Tools development version
Tools 0.4.5
- Fix how jobby parses slurm stdout/stderr files. (#122, @kopardev, @kelly-sovacool)
- The
run_jobby_on_*scripts are now deprecated in favor of usingjobbydirectly. (#123, @kelly-sovacool) - Use miniforge3 installation on biowulf. (#124, @kelly-sovacool)
Tools 0.4.4
- Fixes for
ccbr_tools install: (#98, @kelly-sovacool)- Correct installation method for
spacesavers2. - Change symlink permissions so group members can edit them.
- Correct installation method for
- Allow nextflow pipelines to specify walltime & memory usage for the slurm submission template. (#116, @kelly-sovacool)
Tools 0.4.3
- Add standalone script to retrieve HPC cluster name. (#93, @kopardev)
module_listnow retrieves the full module version. (#93, @kopardev)
Tools 0.4.2
- Fix
ccbr_tools install: add group write permissions after installation. (#89, @kelly-sovacool) - Updates to
nextflow.run(): (#90, @kelly-sovacool)- Allow pipelines to pass in module versions.
- This is needed for the next version of champagne which will use nextflow v25 for the new output directive, while other pipelines will continue to use the default nextflow module version.
- Ensure
nextflow.run()stops if the nextflow preview errors out.
- Allow pipelines to pass in module versions.
Tools 0.4.1
- Breaking change: Spooker now takes named options rather than positional arguments. (#85, @kelly-sovacool)
- jobby now omits log columns when the format is markdown, even if
--outerris used. (#86, @kopardev) - Minor documentation improvements. (#81, @kelly-sovacool)
Tools 0.4.0
New features
jobbyoverhaul (#59, #75, @kopardev)- Uses
saccountto get slurm job information, which should work for any HPC running slurm. - Has options
--tsv,--json, and--yamlto output the job information in those formats. If not specified, markdown is used. - Can accept a snakemake log file, nextflow log file, or a list of slurm job IDs as input.
- Can include the job stdout/stderr log files (
--outerr) for only failed jobs or all jobs (--include-completed). (#71, @kelly-sovacool)
- Uses
module_listis a new utility to list all loaded modules as JSON or retrieve the version of a specific module. (#63, @kopardev)ccbr_tools installhas new options: (#60, @kelly-sovacool)--typeto specify the type of tool to install (e.g.PythonTool,BashTool,Snakemake, orNextflow).--hpc(e.g.biowulf,frce) to specify the HPC environment for debugging purposes.
spookerupdates:- Now accepts the path to the pipeline CLI as an optional argument. (#69, @kelly-sovacool)
- Now bundles all pipeline information into a single gzipped JSON file. (#71, @kelly-sovacool)
- A tar archive is no longer created, as only one file is now needed.
- The tree & jobby information are included as strings in the JSON file.
- New shared conda environment on biowulf:
py3.11-8for ccbrpipeliner/8. (#79, @kelly-sovacool)
Bug fixes
- Fix
ccbr_tools install: use relative paths for symlinks within the same directory. (#58, @kelly-sovacool)
Tools 0.3.2
- Fix
ccbr_tools install: use the full path to the conda env on biowulf. (#64, @kelly-sovacool)
Tools 0.3.1
- Bug fixes in
ccbr_tools install:- ccbr_tools & ccbr_actions were using incorrect repo names. (#53, @kelly-sovacool)
- absolute paths were not being used for the symlinks. (#55, @kelly-sovacool)
- Minor documentation improvements. (#54, @kelly-sovacool)
Tools 0.3.0
- Allow relaxed version with only major and minor components in
match_semver()withstrict_semver=False. (#49, @kelly-sovacool) - Remove
argsand addrepoparameter toget_latest_release_tag()andget_latest_release_hash(). (#51, @kelly-sovacool)
Tools 0.2.4
- Fix
ccbr_tools.pipeline.nextflow.run: (#46, @kelly-sovacool)- make sure preview loads necessary modules.
- improve stack trace when nextflow command fails.
- New theme templates based on the FNL branding guide: (#47, @kelly-sovacool)
mkdocs-fnlfor websites built with mkdocs material.pkgdown-fnlfor R package websites built with pkgdown.
- Create helper to install software on supported HPCs.
