Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go

Source identity

Field	Recorded value
Source type	GitHub source
Original record	Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go
Date	Not specified in source
Repository	Janady13/missionmatch
Revision	720d47fae476b0d38477ae4401697d372b1ea2bc
Branch	feature/opportunity-to-outcome
Path or archive member	internal/modules/ontology/adapters/corpus_test.go
Original SHA-256	b489635ceeffe84074ff89f0170c44b79ad97a01218662a8c107575468a8348e
Git blob SHA	4b13152b65972c58762310f497a28b6bd711dc64
Reading-copy SHA-256	b489635ceeffe84074ff89f0170c44b79ad97a01218662a8c107575468a8348e

Original source ↗

Every recorded occurrence

These 20 records contain identical original source bytes. Their paths and revisions remain part of the archive.

Original record	Branch or archive	Revision / identifier	Original location
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	feature/opportunity-to-outcome	720d47fae476b0d38477ae4401697d372b1ea2bc	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	collision-deconfliction	4ee5a09972608ed89f8377c6665542d5f450f0a6	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws6-retrieval	6c56e29231ec05735927fa265c97d517499dfeaf	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	m5-price-to-win	41000ad30fff4129fdaf5ec84e15ba14c27d4890	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws1-ontology-sync	b596f449a32a1016fbd403c36505cc00a1ac9ee6	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	master	1e9249840a4302ec82e123b9c9cb6f75badea3e4	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws8-teaching-loop	c823f0fb6bb2aba0f39001eb6de96f4000d9321f	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	mag-substrate	b82c7100a492dec67dfd1eda6b3f53def859be35	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws3-clearance	4b8e7ce594c068621867f9c9deed0ec64ae6c553	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	phase2-eval-fairness	d361af254701975fe239927656ababb56374655e	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	time-to-fill-survival	b66e1e505863d4f17741a4b9ede73509e45bd6c6	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	graphrag-foundation	b0ddeb9f633fd138d09309d1a17b7d13bc3a67f7	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws9-bench-ingestion	f585a9ef32fa554663ce1fba9a414a54b6e49cf0	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws2-alias-relationship	e4dada4601c6eb101830f9d223a9655dfa120616	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	presentation-readiness	eb574acaa29d71ce4d0f852572e87c275c79fdb7	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	vendor-intelligence	fe9796a19c261951d3f2565206c0c1b5ac294465	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	ws7-federal-signal	6a9f6ce9cfb13345ffb47770d4960e7c2d9c17c2	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	phase2-connectors	e038d4c687aec16f8bd9bc4d907eea14d91f8704	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	feature/aioe-upgrade	2178e7298d0e5d9af83afa12402fd587f0ba6ed8	Open original
Janady13/missionmatch / internal/modules/ontology/adapters/corpus_test.go	capture-readiness	bae6af7cb2997e819174a9e9cfff5e16067a1975	Open original

Recovered source transcript

Download complete source text ↓59 lines · 1.6 KB

package adapters

import (
	"context"
	"testing"

	"github.com/missionworkforce/missionmatch/internal/modules/ontology"
	"github.com/missionworkforce/missionmatch/internal/platform/extraction"
)

type stubText struct{ blobs []string }

func (s stubText) Texts(_ context.Context, yield func(string) error) error {
	for _, b := range s.blobs {
		if err := yield(b); err != nil {
			return err
		}
	}
	return nil
}

type stubExtractor struct{ perText map[string][]string }

func (s stubExtractor) ExtractResume(_ context.Context, _ string) (extraction.RawExtraction, error) {
	return extraction.RawExtraction{}, nil
}

func (s stubExtractor) ExtractRequirement(_ context.Context, text string) (extraction.RawExtraction, error) {
	return extraction.RawExtraction{Skills: s.perText[text]}, nil
}

func TestCorpusYieldsDedupedSkills(t *testing.T) {
	blob1 := "Seeking a SIGINT analyst with Splunk and AWS experience."
	blob2 := "Cloud engineer with AWS and Kubernetes."
	ex := stubExtractor{perText: map[string][]string{
		blob1: {"AWS", "Splunk"},
		blob2: {"AWS", "Kubernetes"},
	}}
	src := NewCorpusSourceWith(stubText{blobs: []string{blob1, blob2}}, ex)

	seen := map[string]int{}
	err := src.Fetch(context.Background(), func(r ontology.VocabRecord) error {
		if r.Kind != ontology.KindSkill {
			t.Fatalf("unexpected kind %q", r.Kind)
		}
		seen[r.Term]++
		return nil
	})
	if err != nil {
		t.Fatalf("fetch: %v", err)
	}
	if seen["AWS"] != 1 {
		t.Fatalf("AWS should be deduped to 1, got %d", seen["AWS"])
	}
	if seen["Splunk"] != 1 || seen["Kubernetes"] != 1 {
		t.Fatalf("expected Splunk and Kubernetes once each, got %v", seen)
	}
}