kasperl 0.0.3

seppl-based Python library with generic plugins for pipelines.

seppl-based (https://github.com/waikato-datamining/seppl) Python library with generic plugins for pipelines and building blocks for creating command-line tools for pipelines.

Changelog

0.0.3 (2026-05-05)

• the load_pipeline method now automatically expands environment variables ($name or ${name}, under Windows also %name%) to have the same flexibility as when explicitly specifying the command-line arguments

0.0.2 (2026-04-22)

• the locate_file method now supports looking for files with different image/annotation prefixes

• added croniter dependency

• added the cron dummy reader which outputs a string according to the provided execution expression (e.g., every 10 seconds or every 5 minutes on workdays)

• load_pipeline removes comments now

• added the log-data filter for logging information about the data passing through

• the set-metadata filter now expands placeholders in the value if of type ‘string’

• the rename filter now allows applying regexp/group expansion to the name

• added the count-data filter for counting data items

• added BytesSupporter mixin

• added the shell-exec reader for executing arbitrary external commands

• added the get-metadata filter to extract field values from the meta-data

• added the sleep filter for waiting specified number of seconds before forwarding data

• added –log_execution_time flag to sub-process, tee, trigger

• parse_conversion_args and perform_pipeline_execution now support default placeholders HOME/CWD/TMP

• generators csv-file, dirs, files and text-file now support placeholders in their paths

• added log-placeholder filter, which can be used for logging placeholder values

• added sanitize-name filter, for removing unwanted characters from file names

• added more logging info to to-text-file writer

• using seppl.variables now

• requiring seppl>=0.3.1 now

0.0.1 (2025-10-31)

• initial release