Janady13/missionmatch / internal/platform/sourcing/pdl.go

Source identity

Field	Recorded value
Source type	GitHub source
Original record	Janady13/missionmatch / internal/platform/sourcing/pdl.go
Date	Not specified in source
Repository	Janady13/missionmatch
Revision	720d47fae476b0d38477ae4401697d372b1ea2bc
Branch	feature/opportunity-to-outcome
Path or archive member	internal/platform/sourcing/pdl.go
Original SHA-256	cceb4e4c8fb5c799b7ca792df35e2344188d10966a13f14ee4462be65d827367
Git blob SHA	4fd29c1c83bed8b3ba5b6c8224a5e774b5bc0d0d
Reading-copy SHA-256	cceb4e4c8fb5c799b7ca792df35e2344188d10966a13f14ee4462be65d827367

Original source ↗

Every recorded occurrence

These 36 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/sourcing/pdl.go	feature/opportunity-to-outcome	720d47fae476b0d38477ae4401697d372b1ea2bc	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	per-tenant-routing	9bde0f9a74b89d01980fa658e5253acda7528d0c	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	collision-deconfliction	4ee5a09972608ed89f8377c6665542d5f450f0a6	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws6-retrieval	6c56e29231ec05735927fa265c97d517499dfeaf	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-a-shared-core	e0de03ac3a1ec826bd15d0e048e3b02baa23eeff	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	m5-price-to-win	41000ad30fff4129fdaf5ec84e15ba14c27d4890	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-c-slatematch	5cd60c918301b46ecd6ed3e03e20c395da0062e2	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws1-ontology-sync	b596f449a32a1016fbd403c36505cc00a1ac9ee6	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	master	1e9249840a4302ec82e123b9c9cb6f75badea3e4	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws8-teaching-loop	c823f0fb6bb2aba0f39001eb6de96f4000d9321f	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	w3-ai-staffing-recommendations	43d56387270ef4da4cabd794d97ca6a49909bc7d	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	mag-substrate	b82c7100a492dec67dfd1eda6b3f53def859be35	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws3-clearance	4b8e7ce594c068621867f9c9deed0ec64ae6c553	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-d-command-center	255e7d1d420abf5e277d0cd8c9b14f6f72b6acb9	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	phase2-eval-fairness	d361af254701975fe239927656ababb56374655e	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-e-dataflow-deploy	433815a042f9741411e60fa553fe72143e9b10db	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ruthless-debug-sweep	b7d742f8f032984be3817efffa3eb8b13b651c1f	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	time-to-fill-survival	b66e1e505863d4f17741a4b9ede73509e45bd6c6	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ai-intelligence-layer	5d5ddff70393a197c5f54a45280bb5dd42b24755	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	graphrag-foundation	b0ddeb9f633fd138d09309d1a17b7d13bc3a67f7	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-a-opportunity-intelligence	1675031db74316e100782d063e9eda1a597fe3a1	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	deploy-and-ops	bdacd9d0b4d5998d25fc74bf4007f4c3ec538c7a	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws9-bench-ingestion	f585a9ef32fa554663ce1fba9a414a54b6e49cf0	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws2-alias-relationship	e4dada4601c6eb101830f9d223a9655dfa120616	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	presentation-readiness	eb574acaa29d71ce4d0f852572e87c275c79fdb7	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	vendor-intelligence	fe9796a19c261951d3f2565206c0c1b5ac294465	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-b-screenmatch	09015496d565a172bcaa15cfe9a8e4b072b13df0	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ws7-federal-signal	6a9f6ce9cfb13345ffb47770d4960e7c2d9c17c2	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-e-commercial-stubs	4437da660539284a400ee380b9b427a300c369ee	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	ai-brain-live	aac32fc49c878cc5bca099459307611562d0bac1	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-d-missionmatch	3be46ba89edb737182075a2b320509b42dc916f8	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	phase2-connectors	e038d4c687aec16f8bd9bc4d907eea14d91f8704	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	workstream-b-sam-contract-awards	d5101cceb98e70000cebb31d3e7ac73d747d3a6d	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	w0-aedt-compliance	39df94f92b2a33f852481fe1954dabed841524bb	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	feature/aioe-upgrade	2178e7298d0e5d9af83afa12402fd587f0ba6ed8	Open original
Janady13/missionmatch / internal/platform/sourcing/pdl.go	capture-readiness	bae6af7cb2997e819174a9e9cfff5e16067a1975	Open original

Recovered source transcript

Download complete source text ↓174 lines · 4.3 KB

package sourcing

import (
	"bytes"
	"context"
	"encoding/json"
	"errors"
	"fmt"
	"io"
	"net/http"
	"strings"
	"time"
)

// PDLBase is the People Data Labs v5 person search endpoint.
const PDLBase = "https://api.peopledatalabs.com/v5/person/search"

