8.1.3 (2024-01-15)#

Bug Fixes#

  • bug with preemptible rules (#2616) (c6d7141)

  • do not pass snakefile as metadata when wms monitor flag is used (#2573) (13b3205)

  • use default group settings if not execution workflow (fixes attribute error occurring with –report) (#2617) (21e9964)

8.1.2 (2024-01-12)#

Bug Fixes#

  • local mtime handling in case of storage plugins and cleaner error message for parallel storage retrieval (#2611) (880b264)

  • Migrate away from deprecated pulp API (#2610) (fb26640)

8.1.1 (2024-01-11)#

Bug Fixes#

  • deduplicate input files before retrieval from storage (#2600) (37cf475)

8.1.0 (2024-01-08)#


Bug Fixes#

  • add mamba to docker image (eb0c884)

  • correctly report lineno (#2584) (967a0d7)

  • move apptainer into separate env in docker image (94e9e2c)

  • single line f-string format error in py3.12 (#2588) (87c06c0)


8.0.1 (2023-12-21)#

Bug Fixes#

  • remove bash completion entrypoint (no longer supported, was too slow to be usable anyway) (922b53a)


  • fix cli options rendering (264c1a9)

  • fixes in migration guide (f8adefa)

8.0.0 (2023-12-20)#


Snakemake 8 marks the beginning of decomposing Snakemake into a framework of plugins. This enables the democratization of method development within the Snakemake ecosystem. We start with plugins for storage and execution backends. In the future, there will be plugins for the scheduling, metadata, software deployment, reporting, and many more. This way, it will be possible to easily launch and explore new developments in workflow management and reproducible data analysis without the need to get your work merged into the main codebase of Snakemake and also without the need to develop a new workflow management system as a proof of concept.

In detail, Snakemake 8 introduces the following changes. Unfortunately it was unavoidable to break some usages (we apologize). Nevertheless, we tried to ensure that every removed or modified feature has been replaced with an equivalent reimplementation, as outlined in our migration docs. While Snakemake 8 has an even more thorough testing framework than any release before, and while it has been quite heavily tested in practice by us, you might initially experience bugs and glitches for which we want to apologize beforehand. We think that the massive codebase improvements are worth it in the long run, and hope that everything goes well. As always, any pull requests with test cases and pointers to bugs are more than welcome.

Detailed breaking changes#

  • removed the long time ago deprecated support for dynamic, version, and subworkflow (see the migration docs)

  • migrated old remote providers into storage plugins (see the migration docs)

  • migrated execution backends into plugins, including a change in the respective command line interfaces (see the migration docs)

  • deprecates --use-conda and --use-singularity in favor of --software-deployment-method conda or --software-deployment-method apptainer and --software-deployment-method conda apptainer (see the migration docs)

  • profile support is now versioned, such that different profiles can be written for different minimum Snakemake versions (see the migration docs)

  • redesigned Snakemake API. It now uses a modern, dataclass based approach (see the migration docs)


  • add ability to inject conda environments into running Snakefile (#2479) (6140e29)

  • add functionality for deploying sources if no shared FS is assumed (#2486) (76eac3c)

  • add option to control software deployment mode (shared or non shared FS) (#2525) (04ec2c0)

  • allow detailed configuration of shared FS usage (#2528) (0d34be9)

  • allow environment variables in string values of profile (e.g. paths may now contain elements like $USER). (58dc70c)

  • allow python expressions in –set-resources (#2521) (022a31e)

  • allow to set latency_wait in executor test suite (c0bca0b)

  • automatically upload workflow sources to default storage provider if no shared FS is used (a450c49)

  • Faster ci test setup (#2489) (4798e8a)

  • implement precommand (#2482) (ff0f979)

  • redesigned Snakemake API. It now uses a modern, dataclass based approach (#2403) (2be3bfa)

  • support for external executor plugins (#2305) (c9eaa4e)

  • version specific profile config files (profile/config.v8+.yaml with profile/config.yaml as fallback that matches any version) (#2498) (47e5811)

Bug Fixes#

  • adapt to changes in snakemake-interface-executor-plugins (635c68a)

  • add storage provider args to deploy sources command (67178e3)

  • add testcase for script directive to work with Python 3.7 and corresponding fix. (0b4ae2e)

  • allow pepfile and pepschema to take pathlib (#2546) (ca91661)

  • also inherit rule proxies if there is no rulename modifier specified in a use rule statement (#2440) (1570289)

  • assume at most 8GB memory for default resources. This way, we avoid exploding memory requirements for large input files that are very unlikely to be put entirely into memory by any tool. (11c2ecc)

  • comparison to float in scheduler (ef44d84)

  • detect job paths that leave and then enter a group. Such paths are invalid because then the group depends on itself. (#2527) (5383a4d)

  • ensure that auto deployment of default storage provider works in containers with read-only root home. (1a347ff)

  • ensure that log and benchmark files are uploaded to storage as well (#2545) (6aabb5d)

  • ensure that targetjob is always forced. This fixes a bug causing run-directive rules to not being executed even when enforced via e.g. -R. (#2448) (b2a60d5)

  • fix cache handling and unlock handling (2f4d5e1)

  • fix nargs definition for –deploy-sources (fc252c8)

  • fix path handling when detective profiles (fe63881)

  • fix storage handling on windows by converting all paths to posix paths (#2519) (7864a76)

  • handle different f-string tokens in py3.12 (#2485) (f2c7613)

  • handle storage for local jobs; add test case (6d978ef)

  • handling of group jobs when obtaining temp input files (71be1de)

  • import (#2402) (2c831f1)

  • improved error handling for storage upload; fixed bugs caused by outdated calls to IOFile.exists(). (720bb84)

  • improved error messages in case of invalid storage queries (9671fd0)

  • in addition to localrules statement, infer that job is local if it has any input or output file that is marked as local (#2541) (e8b682b)

  • only deactivate conda inject envs upon workflow tear down (#2503) (e6dfdd4)

  • Panoptes –wms-monitor-arg (#2444) (98d2bdf)

  • proper reuse of rule proxies when importing several times from the same module (#2404) (e867dda)

  • Restore backward compatibility for Google Life Sciences executor (#2461) (5e3a464)

  • shadow “full” mode ignore symlinks (#2516) (1d58120)

  • show failed logs in executor testcases (92f7bf4)

  • Slack log service (#2537) (26eb4ba)

  • sort report (sub-)categories in lexicographical order (#2449) (d0705ad)

  • update minimum snakemake-interface-storage-plugins version (0ef7226)

  • use temporary directory (faster, more likely local, always writable) for persistence and source cache in case of remote execution without shared fs (#2502) (c8fa7ba)

  • wait for logs before showing them on error (a4ff328)


  • document name directive with example (#2534) (cce5551)

  • fix syntax in cluster example (#2460) (64e9645)

  • notes on arm based machines in tutorial docs (0586f04)

  • rust: Fix typo on rust-script version (#2488) (a79dd94)

7.32.4 (2023-08-18)#

Bug Fixes#

  • always sort report (sub-)categories in lexicographical order

  • also inherit rule proxies if there is no rulename modifier specified in a use rule statement

  • ensure that targetjob is always forced. This fixes a bug causing run-directive rules to not being executed even when enforced via e.g. -R.

7.32.3 (2023-08-07)#

Bug Fixes#

  • fix bug occuring when using multiple use rule statements in combination with the rules object for referring to output of already defined rules.

7.32.2 (2023-08-07)#

Bug Fixes#

  • unnecessary set Snakefile in AzBatch executor (#2397) (78e6d6e)

7.32.1 (2023-08-05)#

Bug Fixes#

  • add missing spaces between lines that get concatenated. (#2268) (7238458)

  • better message about profile usage upon execution (#2391) (cf8aea5)

  • do not overwrite default resources setting in azure batch executor (#2395) (4aef3b9)

  • updating of non-dict config values gives error (#2364) (b33aeec)

  • wrong rule names when nesting module imports (#1817) (65c79a4)


  • basics.rst: suggest VS Code instead of deprecated Atom as IDE (#2368) (1357316)

7.32.0 (2023-08-03)#


  • add support for Kubernetes service account name spec (#2254) (3370426)

Bug Fixes#

  • Enable values with an = sign in default_resources (#2340) (c1c9229)

  • Escape workdir paths for potential spaces in paths (#2196) (9261f7e)

  • ga4gh executor resources (#2042) (ad6eaef)

  • print exceptions when job is not a shell job (#2385) (8a37b85)

  • remote-azblob-sasToken-Authorization (#1800) (bc854a7)

  • wms-monitor now gets data in correct json format (#2347) (7fafa7a)


7.31.1 (2023-08-02)#

Bug Fixes#

  • require python >=3.7 again (the python 3.9 dependency was unnecessary) (#2372) (0d0e9c4)


  • update add minimum Python version bump (#2370) (48e934d)

7.31.0 (2023-07-26)#


  • Add support for Google Service Accounts and GCE VM network configuration (#2318) (2b754aa)

7.30.2 (2023-07-20)#

Breaking changes#

  • Bump minimum Python version from 3.7 to 3.9 (#2369) (4608163)

Bug Fixes#

  • do not allow setting benchmark and between-workflow caching for the same rule. The reason is that when the result is taken from cache, there is no way to fill the benchmark file with any reasonable values. (#2335) (e2d64fa)

  • ensure lazy evaluation of resource functions/callables (this also entails, for now, a removal of the thread statistics in the yellow job stats table); further, added some clarifying sentences about resource function evaluation to the docs (#2356) (4c591b7)

  • handle non-PEP440 versions of apptainer/singulariy (#2337) (dea6ba8)

  • remote GS builds too many inventories; io:collect_mtime always uses uncached mtime (#2266) (bad9115)

  • Solve apptainer version issue (#2333) (a876e0f)

  • SyntaxWarnings due to non-raw regex pattern strings (#2359) (a08c0b0)


  • clarify minimum Snakemake version for profiles (86dc277)

  • clarify the channel priority in environment definition deployment.rst (#2352) (76aa964)

  • fix typo (stackoverflow issue) (#2365) (f770984)

  • note on using checkpoint mechanism only for input function, not for params or resources. (#2353) (4be2f9d)

7.30.1 (2023-06-28)#

Bug Fixes#

7.30.0 (2023-06-28)#


  • allow profiles to be YTE templates; adapt to eido 2.0 (#2325) (67d9ff2)

7.29.0 (2023-06-21)#


  • introduce –workflow-profile for additional workflow specific profiles that overwrite global profiles; add ability to define key-value CLI flags like –set-threads or –set-resources as multi-level dictionaries in profile config yaml files (#2310) (9675c17)

Bug Fixes#

  • addressing #2197 by allowing 256 character account names in slurm (#2198) (ab58c65)

  • removed distutils from snakemake (#2312) (9b8c362)

