feat(spur-cli): honor SRUN_/SALLOC_/SLURM_ env-var defaults for srun …

[biluriuday]

…and salloc

Summary

Adds environment-variable defaults for srun and salloc, aligning Spur with Slurm's precedence model (closes #279). Also fixes --gres CLI parsing to match Slurm semantics.

Precedence for every affected flag is now: CLI arg > SPUR_* native twin > SLURM_*/SRUN_*/SALLOC_* > built-in default.

What changed

• New shared helper module (env_defaults.rs) — apply_str/apply_string/apply_flag/apply_csv resolve multi-alias env vars with correct CLI precedence, detected via ArgMatches::value_source. Includes a shared EnvGuard test fixture and unit tests.
• srun — parses into ArgMatches, then applies SLURM_* (plus genuine SRUN_* where Slurm uses them) and SPUR_* twins as defaults. This lets srun inside an allocation inherit SLURM_NTASKS/SLURM_CPUS_PER_TASK/SLURM_NNODES, so a step sizes to the allocation instead of the bare -n1 -c1 defaults.
• salloc — same pattern for SALLOC_*/SPUR_*. Stays Slurm-faithful: --nodes, --ntasks, --cpus-per-task, and --job-name have no env var in real Slurm, so Spur ignores them too.
• --gres fix (srun + salloc) — added value_delimiter = ',' + overrides_with = "gres" so a comma-list (--gres=gpu:1,fpga:2) expands to multiple GRES and a repeated --gres replaces (last wins), matching Slurm and Spur's own sbatch. Previously the CLI kept comma-lists as one un-parseable element and accumulated repeated flags.

Design notes

• Clap's #[arg(env=...)] supports only one env var per flag, so multi-alias resolution is done in a post-parse step keyed off value_source (keeps CLI-set detection accurate even with overrides_with).
• Env-dependent tests run serially (serial_test) and use EnvGuard, which clears all SPUR_/SLURM_/SALLOC_/SRUN_-prefixed vars on construction/drop, keeping them independent of the runner's environment and of each other.

Testing

• cargo clippy -p spur-cli --all-targets — clean.
• cargo test -p spur-cli — all 151 tests pass (new unit tests cover defaults, CLI override, alias precedence, numeric/bool/CSV parsing, gres comma-split, and last-wins).
• Bare-metal cluster e2e (3 nodes, non-Raft): verified on the deployed binaries:
  • salloc: SALLOC_PARTITION env default applied; CLI -p overrides it.
  • srun: SLURM_NTASKS env default applied to submitted job; CLI -n overrides it; SPUR_NTASKS takes precedence over the SLURM_NTASKS twin.

Known gap / follow-up

sbatch still resolves env defaults via clap's single SBATCH_* attribute per flag and does not yet honor SPUR_* twins (e.g. SPUR_PARTITION), unlike srun/salloc. Unifying sbatch onto env_defaults is deferred to a follow-up PR since it must preserve sbatch's CLI > env > #SBATCH directive precedence.

[Copilot]

Pull request overview

Adds Slurm-compatible environment-variable defaulting for srun/salloc (with SPUR_* native twins and CLI precedence), and updates --gres parsing to match Slurm’s comma-splitting + last-wins behavior.

Changes:

• Introduces env_defaults.rs helpers to apply ordered env-var aliases only when a flag wasn’t set on the CLI (via ArgMatches::value_source).
• Updates srun and salloc to parse into ArgMatches, then apply env defaults post-parse.
• Fixes --gres to split comma-lists and make repeated --gres replace previous values (last wins).

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
crates/spur-cli/src/srun.rs	Switches to matches-based parsing, applies env defaults, and updates --gres parsing semantics.
crates/spur-cli/src/salloc.rs	Switches to matches-based parsing, applies env defaults, and updates --gres parsing semantics.
crates/spur-cli/src/main.rs	Registers the new shared env-defaults module.
crates/spur-cli/src/env_defaults.rs	Adds shared env resolution helpers and env-isolation test guard + unit tests.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 98.15838% with 10 lines in your changes missing coverage. Please review.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #421      +/-   ##
==========================================
+ Coverage   70.86%   71.21%   +0.34%     
==========================================
  Files         140      141       +1     
  Lines       41849    42390     +541     
==========================================
+ Hits        29656    30185     +529     
- Misses      12193    12205      +12     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.