Releases: chaoss/grimoirelab-elk
Releases · chaoss/grimoirelab-elk
0.103.0-rc.3
grimoire-elk 0.103.0-rc.3 - (2022-08-23)
New features:
- Configurable study aliases
Before this fix, the name of the study aliases was hard coded. From
now on, these values can be configured by passing them as parameters. - Entry points to support enriching plugin mechanism
This plugin mechanism allows ELK to use third-party enriching
backends. Developers only need to include the next snippet in their
package configuration and ELK will import them as any other core
backend.entry_points={"grimoire_elk": "mybackend = package.subpackage:get_connectors"},
NOTE: In the previous
example,get_connectors
is a function that returns the backend
classes needed to enrich a datasource. Take the next snipped as an
example:def get_connectors(): return {"backend": [MyBackend, MyBackendOcean, MyBackendEnrich, MyBackendCommand]}
Bug fixes:
- [raw/slack] Handle fields in "data.file"
Avoid the 'Failed to insert data to ES' error when a document contains
at least one immense term in 'data.file' (whose UTF8 encoding is
longer than the max length 32766).
0.103.0-rc.2
grimoire-elk 0.103.0-rc.2 - (2022-07-22)
New features:
- Entry points to support enriching plugin mechanism
This plugin mechanism allows ELK to use third-party enriching
backends. Developers only need to include the next snippet in their
package configuration and ELK will import them as any other core
backend.entry_points={"grimoire_elk": "mybackend = package.subpackage:get_connectors"},
NOTE: In the previous
example,get_connectors
is a function that returns the backend
classes needed to enrich a datasource. Take the next snipped as an
example:def get_connectors(): return {"backend": [MyBackend, MyBackendOcean, MyBackendEnrich, MyBackendCommand]}
- Configurable study aliases
Before this fix, the name of the study aliases was hard coded. From
now on, these values can be configured by passing them as parameters.
Bug fixes:
- [raw/slack] Handle fields in "data.file"
Avoid the 'Failed to insert data to ES' error when a document contains
at least one immense term in 'data.file' (whose UTF8 encoding is
longer than the max length 32766).
0.103.0-rc.1
grimoire-elk 0.103.0-rc.1 - (2022-07-21)
New features:
- Entry points to support enriching plugin mechanism
This plugin mechanism allows ELK to use third-party enriching
backends. Developers only need to include the next snippet in their
package configuration and ELK will import them as any other core
backend.entry_points={"grimoire_elk": "mybackend = package.subpackage:get_connectors"},
NOTE: In the previous
example,get_connectors
is a function that returns the backend
classes needed to enrich a datasource. Take the next snipped as an
example:def get_connectors(): return {"backend": [MyBackend, MyBackendOcean, MyBackendEnrich, MyBackendCommand]}
- Configurable study aliases
Before this fix, the name of the study aliases was hard coded. From
now on, these values can be configured by passing them as parameters.
Bug fixes:
- [raw/slack] Handle fields in "data.file"
Avoid the 'Failed to insert data to ES' error when a document contains
at least one immense term in 'data.file' (whose UTF8 encoding is
longer than the max length 32766).
0.102.0
GrimoireELK 0.102.0 - (2022-06-24)
Bug fixes:
- Poetry build version updated
The current version of thepoetry
builder was old and caused some
problems when installing packages in developer mode. With this new
version, it's possible to install ELK in developer or editable mode
without usingsetuptools
files (i.e.setup.cfg
andsetup.py
).
Others:
- GrimoireLab dependencies update
Dependencies have been updated to use the latest versions of
GrimoireLab packages:- grimoirelab-toolkit >= 0.3
- perceval >= 0.19
- perceval-mozilla >= 0.3
- perceval-opnfv >=0.2
- perceval-puppet >= 0.2
- perceval-weblate >= 0.2
- graal >= 0.3
- cereslib >= 0.3
0.101.1
GrimoireELK 0.101.1 - (2022-06-03)
Bug fixes:
- Pair programming regular expression improvement
The regular expression to detect pair programming authors in git
datasource has been improved adding more connectors. The following
list shows the current connectors: -[aA][nN][dD]
-&
-+
- Fix fetching authors with min and max dates
The old query only returns 10000 items due to ElasticSearch and
OpenSearch now usecomposite
aggregation to paginate all buckets.
0.101.0
GrimoireELK 0.101.0 - (2022-06-03)
New features:
- Anonymize parameters
A new API functionanonymize_params(params)
allows to anonymize a
list of parameters that developers might consider secrets. It will
only take effect for those parameters which their names are defined on
SECRET_PARAMETERS
constant. - Dependencies file for development
Include a new requirements.txt file to install the development version
from the repositories. - Support to add identities to SortingHat from Enrich class
New methodsadd_identities
andadd_identity
are available on
Enrich
class to add new identities to aSortingHat
database.
Libraries usingelk
won't need to usesortinghat
library to store
new identities while enriching items. - Collect and enrich confluence by spaces
Users can collect and enrich Confluence data from specific spaces
only, adding the option--spaces=[SPACE1, SPACE2, ...]
to their
projects.json
. The following example shows how to collect and to
enrich data from only theDEV
andSUPP
spaces only.{ "project": { "confluence": [ "http://example.com --spaces=[DEV, SUPP]" ] } }
- Add demographic study to mattermost
Add the demographic study to the mattermost data source.
Bug fixes:
- Skip sources with wrong arguments
Sources won't interrupt collection or enrichment processes when any of
their arguments are invalid. Instead, they will be skipped. - Git fix 'is_git_commit_multi_author'
This fixes theis_git_commit_multi_author
field when the commit has
more than one author andpair-programming
is enabled. This is
important for calculatingpair_programming_*
values. - SortingHat as a required package
SortingHat is required for running grimoireelk but it was set as an
extra dependency. From now on sortinghat is always installed with
grimoireelk. - Confluence credentials not stored in raw indexes
Credentials for Confluence datasource were stored in raw indices as
part of the URL. For now on, credentials that are part of URLs are
removed before any data is stored.
Feature removals:
- Drop Python 3.6 support
Python 3.6 reached the end of life at the end of 2021. This means it
won't receive new updates or patches to fix security issues.
Therefore, this package will only work with Python >= 3.7 from now on. - FINOS Meeting backend removed
FINOS Meeting backend is not supported anymore. The Perceval backend
was archived in Bitergia repository so it shouldn't be part of the
core anymore because it's software not maintained.