  • Update to move “file” param to “print” (#2291) (92352b6)

7.28.3 (2023-06-16)#

Bug Fixes#

  • Detect pandas availability to select serializer (#2300) (e08a771)

Performance Improvements#

  • avoid superflous mtime checks when the same file is referred to by multiple jobs (#2284) (eb6e2e1)


  • update docs for azbatch and dockerhub ref (#2298) (908dbf1)

7.28.2 (2023-06-13)#

Bug Fixes#

  • fix pandas import handling in metadata persistence (27f7b40)

7.28.1 (2023-06-11)#

Bug Fixes#

  • Bump yte from >=1.0,<2.0 to >=1.5.1,<2.0 (#2275) (8c0b34f)

  • remove superfluous dependency (aad61a0)

7.28.0 (2023-06-11)#


7.27.0 (2023-06-07)#


  • Allow the environment variable SNAKEMAKE_CONDA_PREFIX to be present without –use-conda (#2263) (e4eba8d)

Bug Fixes#

  • adapt linting rule to Python 3.11 (a3a5c58)

7.26.0 (2023-05-22)#


  • allow config files to be processed with YTE (#2269) (8e1c22f)

7.25.4 (2023-05-12)#

Bug Fixes#

  • fix scrolling behavior in landing page of report for large workflows (63c0c31)

  • report spacing (f3954b3)


7.25.3 (2023-05-03)#

Bug Fixes#

  • fix missed wildcard constraints when using local rule inheritance (#2242) (8e94785)

7.25.2 (2023-04-28)#

Bug Fixes#

  • Fix inconsistencies between detailed summary and normal summary (#2218) (d903123)

  • Fix race condition when creating lock directory (#2225) (66ea4d1)

  • quote paths given to singularity in order to ensure that it does not fail when paths contain whitespace (#2190) (a572fb7)


  • added changelog info for >v7.19.1 parsing error of “hh:mm:ss” time format in runtime resource (#2189) (2889f38)

  • update misc/vim/Readme with info for packer.nvim (#2095) (32166a7)

  • Update workflow syntax with priority directive (#2188) (af10db5)

7.25.1 (2023-04-28)#

Bug Fixes#

  • allow log directive in default target rule (#2191) (86e9624)

  • only consider global wildcard_constraints from the same module (#2235) (c412b71)

  • Use attribute to fill rule field in summary (#2217) (837c3fd)


7.25.0 (2023-03-23)#


Bug Fixes#

  • always make sure that the original path of source cached files is properly passed into metadata persistence records (#2179) (8bacbd0)

  • slurm batch job status queries (#2167) (0bb69e4)


  • Change snakemake-tutorial download link to always be the latest (#2183) (ae8a8f4)

  • fix typos in –help (#2182) (09f0cbe)

  • Improve error message when rule contains multiple run/shell/script/notebook/wrapper/template_engine/cwl keywords (#2186) (cd5a3c4)

7.24.2 (2023-03-14)#

Bug Fixes#

  • fix index out of bounds error raised by usage of workflow.source_path called from input or params functions (thanks @AKBrueggemann) (#2170) (cf8e6e8)

  • limit length of failed logs decorations (#2125) (6fc9243)

  • raise error if callable is passed to expand. (#2171) (1f28476)

  • rounding for batch calculation (#2064) (cbdbf9b)

7.24.1 (2023-03-09)#

Bug Fixes#

  • better job status queries for slurm executor (#2136) (a4df38c)

  • get python version for script environment in a backwards compatible way that works down to python 2.7 (#2161) (44e59b9)

  • prevents DeprecationWarning caused by using old draft of json schema (#2152) (9791ffb)

Performance Improvements#

  • Gfal2 remote provider using gfal2-python instead of gfal2-utils. (#2128) (0b9bfe5)


7.24.0 (2023-03-01)#


  • limit the number of input/output files in job properties (#2149) (d93f091)

Bug Fixes#

  • #2130 by patching the protect() method so the path of files in subdirectories is properly resolved during write-protection (#2131) (1a754fd)

  • sre_constants import because of deprecation (#2139) (3b326db)

  • ensure user and group rw permissions for metadata files and source cache (#2132) (cc51faa)

  • is_run error with local, group jobs (#2133) (31bfcd5)

  • require toposort >= 1.10 (#2145) (3cb54b8)


7.23.1 (2023-02-18)#

Bug Fixes#

7.23.0 (2023-02-18)#


  • changed report layout to display menu always left of the results. For fullscreen, one can still hide the menu, which leads to automatic growth of the results (#2116) (d771b1b)

  • Publish docker images for amd64 & arm64 (#2105) (4c898f5)

Bug Fixes#

  • use text/markdown for long_description_content_type (#2112) (0241075)

Performance Improvements#

  • Improve execution speed of cleanup_workdir (in dag) (#2103) (1fbc5f5)

7.22.0 (2023-02-12)#


Bug Fixes#

  • assume shared filesystem by default when running with –flux (#2075) (4bec2fd)

  • properly handle NA values for paramspaces (#2098) (6b6a880)

7.21.0 (2023-01-30)#


  • ability to encode paramspaces into a single wildcard, via the newly introduced single_wildcard argument of Paramspace. (#2069) (728ab3c)

  • allow input, output, and params to be used in functions passed to report mark arguments (#2081) (93ff8b6)

Bug Fixes#

  • more robust encoding of params in persistent metadata storage. This way, pandas parameters do not lead to spurious rerun triggers. (#2080) (106a4c3)

  • more robust parsing of sacct output in slurm executor (#2036) (fe651f8)

  • Postprocess job groups in toposorted order for correct touch times (#2073) (10b5849)

7.20.0 (2023-01-18)#


  • add tes token (#1966) (59a8fa0)

  • Add token auth to GitLab/GitHub hosting providers (#1761) (e03a3b4), closes #1301

  • allow for human friendly resource definitions (e.g. mem=”5GB”, runtime=”1d”) which deprecates slurm constrained time format (e.g. runtime=”hh:mm:ss”) (#1861) (24610ac) (#2154)

Bug Fixes#

  • :bug: - fix hyperlink (#2046) (9519d31)

  • Catch missing error stream in Slurm executor (#2063) (c21fc7e)

  • correctly parse empty values in config cli (#2032) (1b0689d)

  • Correctly parse UserDicts in executors (#2016) (e3926fa)

  • Fix handling of –jobs in no-exec state (#2029) (e8e8222)

  • make --show-failed-logs handle empty log files (#2039) (683c6f2), closes #2023

  • make python version check more robust (#2058) (e685621)

  • parsing error when last line is comment (#2054) (a928dd4)

  • prevent overriding of retries when set to 0 (#2053) (a328f3e)

  • propagate attempt count from group to subjobs (#2052) (da3f1c0)

  • remove overflow from rulegraph div in report (9a0aaa7)

  • skip type checks of missing dir in touch mode (#2051) (ae00c25)

  • slurm default_resources quoting (#2043) (47d3fc3)

  • Update list of python versions in classifiers (#2020) (7a98100)

  • use short argument name for --chdir for compatibility with Slurm <=v17 (#2040) (a9ed3ec)

  • human friendly resource definitions introduce inability to parse slurm specific time format (e.g. “hh:mm:ss”). New time format (e.g. “1d”) adds portability among various job schedulers and clusters (#2154)


7.19.1 (2022-12-13)#

Bug Fixes#

  • improved default resources parsing (also allowing to deactivate a default resource via setting it to None) (#2006) (e6cdb32)


7.19.0 (2022-12-13)#


  • add keyword to gridftp remote provide to specify the number or disable usage of multiple data stream (#1974) (3e6675d)

  • provide information about temp, pipe, and service files in –summary (#1977) (c7c7776)

  • native SLURM support (–slurm, see docs) (#1015) (c7ea059)

Bug Fixes#

  • avoid logfile writing in case of dryrun; better hints in case of incomplete checkpoints (#1994) (a022705)

  • handle case where zenodo deposition does not return files (#2004) (b63c4a7)

  • issue #1882 WorkflowError: Metadata can’t be created as it already exists (Windows) (#1971) (d4484e6)

  • json validation error with markdown cells (#1986) (6c26f75)

7.18.2 (2022-11-10)#

Bug Fixes#

  • Change ratelimiter dependency to throttler (#1958) (50b8f16)

  • fixed problem with leaked modifications when inheriting multiple times from the same rule (#1957) (2475cbc)

  • forwarding –keep-incomplete to cluster executor (#1951) (2894c7d)

  • show input files on job error (#1949) (ad21631)

7.18.1 (2022-11-03)#

Bug Fixes#

  • regression ValueError introduced with 7.17.2 (#1947) (53a4fca)

7.18.0 (2022-10-31)#


  • first try to match output files against input files while persisting wildcard values from the consuming job. This can dramatically reduce ambiuity problems. Thanks to @descostesn! (#1939) (d093907)

7.17.2 (2022-10-28)#

Bug Fixes#

  • Consider source cache when setting search path for python scripts. This allows to import from Python modules next to scripts while deploying the workflow as a snakemake module, even from remote locations. (#1940) (27be1d4)

7.17.1 (2022-10-28)#

Bug Fixes#

  • change source cache entries to keep the original name and folder structure, such that imports from e.g. scripts also work with remote modules (if specified as additional input files with workflow.source_path) (#1936) (c34f3f6)

7.17.0 (2022-10-27)#


  • allow to define the cache mode per rule (this enables to exclude software envs from the caching hash value, which can be handy e.g. for download rules where the software version does not affect the result) (#1933) (715e618)

Performance Improvements#

  • cached os.pathconf() call in _record_path() (#1920) (551badb)

7.16.2 (2022-10-26)#

Bug Fixes#

  • fix false rerun triggering downstream of checkpoints due to spurious parameter, code or software env changes (638ea86)

  • remove redundant dot in expand call in multiext documentation (#1921) (278beaa)

7.16.1 (2022-10-18)#

Bug Fixes#

  • conda create –no-shortcuts absent on Linux/MacOS (regression from #1046) (#1916) (8a86a1e)

  • fix typo in line display of exceptions (#1912) (55e38a6)

7.16.0 (2022-10-14)#


Bug Fixes#

  • allow report generation to handle pathlib objects (#1904) (7c34656)

  • fix false reruns after checkpoints (#1907) (dc5af12)

7.15.2 (2022-10-08)#

Bug Fixes#

  • Comparison of rules and non-rule instances (#1894) (bf01ece)

  • delay evaluation of tmpdir to actual job execution, and not submission. This way, tmpdir can be dependent on the node context. (#1860) (4203556)

  • ensure that rule name string instead of object is passed to tabulate package (#1898) (f9ff157)

  • issue 1846 (#1888) (da2dfbd)

  • lexicographically sorted rule display with –list, and trimmed rule docstrings (#1880) (32128ae)

