feat(batch): upfront target validation (v0.2.2)

[aborruso]

Summary

• batch ora valida subito che --to sia un target supportato per --from, prima ancora di leggere il file CSV
• Errore chiaro con hint: Run 'openverto targets <from>' to see valid options
• Bump versione 0.2.1 → 0.2.2

Test plan

• test_batch_rejects_invalid_target: EPSG 32633 rifiutato come target di 23033, exit code ≠ 0, messaggio corretto
• test_batch_accepts_valid_target_combination: percorso valido 23033→6706 non bloccato dalla validazione
• pytest tests/test_offline.py → 19 passed

🤖 Generated with Claude Code

[Copilot]

Pull request overview

This PR introduces an upfront validation step in the batch CLI command to ensure --to is a supported target for the given --from before processing the input CSV, adds offline tests to cover the new behavior, and bumps the project version to v0.2.2.

Changes:

• Add upfront --from/--to compatibility validation in openverto batch, with a clearer error message and hint.
• Add two offline CLI tests covering rejection/acceptance of target combinations.
• Bump version to 0.2.2 and document the release in LOG.md.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
src/openverto/cli.py	Adds upfront target validation to batch before reading the CSV.
tests/test_offline.py	Adds offline CLI tests for invalid/valid --from/--to pairs.
pyproject.toml	Version bump 0.2.1 → 0.2.2.
LOG.md	Changelog entry for v0.2.2.

---

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

[greptile-apps]

Greptile Summary

This PR adds an upfront validation step to the batch command that checks whether --to is a valid conversion target for --from before reading the CSV file, giving users a clear error message with a hint to run openverto targets <from>. The version is bumped from 0.2.1 to 0.2.2.

• batch now calls lib_targets(in_e, with_live=False) immediately after parsing EPSG codes and exits with a descriptive error if the target is not in the valid set, avoiding a wasted CSV read.
• Two new offline tests verify the rejection path (EPSG 32633 refused for source 23033) and the acceptance path (23033→6706 passes validation and proceeds to a faked network call).

Confidence Score: 4/5

Safe to merge; the change is small, well-tested, and does not affect the conversion path.

The validation logic is correct — _die always raises typer.Exit, so valid_targets is never actually unbound at runtime. The only blemish is that _die is typed -> None rather than -> NoReturn, which leaves the variable looking possibly-unbound to static checkers and creates a latent fragility if _die is ever changed.

src/openverto/cli.py — the _die return annotation.

Important Files Changed

Filename	Overview
src/openverto/cli.py	Adds upfront target validation in batch using lib_targets(in_e, with_live=False); logic is correct but valid_targets appears possibly-unbound to static checkers because _die lacks a NoReturn annotation.
tests/test_offline.py	Two new offline tests cover the rejection (invalid target) and acceptance (valid target, faked network) paths; both are well-scoped and use monkeypatching correctly.
pyproject.toml	Version bumped from 0.2.1 to 0.2.2; no other changes.
LOG.md	Changelog entry added for v0.2.2 describing the new upfront validation and tests.

_{Reviews (1): Last reviewed commit: "fix(batch): use with_live=False for upfr..." | Re-trigger Greptile}

[greptile-apps]

Potentially-unbound variable visible to static analysis

valid_targets is only assigned inside the try block. If KeyError is raised and _die is called, execution never reaches the if statement below — but _die is typed -> None, not -> NoReturn. Static type checkers (mypy/pyright) will therefore report valid_targets as possibly unbound on the next line, and any future change to _die that makes it return normally (e.g. for testing) would produce a NameError. Annotating _die as -> NoReturn (or restructuring with an else block) is the minimal fix.