The format is based on Keep a Changelog.
- liftet upperbound < 2.0 on pandas dependency
- tagpack without any tags does not raise an validation error anymore
- update coinaddrvalidator dep to avoid error on install
- support for python 3.11
- fix cluster mapping errors on invalid addresses. Instead just skip them.
- avoid errors with numpy 2.0 setting version restriction to < 2.0
- avoid install errors by setting proper python version upper bound to 3.10
- clustermappings for eth did not work
- add new concept for tokens (defi_token)
- allow all concepts to be used in category field (align category and concepts)
- switched to calver
- add cluster mapping for tron
- new --create-db flag for init and sync which tries to automatically create the database if it does not exist
- Bump cassandra driver version 3.27, import lz4 to enable cassandra compression
- setup automatic pypi publish with github actions
- cluster mapping does not use new rerun-cluster-mapping-with-env env
- error on insert when no tagpacks are loaded
- parallel tagpack insert (-n-workers parameter, default 1 worker)
- new --rerun-cluster-mapping-with-env flag on sync command to update all cluster mappings
- old concepts (entity and abuse) have precedence over new concepts (concepts.yaml)
- new db field concepts (list) to store multiple categories with a tag (requires db-resync)
- fix bug in ks_map handling
- fix field not found error in keyspace check
- handle if keyspace is not found gracefully
- new confidence score forensic_investigation (70), #87
- cluster mapping can now use gs-lib config for import (--use-gs-lib-env option) #84
- Sync command has option to directly run cluster mapping via gs-lib (--run-cluster-mapping-with-env) #84
- better handling for large yaml tagpack files #85
- Support for altering postgres config params in docker-compose setup #83
- Using cSafeLoader for yaml files when possible for better performance
- Deprecate is_public in config.yaml, in favor of cmd flag #82
- Fixed compatibilty issues with python 3.8 (importlib.files)
- Full support for supporting ActorPacks #41
- actor pack validation and insertion for the extensive actor pack in public repository
- interactive process of adding suitable actors to existing tagpacks
- calculation of data quality measures
- Auto update data feature to streamline insertion of updates
- clone tagpack repositories and insert all tags and actors in the TagStore #73
- new confidence scores unknown (5) heuristic (10), ledger_immanent (100)
- add --by-currency option for tagstore composition query
- add --csv option for selected commands
- add tag-version pseudo target to Makefile
- Harmonised command structure #59
- Moved TagPack documentation to GraphSense public tagpacks repo
- Removed unnecessary columns in cluster mapping table #45
- ETH addresses are normalized to lower-case before inserting them to the DB #39
- Parallel import of cluster mapping #4
- Improve
cluster_defining_tags_by_frequency_and_maxconfidence
view
- Command to show tagstore content composition: listing of creators and their contributions
- Command to calculate tag quality measures #49
- Address validation for currencies supported by
coinaddrvalidator
library #22 - Confidence scoring handling: ingest confidence scores from local file #35
- Add
colorama
dependency to disable coloring on file redirect
- Support for PostgreSQL connection pooling
- Removal of duplicate tags
- Fix cluster mapping of ETH addresses without external txs
- consolidate tagpack-level properties
conda
setup
- Optional prefix for tagpack
- View for tag count by cluster
- View for cluster defining tags by frequency and maxconfidence
- Bug with
--add_new
flag
- Improved ingest: improved schema validation feedback for user, remove duplicates in tagpacks
- Generate tagpack base URI automatically (no longer necessary to keep config.yaml up-to-date)
- Updated confidence.csv
- YAML file inclusion to share a header file between multiple tagpack files
- Encourage clean tagpack repository when ingesting tagpacks (can be disabled with command line options
--no_strict_check
and--no_git
) - URI field in TagStore database to support backlink to tagpack repository in the dashboard
- Option to ingest all taxonomies at once instead of one taxonomy at a time
- Command-line option
--config
to supply path to config file - Command-line option
--force
to force re-ingest if tagpack already is present in database - Command-line option
--add_new
to skip over ingested tagpacks and ingest only new ones
- Tagstore design: migrate tag handling from Cassandra keyspaces to external tag store(s)
- Tagpack validation of confidence value, which is now categorical instead of numerical
- Check for duplicate entries
- Database view for tag statistics
- Entity-related components
- Bug in tag validation
- Cersions of third party library dependencies
- Additional schema fields
- Command line tool usability fixes
- Support for dev version numbers
- Switched to GitHub action workflows
- Added support for entity tags
- Move TagPack documentation to TagPack rep
- Update package dependencies
- Add additional TagPack validation tests
- Add support for Entity Tags
- Removed unnecessary lookup tables
- Fixed PEP8 warnings (
flake8
)
- Separated TagPack Management Tool from public TagPacks
- Refactored scripts into TagPack Management tool
- Re-implemented validation and ingest procedures
- Support for ingestion / validation of remote taxonomy concepts
- New tagpacks
- Abuses field
- Splitted config from schema
- Improved argparse
- Renaming categories
- Jupyter notebooks
- Tagpacks: walletexplorer, ransomware, sextortion (Talos), miners
- Schema creation, validate and ingest scripts
- Documentation, License, etc.