Performance Improvements#

  • Average NamedList getitem performance improvement (#1825) (10451b7)

7.15.1 (2022-10-04)#

Bug Fixes#


7.15.0 (2022-10-04)#


Bug Fixes#


7.14.2 (2022-09-26)#

Bug Fixes#

  • reduce resource requirements for kubernetes tests (#1876) (cb4b78a)

7.14.1 (2022-09-23)#

Bug Fixes#

  • allocation of local ssds in k8s tests (#1870) (d0de4dc)

  • allow script directive to take pathlib Path (#1869) (12cdc96)

  • catch errors in remote.AUTO provider list (#1834) (c613ed2)

  • consistently use text output in conda shell commands and various little fixes for failing test cases due to conda package changes (#1864) (4234fe7)

  • declare associative arrays (#1844) (90ae449)

  • fix falsely triggered reruns if input files are obtained via workflow.source_path() (#1862) (2dc2e6a)

  • fixed typos (#1847) (a1e49b6)

  • k8s container volume mounts as list (#1868) (5c54df3)

  • None type error when invoking Workflow object manually (#1731) (dc45ccb)

  • request disk_mb resource from k8s (#1858) (f68f166)

  • respect shebang lines in post-deploy scripts (see deployment docs) (#1841) (c26c4b6)

7.14.0 (2022-08-27)#


  • add support for bash scripts in the script directive (beyond small shell commands) (#1821) (c4cf8fd)


7.13.0 (2022-08-25)#


  • add gitfile option to make it possible to use local git repos when importing modules (#1376) (1a3b91f)

Bug Fixes#

  • allow to use {wildcards} for group jobs in cluster config (#1555) (f0ec73d)

  • avoid “Admin” prompt when using conda on windows (#1046) (552fadf)

  • handle benmark bug that arise with singularity (#1671) (10ef7c4)

  • Open Snakefile for reading with explicit encoding specified (#1146) (ec1d859)

  • remove superfluous comma causing TypeError in conda-frontend error message (#1804) (87b013c)


  • explain SNAKEMAKE_PROFILE environment variable (2b32bba)

  • update contribution docs (09a5595)

7.12.1 (2022-08-09)#

Bug Fixes#

  • Fix case of multiple scattergather processes (#1799) (417aad4)

  • more comprehensive error reporting for RuleExceptions (#1802) (1cd9512)

7.12.0 (2022-07-29)#


Bug Fixes#

  • Fix technical bugs in resource-scope documentation (#1784) (878420c)

  • move max_status_checks_per_second attribute setting before the wait thread of cluster backends is started to avoid missing attribute errors (#1775) (a48e9d0)

7.11.0 (2022-07-27)#


  • improved resource handling in groups and ability to define resource scopes (global or per node), see docs and –help (#1218) (a8014d0)

Bug Fixes#

  • fixed conda frontend detection and checking to also work with latest mambaforge (#1781) (225e68c)

7.10.0 (2022-07-26)#


  • Support conda environment definitions to be passed as function pointers, similar to input, params, and resources (#1300) (6f582f1)

Bug Fixes#

  • fix regression in workflow source acquisition of google life science executor (#1773) (c07732e)

  • limit filename length of temporary files generated by the persistence backend (metadata, incomplete markers, etc.) (#1780) (59053e7)

7.9.0 (2022-07-19)#


  • make it possible to exclude rules that will be imported when using ‘use rule’ statement (#1717) (d9e0611)

Bug Fixes#

  • add lock free mechanism for avoiding race conditions when writing persistence information; consider corrupt metadata records as non-existent (#1745) (71fe952)

  • conda python interpreter path on Windows (#1711) (155c9d6)

  • ensures that REncoder also checks for numpy.bool_ in encode_value (#1749) (10a6e1d)

  • Move quiet default after profile parsing (#1764) (6ade76d)

7.8.5 (2022-06-30)#


  • fix long description type for pypi (set to markdown) (d8d9b8f)

7.8.4 (2022-06-30)#

Bug Fixes#

  • only display a warning in case of non-strict channel priorities (#1752) (b84fa33)

  • pass triggers and resources to subworkflow (#1733) (fa7fb75)

  • add pyproject.toml to use setuptools features (#1725) (454bfd1)


7.8.3 (2022-06-20)#

Bug Fixes#

  • allow apptainer as a successor to singularity. (#1706) (bcbdb0b)

  • improved provenance trigger info (#1720) (29d959d)

  • small changes to make docs checkpoint example functional (#1714) (1d4909e)

7.8.2 (2022-06-08)#

Bug Fixes#

  • fixed bug in needrun computation of jobs downsteam of checkpoints (#1704) (c634b78)

7.8.1 (2022-05-31)#

Bug Fixes#

  • handling of remaining jobs when using –keep-going (#1693) (87e4303)

  • more robust calculation of number of jobs until ready for execution (#1691) (fdfc717)

  • propagate rerun trigger info to cluster jobs; fix a bug leading to software stack trigger generating false positives in case of conda environments; fixed display of info message in case of provenance triggered reruns (#1686) (503c70c)

  • set channel priority in container system wide (#1690) (41175b3)

7.8.0 (2022-05-24)#


  • automatically rerun jobs if parameters, code, input file set, or software stack changed (thanks to @cclienty and @timtroendle). This also increases performance of DAG building by handling job “needrun” updates level wise, while avoiding to perform a full check for those jobs that are already downstream of a job that has been determined to require a rerun. (#1663) (4c11893)

  • enable the definition of conda pin files in order to freeze an environment. This can drastically increase the robustness because it allows to freeze an environment at a working state. (#1667) (53972bf)

Bug Fixes#

  • fail with error if conda installation is not set to strict channel priorities (#1672) (f1ffbf2)

  • fix errors occurring when refering to input func via rules..input (#1669) (28a4795)

  • parsing error when combining single line directive with multi-line directive in use rule statements (#1662) (26e57d6)

7.7.0 (2022-05-16)#


  • add flag ensure that allows to annotate that certain output files should be non-empty or agree with a given checksum (#1651) (76f69d9)

  • for small files, compare checksums to determine if job needs to run if input file is newer than output file (#1568) (1ae85c6)

  • LockException (#1276) (f5e6fa6)

  • new directive “retries” for annotating the number of times a job shall be restarted after a failure (#1649) (c8d81d0)

Bug Fixes#


  • singularity sometimes uses system /tmp explanation (#1588) (170c1d9)

7.6.2 (2022-05-06)#

Bug Fixes#

  • fixed permission issues when using zenodo remote provider to access restricted depositions (#1634) (510f534)

7.6.1 (2022-05-04)#

Bug Fixes#

  • check for skipped rules in case of local rule inheritance (#1631) (9083ac1)

7.6.0 (2022-05-03)#


  • enable restricted access support in zenodo remote provider (#1623) (692caf9)

Bug Fixes#

  • avoid erroneous too early deletion of parent directories in case of failed jobs (thanks to @SichongP). (#1601) (b0917e6)

  • ensure that rule inheritance considers the same globals and other settings as parent module (#1621) (104cab9)

  • issue 1615 - Switch formatting condition for dictionary (#1617) (0771062)

  • multiext prefix computation in case it is used within a module that defines an additional prefix (#1609) (fc6dfc6)

  • remove redundant print (#1608) (cc7e0e3)

7.5.0 (2022-04-26)#


Bug Fixes#

  • properly use configfiles specified via CLI also if configfile specified via configfile directive is not present (1e0649a)


7.4.0 (2022-04-22)#


  • Allow paramspace to separate filename params with custom separator (#1299) (8236e80)

Bug Fixes#

7.3.8 (2022-04-06)#

Bug Fixes#

  • support multiple input files for template_engine rules (#1571) (aee7cf2)

7.3.7 (2022-04-05)#

Bug Fixes#

  • allow labels function to return None (#1565) (fef74d6)

  • do not wrap whitespace in result info headers of reports (653d0d0)

  • fixed detection of norun rules inside of modules (#1566) (d2223d4)

  • properly use retry mechanism in source cache (#1564) (624a83d)

7.3.6 (2022-04-02)#

Bug Fixes#

  • always recalculate job resources before job is scheduled as input might have changed or not have been present initially (#1552) (44aacdb)

  • fixed handling of input functions and unpack when using the prefix setting of module definitions (#1553) (d561e04)

  • fixed parsing of subsequent use rule statements directly beneath each other (#1548) (77d5a08)

  • fix spurious missing file errors when using google storage (#1541) (1b3ede1)

  • proper error message if resource types do not match (#1556) (1112321)

  • quote workdir in job exec prefix to allow to spaces in the workdir (#1547) (c3a593e)

  • report error and possible cause if metadata cleanup fails (#1554) (6866134)

7.3.5 (2022-03-31)#

Bug Fixes#

  • do not remove existing temp files in case of dryrun (#1543) (e820f97)

  • fixed bug in missing input file handling for cluster jobs (#1544) (40e2eb2)


  • explain automatic decompression strategy for http remote provider (e6826b6)

7.3.4 (2022-03-30)#

Bug Fixes#

  • better error messages in case of missing files after latency period (#1528) (5b394c0)

  • correct handling of exceptions in input functions that are generators (#1536) (d9a56aa)

  • obtaining conda prefix when using in combination with singularity (#1535) (99b22d3)

  • proper error message in case of missing git when checking for source files (#1534) (92887a3)

  • throw error message in case of target rule that depends on a pipe. (#1532) (b9e9a7e)


7.3.3 (2022-03-28)#

Bug Fixes#

  • better error message in case of failing to create conda env (#1526) (e7a461c)

  • fix singularity logging messages causing conda fail (#1523) (7797595)

  • more robust handling of incompletely evaluated parameters (any interaction with them will result in a string now). (#1525) (3d4c768)


  • details on benchmarked results (64fea09)

7.3.2 (2022-03-25)#

Bug Fixes#

  • fixed code change detection (#1513) (67298c6)

  • modify dag and workflow display in report to also work for big DAGs (#1517) (1364dfb)


  • Clarify the use of conda with notebook directive (#1515) (aefb1eb)

7.3.1 (2022-03-23)#

Bug Fixes#

  • add about page to report, including embedded packages and licenses (#1511) (142a452)

  • in google live science backend, save multiple logs per rule name and overwrite existing logs (#1504) (9e92d63)

  • in rules from imported modules, exclude modified paths from module prefixing (#1494) (1e73db0)

  • Replaced pathlib relative_to with os.relpath (#1505) (dc65e29)

  • update for minimum of Python 3.7 (#1509) (62024e2)

7.3.0 (2022-03-21)#


  • Support for machine_type for kubernetes executor (#1291) (12d6f67)

Bug Fixes#

  • always wait for input files before starting jobs, also upon local execution and within group jobs. This should add further robustness against NFS latency issues. (#1486) (cab2adb)

