Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
5208cbd
feat(portfolio): add bilingual recruiter-facing evidence readout copy
Alexisvop Oct 24, 2026
c2875a6
feat(portfolio): showcase one gain, one regression and one unchanged …
Alexisvop Oct 24, 2026
7b8ed06
style(portfolio): add accessible responsive recruiter case-study cards
Alexisvop Oct 24, 2026
e1eae20
test(portfolio): verify readout evidence values, deep links and EN/ES…
Alexisvop Oct 24, 2026
9997328
docs: add evidence-based recruiter tour and clarify Cloudflare verifi…
Alexisvop Oct 24, 2026
e691e34
docs(portfolio): describe transparent case-study summaries and proven…
Alexisvop Oct 24, 2026
958ca5f
docs: add bilingual, traceable outcomes for technical reviewers
Alexisvop Oct 24, 2026
e0503b9
docs: distinguish historic acceptance candidate from current public r…
Alexisvop Oct 24, 2026
8dda73d
docs: reconcile repository publication state with merged PR #2
Alexisvop Oct 24, 2026
7bd06fa
docs: clarify website report in historical repository publication plan
Alexisvop Oct 24, 2026
a45707f
docs: annotate historical release checklist with reported hosting
Alexisvop Oct 24, 2026
bf896e0
docs: annotate historical audit without rewriting initial findings
Alexisvop Oct 24, 2026
56437cb
docs: mark production readiness evidence as archived pre-merge snapshot
Alexisvop Oct 24, 2026
6da434f
docs: record current publication and bilingual recruiter presentation
Alexisvop Oct 24, 2026
4e2f69f
feat(portfolio): add honest recruiter-oriented social previews
Alexisvop Oct 24, 2026
a5fca18
style(portfolio): prevent narrow-phone histogram metric overflow
Alexisvop Oct 24, 2026
d364242
docs(demo): add evidence-backed 90-second recruiter walkthrough
Alexisvop Oct 24, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion PUBLICATION_PLAN.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Historical independent public repository publication plan

**Completed — do not repeat repository creation, Git initialization, the initial commit or the initial push.** The owner-authorized public release is available at [Oracle-Performance-Analyzer](https://github.com/Alexisvop/Oracle-Performance-Analyzer), on main, with independent root commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e. The [initial GitHub Actions run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) completed successfully. Website deployment remains pending and is not authorized.
**Completed — do not repeat repository creation, Git initialization, the initial commit or the initial push.** The owner-authorized public release is available at [Oracle-Performance-Analyzer](https://github.com/Alexisvop/Oracle-Performance-Analyzer), on main, with independent root commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e. The [initial GitHub Actions run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) completed successfully. The owner now reports a Cloudflare Pages website, but the live revision and settings remain unverified (HTTP 403 to automated checks); this plan confers no new deployment authorization.

The original pre-publication plan below is retained as historical context. Its proposed name, early 404 lookup, pending URL, authorization gates and unexecuted-command statements describe the preparation stage before publication. They are not the current publication status and confer no new authorization. The example commit message below was a proposal; the actual initial message was feat: initial public release of Oracle SQL Performance Analyzer.

Expand Down
2 changes: 1 addition & 1 deletion PUBLIC_RELEASE_AUDIT.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Historical Hardened public-edition release audit

**Historical pre-publication snapshot.** The body below records local preparation before the initial public release; its pending-publication and unexecuted-hosted-CI statements apply to that stage. The [public repository](https://github.com/Alexisvop/Oracle-Performance-Analyzer) has since been published at commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e and its [initial hosted CI run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) passed. Deployment remains pending and is not authorized. The original observations and decisions below are retained for traceability.
**Historical pre-publication snapshot.** The body below records local preparation before the initial public release; its pending-publication and unexecuted-hosted-CI statements apply to that stage. The [public repository](https://github.com/Alexisvop/Oracle-Performance-Analyzer) has since been published at commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e and its [initial hosted CI run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) passed. The owner now reports an existing Pages URL, but automated live verification returned HTTP 403. The original audit below remains a historical pre-publication record and authorizes no new deployment. The original observations and decisions below are retained for traceability.