// ErrThrottled is returned on a 429 from the sourcing provider.
var ErrThrottled = errors.New("sourcing: rate limited (429)")

// PDL is the People Data Labs Provider. includeContact controls whether email and phone are kept.
type PDL struct {
	key            string
	base           string
	includeContact bool
	http           *http.Client
}

// NewPDL builds the provider. An empty base uses the public endpoint.
func NewPDL(key, base string, includeContact bool) *PDL {
	if strings.TrimSpace(base) == "" {
		base = PDLBase
	}
	return &PDL{key: key, base: base, includeContact: includeContact, http: &http.Client{Timeout: 45 * time.Second}}
}

func (p *PDL) Name() string { return "People Data Labs" }

// Search runs one person search and normalizes the results.
func (p *PDL) Search(ctx context.Context, q SearchQuery) ([]SourcedPerson, error) {
	size := q.Size
	if size < 1 {
		size = 10
	}
	payload, err := json.Marshal(map[string]any{"query": q.ESQuery, "size": size})
	if err != nil {
		return nil, fmt.Errorf("sourcing: marshal request: %w", err)
	}
	req, err := http.NewRequestWithContext(ctx, http.MethodPost, p.base, bytes.NewReader(payload))
	if err != nil {
		return nil, errors.New("sourcing: failed to build request")
	}
	req.Header.Set("X-Api-Key", p.key)
	req.Header.Set("Content-Type", "application/json")

	resp, err := p.http.Do(req)
	if err != nil {
		return nil, errors.New("sourcing: request failed")
	}
	defer func() { _ = resp.Body.Close() }()

	if resp.StatusCode == http.StatusTooManyRequests {
		return nil, ErrThrottled
	}
	if resp.StatusCode != http.StatusOK {
		snippet, _ := io.ReadAll(io.LimitReader(resp.Body, 512))
		return nil, fmt.Errorf("sourcing: status %d: %s", resp.StatusCode, strings.TrimSpace(string(snippet)))
	}

	var r pdlResponse
	if err := json.NewDecoder(resp.Body).Decode(&r); err != nil {
		return nil, fmt.Errorf("sourcing: decode response: %w", err)
	}

	out := make([]SourcedPerson, 0, len(r.Data))
	for _, raw := range r.Data {
		sp, ok := p.normalize(raw)
		if !ok {
			continue
		}
		out = append(out, sp)
	}
	return out, nil
}

// normalize uses map based extraction because PDL data is messy: location_name can be a boolean,
// inferred_years_experience can be a float, and contact arrays vary. Type tolerant helpers keep a
// single odd field from dropping an otherwise good record.
func (p *PDL) normalize(raw json.RawMessage) (SourcedPerson, bool) {
	var m map[string]any
	if err := json.Unmarshal(raw, &m); err != nil {
		return SourcedPerson{}, false
	}
	id := strings.TrimSpace(asString(m["id"]))
	name := strings.TrimSpace(asString(m["full_name"]))
	if id == "" || name == "" {
		return SourcedPerson{}, false
	}
	sp := SourcedPerson{
		ExternalID: id,
		FullName:   name,
		JobTitle:   strings.TrimSpace(asString(m["job_title"])),
		Role:       strings.TrimSpace(asString(m["job_title_role"])),
		Location:   strings.TrimSpace(asString(m["location_name"])),
		Skills:     asStringSlice(m["skills"]),
		Years:      asInt(m["inferred_years_experience"]),
		Raw:        raw,
	}
	if p.includeContact {
		sp.Emails = emailAddresses(m["emails"])
		sp.Phones = asStringSlice(m["phone_numbers"])
	}
	return sp, true
}

type pdlResponse struct {
	Status int               `json:"status"`
	Total  int               `json:"total"`
	Data   []json.RawMessage `json:"data"`
	Error  json.RawMessage   `json:"error"`
}

func asString(v any) string {
	if s, ok := v.(string); ok {
		return s
	}
	return ""
}

func asInt(v any) int {
	switch n := v.(type) {
	case float64:
		if n > 0 {
			return int(n)
		}
	case string:
		var i int
		if _, err := fmt.Sscanf(n, "%d", &i); err == nil && i > 0 {
			return i
		}
	}
	return 0
}

func asStringSlice(v any) []string {
	arr, ok := v.([]any)
	if !ok {
		return nil
	}
	out := make([]string, 0, len(arr))
	for _, e := range arr {
		if s, ok := e.(string); ok && s != "" {
			out = append(out, s)
		}
	}
	return out
}

func emailAddresses(v any) []string {
	arr, ok := v.([]any)
	if !ok {
		return nil
	}
	var out []string
	for _, e := range arr {
		if m, ok := e.(map[string]any); ok {
			if a := asString(m["address"]); a != "" {
				out = append(out, a)
			}
		}
	}
	return out
}