  • cleaned up and rewritten execution backend structure, (fixing #1475, #860, #1007, #1008) (PR #1491) (e87cc97)

  • do not skip local conda env creation per se when having no shared FS, because it is still needed for local jobs. Instead, decide for each env whether it is needed locally or not. (#1490) (3f03c5d)

  • fixed temp file deletion for group jobs (#1487) (d030443)

  • improve robustness when retrieving remote source files, fixed usage of local git repos as wrapper prefixes (in collaboration with @cokelaer and @Smeds) (#1495) (e16531d)

  • mtime inventory for google storage was accidentally setting a float instead of a proper mtime object (#1484) (7c762c7)

  • render empty caption if nothing defined in report flag (013a6e8)


7.2.1 (2022-03-14)#

Bug Fixes#

  • add missing report.templates.components module to (cb4e3fe)


  • add install info of development (git) version to docs (#1477) (2a2d6cd)

7.2.0 (2022-03-13)#


  • improved reports: more interactive and modern interface, ability to define a label based representation of files (#1470) (d09df0c)

Bug Fixes#

  • always deploy conda envs in main process when assuming a shared file system (fixes issue #1463) (#1472) (79788eb)

  • do not wait for named or containerized conda envs (#1473) (6b1d09c)

  • implement lock-free source file caching. This avoids hangs on network file systems like NFS. (#1464) (9520e98)

7.1.1 (2022-03-07)#

Bug Fixes#

  • quote jobid passed to status script to support multi-cluster Slurm setup (#1459) (0232201)

7.1.0 (2022-03-04)#


  • Zenodo remote provider for transparent storage on and retrieval from Zenodo (#1455) (4586ef7)

Bug Fixes#

  • disable mtime retrieval from github api for now. This quickly exceeds rate limits. (1858bb9)

  • display change warnings only for jobs that won’t be executed otherwise (086f60f)

  • work around segfault with >100 jobs in google life sciences backend (#1451) (2c0fee2)

7.0.4 (2022-03-03)#

Bug Fixes#

  • more details on input and output exceptions (missing input, protected output, etc.) (#1453) (8d64af2)

7.0.3 (2022-03-02)#

Bug Fixes#

  • fix a bug leading to duplicate conda env initializations; fix display of jobs and output files with changes (994b151)

  • preserve empty names input or output file lists in params or resource functions (0d19ab0)

  • remove accidental pdb statement (9c935f1)

  • remove deprecated and add missing arguments to internal functions (93a7e39)

7.0.2 (2022-03-01)#

Bug Fixes#

  • add local marker for input files in cufflinks example. fixes issue #1362 (90bc88b)

  • failure to properly apply default remote prefix in combination with the unpack marker (#1448) (82666f1)

  • set mtime for cached source files [WIP] (#1443) (dd27209)

  • small bug in snakemake.executors (#1440) (6e64292)


  • fix list display in docs (3724367)

  • fix list display in docs (2dd0e91)

  • Fix typo and grammar mistake in scatter-gather section. (#1441) (f218aaa)

7.0.1 (2022-02-26)#

Bug Fixes#

  • avoid incomplete remote files in case of errors and automatically retry download and upload (#1432) (8fc23ed)

  • do not apply module prefix in case of remote files (5645b3f)

  • do not require –cores or –jobs to be set when –cleanup-metadata is used. (#1429) (9c73907)

  • more robust place for runtime source file cache (#1436) (2681f6f)

  • provide details on error when failing to evaluate default resources (#1430) (04f39a9)

  • provide proper error when using immediate submit in combination with checkpoint jobs. (#1437) (865cf0f)


  • explain relative path interpretation (#1428) (add9a05)

  • Fix problems with code blocks and broken internal link. (#1424) (5d4e7d8)

  • temaplte rendering examples and available variables (#1431) (5995e9e)

  • update copyright year (#1427) (6b9f5da)

7.0.0 (2022-02-23)#


  • require at least Python 3.7 (fd5daae)


  • adding service jobs, i.e. the ability to define jobs that provide a resource for consumers (like a shared memory device or a database), and will be automatically terminated by Snakemake once all consumers are finished. (see docs, #1413) (a471adb)

  • support for group local jobs by enabling optional groupid consideration in input functions (see docs, #1418) (5d45493)

  • Adding –cluster-cancel and –cluster-cancel-nargs (#1395) (0593de1)

  • cluster sidecar (#1397) (b992cd1)

  • template rendering integration (yte and jinja2) (#1410) (e1cbde5)

Bug Fixes#

  • bug in pipe group handling that led to multiple assignments of the same group id to different groups; bug that accidentally added already running groups of the list of ready jobs (issue #1331) (#1332) (1a9b483)

  • display wrapper or external script code in report #1393 (#1404) (a007bd1)

  • do not pass SNAKEMAKE_PROFILE into cluster-submit (#1398) (#1407) (7189183)

  • issue with duplicated prefix for checkpoints on cloud (#1294) (8ed0c8c)

  • keep flags with apply_wildcards on cloned IOFile (#1416) (23c943f)

  • remove raise that limits using –config with dicts (#1341) (bd65057)

  • Repair MREs from #823 (#1203) (b007979)

  • warn on non-file-modification-date changes like params, code, or input files (#1419) (b5f53f0)

6.15.5 (2022-02-09)#

Bug Fixes#

  • convert conda env to string before checks (#1382) (7a8da9f)

  • fix pepfile handling in case of module usage (#1387) (f097a76)

6.15.4 (2022-02-09)#

Bug Fixes#

  • fix issue when generating unit tests for rules with directory output (#1385) (7db614f)


  • fix tutorial setup instructions for MacOS. (#1383) (b57b749)

6.15.3 (2022-02-07)#

Bug Fixes#

  • skip global report caption when using a module (#1379) (a755cee)

6.15.2 (2022-02-05)#

Bug Fixes#

  • avoid mutable default argument (#1330) (978cc93)

  • don’t raise WorkflowError when entry is empty (#1368) (1fc6f7b)

  • fix assertion error in conda env file spec when applying wildcards (thanks @ddesvillechabrol) (#1377) (6200652)

  • fix None type error when invoking Workflow object manually (#1366) (fca3895)

  • XRootDHelper.exists supports non posix filesystem (object store) (#1348) (7a3ad2f)


6.15.1 (2022-01-31)#

Bug Fixes#

  • consider post-deploy script for env hashing (#1363) (d50efd9)

6.15.0 (2022-01-29)#


  • adding default_target directive for declaring default target rules that are not the first rule in the workflow. (#1358) (638ec1a)

Bug Fixes#

  • Draft notebook filename with wildcards and params. (#1352) (11d4dc8)

  • proper error message when defining cache eligibility for rules with multiple output files and no multiext declaration. (#1357) (47b5096)


  • Command line arguments for configuration files (#1343) (ad8aaa4)

  • fix broken link in executor_tutorial/tutorial.rst (#1360) (c9be764)

6.14.0 (2022-01-26)#



6.13.1 (2022-01-11)#

Bug Fixes#

  • –conda-frontend value not passed on to cluster jobs (#1317) (df46ddb)

  • atomic job error display (#1326) (aa2c265)

  • fix source cache handling for remote source files retrieved via github() or gitlab() tags. (#1322) (6e2ecd2)

  • typos in code examples (#1324) (60010e4)

6.13.0 (2021-12-21)#


  • allow prefix definition in module statements (#1310) (29e6540)

6.12.3 (2021-12-09)#

Bug Fixes#

  • fixed display of any exceptions and errors from within a workflow definition (23d40d9)

6.12.2 (2021-12-07)#

Bug Fixes#

  • rule inheritance within modules (did previously lead to key errors) (#1292) (603e0a8)


  • Fix typo in rules.rst (—draft-notebook) (#1290) (f5c42cf)

6.12.1 (2021-11-29)#

Bug Fixes#

  • set default number of nodes to 1 in test cases (#1288) (f6e12b4)

6.12.0 (2021-11-29)#


  • add flag –draft-notebook for generating a skeleton notebook for manual editing (e.g. in VSCode). (#1284) (d279322)

Bug Fixes#


6.11.1 (2021-11-26)#

Bug Fixes#

  • provide temporary IPYTHONDIR for notebook execution in order to avoid race conditions in upon execution of multiple notebooks at the same time. (#1280) (4d70da1)


  • move psutil import into benchmark methods to avoid needing it as a dependency for doc building (6ffe38d)

  • require sphinx>=3 (1773875)

  • skip lazy property (2883718)

6.11.0 (2021-11-25)#


  • fail with an error if snakemake cannot write job metadata. (#1273) (cd968cd)

Bug Fixes#

  • Adds fixes for the first two MREs in #823 (#1215) (cfd2f89)

  • env file usage after changes to source file handling (inspired by #1233 and #1211). (#1236) (3ac8e85)

  • fixed code change detection when using modules (#1264) (b571e09)

  • handle config file extension/overwriting more explicitly (#1251) (d0a7bf2)

  • Issue #1253 (problems editing Jupyter Notebooks) (#1255) (3398ddf)

  • more informative nothing to be done message (#1234) (368d265)

  • only consider context of shell command for technical switches if called from snakemake rules. (#1213) (4816a58)

  • R encoding of pathlib.Path objects (#1201) (bd516e9)

  • Use ‘snakemake.utils.update_config’ instead of ‘dict.update’ (#1126) (2658027)

6.10.0 (2021-10-21)#


  • Add more informative errors when evaluation of --default-resources fails (#1192) (b3c4e68)

Bug Fixes#

  • add quotes to each item of the wait_for_files list (#1160) (72856ed)

  • caching process (#1225) (0825a29)

  • enable usage of job grouping in GLS (#1054) (d243c22)

  • Only –bind Snakemake when we’re working with a Python script (#1206) (1d79f62)

  • run dependencies with non-existent ancient files before the consuming job (#1202) (84d1f64), closes #946

  • status cmd repeats until killed by 11 different signals (#1207) (8b28b57)

  • typo in sourcecache use (#1229) (8b54bc5)

  • wms monitor arg parsing now accepts any kind of value (#1181) (313de93)


  • Clarification of –cluster-stats docs & elaborating on the situation where job ids are not passed to the status script (#1221) (ed0e4a2)

  • Combine CHANGELOG.rst with (#1228) (19f5a43)

  • Mention required openssl dep for rust-script (#1216) (fc8c5f6)

  • Unpin docutils version (#1230) (15a82bf)

6.9.1 (2021-09-30)#

Bug Fixes#

  • fix function call when creating report and hashes for between workflow caching (#1198) (a4f6836)

6.9.0 (2021-09-29)#


  • autoconvert Path objects to str when passing to R or Julia scripts (80ec513)

Bug Fixes#

  • fix source retrieval during between workflow caching and report generation (2394ca4)

6.8.2 (2021-09-29)#

Bug Fixes#

  • fix path returned by get_source() (ee05315)

