Release Notes for 0.1.0a6

New Features

tighten Command (#40)

• Ratchets down existing exact-path rules as files shrink
• Removes rules for files now under the threshold
• Never adds new rules or raises limits

Breaking Changes

Reworked baseline Command (#38)

• Now resets all exact-path rules to match current file sizes
• Adds rules for new violations, updates existing rules to current sizes, removes rules for compliant files
• Removed --allow-growth flag — baseline now always reflects reality

Updated relax Command (#39)

• Renamed --buffer to --extra (alias kept for compatibility)
• Default extra changed from 100 to 0

Bug Fixes

Fixed Glob Pattern Matching (#35)

• * now correctly stays within path segments
• ** matches across directories

Fixed relax Exit Code (#36)

• Now returns success when no changes needed (was incorrectly returning failure)

Fixed Output Formatting (#42, #43)

• Unified output formatting across baseline/tighten/relax commands
• Consistent change indicators: + added, ~ updated, - removed
• Fixed sorting order

What's Changed

• Add release scripts by @jakekaplan in #33
• Fix glob pattern matching by @jakekaplan in #35
• fixed relax exit code by @jakekaplan in #36
• Update baseline by @jakekaplan in #38
• add tighten command by @jakekaplan in #40
• update relax by @jakekaplan in #39
• fix tighten and relax by @jakekaplan in #41
• make output consistent by @jakekaplan in #42
• fix sorting, updated output by @jakekaplan in #43
• Centralize and refactor helpers by @jakekaplan in #44
• prep-0.1.0a7 by @jakekaplan in #45

Full Changelog: v0.1.0-alpha.6...v0.1.0-alpha.7

Release Notes for 0.1.0a6

New Features

JSON Output Format (#26)

• Added --output-format json flag to check command for CI/tooling integration
• Includes violations, summary stats, skip warnings, and walk errors

Breaking Changes

Renamed accept-defeat to relax (#30)

• loq accept-defeat → loq relax (same functionality, friendlier name)

What's Changed

• fix accept defeat format by @jakekaplan in #18
• Output format by @jakekaplan in #26
• readme updates by @jakekaplan in #27
• readme updates by @jakekaplan in #28
• more readme cleanup by @jakekaplan in #29
• Rename accept-defeat to relax by @jakekaplan in #30
• clearer naming by @jakekaplan in #31
• prep-0.1.0a6 by @jakekaplan in #32

Full Changelog: v0.1.0-alpha.5...v0.1.0-alpha.6

New Features

accept-defeat command

(#15) Need to ship while files are still too big? The new accept-defeat command creates exact-path rules for files currently failing checks, giving you a buffer to work with:

# File is 847 lines, limit is 500
loq check src/legacy_monster.rs
# ✖  847 > 500   src/legacy_monster.rs

# Accept defeat with default 100-line buffer
loq accept-defeat

# loq.toml now has:
# [[rules]]
# path = "src/legacy_monster.rs"
# max_lines = 947

Options:

• loq accept-defeat src/file.rs — only accept defeat on specific files
• loq accept-defeat --buffer 50 — add 50 lines instead of the default 100

This differs from baseline which ratchets limits down over time. accept-defeat is for when you need to unblock CI right now and will fix the file later.

Full Changelog: v0.1.0-alpha.4...v0.1.0-alpha.5

What's Changed

• small agents.md updates by @jakekaplan in #9
• README.md updates by @jakekaplan in #10
• Add CI benchmarks by @jakekaplan in #11
• written in rust by @jakekaplan in #13
• Accept defeat :( by @jakekaplan in #15
• Improve readme by @jakekaplan in #16
• prep alpha5 by @jakekaplan in #17

Full Changelog: v0.1.0-alpha.4...v0.1.0-alpha.5

loq v0.1.0-alpha.4 Release Notes

Bug Fixes

Exclude patterns now work correctly (#5)

Exclude patterns in config were being ignored. Files in excluded directories (like docs/**) were still being checked and reported as violations. This is now fixed - exclude patterns are properly respected during file walking.

Baseline respects pattern-based rules (#4)

loq baseline was ignoring pattern-based rules and only comparing against default_max_lines. For example, if you had tests/** set to 1000 lines, baseline would still add explicit rules for test files over 500 lines (the default). Now baseline evaluates each file against its effective limit, only adding overrides when files exceed their pattern-matched threshold.

New Features

Custom fix guidance (#2)

New fix_guidance config field lets you customize the guidance message shown when files exceed their limits:

fix_guidance = "Split large files or request an exception in #eng-standards"

--allow-growth flag for baseline (#6)

New --allow-growth flag on loq baseline sets limits to the current line count even when files already have a configured limit. Without this flag, baseline preserves existing limits. With it, limits grow to match reality - useful when you've intentionally grown a file and want to update its limit without manual editing.

# File has 600 lines but limit is 500
loq baseline              # Keeps limit at 500 (existing limit preserved)
loq baseline --allow-growth  # Updates limit to 600

Pre-commit hook support

Added .pre-commit-hooks.yaml for easy integration with pre-commit:

- repo: https://github.com/jakekaplan/loq
  rev: v0.1.0-alpha.4
  hooks:
    - id: loq

Full Changelog: v0.1.0-alpha.3...v0.1.0-alpha.4

What's Changed

• Add fix_guidance config field by @jakekaplan in #3
• Rename --relax to --allow-growth by @jakekaplan in #6
• bump to 0.1.0a4 by @jakekaplan in #7
• fix license by @jakekaplan in #8

Full Changelog: v0.1.0-alpha.3...v0.1.0-alpha.4

Full Changelog: v0.1.0-alpha.2...v0.1.0-alpha.3

Full Changelog: v0.1.0-alpha.2...v0.1.0-alpha.3

Full Changelog: v0.1.0-alpha.1...v0.1.0-alpha.2

Full Changelog: v0.1.0-alpha.1...v0.1.0-alpha.2

Full Changelog: https://github.com/jakekaplan/loq/commits/v0.1.0-alpha.1