This is local agent verification, not independent ChatGPT approval or permission to publish. No Git initialization, commit, remote repository creation, push, visibility change, deployment, Oracle connection/mutation or container operation was performed.

Expand Down
2 changes: 1 addition & 1 deletion PUBLIC_RELEASE_CHECKLIST.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
# Historical Hardened public-edition checklist

**Historical pre-publication snapshot.** The body below records local preparation before the initial public release; its pending-publication and unexecuted-hosted-CI statements apply to that stage. The [public repository](https://github.com/Alexisvop/Oracle-Performance-Analyzer) has since been published at commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e and its [initial hosted CI run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) passed. Deployment remains pending and is not authorized. The original observations and decisions below are retained for traceability.
**Historical pre-publication snapshot.** The body below records local preparation before the initial public release; its pending-publication and unexecuted-hosted-CI statements apply to that stage. The [public repository](https://github.com/Alexisvop/Oracle-Performance-Analyzer) has since been published at commit fa7237f071fcc0a6996b986835b2681e0ecc0b2e and its [initial hosted CI run](https://github.com/Alexisvop/Oracle-Performance-Analyzer/actions/runs/39892553663) passed. The owner now reports an existing Pages URL, but automated live verification returned HTTP 403. The original checklist below remains a historical pre-publication record; it authorizes no new deployment. The original observations and decisions below are retained for traceability.

GO for local completion and independent private review. NO-GO for repository creation, publication or deployment until independent review and separate owner authorization. A verdict is not permission.

Expand Down
4 changes: 2 additions & 2 deletions PUBLIC_REPOSITORY.json
Original file line number Diff line number Diff line change
@@ -1,12 +1,12 @@
{
"url": "https://github.com/Alexisvop/Oracle-Performance-Analyzer",
"proposed_name": "Oracle-Performance-Analyzer",
"status": "Public source published; website reported published by owner; final branch integration pending",
"status": "Public repository and PR #2 merged on main; Cloudflare website owner-reported, live revision not independently verified",
"publication_authorized": true,
"portfolio_source_mode": "reviewed same-origin static assets",
"name_check": "Public repository existence and PUBLIC visibility verified; the earlier 404 was a historical pre-publication availability check.",
"website_deployed": true,
"deployment_authorized": false,
"website_url": "https://oracle-performance-analyzer.pages.dev/",
"website_verification": "Owner-reported publication; automated HTTPS requests returned HTTP 403; deployed revision and Cloudflare settings NOT VERIFIED"
"website_verification": "Owner-reported publication at the recorded URL; automated HTTPS retrieval returned HTTP 403. Main and frontend CI pass, but hosted revision, response headers and Cloudflare project settings are NOT independently verified."
}
6 changes: 6 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,12 @@ A reproducible database performance engineering laboratory with a bilingual Reac

> OSPA is an independent educational and engineering project. It is not an official Oracle product and is not affiliated with or certified by Oracle Corporation. It does not replace Oracle Real Application Testing, SQL Performance Analyzer or SQL Tuning Advisor.

## Recruiter quick tour

The [bilingual public portfolio](https://oracle-performance-analyzer.pages.dev/) now opens with a three-case technical readout using the **recorded Phase 4** measurements: **E01** email lookup (observed median **53.17% lower** with an index), **E03** date-range query (observed median **25.24% higher** despite a chosen index), and **E04** status histogram (captured estimated rows **2,500 → 2,500**, without runtime cursor verification). These are descriptive, small-dataset observations, not general tuning guarantees. Each card links to the underlying inspected result and original public evidence. See [methodology](docs/benchmark-methodology.md) and the [recruiter summary](docs/portfolio-summary.md).

**Deployment boundary:** the owner has reported the Cloudflare Pages site as published. Automated retrieval still returned HTTP 403, so its **deployed revision, current headers and hosted execution are not independently verified**. The latest GitHub CI results validate the source branch, not the public hostname. Updating GitHub files in this PR does **not** itself authorize or confirm a Cloudflare deployment.