6.8.1 (2021-09-24)#

Bug Fixes#

  • async_run to allow nested event loops. (#1170) (5dc6bbd)

  • merging of pipe groups when multiple rules are chained together via pipes (#1173) (de91d2c)

  • potential memory corruption caused by Google storage objects accessed from different threads (#1174) (41a5071)

Performance Improvements#

  • more extensive caching of source files, including wrappers. (#1182) (bdb75f8)


6.8.0 (2021-09-06)#


Bug Fixes#

  • AmbiguousRuleException bug caused by weak ordering of rules (#1124) (7f54c39)

  • Bugfix tes add files (#1133) (8892bf2)

  • Disable Persistence cache for snakemake jobs (#1159) (7110f9d)

  • efficient job status checking when using DRMAA API (this should yield much better parallelization and performance when using –drmaa) (#1156) (ac004cb)

  • improved error handling for cluster status scripts and smarter job selector choice in case of cluster submission (use greedy for single jobs). (#1142) (48d2dd9)

  • Initialize assignments dictionary when setting rule-based resources (#1154) (68c13fd)

  • key error when handling FileNotFoundError in input functions. (#1138) (d25f04d)

  • linting of remote snakefiles (#1131) (2104e10)

Performance Improvements#

  • improve job selection performance in case of potential ambiguity that is resolved by comprehensive ruleorder statements. (#1147) (921f4f7)

6.7.0 (2021-08-12)#


  • Add support for rust scripts (enabling directly integrated ad-hoc robust high performance scripting) (#1053) (f0e8fa2)

Bug Fixes#

  • Ga4gh tes bugfixes (#1127) (af21d6c)

  • improved display of percentage of done jobs (1fee8c0)

  • improved error message in case of target rule misspecification (83b1f5b)


  • fix contributing executors link (#1112) (4bb58d1)

  • Fix typo in file path in remote files documentation (#1110) (9ce294f)

6.6.1 (2021-07-19)#

Bug Fixes#

  • avoid superfluous calls of conda info that have slowed down Snakemake since 6.4.1. (#1099) (e990927)

6.6.0 (2021-07-16)#


  • Allow to mark all output files as temp with –all-temp (#1097) (0ac3b38)

6.5.5 (2021-07-16)#

Bug Fixes#

6.5.4 (2021-07-16)#


  • Fixed –touch in combination with temp files (issue #1028) (@johanneskoester, @iromeo).


  • Fix syntax error in docs/ and update sphinx.ext.napoleon import (#1084) (3e3fac2)

  • Improved pepfile (pepschema) documentation (@stolarczyk).

[6.5.3] - 2021-07-06#

  • Fixed a bug occuring when using –resources in the command line interface (@johanneskoester).

  • Minor improvements in the docs (@johanneskoester).

[6.5.2] - 2021-07-02#

  • Create directory pointed to by tmpdir resource if it does not yet exist (@johanneskoester).

  • Use a single core again in dryrun if –cores is not specified (@johanneskoester).

  • Bugfix for FTP remote provider (@jmeppley).

  • Improved documentation (@corneliusroemer).

[6.5.1] - 2021-06-24#

  • Extended best practices document (@johanneskoester)

  • Restore -j all behavior for local execution as a (deprecated) way of running Snakemake on all cores. Recommended now: --cores all (@johanneskoester).

  • Improved handling and better error messages for checkpoints (@johanneskoester).

[6.5.0] - 2021-06-22#

  • Allow to set the default profile via the environment variable $SNAKEMAKE_PROFILE.

  • There is a new default resource tmpdir (by default reflects the system setting), which is automatically used for temporary files by shell commands and scripts which properly consider the usual environment variables like $TMP, $TEMP, $TMPDIR (@johanneskoester).

  • The CLI flags –jobs and –cores are now separated, with –cores being responsible for local cores and global cores in the cluster case, and –jobs being responsible for number of jobs. Still -j and –jobs works as a fallback for local execution (@johanneskoester).

  • Added the ability to overwrite resources via –set-resources (@johanneskoester).

  • Various fixes for Windows execution (@melund).

  • Fixed a bug with fractional resources (@johanneskoester).

  • Fixed timeouts and other issues in google life science backend (@johanneskoester).

  • Fixed a bug with missing conda frontend definitions in subworkflows (@johanneskoester).

  • Skip envvar checking during linting (@johanneskoester).

  • Fixed a bug causing container images in modules to be ignored (@johanneskoester).

[6.4.1] - 2021-05-27#

  • Fixed bug in workflow.source_path() that occurred with modules included from remote locations (@johanneskoester).

  • Inform cluster jobs about conda/mamba/activate path such that they don’t need to determine this themselves (@johanneskoester).

[6.4.0] - 2021-05-20#

  • Improvements in the docs (resource usage, best practices, remote files) (@johanneskoester, @admorris).

  • functions given to params can now safely open input files generated by previous rules. If they are not present, TBD will be displayed and function will be reevaluated immediately before the job is executed (i.e. when files are present) (@ASLeonard).

  • Connection pool for SFTP and FTP remote files, increasing download performance (@jmeppley).

  • Require correct minimum version of smart_open (@Redmar-van-den-Berg).

  • Added workflow.source_path(path), allowing to get the correct path relative to the current Snakefile, even when Snakefile is included via URL (@johanneskoester).

  • Fixed bugs in module system (@johanneskoester, @dlaehnemann).

  • Fixed issue with checkpoints and ruleorder where phantom dependencies are not properly removed from the DAG (@jmeppley, @johanneskoester).

  • Disable tibanna behavior that opens a browser window for each job (@nigiord).

  • Allow Paramspace(..., filename_params="*"), meaning that all columns of the paramspace will be encoded into the filename (@kpj).

  • Avoid PATH modification in cluster jobs (@johanneskoester).

  • For large sets of input files, pass files to wait for (FS latency) as a file instead of command line args (@kpj, @epruesse).

[6.3.0] - 2021-04-29#

  • Changed behavior of workflow.snakefile to always point to the current file instead of the main Snakefile (also in case of includes and modules) (@johanneskoester).

  • Fixed a typo in an error message (@nikostr).

[6.2.0] - 2021-04-22#

  • Support for integration of foreign workflow management systems by introducing a handover directive that passes on all resources to a particular rule (which can then invoke another workflow management system). See the docs (“Integrating foreign workflow management systems”) (@johanneskoester).

  • Behavior improvement for temp handling of checkpoint rules (@epruesse).

  • Several improvements in the docs (@johanneskoester).

[6.2.1] - 2021-04-20#

  • Fixed a minor bug in the linter.

[6.2.0] - 2021-04-20#

  • Fixed several glitches in paramspace implementation (handling of bools, returning scalar values) (@kpj).

  • Fixed bugs in module implementation (@dlaehnemann, @johanneskoester).

  • Fall back to greedy scheduling solver if ILP solver needs more than 10 sec (@johanneskoester).

[6.1.1] - 2021-04-07#

  • Fixed several small bugs of the new module system (@johanneskoester, @dlaehnemann).

  • Fixed archive based conda deployment (@johanneskoester).

  • Better handling of download and target attributed in the interactive report (@johanneskoester).

[6.1.0] - 2021-04-01#

  • Snakemake now uses mamba as the default conda frontend (which can be overwritten by specifying to use conda via the –conda-frontend flag) (@johanneskoester).

  • Profiles using –cluster option can now handle relative submit script paths in combination with arguments (@kdm9).

  • New AutoRemoteProvider, which infers the type of remote file protocol from the given URL (@kpj).

  • When using global container directive, container usage can be deactivated on a per rule base (@bilke).

  • Bugfixes for checkpoint handling (@johanneskoester).

  • Bugfixes for the module system (@johanneskoester, @dlaehnemann).

  • Various improvements for the tutorial.

[6.0.5] - 2021-03-11#

  • Fix bug (introduced with 6.0) when handling of HTML directories in report (@johanneskoester).

[6.0.4] - 2021-03-11#

  • Various textual improvements in the tutorial (@dlaehnemann).

[6.0.3] - 2021-03-08#

  • No longer use a shortened hash for naming conda environments in .snakemake/conda (@johanneskoester).

  • Various little updates to the docs (@johanneskoester).

[6.0.2] - 2021-03-03#

  • Fix race condition in conda checking code (@johanneskoester).

[6.0.1] - 2021-03-03#

  • Restored Python 3.5 compatibility by removing f-strings (@mbhall88)

  • Fix rendering issue in the docs.

  • Add gitpod dev environment and gitpod environment for the tutorial.

[6.0.0] - 2021-02-26#

[5.32.2] - 2021-02-11#


  • Fixed infinite loading of results in Snakemake reports (@FelixMoelder)

[5.32.1] - 2021-02-08#


  • Improved warning on wildcard constraints (@jheuel)

  • Improved logging from the new scheduler implementation (@johanneskoester)

  • Restored Python 3.5 compatibility by removing f-strings (@mbhall88)

  • Snakemake now automatically adds a global wildcard constraint for {scatteritem}, when scatter/gather support is used.

  • The zip variant of Snakemake reports is now compressed (@FelixMoelder).

  • Improved docs (@ctb).

  • Make output file removal in cluster mode more robust (@sebschmi).

[5.32.0] - 2021-01-15#


  • Handle accidental use of GLS backend with singularity (@vsoch).

  • Improved and extended WMS-monitor implementation (@vsoch).

  • Display index and total count in {scatteritem} when using the scatter-gather helper (@johanneskoester).

  • Fixed problems with jobid display when handling checkpoint updates (@johanneskoester, @jmeppley).

  • Fixed bug when checking for directory containment of output files (@jmeppley).

  • Implement –no-subworkflows treatment in combination with –cluster (@goi42).

[5.31.1] - 2020-12-21#


  • added wget again to the container image

[5.31.0] - 2020-12-21#


  • The Paramspace helper for automatically exploring parameter spaces given as Pandas dataframes.

  • A new directive name: for setting rule names from variables.


  • Various small bug fixes for scheduling and checkpoint handling.

  • Automatically block R_LIBS, PYTHONPATH, PERL5LIB, and PERLLIB when using conda with –use-conda. This behavior can be deactivated with –conda-not-block-envvars.

  • Update container image to latest singularity.

[5.30.2] - 2020-12-16#


  • Fix permission issues with jobscripts on some systems (@Phhere).

  • Added notes on WSL to the tutorial (@RomainFeron).

  • Scheduler fixes (@johanneskoester).

  • Fixed a bug in checkpoint handling that led to hanging workflow execution (@jmeppley).

  • Pass cluster nodes to subworkflows (@votti).

  • Fix start time recording in metadata (@lparsons).

  • Fix time retrieval in reports (@johanneskoester).

  • Fix error when returning a Path from an input function (@sappjw).

  • Extending monitoring docs with some notes about future api changes (@vsoch).

