Terminal

Default format. Designed for humans reading a shell, and for CI logs where someone scrolls back to find what broke.

Sample output

ℹ [lazy_class] cha-core/src/cache.rs:8:7-8:16 Class `FileEntry` has only 0 method(s) and 8 lines, consider Inline Class
  → suggested: Inline Class
ℹ [lazy_class] cha-core/src/cache.rs:19:7-19:20 Class `FindingsEntry` has only 0 method(s) and 4 lines, consider Inline Class
  → suggested: Inline Class
ℹ [lazy_class] cha-core/src/cache.rs:26:7-26:16 Class `CacheMeta` has only 0 method(s) and 4 lines, consider Inline Class
  → suggested: Inline Class
ℹ [lazy_class] cha-core/src/cache.rs:298:9-298:19 Class `FileStatus` has only 0 method(s) and 6 lines, consider Inline Class
  → suggested: Inline Class
ℹ [data_class] cha-core/src/cache.rs:8-15 Class `FileEntry` has 4 fields but no behavior methods, consider Move Method
  → suggested: Move Method, Encapsulate Field

26 issue(s) found (0 error, 0 warning, 26 hint). (showing top 5)

Health scores:
  C cha-core/src/cache.rs (~60min debt)

Tech debt: ~1h | A:0 B:0 C:1 D:0 F:0

(Captured with cha analyze --format terminal --top 5 cha-core/src/cache.rs.)

Each finding is one line: <icon> [smell] <path>:<line>[:<col>][-<endline>[:<endcol>]] <message>. Icons are ✗ for error, ⚠ for warning, ℹ for hint. The optional → suggested: … line below names refactorings the detector recommends.

When to use it

  • Local development — running cha analyze directly in a terminal.
  • CI log output where humans (not parsers) read the result.
  • Pre-commit hooks: short, scannable, easy to spot in a wall of git output.
  • Quick triage with --top 10 to see the highest-priority issues first.

Notes / Gotchas

  • When the same smell fires more than 5 times in a single run, the default view aggregates: it prints the smell name once with an occurrence count and shows the top 3 most severe locations followed by … and N more (use --all to show all). Pass --all to expand every group.
  • --top N truncates after sorting by risk score and adds (showing top N) to the summary.
  • The summary line is always printed; an empty run produces No issues found..
  • Health scores and tech-debt totals appear after the findings, derived from the same data.