builtbykev/vyndr
1
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity

Session 46: Grade card intel + name normalization + pitchers (2122 tests)

Browse Source 
Three focused P1 fixes on the Session-45 snapshot model.

- Grade card intel ROOT CAUSE: gameLogService is NBA/WNBA-only (offline Python),
  so MLB props never got l5_avg/l20_avg and buildIntelFields returned {}. Wired
  MLB game logs into featureCache.gameLogFeatures via mlbStatsAdapter.getPlayerStats
  (pure mlbGameLogFeatures + MLB stat_type->field map). buildIntelFields gained
  playerStats/projection fallbacks for partial intel.
- Player name normalization: src/utils/playerName.js (+ web/src/lib copy):
  normalizeName -> {display,key}. Strips periods, de-dots suffix, accent-folds
  the key. Applied in snapshotService grouping, slateAdapter grade index +
  player-strip merge (variants collapse, longest name shown), and
  playerIntelService. "A.J. Ewing"/"AJ Ewing" + "Jazz Chisholm"/"Jr." now merge.
- MLB starting pitchers: new GET /api/schedule/:sport/pitchers (probablePitchers
  service wrapping mlbStatsAdapter.getScheduleWithPitchers + best-effort ERA).
  Slate fetches it, builds a team->pitcher map (full name + mascot match),
  attaches pitchers to MLB GameCardData. + Next proxy.

Backend 2100 -> 2122 tests (+22), 176 suites. Web build clean (exit 0).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Kev
2026-06-18 23:56:26 -04:00
parent f8b120c0aa
commit c8fc9f577e
20 changed files with 608 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/services/intelligence/analyzeViaEngine1.js
+19 -6
View File
@@ -296,18 +296,31 @@ function matchupGradeFromRank(rank) {
* yield the fallback. The archetype strip lights up once the snapshot pipeline
* (Session 44) feeds per-player season lines into the grade response.
*/
function buildIntelFields(features = {}) {
const firstFinite = (...vals) => vals.find((v) => Number.isFinite(v));
function buildIntelFields(features = {}, opts = {}) {
const out = {};
const round1 = (n) => Math.round(n * 10) / 10;
if (Number.isFinite(features.l20_avg)) out.season_avg = round1(features.l20_avg);
else if (Number.isFinite(features.season_avg)) out.season_avg = round1(features.season_avg);
if (Number.isFinite(features.l10_avg)) out.last10_avg = round1(features.l10_avg);
else if (Number.isFinite(features.l5_avg)) out.last10_avg = round1(features.l5_avg);
// Resilience (Session 46): fall back to a caller-supplied playerStats bundle
// and the model projection when the feature vector is sparse. Partial intel
// beats none — we add only the fields we can actually back with a number.
const ps = opts.playerStats || {};
const proj = Number.isFinite(opts.projection) ? opts.projection : undefined;
const form = computeFormScore(features);
const seasonAvg = firstFinite(features.l20_avg, features.season_avg, ps.season_avg, proj);
if (seasonAvg != null) out.season_avg = round1(seasonAvg);
const last10 = firstFinite(features.l10_avg, features.l5_avg, ps.last10_avg);
if (last10 != null) out.last10_avg = round1(last10);
let form = computeFormScore(features);
if (form == null && Number.isFinite(ps.form)) form = Math.round(ps.form);
if (form != null) out.form = form;
if (Number.isFinite(features.usage_rate)) out.usage = `${round1(features.usage_rate)}%`;
else if (Number.isFinite(features.minutes_per_game)) out.usage = `${Math.round(features.minutes_per_game)} min`;
else if (ps.usage) out.usage = String(ps.usage);
const matchup = matchupGradeFromRank(features.opp_rank_stat);
if (matchup) out.matchup_grade = matchup;
if (Number.isFinite(features.rest_days)) out.rest = `${features.rest_days}d rest`;