[5.30.0] - 2020-11-23#


  • Benchmarks now also report CPU time (@natir).


  • Fixed a reauthentication bug in Kubernetes support (@haizi-zh).

[5.29.0] - 2020-11-19#


  • Fixed several bugs in reports and scheduler.

  • Remove automatic (but buggy) encoding of csv/tsv files into HTML tables in the report (we will soon have a better alternative).

  • Fixed bug in kubernetes executor occurring with large source files.

[5.28.0] - 2020-11-12#


  • Execution backend for GA4GH TES (task execution scheduler) an abstraction layer for various cluster and cloud queuing systems (@svedziok, @uniqueg).

  • script, notebook, wrapper and cwl directives now permit to use wildcards and params for composing paths (@johanneskoester).


  • Restored compatibility with Python 3.5 and 3.6 (@cclienti).

  • Various usability bug fixes (@goi43, @johanneskoester, @dcroote).

  • Better and more secure parsing of values when using –config (@bingxiao).

[5.27.4] - 2020-11-03#


  • Further speed improvements for DAG computation.

  • Fixed metadata migration errors occuring with long output file paths.

  • Add WorkflowHub specifications to the docs.

  • Fix group assignments.

[5.27.3] - 2020-10-30#


  • Added missing files to source distribution.

[5.27.2] - 2020-10-30#


  • DAG computation runtime has been improved by orders of magnitude, it is linear in the number of jobs now (@mhulsmann, @johanneskoester).

  • Stat calls have been dramatically reduced and are now performed in parallel (@johanneskoester).

  • Scheduler fixes (@FelixMoelder).

  • Directory support and other fixes for Google Life Sciences backend (@vsoch, @millerdz).

  • Support for panoptes monitor server (@fgypas).

  • Extended pathlib support (@mbhall88).

  • Vim plugin improvements (@troycomi).

  • Prevent jobs being rerun when input files are marked as ancient and another job in the DAG creates them.

  • Fixed –list-code-changes for included rules (@jbloom).


  • Syntax highlighting for nano (@baileythegreen).

[5.26.1] - 2020-10-01#


  • Use coin ILP solver for scheduling by default (GLPK has bugs that can cause it to fail in certain situations).

  • If coin is not available, fall back to greedy scheduler.

[5.26.0] - 2020-09-30#


  • Flag –max-inventory-time for setting maximum time spend on creating file inventory.

  • Flag –scheduler-ilp-solver for defining which solver to use for the ILP scheduler.


  • Fixed various bugs with the new scheduler (@FelixMoelder).

  • Fixed bug causing certain parameters not to be passed to the cluster (–set-scatter, –scheduler, –set-threads).

  • Updated docs and fixed of google backend (@vsoch).

  • Display jupyter notebook code in reports.

  • Improved scheduler behavior in order to directly remove temporary files if possible.

[5.25.0] - 2020-09-18#


  • Simplified and more configurable support for scatter-gather processes (see docs).

  • Fully configurable DAG partitioning by grouping jobs at the command line. This should provide a vast additional improvement to scalability in cluster and cloud settings.


  • Depend on latest pulp, thereby enable Python >=3.8 compatibility again.

  • Fixes for snakefile handling in google life sciences backend (@vsoch).

[5.24.2] - 2020-09-15#


  • Fixed a bug in the linter that caused a false warning when using resources in shell commands.

[5.24.1] - 2020-09-13#


  • Depend on pulp < 2.0, which includes the default coin cbc solver for all platforms.

[5.24.0] - 2020-09-09#


  • Preemtion support for google cloud backend (@vsoch).


  • Fixed compatibility issues in new scheduler code (@dtrodrigues and @johanneskoester).

  • Improved error messages (@Sam-Tygier, @terrycojones)

  • Various small bug fixes.

  • Improved profile documentation (@johanneskoester).

[5.23.0] - 2020-08-24#


  • Support for workflow configuration via portable encapsulated projects (PEPs,

  • A new ILP based default scheduler now ensures that temporary files are deleted as fast as possible (@FelixMoelder, @johanneskoester).


  • Fixed bug in modification date comparison for files in google storage (@vsoch).

  • Various small documentation improvements (@dcroote, @erjel, @dlaehnemann, @goi42).

[5.22.1] - 2020-08-14#


  • Fixed a missing dependency for google storage in cloud execution.

[5.22.0] - 2020-08-13#


  • Added short option -T for CLI parameter --restart-times (@mbhall88).


  • Various small fixes for google storage and life sciences backends (@vsoch).

[5.21.0] - 2020-08-11#


  • Added default-remote-provider support for Azure storage (@andreas-wilm).

  • Various small bug fixes and documentation improvements.

[5.20.1] - 2020-07-08#


  • Fixed a bug that caused singularity args to be not passed on correctly when using script or conda.

[5.20.0] - 2020-07-08#


  • Exceptions in input functions are now handled in a smarter way, by choosing alternative paths in the DAG if available.

  • Debugging dag creation (–debug-dag) now gives more hints if alternative DAG paths are chosen.

  • Fixes for XRootD remote file implementation.

  • Improved CLI documentation.

  • Improved docs.

  • Various minor bug fixes.

  • Restored Python 3.5 compatibility.

  • Speed improvements for workdir cleanup.

  • Allow Path objects to be passed to expand.

[5.19.3] - 2020-06-16#


  • Performance improvements for DAG generation (up to 7x in the google cloud, anything from a little to massive in a cluster, depending on the overall filesystem performance).

  • Made harcoded bucket in google cloud executor configurable.

  • Improved speed of –unlock command.

[5.19.2] - 2020-06-04#


  • Fixed a bug in script and wrapper directives. Tried to decode a str.

[5.19.1] - 2020-06-03#


  • Fixed an issue with the parameter linting code, that could cause an index out of bounds exception.

[5.19.0] - 2020-06-02#


  • The multiext function now allows arbitrary file extensions (no longer required to start with a “.” (thanks to @jafors)

  • The include directive can now also take a Pathlib Path object (thanks to @mbhall88).


  • Jupyter notebook integration no longer automatically starts a browser.

  • Empty directories are cleaned up after workflow execution.

  • Fixed directory handling: no longer fail if the same job writes both a dir and a contained file.

  • Linter now recommends using spaces only for indentation.

  • Persistence dir “aux” has been renamed to “auxilliary” in order to make windows happy.

  • Linter now distinguishes awk syntax from regular variable usage.

  • Various bug fixes for Windows (thanks to @melund).

[5.18.0] - 2020-05-21#


  • Native Google Cloud support via the (despite the name generic) lifesciences API.

  • Ability to optionally exchange the conda frontend to mamba (faster and sometimes more correct) instead of conda.


  • Improved notebook integration experience, with various removed bugs and pitfalls.

  • Auto-retry google storage API calls on transient or checksum errors.

[5.17.0] - 2020-05-07#


  • –envvars flag for passing secrets to cloud executors


  • Wider thumbnail dialogs in report.

  • Updated installation instructions.

  • Various small kubernetes bug fixes.

  • Bug fix for iRods remote files.

[5.16.0] - 2020-04-29#


  • Interactive jupyter notebook editing. Notebooks defined by rules can be interactively drafted and updated using snakemake –edit-notebook (see docs).


  • Fixed group resource usage to occupy one cluster/cloud node.

  • Minor bug fixes.

[5.15.0] - 2020-04-21#


  • The resource directive can now take strings, e.g. for defining a GPU model (see docs). This will e.g. be used for upcoming updates to cloud executors.

  • More extensive conda cleanup with –conda-cleanup-packages, meant for CI usage.

  • Further polish for reports.

[5.14.0] - 2020-04-08#


  • Redesigned HTML reports, with improved interface and performance.

  • For big data, HTML reports can now be stored as ZIP, where files are not anymore embedded but rather are stored in an auxilliary folder, such that they don’t have to be in memory during report rendering.

  • Added subcategories to report (see docs).

  • Fixed a bug linter, leading to only one rule or snakefile to be linted.

  • Breaking change in CLI: added flags –conda-cleanup-envs and –conda-cleanup-pkgs, removed flag –cleanup-conda.

  • Fixed scheduling of pipe jobs, they are now always scheduled, fixing a hangup.

  • Corrected quoting of shell command for cluster submission.

[5.13.0] - 2020-03-27#


  • Allow to flag directories for inclusion in the report.


  • Fixed hash computation for –cache in case of positional params arguments.

  • Automatically restrict thread usage of linear algebra libraries to whatever is specified in the rule/job.

[5.12.3] - 2020-03-24#


  • Various minor bug fixes.

[5.12.2] - 2020-03-24#


  • Further improved linter output.

[5.12.1] - 2020-03-24#


  • Linter fixes

[5.12.0] - 2020-03-24#


  • Fixed the ability to supply functions for the thread directive.

  • Improved error messages for caching.


  • A new “cache: true” directive that allows to annotate between workflow caching eligibility for rules in the workflow.

[5.11.2] - 2020-03-19#


  • Fixed a spurious error message complaining about missing singularity image if –use-singularity is not activated.

[5.11.1] - 2020-03-16#


  • Fixed a KeyError bug when executing a workflow that defines containers without –use-singularity.

[5.11.0] - 2020-03-16#


  • Fixes for environment modules and tibanna-based AWS execution.

  • Fixes for –default-resources defaults.

  • –cores is now a mandatory argument!

  • Automatic checksum validation for google storage.


  • Azure storage authentication via SAS

  • A generic container directive that will in the future allow for other backends than just singularity. This deprecates the singularity directive, which will however stay functional at least until the next major release.

  • envvars directive for asserting environment variable existence. See docs.

  • support for AWS spot instances via –tibanna-config spot=true.

  • Automatic code quality linting via –lint.

[5.10.0] - 2020-01-20#


  • Jupyter notebook integration, see docs. This enables interactive development of certain data analysis parts (e.g. for plotting).

  • Ability to overwrite thread definitions at the command line (--threads rulename=3), thereby improving scalability.

  • Requester pays configuration for google storage remote files.

  • Add keyword allow_missing to expand function, thereby allowing partical expansion by skipping wildcards for which no keywords are defined.


  • Various bug fixes, e.g. for between workflow caching and script execution.

[5.9.1] - 2019-12-20#


  • Added a missing module.

[5.9.0] - 2019-12-20#


  • Support for per-rule environment module definitions to enable HPC specific software deployment (see docs).

  • Allow custom log handler defitions via –log-handler-script (e.g. post errors and progress to a slack channel or send emails).

  • Allow setting threads as a function of the given cores (see docs).


  • Various minor fixes.

[5.8.2] - 2019-12-16#


  • Implemented a multiext helper, allowing to define a set of output files that just differ by extension.


  • Fixed a failure when caching jobs with conda environments.

  • Fixed various minor bugs.

  • Caching now allows to cache the output of rules using multiext.

