Janady13/missionmatch / internal/platform/missionclass/missionclass.go

Source identity

Field	Recorded value
Source type	GitHub source
Original record	Janady13/missionmatch / internal/platform/missionclass/missionclass.go
Date	Not specified in source
Repository	Janady13/missionmatch
Revision	720d47fae476b0d38477ae4401697d372b1ea2bc
Branch	feature/opportunity-to-outcome
Path or archive member	internal/platform/missionclass/missionclass.go
Original SHA-256	0d4708ffcb79cfe99d254354af377020f8e339bf927486aa52ac72681f95c18b
Git blob SHA	ba519d2208e3a8d64ce9acfeb0df9813535b1d24
Reading-copy SHA-256	0d4708ffcb79cfe99d254354af377020f8e339bf927486aa52ac72681f95c18b

Original source ↗

Every recorded occurrence

These 15 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/platform/missionclass/missionclass.go	feature/opportunity-to-outcome	720d47fae476b0d38477ae4401697d372b1ea2bc	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	collision-deconfliction	4ee5a09972608ed89f8377c6665542d5f450f0a6	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	m5-price-to-win	41000ad30fff4129fdaf5ec84e15ba14c27d4890	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	master	1e9249840a4302ec82e123b9c9cb6f75badea3e4	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	ws8-teaching-loop	c823f0fb6bb2aba0f39001eb6de96f4000d9321f	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	mag-substrate	b82c7100a492dec67dfd1eda6b3f53def859be35	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	time-to-fill-survival	b66e1e505863d4f17741a4b9ede73509e45bd6c6	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	graphrag-foundation	b0ddeb9f633fd138d09309d1a17b7d13bc3a67f7	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	ws9-bench-ingestion	f585a9ef32fa554663ce1fba9a414a54b6e49cf0	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	presentation-readiness	eb574acaa29d71ce4d0f852572e87c275c79fdb7	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	vendor-intelligence	fe9796a19c261951d3f2565206c0c1b5ac294465	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	ws7-federal-signal	6a9f6ce9cfb13345ffb47770d4960e7c2d9c17c2	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	phase2-connectors	e038d4c687aec16f8bd9bc4d907eea14d91f8704	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	feature/aioe-upgrade	2178e7298d0e5d9af83afa12402fd587f0ba6ed8	Open original
Janady13/missionmatch / internal/platform/missionclass/missionclass.go	capture-readiness	bae6af7cb2997e819174a9e9cfff5e16067a1975	Open original

Recovered source transcript

Download complete source text ↓127 lines · 4.5 KB

// Package missionclass is the pure deterministic mission and labor classifier. It tags federal record
// text with structured mission areas (the missionclass.Codes vocabulary) using a keyword gazetteer, the
// WS2 alias layer, and a curated PSC and NAICS crosswalk, each inference carrying the mechanism and the
// evidence phrase so every tag is traceable. The package is pure: no clock, no randomness, no network,
// and no DB, so its output is byte stable for fixed input. Mission tags are NEVER a scoring input and
// never reference the eeo schema.
package missionclass

import (
	"regexp"
	"sort"
	"strings"
)

// Mechanism records HOW a mission was inferred, so every tag is traceable (the WS7 acceptance criterion).
const (
	MechKeyword = "keyword"
	MechAlias   = "alias"
	MechPSC     = "psc"
	MechNAICS   = "naics"
)

// Hit is one mission inference with its provenance.
type Hit struct {
	Code      string
	Mechanism string
	Evidence  string // the matched phrase or code
}

type form struct {
	code string
	re   *regexp.Regexp
	text string // canonical evidence phrase for this form
}

// wbForm compiles a case insensitive, word boundary anchored matcher for a phrase. Any single space in a
// multi word phrase becomes \s+, so a line break or a double space between tokens still matches. The
// phrase is regex quoted, so a literal containing a metacharacter is matched safely.
func wbForm(code, phrase string) form {
	return form{code: code, text: phrase,
		re: regexp.MustCompile(`(?i)\b` + strings.ReplaceAll(regexp.QuoteMeta(phrase), " ", `\s+`) + `\b`)}
}

// missionForms is iterated in Codes order. Keep in sync with ai/mission.go missionVocabulary.
var missionForms = []form{
	wbForm("SIGINT", "sigint"), wbForm("SIGINT", "signals intelligence"),
	wbForm("GEOINT", "geoint"), wbForm("GEOINT", "geospatial intelligence"),
	wbForm("HUMINT", "humint"), wbForm("HUMINT", "human intelligence"),
	wbForm("MASINT", "masint"), wbForm("MASINT", "measurement and signature intelligence"),
	wbForm("OSINT", "osint"), wbForm("OSINT", "open source intelligence"),
	wbForm("SAT_C2", "satellite c2"), wbForm("SAT_C2", "satellite command and control"),
	wbForm("CDS", "cross domain solution"), wbForm("CDS", "cross domain solutions"),
	wbForm("COMSEC", "comsec"), wbForm("COMSEC", "communications security"),
	wbForm("TEMPEST", "tempest"),
	wbForm("EW", "electronic warfare"),
}

// ClassifyKeyword returns one Hit per mission code whose keyword form appears in text, in Codes order.
func ClassifyKeyword(text string) []Hit {
	if strings.TrimSpace(text) == "" {
		return nil
	}
	found := map[string]Hit{}
	for _, f := range missionForms {
		if _, ok := found[f.code]; ok {
			continue
		}
		if f.re.MatchString(text) {
			found[f.code] = Hit{Code: f.code, Mechanism: MechKeyword, Evidence: f.text}
		}
	}
	return ordered(found)
}

// AliasIndex maps a lowercased mission bearing phrase to a mission code. It is built from the curated WS2
// alias rows and the category 'mission_area' skills, so the alias layer (not a cosine threshold) carries
// acronym and synonym knowledge. Phrases are matched on a word boundary, whitespace tolerant.
type AliasIndex struct {
	forms []form
}

// NewAliasIndex builds an AliasIndex from a phrase to code map. The forms are appended in sorted key
// order so the build is deterministic and the resulting index iterates the same way every run.
func NewAliasIndex(phraseToCode map[string]string) AliasIndex {
	keys := make([]string, 0, len(phraseToCode))
	for k := range phraseToCode {
		keys = append(keys, k)
	}
	sort.Strings(keys) // deterministic build order
	ai := AliasIndex{}
	for _, k := range keys {
		ai.forms = append(ai.forms, wbForm(phraseToCode[k], k))
	}
	return ai
}

// ClassifyAlias returns one Hit per mission code whose alias phrase appears in text, in Codes order.
func ClassifyAlias(text string, idx AliasIndex) []Hit {
	if strings.TrimSpace(text) == "" {
		return nil
	}
	found := map[string]Hit{}
	for _, f := range idx.forms {
		if _, ok := found[f.code]; ok {
			continue
		}
		if f.re.MatchString(text) {
			found[f.code] = Hit{Code: f.code, Mechanism: MechAlias, Evidence: f.text}
		}
	}
	return ordered(found)
}

// ordered returns the hits sorted by the canonical Codes order, for byte stable output.
func ordered(m map[string]Hit) []Hit {
	rank := map[string]int{}
	for i, c := range Codes {
		rank[c] = i
	}
	out := make([]Hit, 0, len(m))
	for _, h := range m {
		out = append(out, h)
	}
	sort.Slice(out, func(i, j int) bool { return rank[out[i].Code] < rank[out[j].Code] })
	return out
}