- usage:
ccbr_tools install TOOL_NAME VERSION_TAG
- usage:
Tools 0.2.3
- Output ccbrpipeliner module version in spooker metadata. (#43, @kelly-sovacool)
- Spooker now correctly outputs metadata as a yaml file. (#43, @kelly-sovacool)
- Improvements to
ccbr_tools.pipeline.nextflow.run: (#44, @kelly-sovacool)- Use
-resumeby default and turn it off with--forceall. - Use
--outputoption. - Run
-previewbefore launching the pipeline with slurm. - When running on biowulf, try adding spooker to the PATH if it’s not available.
- Use
Tools 0.2.2
- Fix bug where spooker failed when more than 2 arguments were passed. (#41, @kelly-sovacool)
Tools 0.2.1
- Spooker update: accept pipeline version as an optional third positional argument. (#39, @kelly-sovacool)
- Bump cffconvert version for compatibility with nf-schema. (#38, @kelly-sovacool)
Tools 0.2.0
- new commands:
ccbr_tools send-emailfor sending emails from the command line. (#26, @kelly-sovacool)- With new helper function:
send_email.send_email_msg(). - Works when run from biowulf.
- With new helper function:
ccbr_tools quarto-addto add quarto extensions from this package. (#30, @kelly-sovacool)- Includes new format
fnlfor our documentation websites.
- Includes new format
- new functions for creating a contributors page for documentation websites:
github.print_contributor_images(). (#27, @kelly-sovacool) - new script from
CCBR/TaskManagement:github_milestones.sh. (#29, @kelly-sovacool) - documentation improvements:
- fix docstrings rendering – use Google style. (#25, @kelly-sovacool)
- overhaul navigation structure of docs website. (#28, @kelly-sovacool)
- style the website to follow FNL branding guidelines. (#30, @kelly-sovacool)
- miscellaneous minor improvements. (#32, @kelly-sovacool)
- bug fixes:
- include data files in package installation for
homologfinder. (#31, @kelly-sovacool)
- include data files in package installation for
Tools 0.1.4
- fix copy location for spook. (@kopardev)
Tools 0.1.3
- fix shared SIF cache directory spelling for biowulf. (#23, @kelly-sovacool)
Tools 0.1.2
- use major & minor version for docs website subdirectories. (#15, @kelly-sovacool)
- fig bug where
nextflow.run()did not import the correct HPC modules. (#20, @kelly-sovacool) - fix bug in
_get_file_mtime(). (#21, @kelly-sovacool)
Tools 0.1.1
- fix: don’t add extra newline to command stdout/stderr for
shell_run()andexec_in_context(). (#10, @kelly-sovacool) - minor docuemntation improvements. (#12, @kelly-sovacool)
Tools 0.1.0
The Tools repository is now restructured as a Python package. All previous python scripts which included command line utilities have been moved to src/, and all other scripts have been moved to scripts/. In both cases, they are available in the path when the package is installed.
Functions which were part of both XAVIER and RENEE are available for re-use in other bioinformatics pipelines for tasks such as determining the HPC environment, retrieving available genome annotations, and printing citation and version information. Explore the ccbr_tools reference documentation for more information: https://ccbr.github.io/Tools/reference/
CLI Utilities
Command-line utilities in CCBR Tools.
ccbr_toolsgb2gtfhfintersectjobbyjobinfopeek
Run a command with --help to learn how to use it.
External Scripts
Additional standalone scripts for various common tasks in scripts/ are added to the path when this package is installed. They are less robust than the CLI Utilities included in the package and do not have any unit tests.
add_gene_name_to_count_matrix.Raggregate_data_tables.Rargparse.bashcancel_snakemake_jobs.shcreate_hpc_link.shextract_value_from_json.pyextract_value_from_yaml.pyfilter_bam_by_readids.pyfilter_fastq_by_readids_highmem.pyfilter_fastq_by_readids_highmem_pe.pygather_cluster_stats.shgather_cluster_stats_biowulf.shget_buyin_partition_list.bashget_slurm_file_with_error.shgsea_preranked.shkaryoploter.Rmake_labels_for_pipeliner.shrawcounts2normalizedcounts_DESeq2.Rrawcounts2normalizedcounts_limmavoom.Rrun_jobby_on_nextflow_logrun_jobby_on_nextflow_log_full_formatrun_jobby_on_snakemake_logrun_jobby_on_snakemake_log_full_formatspookerwhich_vpn.sh
Tools 0.0.1
This tag marks the repository state from before refactoring it as a python package.