[5.8.1] - 2019-11-15#


  • Fixed a bug by adding a missing module.

[5.8.0] - 2019-11-15#


  • Blockchain based caching between workflows (in collaboration with Sven Nahnsen from QBiC), see the docs.

  • New flag –skip-cleanup-scripts, that leads to temporary scripts (coming from script or wrapper directive) are not deleted (by Vanessa Sochat).


  • Various bug fixes.

[5.7.4] - 2019-10-23#


  • Various fixes and adaptations in the docker container image and the test suite.

[5.7.1] - 2019-10-16#


  • Ability to print log files of failed jobs with –show-failed-logs.


  • Fixed bugs in tibanna executor.

  • Fixed handling of symbolic links.

  • Fixed typos in help texts.

  • Fixed handling of default resources.

  • Fixed bugs in azure storage backend.

[5.7.0] - 2019-10-07#


  • Fixed various corner case bugs. Many thanks to the community for pull requests and reporting!

  • Container execution adapted to latest singularity.


  • First class support for Amazon cloud execution via a new Tibanna backend. Thanks to Soo Lee from Harvard Biomedical Informatics!

  • Allow multiple config files to be passed via the command line.

  • A new, more detailed way to visualize the DAG (–filegraph). Thanks to Henning Timm!

  • Pathlib compatibility added. Input and output files can now also be Path objects. Thanks to Frederik Boulund!

  • New azure storage remote provider. Transparently access input and output files on Microsoft Azure. Thanks to Sebastian Kurscheid!

[5.6.0] - 2019-09-06#


  • Fix compatibility with latest singularity versions.

  • Various bug fixes (e.g. in cluster error handling, remote providers, kubernetes backend).


  • Add –default-resources flag, that allows to define default resources for jobs (e.g. mem_mb, disk_mb), see docs.

  • Accept --dry-run as a synonym of --dryrun. Other Snakemake options are similarly hyphenated, so other documentation now refers to --dry-run but both (and also -n) will always be accepted equivalently.

[5.5.4] - 2019-07-21#


  • Reports now automatically include workflow code and configuration for improved transparency.

[5.5.3] - 2019-07-11#


  • Various bug fixes.

  • Polished reports.

[5.5.2] - 2019-06-25#


  • Various minor bug fixes in reports.

  • Speed improvements when using checkpoints.

[5.5.1] - 2019-06-18#


  • Improved report interface. In particular for large files.

  • Small TSV tables are automatically rendered as HTML with datatables.

  • Be more permissive with Snakefile choices: allow “Snakefile”, “snakefile”, “workflow/Snakefile”, “workflow/snakefile”.

[5.5.0] - 2019-05-31#


  • Script directives now also support Julia.


  • Various small bug fixes.

[5.4.5] - 2019-04-12#


  • Fixed a bug with pipe output.

  • Cleaned up error output.

[5.4.4] - 2019-03-22#


  • Vastly improved performance of HTML reports generated with –report, via a more efficient encoding of dara-uri based download links.

  • Tighter layout, plus thumbnails and a lightbox for graphical results in HTML reports.

  • Bug fix for pipe groups.

  • Updated docs.

  • Better error handling in DRMAA executor.

[5.4.3] - 2019-03-11#


  • More robust handling of conda environment activation that should work with all setups where the conda is available when starting snakemake.

  • Fixed bugs on windows.

[5.4.2] - 2019-02-15#


  • Fixed a bug where git module cannot be imported from wrapper.

[5.4.1] - 2019-02-14#


  • Warning when R script is used in combination with conda and R_LIBS environment variable is set. This can cause unexpected results and should be avoided.


  • Improved quoting of paths in conda commands.

  • Fixed various issues with checkpoints.

  • Improved error messages when combining groups with cluster config.

  • Fixed bugs in group implementation.

  • Fixed singularity in combination with shadow.

[5.4.0] - 2018-12-18#


  • Snakemake now allows for data-dependent conditional re-evaluation of the job DAG via checkpoints. This feature also deprecates the dynamic flag. See the docs.

[5.3.1] - 2018-12-06#


  • Various fixed bugs and papercuts, e.g., in group handling, kubernetes execution, singularity support, wrapper and script usage, benchmarking, schema validation.

[5.3.0] - 2018-09-18#


  • Snakemake workflows can now be exported to CWL via the flag –export-cwl, see the docs.


  • Fixed bug in script and wrapper execution when using --use-singularity --use-conda.

  • Add host argument to S3RemoteProvider.

  • Various minor bug fixes.

[5.2.4] - 2018-09-10#


  • New command line flag –shadow-prefix


  • Fixed permission issue when using the script directive. This is a breaking change for scripts referring to files relative to the script directory (see the docs).

  • Fixed various minor bugs and papercuts.

  • Allow URL to local git repo with wrapper directive (git+file:///path/to/your/repo/path_to_file@@version)

[5.2.2] - 2018-08-01#


  • Always print timestamps, removed the –timestamps CLI option.

  • more robust detection of conda command

  • Fixed bug in RMarkdown script execution.

  • Fixed a bug in detection of group jobs.

[5.2.0] - 2018-06-28#


  • Directory outputs have to marked with directory. This ensures proper handling of timestamps and cleanup. This is a breaking change. Implemented by Rasmus Ågren.

  • Fixed kubernetes tests, fixed kubernetes volume handling. Implemented by Andrew Schriefer.

  • jinja2 and networkx are not optional dependencies when installing via pip.

  • When conda or singularity directives are used and the corresponding CLI flags are not specified, the user is notified at the beginning of the log output.

  • Fixed numerous small bugs and papercuts and extended documentation.

[5.1.5] - 2018-06-24#


  • fixed missing version info in docker image.

  • several minor fixes to EGA support.

[5.1.4] - 2018-05-28#


  • Allow category to be set.


  • Various cosmetic changes to reports.

  • Fixed encoding issues in reports.

[5.1.3] - 2018-05-22#


  • Fixed various bugs in job groups, shadow directive, singularity directive, and more.

[5.1.2] - 2018-05-18#


  • Fixed a bug in the report stylesheet.

[5.1.0] - 2018-05-17#


  • A new framework for self-contained HTML reports, including results, statistics and topology information. In future releases this will be further extended.

  • A new utility snakemake.utils.validate() which allows to validate config and pandas data frames using JSON schemas.

  • Two new flags –cleanup-shadow and –cleanup-conda to clean up old unused conda and shadow data.


  • Benchmark repeats are now specified inside the workflow via a new flag repeat().

  • Command line interface help has been refactored into groups for better readability.

[5.0.0] - 2018-05-11#


  • Group jobs for reduced queuing and network overhead, in particular with short running jobs.

  • Output files can be marked as pipes, such that producing and consuming job are executed simultaneously and interfomation is transferred directly without using disk.

  • Command line flags to clean output files.

  • Command line flag to list files in working directory that are not tracked by Snakemake.


  • Fix of –default-remote-prefix in case of input functions returning lists or dicts.

  • Scheduler no longer prefers jobs with many downstream jobs.

[4.8.1] - 2018-04-25#


  • Allow URLs for the conda directive. # Changed

  • Various minor updates in the docs.

  • Several bug fixes with remote file handling.

  • Fix ImportError occuring with script directive.

  • Use latest singularity.

  • Improved caching for file existence checks. We first check existence of parent directories and cache these results. By this, large parts of the generated FS tree can be pruned if files are not yet present. If files are present, the overhead is minimal, since the checks for the parents are cached.

  • Various minor bug fixes.

[4.8.0] - 2018-03-13#


  • Integration with CWL: the cwl directive allows to use CWL tool definitions in addition to shell commands or Snakemake wrappers.

  • A global singularity directive allows to define a global singularity container to be used for all rules that don’t specify their own.

  • Singularity and Conda can now be combined. This can be used to specify the operating system (via singularity), and the software stack (via conda), without the overhead of creating specialized container images for workflows or tasks.

[4.7.0] - 2018-02-19#


  • Speedups when calculating dry-runs.

  • Speedups for workflows with many rules when calculating the DAG.

  • Accept SIGTERM to gracefully finish all running jobs and exit.

  • Various minor bug fixes.

[4.6.0] - 2018-02-06#


  • Log files can now be used as input files for other rules.

  • Adapted to changes in Kubernetes client API.

  • Fixed minor issues in –archive option.

  • Search path order in scripts was changed to fix a bug with leaked packages from root env when using script directive together with conda.

[4.5.1] - 2018-02-01#


  • Input and output files can now tag pathlib objects. # ## Changed

  • Various minor bug fixes.

[4.5.0] - 2018-01-18#


  • iRODS remote provider # ## Changed

  • Bug fix in shell usage of scripts and wrappers.

  • Bug fixes for cluster execution, –immediate-submit and subworkflows.

[4.4.0] - 2017-12-21#


  • A new shadow mode (minimal) that only symlinks input files has been added.


  • The default shell is now bash on linux and macOS. If bash is not installed, we fall back to sh. Previously, Snakemake used the default shell of the user, which defeats the purpose of portability. If the developer decides so, the shell can be always overwritten using shell.executable().

  • Snakemake now requires Singularity 2.4.1 at least (only when running with –use-singularity).

  • HTTP remote provider no longer automatically unpacks gzipped files.

  • Fixed various smaller bugs.

[4.3.1] - 2017-11-16#


  • List all conda environments with their location on disk via –list-conda-envs.


  • Do not clean up shadow on dry-run.

  • Allow R wrappers.

[4.3.0] - 2017-10-27#


  • GridFTP remote provider. This is a specialization of the GFAL remote provider that uses globus-url-copy to download or upload files. # ## Changed

  • Scheduling and execution mechanisms have undergone a major revision that removes several potential (but rare) deadlocks.

  • Several bugs and corner cases of the singularity support have been fixed.

  • Snakemake now requires singularity 2.4 at least.

[4.2.0] - 2017-10-10#


  • Support for executing jobs in per-rule singularity images. This is meant as an alternative to the conda directive (see docs), providing even more guarantees for reproducibility.


  • In cluster mode, jobs that are still running after Snakemake has been killed are automatically resumed.

  • Various fixes to GFAL remote provider.

  • Fixed –summary and –list-code-changes.

  • Many other small bug fixes.

[4.1.0] - 2017-09-26#


  • Support for configuration profiles. Profiles allow to specify default options, e.g., a cluster submission command. They can be used via ‘snakemake –profile myprofile’. See the docs for details.

  • GFAL remote provider. This allows to use GridFTP, SRM and any other protocol supported by GFAL for remote input and output files.

  • Added –cluster-status flag that allows to specify a command that returns jobs status. # ## Changed

  • The scheduler now tries to get rid of the largest temp files first.

  • The Docker image used for kubernetes support can now be configured at the command line.

  • Rate-limiting for cluster interaction has been unified.