## Engineering scope

OSPA investigates SQL plans, selectivity, B-tree indexes, optimizer statistics and descriptive benchmarking. The Python CLI uses python-oracledb Thin mode with an approved query catalog and explicit safety guards. Offline reports and the website need no Oracle connection.
Expand Down
2 changes: 1 addition & 1 deletion docs/acceptance.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,4 +4,4 @@ The laboratory supports deterministic synthetic data, approved SQL/correctness c

Release acceptance additionally requires: complete reviewed file allowlist and hash manifest; no private credentials/context/history; exact retained numerical observations; satisfied third-party notices; independently installed offline tests/builds; local source/evidence links; and explicit future publication approval. Current outcomes belong to the release checklist, not historical stage summaries.

Group C and Engineering Challenges are not implemented. No live Oracle test, deployment, repository creation or public-source URL exists for this candidate. [Security](security.md), [development](development.md), [provenance](provenance.md).
Group C and Engineering Challenges are not implemented. The initial pre-publication candidate did not create a repository, publish a site, deploy or execute live Oracle tests. The public source repository now exists, and the owner reports a Pages site; hosted revision verification remains blocked by HTTP 403. Those later publication facts do not change the historical tests or authorize new Oracle/Cloudflare operations. [Security](security.md), [development](development.md), [provenance](provenance.md).
12 changes: 12 additions & 0 deletions docs/demo.md
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,18 @@ Install from the root [README](../README.md), run the offline unit tests, list t

Inspect the generated results/phase5/report.md using the [offline reproducibility instructions](development.md#workflow-a--offline-verification) after generation and compare medians/IQR/p95 against [raw public evidence](evidence/README.md). Build the EN/ES [portfolio](../portfolio/README.md), explore baseline/index/storage views and inspect estimated-plan disclosures. Source links open approved local public assets.

## Recruiter walkthrough — static portfolio (no database access)

Open the [public portfolio](https://oracle-performance-analyzer.pages.dev/)
(owner-reported deployment; live revision not independently verified) and select
EN or ES. Begin with the three-case readout: E01 email lookup (**−53.17%**
median client latency), E03 date range (**+25.24%** with an index range scan)
and E04 histogram (**2,500 → 2,500 E-Rows**, an unchanged estimate—not actual
operator row counts). Follow each case's **Inspect result** and evidence link.
Then explore the fuller Group A, baseline and storage views and the recorded
statistics restoration. These are historical small-profile observations; the
website executes no Oracle SQL, and this walkthrough requires no credentials.

## Flow B — Live Oracle laboratory demonstration (rootless Podman on Arch Linux)

This is a separately authorized lab flow, not a command to start or modify an existing container. Inspect the lab identity and data first; never reuse historical authorization or execute arbitrary SQL. A guarded, read-only catalog/estimated-plan demonstration is described by the CLI help and [query catalog](query-catalog.md). Mutating index/statistics experiments require new scoped approval and exact restoration checks.
Expand Down
18 changes: 18 additions & 0 deletions docs/engineering-log.md
Original file line number Diff line number Diff line change
Expand Up @@ -3,3 +3,21 @@
This is a sanitized public summary, not the private operational journal. Phases 1–3 established guarded connectivity, deterministic synthetic schema/data, approved queries, independent correctness and estimated plans. Phase 4 retained baseline/index/histogram observations, unfavorable outcomes and exact recovery. Phase 5 established offline reports, integrity checks, tests and CI.

The public edition excludes internal prompts, personal context, host diagnostics and Git history. Metadata derivatives are explicitly mapped; measurements and original private records are unchanged. No consumed authorization grants a future operation. [Evidence provenance](evidence/PROVENANCE.json), [methodology](benchmark-methodology.md), [acceptance](acceptance.md).

