cli/pkg/collect/excavate_test.go
Snider f2bc912ebe feat: infrastructure packages and lint cleanup (#281)
* ci: consolidate duplicate workflows and merge CodeQL configs

Remove 17 duplicate workflow files that were split copies of the
combined originals. Each family (CI, CodeQL, Coverage, PR Build,
Alpha Release) had the same job duplicated across separate
push/pull_request/schedule/manual trigger files.

Merge codeql.yml and codescan.yml into a single codeql.yml with
a language matrix covering go, javascript-typescript, python,
and actions — matching the previous default setup coverage.

Remaining workflows (one per family):
- ci.yml (push + PR + manual)
- codeql.yml (push + PR + schedule, all languages)
- coverage.yml (push + PR + manual)
- alpha-release.yml (push + manual)
- pr-build.yml (PR + manual)
- release.yml (tag push)
- agent-verify.yml, auto-label.yml, auto-project.yml

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* feat: add collect, config, crypt, plugin packages and fix all lint issues

Add four new infrastructure packages with CLI commands:
- pkg/config: layered configuration (defaults → file → env → flags)
- pkg/crypt: crypto primitives (Argon2id, AES-GCM, ChaCha20, HMAC, checksums)
- pkg/plugin: plugin system with GitHub-based install/update/remove
- pkg/collect: collection subsystem (GitHub, BitcoinTalk, market, papers, excavate)

Fix all golangci-lint issues across the entire codebase (~100 errcheck,
staticcheck SA1012/SA1019/ST1005, unused, ineffassign fixes) so that
`core go qa` passes with 0 issues.

Closes #167, #168, #170, #250, #251, #252, #253, #254, #255, #256

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
2026-02-04 11:34:43 +00:00

202 lines
4.7 KiB
Go

package collect
import (
"context"
"fmt"
"testing"
"github.com/host-uk/core/pkg/io"
"github.com/stretchr/testify/assert"
)
// mockCollector is a simple collector for testing the Excavator.
type mockCollector struct {
name string
items int
err error
called bool
}
func (m *mockCollector) Name() string { return m.name }
func (m *mockCollector) Collect(ctx context.Context, cfg *Config) (*Result, error) {
m.called = true
if m.err != nil {
return &Result{Source: m.name, Errors: 1}, m.err
}
result := &Result{Source: m.name, Items: m.items}
for i := 0; i < m.items; i++ {
result.Files = append(result.Files, fmt.Sprintf("/output/%s/%d.md", m.name, i))
}
if cfg.DryRun {
return &Result{Source: m.name}, nil
}
return result, nil
}
func TestExcavator_Name_Good(t *testing.T) {
e := &Excavator{}
assert.Equal(t, "excavator", e.Name())
}
func TestExcavator_Run_Good(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
cfg.Limiter = nil
c1 := &mockCollector{name: "source-a", items: 3}
c2 := &mockCollector{name: "source-b", items: 5}
e := &Excavator{
Collectors: []Collector{c1, c2},
}
result, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.True(t, c1.called)
assert.True(t, c2.called)
assert.Equal(t, 8, result.Items)
assert.Len(t, result.Files, 8)
}
func TestExcavator_Run_Good_Empty(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
e := &Excavator{}
result, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.Equal(t, 0, result.Items)
}
func TestExcavator_Run_Good_DryRun(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
cfg.DryRun = true
c1 := &mockCollector{name: "source-a", items: 10}
c2 := &mockCollector{name: "source-b", items: 20}
e := &Excavator{
Collectors: []Collector{c1, c2},
}
result, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.True(t, c1.called)
assert.True(t, c2.called)
// In dry run, mockCollector returns 0 items
assert.Equal(t, 0, result.Items)
}
func TestExcavator_Run_Good_ScanOnly(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
c1 := &mockCollector{name: "source-a", items: 10}
var progressMessages []string
cfg.Dispatcher.On(EventProgress, func(e Event) {
progressMessages = append(progressMessages, e.Message)
})
e := &Excavator{
Collectors: []Collector{c1},
ScanOnly: true,
}
result, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.False(t, c1.called, "Collector should not be called in scan-only mode")
assert.Equal(t, 0, result.Items)
assert.NotEmpty(t, progressMessages)
assert.Contains(t, progressMessages[0], "source-a")
}
func TestExcavator_Run_Good_WithErrors(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
cfg.Limiter = nil
c1 := &mockCollector{name: "good", items: 5}
c2 := &mockCollector{name: "bad", err: fmt.Errorf("network error")}
c3 := &mockCollector{name: "also-good", items: 3}
e := &Excavator{
Collectors: []Collector{c1, c2, c3},
}
result, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.Equal(t, 8, result.Items)
assert.Equal(t, 1, result.Errors) // c2 failed
assert.True(t, c1.called)
assert.True(t, c2.called)
assert.True(t, c3.called)
}
func TestExcavator_Run_Good_CancelledContext(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
ctx, cancel := context.WithCancel(context.Background())
cancel() // Cancel immediately
c1 := &mockCollector{name: "source-a", items: 5}
e := &Excavator{
Collectors: []Collector{c1},
}
_, err := e.Run(ctx, cfg)
assert.Error(t, err)
}
func TestExcavator_Run_Good_SavesState(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
cfg.Limiter = nil
c1 := &mockCollector{name: "source-a", items: 5}
e := &Excavator{
Collectors: []Collector{c1},
}
_, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
// Verify state was saved
entry, ok := cfg.State.Get("source-a")
assert.True(t, ok)
assert.Equal(t, 5, entry.Items)
assert.Equal(t, "source-a", entry.Source)
}
func TestExcavator_Run_Good_Events(t *testing.T) {
m := io.NewMockMedium()
cfg := NewConfigWithMedium(m, "/output")
cfg.Limiter = nil
var startCount, completeCount int
cfg.Dispatcher.On(EventStart, func(e Event) { startCount++ })
cfg.Dispatcher.On(EventComplete, func(e Event) { completeCount++ })
c1 := &mockCollector{name: "source-a", items: 1}
e := &Excavator{
Collectors: []Collector{c1},
}
_, err := e.Run(context.Background(), cfg)
assert.NoError(t, err)
assert.Equal(t, 1, startCount)
assert.Equal(t, 1, completeCount)
}