-e243-47d2-aa85-8b200c12d86f'
Logging information about the creation of tokens enables you to track where the token came from and which user it belonged to, even after it was deleted.
-
A new check
shoulders_position
has been added to theface_quality
andiso
sections, which determines the most predominant state of the shoulder position from the following:- non-parallel
- parallel
- hidden
-
The execution of the "get faces" request with all
target
fields has been sped up. -
The memory usage by the Task service has been reduced in some cases of using the "Cross-matching" task.
Fixed errors
-
Fixed an error where the script
db_create.py
didn't work for non-default configuration values of the Configurator service. -
Fixed an error where the cache warmed up when starting the Python Matcher service with the "DESCRIPTORS_CACHE" setting turned off.
-
Fixed an error where the Python Matcher service started and continued to work when some workers stopped with an error.
A "worker" is the Python Matcher service process that handles incoming (HTTP) requests.
-
Fixed behavior that could result in the deletion of one or all new event subscriptions if the new subscription was created and the old one was deleted at the same time.
-
Fixed an error where using EXIF data evaluation (parameter
use_exif_info
) could lead to an error of the formInternal server error
.
API (v.6.14.2 - v.6.15.0)#
Improvements#
-
The service allows store an arbitrary files as object now. Following object types was allowed for storing before:
json
,text
,zip
,pdf
.The service accepts a request for storing a file:
POST:/objects
. The file bytes should be set as request body andContent-Type
header should be contains a file mimetype.See create object , get object .
-
A request response to resource
GET:/objects/{object_id}
contains Content-Disposition header now. This header contains an attachment object filename. This feature allows browsers to save objects to PC as files with correct extensions. Filename is generated based onobject_id
and objectmimetype
.See get object .
-
An
Accept
header is ignored for a requestGET:/objects/{object_id}
now. Service returned error12023, Content type is unacceptable
with status code 406 before if object content type did not match with an object mimetype.See get object .
-
Add record to log that the account made the request to endpoints.
- Add
account_id
to record if requests with authorization:
Request invoked by user (account_id: '270531af-e52e-4538-9181-628d9900a0db')
- Add
account_id
withnull
value if request without authorization:
Request invoked by user (account_id: null)
- Add
token_id
to record if authorization with token:
Request invoked by user (account_id: '270531af-e52e-4538-9181-628d9900a0db' token_id: 'd57e16f5-e243-47d2-aa85-8b200c12d86f')
- Add
-
Service dependencies were updated.
-
A
shoulders_position
check was added to iso and face quality estimations.See perform verification , createVerifier , createHandler , generate event , iso , ws resources for details.
Faces (v.4.8.5 - v.4.8.6)#
Improvements#
-
Service dependencies were updated.
-
A getting faces with all targets was speed up.
Image Store (v.3.8.5 - v.3.9.0)#
Improvements#
-
The service allows store an arbitrary files as object now. Following object types was allowed for storing before:
json
,text
,zip
,pdf
.The service accepts two request for storing a file:
POST:/buckets/{bucket}/objects
orPUT:/buckets/{bucket}/objects/{object_id}
. The file bytes should be set as request body andContent-Type
header should be contains a file mimetype.See put object , create object , get object .
-
A request response to resource
GET:/buckets/{bucket}/objects/{object_id}
contains Content-Disposition header now. This header contains an attachment object filename. This feature allows browsers to save objects to PС as files with correct extensions. Filename is generated based onobject_id
and objectmimetype
.See get object .
-
An
Accept
header is ignored for a requestGET:/buckets/{bucket}/objects/{object_id}
now. Service returned error12023, Content type is unacceptable
with status code 406 before if object content type did not match with an object mimetype.See get object .
-
Service dependencies were updated.
Accounts (v.0.1.5 - v.0.1.6)#
Improvements#
-
Service dependencies were updated.
-
Add record to log that the account made the request crate token .
User with account_id: '270531af-e52e-4538-9181-628d9900a0db' create token: 'd57e16f5-e243-47d2-aa85-8b200c12d86f'
Tasks (v.3.15.2 - v.3.15.3)#
Improvements#
-
Service dependencies were updated.
-
Filtering by null was added for several events filters:
-
meta
-
source
-
emotion
-
mask
-
ethnic_group
-
liveness
-
gender
-
apparent_gender
-
headwear_state
-
sleeve_length
-
headwear_apparent_colors
-
upper_clothing_colors
-
lower_garment_type
-
lower_garment_colors
-
shoes_apparent_color
-
backpack_state
-
city
-
district
-
street
-
house_number
-
area
-
geo_position
-
track_id
Actual for next tasks:
-
Updated internal cross-matching task mechanism: reduced memory usage in most cases cross matching task .
Events (v.4.8.5 - v.4.8.6)#
Improvements#
- Service dependencies were updated.
Fixed bugs#
- Add missing
headwear_apparent_color
nullable filter for get stats
Configurator (v.2.1.10 - v.2.1.16)#
Improvements#
- Service dependencies were updated.
Fixed bugs#
- Fix running
base_scripts/db_create.py
script with non-default config.
Sender (v.2.6.5 - v.2.6.6)#
Improvements#
-
Service dependencies were updated.
-
A
shoulders_position
check was added to iso and face quality estimations. See /ws for details.
Admin (v.5.2.2 - v.5.2.3)#
Improvements#
- Service dependencies were updated.
Licenses (v.0.6.8 - v.0.6.13)#
Improvements#
- Service dependencies were updated.
Handlers (v.2.17.2 - v.2.18.0)#
Improvements#
-
Filtering by null was added for candidate events for resources create handler , generate events
-
A
shoulders_position
check was added to iso and face quality estimations.
See perform verification , createVerifier , createHandler , generate event , iso resources for details.
Python Matcher and Python Matcher Proxy (v.1.5.1 - v.1.5.2)#
Improvements#
- Service dependencies were updated.
Fixed bugs#
-
Filtering by missing meta (meta=null) was added for candidate events for resources face matching , body matching , face cross matching and body cross matching .
-
Fail application in case any of workers has died.
-
Issue with warming descriptors cache in case its disabled was fixed.
Backport3 (v.0.8.1 - v.0.8.2)#
Improvements#
- Service dependencies were updated.
Fixed bugs#
- The behavior where creating a new subscription(s) and deleting an old one could be triggered by the service at the same time and potentially result in the loss of one of the new subscriptions has been fixed.
Backport4 (v.1.3.5 - v.1.3.6)#
Improvements#
- Service dependencies were updated.