## 2026-10-23 — Public repository maintenance and recruiter presentation

Public source is available on [GitHub](https://github.com/Alexisvop/Oracle-Performance-Analyzer).
PR #2, the guarded Cloudflare build integration, merged to `main` at
`8b78d2d2910ec49f0a4a3135584714e87cbcacd2`; corresponding offline CI
passed. The owner reports publication at
[oracle-performance-analyzer.pages.dev](https://oracle-performance-analyzer.pages.dev/),
but automated requests return HTTP 403. Public site revision and headers have
not been independently verified, and this record creates no new deployment
or Oracle-operation authorization.

The portfolio now highlights three historical, provenance-backed cases: E01
lower observed median latency, E03 higher observed median despite an index scan,
and E04 unchanged estimated cardinality after histogram collection. English and
Spanish copy is derived exclusively from the existing curated evidence; no
runtime cursor metrics, new database operations or measurements are claimed.
Historical Phase 4 JSON and immutable release manifests remain untouched.
8 changes: 8 additions & 0 deletions docs/final-production-readiness.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,13 @@
# Final production readiness — pre-merge integration report

> **Post-merge status, 2026-10-23:** This is intentionally preserved as the
> **pre-merge report** for PR #2, which was subsequently merged into `main`
> as `8b78d2d2910ec49f0a4a3135584714e87cbcacd2`. GitHub Actions passed
> for that merge commit. The owner reports a Pages deployment, but external
> checks returned HTTP 403; its serving revision, security headers and
> Cloudflare configuration remain **NOT VERIFIED**. The original pending
> statements below describe their dated pre-merge stage, not a current open PR.

## Executive status

OSPA Public Edition is a reproducible synthetic SQL performance laboratory with an offline Python report path and a bilingual static portfolio. The final local integration candidate passes its verification gates. The project is **not closed**: the owner must review and merge the PR, then verify the corresponding stable Cloudflare deployment. This report is a pre-push snapshot; later remote CI observations belong to the PR checks and final review record.
Expand Down
19 changes: 19 additions & 0 deletions docs/portfolio-summary.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,4 +8,23 @@ OSPA is a reproducible synthetic database performance laboratory. It combines ap

OSPA es un laboratorio reproducible de rendimiento SQL con datos sintéticos. Integra consultas aprobadas, validación independiente de resultados, planes estimados, experimentos controlados mediante autorizaciones e informes offline. Conserva resultados favorables y desfavorables; las comparaciones descriptivas no demuestran causalidad. La latencia cliente, el costo estimado y las métricas reales no verificadas se distinguen explícitamente.

## Fast technical readout / Lectura técnica rápida

**EN:** E01 Q01 email-index experiment recorded a **53.17% lower** median client
execute + full-fetch latency. E03 Q07 date-range experiment recorded a **25.24%
higher** median with an INDEX RANGE SCAN. E04 histogram collection left the
captured estimated cardinality at **2,500** rows for the tested binds; this
is not an A-Rows/actual-cursor observation. Each claim is a descriptive Phase 4
small-profile sample result (n = 10 per state), not evidence of causality or
production performance. The [portfolio](https://oracle-performance-analyzer.pages.dev/)
now presents these three outcomes before the deep dive.

**ES:** El experimento E01 Q01 con índice de correo registró una mediana de
latencia cliente **53.17% menor**. E03 Q07 para un rango de fechas registró una
mediana **25.24% mayor** pese al INDEX RANGE SCAN. Tras el histograma E04, la
cardinalidad estimada capturada permaneció en **2,500** filas para los binds
probados; no son filas reales de operadores. Son mediciones descriptivas de
Fase 4 con el perfil pequeño (n = 10 por estado), no pruebas causales ni de
rendimiento en producción.

[Architecture](architecture.md), [results](phase4-results.md), [methodology](benchmark-methodology.md), [source queries](query-catalog.md).
17 changes: 17 additions & 0 deletions portfolio/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,23 @@ The evidence exporter checks 44 input hashes, 23 measurements and eight equivale

[Project README](../README.md), [architecture](../docs/architecture.md), [evidence](../docs/evidence/README.md), [notices](../THIRD_PARTY_NOTICES.md), [security](../docs/security.md).