  • S3 remote provider uses boto3.

  • Resource functions can now use an additional attempt parameter, that contains the number of times this job has already been tried.

  • Various minor fixes.

[4.0.0] - 2017-07-24#


  • Cloud computing support via Kubernetes. Snakemake workflows can be executed transparently in the cloud, while storing input and output files within the cloud storage (e.g. S3 or Google Storage). I.e., this feature does not need a shared filesystem between the cloud notes, and thereby makes the setup really simple.

  • WebDAV remote file support: Snakemake can now read and write from WebDAV. Hence, it can now, e.g., interact with Nextcloud or Owncloud.

  • Support for default remote providers: define a remote provider to implicitly use for all input and output files.

  • Added an option to only create conda environments instead of executing the workflow. # ## Changed

  • The number of files used for the metadata tracking of Snakemake (e.g., code, params, input changes) in the .snakemake directory has been reduced by a factor of 10, which should help with NFS and IO bottlenecks. This is a breaking change in the sense that Snakemake 4.x won’t see the metadata of workflows executed with Snakemake 3.x. However, old metadata won’t be overwritten, so that you can always go back and check things by installing an older version of Snakemake again.

  • The google storage (GS) remote provider has been changed to use the google SDK. This is a breaking change, since the remote provider invocation has been simplified (see docs).

  • Due to WebDAV support (which uses asyncio), Snakemake now requires Python 3.5 at least.

  • Various minor bug fixes (e.g. for dynamic output files).

[3.13.3] - 2017-06-23#


  • Fix a followup bug in Namedlist where a single item was not returned as string.

[3.13.2] - 2017-06-20#


  • The –wrapper-prefix flag now also affects where the corresponding environment definition is fetched from.

  • Fix bug where empty output file list was recognized as containing duplicates (issue #574).

[3.13.1] - 2017-06-20#


  • Fix –conda-prefix to be passed to all jobs.

  • Fix cleanup issue with scripts that fail to download.

[3.13.0] - 2017-06-12#


  • An NCBI remote provider. By this, you can seamlessly integrate any NCBI resouce (reference genome, gene/protein sequences, …) as input file. # ## Changed

  • Snakemake now detects if automatically generated conda environments have to be recreated because the workflow has been moved to a new path.

  • Remote functionality has been made more robust, in particular to avoid race conditions.

  • --config parameter evaluation has been fixed for non-string types.

  • The Snakemake docker container is now based on the official debian image.

[3.12.0] - 2017-05-09#


  • Support for RMarkdown (.Rmd) in script directives.

  • New option –debug-dag that prints all decisions while building the DAG of jobs. This helps to debug problems like cycles or unexpected MissingInputExceptions.

  • New option –conda-prefix to specify the place where conda environments are stored.


  • Benchmark files now also include the maximal RSS and VMS size of the Snakemake process and all sub processes.

  • Speedup conda environment creation.

  • Allow specification of DRMAA log dir.

  • Pass cluster config to subworkflow.

[3.11.2] - 2017-03-15#


  • Fixed fix handling of local URIs with the wrapper directive.

[3.11.1] - 2017-03-14#


  • –touch ignores missing files

  • Fixed handling of local URIs with the wrapper directive.

[3.11.0] - 2017-03-08#


  • Param functions can now also refer to threads. # ## Changed

  • Improved tutorial and docs.

  • Made conda integration more robust.

  • None is converted to NULL in R scripts.

[3.10.2] - 2017-02-28#


  • Improved config file handling and merging.

  • Output files can be referred in params functions (i.e. lambda wildcards, output: …)

  • Improved conda-environment creation.

  • Jobs are cached, leading to reduced memory footprint.

  • Fixed subworkflow handling in input functions.

[3.10.0] - 2017-01-18#


  • Workflows can now be archived to a tarball with snakemake --archive my-workflow.tar.gz. The archive contains all input files, source code versioned with git and all software packages that are defined via conda environments. Hence, the archive allows to fully reproduce a workflow on a different machine. Such an archive can be uploaded to Zenodo, such that your workflow is secured in a self-contained, executable way for the future. # ## Changed

  • Improved logging.

  • Reduced memory footprint.

  • Added a flag to automatically unpack the output of input functions.

  • Improved handling of HTTP redirects with remote files.

  • Improved exception handling with DRMAA.

  • Scripts referred by the script directive can now use locally defined external python modules.

[3.9.1] - 2016-12-23#


  • Jobs can be restarted upon failure (–restart-times). # ## Changed

  • The docs have been restructured and improved. Now available under

  • Changes in scripts show up with –list-code-changes.

  • Duplicate output files now cause an error.

  • Various bug fixes.

[3.9.0] - 2016-11-15#


  • Ability to define isolated conda software environments (YAML) per rule. Environments will be deployed by Snakemake upon workflow execution.

  • Command line argument –wrapper-prefix in order to overwrite the default URL for looking up wrapper scripts. # ## Changed

  • –summary now displays the log files correspoding to each output file.

  • Fixed hangups when using run directive and a large number of jobs

  • Fixed pickling errors with anonymous rules and run directive.

  • Various small bug fixes

[3.8.2] - 2016-09-23#


  • Add missing import in

  • Use threading only in cluster jobs.

[3.8.1] - 2016-09-14#


  • Snakemake now warns when using relative paths starting with “./”.

  • The option -R now also accepts an empty list of arguments.

  • Bug fix when handling benchmark directive.

  • Jobscripts exit with code 1 in case of failure. This should improve the error messages of cluster system.

  • Fixed a bug in SFTP remote provider.

[3.8.0] - 2016-08-26#


  • Wildcards can now be constrained by rule and globally via the new wildcard_constraints directive (see the docs).

  • Subworkflows now allow to overwrite their config file via the configfile directive in the calling Snakefile.

  • A method log_fmt_shell in the snakemake proxy object that is available in scripts and wrappers allows to obtain a formatted string to redirect logging output from STDOUT or STDERR.

  • Functions given to resources can now optionally contain an additional argument input that refers to the input files.

  • Functions given to params can now optionally contain additional arguments input (see above) and resources. The latter refers to the resources.

  • It is now possible to let items in shell commands be automatically quoted (see the docs). This is usefull when dealing with filenames that contain whitespaces.


  • Snakemake now deletes output files before job exection. Further, it touches output files after job execution. This solves various problems with slow NFS filesystems.

  • A bug was fixed that caused dynamic output rules to be executed multiple times when forcing their execution with -R.

  • A bug causing double uploads with remote files was fixed. Various additional bug fixes related to remote files.

  • Various minor bug fixes.

[3.7.1] - 2016-05-16#


  • Fixed a missing import of the multiprocessing module.

[3.7.0] - 2016-05-05#


  • The entries in resources and the threads job attribute can now be callables that must return int values.

  • Multiple --cluster-config arguments can be given to the Snakemake command line. Later one override earlier ones.

  • In the API, multiple cluster_config paths can be given as a list, alternatively to the previous behaviour of expecting one string for this parameter.

  • When submitting cluster jobs (either through --cluster or --drmaa), you can now use --max-jobs-per-second to limit the number of jobs being submitted (also available through Snakemake API). Some cluster installations have problems with too many jobs per second.

  • Wildcard values are now printed upon job execution in addition to input and output files. # ## Changed

  • Fixed a bug with HTTP remote providers.

[3.6.1] - 2016-04-08#


  • Work around missing RecursionError in Python < 3.5

  • Improved conversion of numpy and pandas data structures to R scripts.

  • Fixed locking of working directory.

[3.6.0] - 2016-03-10#


  • onstart handler, that allows to add code that shall be only executed before the actual workflow execution (not on dryrun).

  • Parameters defined in the cluster config file are now accessible in the job properties under the key “cluster”.

  • The wrapper directive can be considered stable. # ## Changed

  • Allow to use rule/job parameters with braces notation in cluster config.

  • Show a proper error message in case of recursion errors.

  • Remove non-empty temp dirs.

  • Don’t set the process group of Snakemake in order to allow kill signals from parent processes to be propagated.

  • Fixed various corner case bugs.

  • The params directive no longer converts a list l implicitly to " ".join(l).

[3.5.5] - 2016-01-23#


  • New experimental wrapper directive, which allows to refer to re-usable wrapper scripts. Wrappers are provided in the Snakemake Wrapper Repository.

  • David Koppstein implemented two new command line options to constrain the execution of the DAG of job to sub-DAGs (–until and –omit-from). # ## Changed

  • Fixed various bugs, e.g. with shadow jobs and –latency-wait.

[3.5.4] - 2015-12-04#


  • The params directive now fully supports non-string parameters. Several bugs in the remote support were fixed.

[3.5.3] - 2015-11-24#


  • The missing remote module was added to the package.

[3.5.2] - 2015-11-24#


  • Support for easy integration of external R and Python scripts via the new script directive.

  • Chris Tomkins-Tinch has implemented support for remote files: Snakemake can now handle input and output files from Amazon S3, Google Storage, FTP, SFTP, HTTP and Dropbox.

  • Simon Ye has implemented support for sandboxing jobs with shadow rules.


  • Manuel Holtgrewe has fixed dynamic output files in combination with multiple wildcards.

  • It is now possible to add suffixes to all shell commands with shell.suffix(“mysuffix”).

  • Job execution has been refactored to spawn processes only when necessary, resolving several problems in combination with huge workflows consisting of thousands of jobs and reducing the memory footprint.

  • In order to reflect the new collaborative development model, Snakemake has moved from my personal bitbucket account to

[3.4.2] - 2015-09-12#


  • Willem Ligtenberg has reduced the memory usage of Snakemake.

  • Per Unneberg has improved config file handling to provide a more intuitive overwrite behavior.

  • Simon Ye has improved the test suite of Snakemake and helped with setting up continuous integration via Codeship.

  • The cluster implementation has been rewritten to use only a single thread to wait for jobs. This avoids failures with large numbers of jobs.

  • Benchmarks are now writing tab-delimited text files instead of JSON.

  • Snakemake now always requires to set the number of jobs with -j when in cluster mode. Set this to a high value if your cluster does not have restrictions.

  • The Snakemake Conda package has been moved to the bioconda channel.

  • The handling of Symlinks was improved, which made a switch to Python 3.3 as the minimum required Python version necessary.

[3.4.1] - 2015-08-05#


  • This release fixes a bug that caused named input or output files to always be returned as lists instead of single files.

[3.4] - 2015-07-18#


  • This release adds support for executing jobs on clusters in synchronous mode (e.g. qsub -sync). Thanks to David Alexander for implementing this.

  • There is now vim syntax highlighting support (thanks to Jay Hesselberth).

  • Snakemake is now available as Conda package.


  • Lots of bugs have been fixed. Thanks go to e.g. David Koppstein, Marcel Martin, John Huddleston and Tao Wen for helping with useful reports and debugging.

See here for older changes.