## Recruiter readout

A visually distinct three-case summary now leads readers from the landing page to
source-backed outcomes: **E01** improved observed median client latency by
53.17%, **E03 / Q07 range** worsened by 25.24% even though the estimated plan
used an index, and the **E04** histogram left the captured estimated E-Rows at
2,500 for the tested bind placeholders. Figures come directly from the existing
`src/data/evidence.json` generated by `scripts/evidence.mjs`; the historical
JSON inputs, hashes, provenance, plans, timing samples and deferred Group C
remain unchanged. All three outcomes are localized into English and Spanish.

The cards link into the original detailed results and locally reviewed evidence.
No new backend, fabricated benchmark or tracking script was introduced. Validate
the new case-study section on mobile and with keyboard navigation before
public deployment. A green local or GitHub build does not verify the inaccessible
hosted Cloudflare origin.

## Prepared Cloudflare Pages settings

Root directory: portfolio; framework: Vite (React); production branch: main; build command: npm run build; output: dist; NODE_VERSION: 26.12.0. The .node-version file pins that runtime and .npmrc disables dependency lifecycle scripts. Keep Cloudflare's automatic dependency installation enabled and use the committed package-lock.json. Local verification uses npm ci.
Expand Down
4 changes: 4 additions & 0 deletions portfolio/index.html
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,10 @@
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<meta name="theme-color" content="#0B1020" />
<meta name="description" content="Explore a reproducible Oracle SQL performance laboratory: verified benchmarks, execution plans, indexing trade-offs, and offline evidence." />
<meta property="og:type" content="website" />
<meta property="og:title" content="OSPA · Oracle SQL Performance Engineering Lab" />
<meta property="og:description" content="Three source-backed SQL tuning cases: a faster lookup, a slower index range scan, and unchanged histogram estimates. Explore reproducible evidence." />
<meta name="twitter:card" content="summary" />
<link rel="icon" type="image/svg+xml" href="%BASE_URL%favicon.svg" />
<title>OSPA — Oracle SQL Performance Analyzer</title>
</head>
Expand Down
20 changes: 20 additions & 0 deletions portfolio/src/App.test.tsx
Original file line number Diff line number Diff line change
Expand Up @@ -91,6 +91,26 @@ describe('Portfolio interactions', () => {
expect(within(screen.getByRole('table')).getAllByRole('row')).toHaveLength(8);
expect(screen.getByText(translations.en.baselineNote)).toBeInTheDocument();
});
it('shows a bilingual, source-backed recruiter readout with favorable and unfavorable cases', async () => {
const user = userEvent.setup();
render(<App />);
expect(screen.getByRole('heading', { name: translations.en.briefTitle })).toBeInTheDocument();
const cards = Array.from(document.querySelectorAll<HTMLElement>('.brief-card'));
expect(cards).toHaveLength(3);
expect(cards[0]).toHaveTextContent(percent(evidence.indexes[0].changePercent));
expect(cards[1]).toHaveTextContent(percent(evidence.indexes[4].changePercent));
expect(cards[2]).toHaveTextContent(
number(evidence.histogram.estimatedRows, 0),
);
expect(within(cards[2]).getByRole('link', { name: translations.en.briefInspect }))
.toHaveAttribute('href', '#histogram');
expect(document.querySelector('#histogram')).not.toBeNull();
for (const card of cards)
expect(within(card).getByRole('link', { name: /Evidence/i })).toHaveAttribute('href', expect.stringContaining('/source/'));
await user.click(screen.getByRole('button', { name: 'ES' }));
expect(screen.getByRole('heading', { name: translations.es.briefTitle })).toBeInTheDocument();
expect(screen.getByText(translations.es.briefCaveat)).toBeInTheDocument();
});
it('opens and closes mobile navigation through accessible controls', async () => {
const user = userEvent.setup();
render(<App />);
Expand Down
Loading
Loading