Bible App Reader

PUBLIC PREVIEW — ACTIVE DEVELOPMENT

The static reader is functional and actively developed. The flexible study workspace, personal Meaning, Study Marks, and My Data remain evolving product surfaces. This repository does not promise a production release or stable API.

Application code, tests, scripts, schemas, and tooling are MIT-licensed. Bundled Bible and study data retains its own source rights and notices. Downstream users must review NOTICE.md and app/data/source-manifest.json before redistribution.

Bible App Reader is a local-first Bible study workspace that runs as a static browser application. It combines multi-translation reading, hover-first supplemental context, Hebrew and Greek Language Study, commentary, cross-references, Strong's lexicons, structured study marks, and portable browser-local data without requiring an account, hosted backend, analytics service, or remote application API.

The default Stable profile preserves the existing experience and storage identities. An explicit local Lab profile (?profile=lab before the hash route) enables complete experimental diagnostics with isolated personal-data, notification, physical-registry, and physical-byte namespaces. Both profiles remain static, local-first, offline-capable browser products and use the same portable bibleapp:user-data version-3 contract.

The app keeps the reader primary while deeper material remains close at hand. Reader words, references, source-language forms, morphology, transliteration marks, and related lexical entries reveal context through hover, keyboard focus, touch, and explicit activation.

What Makes It Different

Capability	Practical value
Hover-first study	Supplemental word, reference, language, and lexical context appears on demand without permanently crowding the reader.
Local-first reading	Reading and research do not depend on a hosted service or account.
Integrated context	Reader text, commentary, outlines, cross-references, Strong's data, and source-language records remain connected in one workspace.
Original-language depth	Hebrew and Greek cards separate source text, transliteration, pronunciation guidance, dictionary form, morphology, glosses, word origin, and related entries.
Structured Study Marks	Favorites and tags can be attached at book, chapter, verse, text-span, and source-token scope.
Portable personal data	Browser-local study state can be exported, imported, and recovered as JSON.
Auditable package	Source manifests, notices, deterministic data tools, package inventory, and verification scripts are included.

Study Experience

Reader and navigation

• Ten bundled English Bible translations.
• Book and chapter navigation for desktop, narrow, and mobile layouts.
• Footnotes, outlines, commentary, parallel passages, cross-references, and verse-scoped actions.
• Reader-to-panel word highlighting and panel history restoration.
• Prose, headings, superscriptions, poetry, and indentation retain their intended presentation.

Hover-first supplemental context

• Reader words can show transient Strong's and language detail without changing the reading location.
• Reference controls can preview a passage before navigation.
• Language and transliteration elements explain letters, marks, and scholarly notation on demand.
• Transient previews do not intentionally replace a locked panel or mutate panel history.
• Pointer interactions have keyboard and practical touch equivalents.

The interaction model and its broader hover, focus, and touch evidence are complete through issues #7 and #16. Focused reference and Language Study preview defects were resolved under issues #39, #40, and #42.

Hebrew and Greek Language Study

• Westminster Leningrad Codex and consonants-only Hebrew records.
• Nestle Greek New Testament 1904 and Scrivener's Textus Receptus 1894 records.
• Source text, transliteration, phonetic spelling, lemma, gloss, morphology, Strong's entries, word origin, and related lexical references.
• Hebrew marks and gematria where applicable.
• Greek letter analysis preserving breathing marks, accents, diaeresis, iota subscript, and other attached marks.
• Lazy verse loading so extended chapter study does not render every card at once.

Study Marks and browser-local data

• Canonical semantic targets for books, chapters, verses, ranges, text spans, source tokens, and source-token spans.
• Favorite remains the canonical favorite assertion, with applicable tags at each supported scope.
• A Study Marks dashboard for reviewing tagged and favorited targets.
• Personal Meaning is separate from Study Marks and applies only to exact canonical source-token identity.
• One My Data surface organized as My study data, Backup and restore, App settings, Local maintenance, and Advanced diagnostics. Advanced diagnostics is collapsed and lazy by default.

Study Marks and personal Meaning remain separate user tools. My Data keeps raw job, package, storage, and capability controls out of the ordinary reader path. Portable exports retain kind bibleapp:user-data and version 3, including sparse legacy compatibility, recovery backups before replacement, and all-or-nothing rejection of malformed imports. Browser-local data is not an account; users should download backups they care about.

Resilience and accessibility

• IndexedDB startup falls back to localStorage when browser storage is blocked or stalls beyond the startup boundary.
• Keyboard-operable study controls and visible focus treatment.
• Pointer, focus, keyboard, and touch support for app-controlled previews.
• Reduced-motion, forced-colors, right-to-left source text, and mobile touch coverage where static or browser verification is practical.
• Tooltips and previews are constrained to the visible panel and viewport.

Flexible Study Workspace and Context

Study information exists at different scopes:

• a word owns lexical, source-language, morphology, and saved-meaning data;
• a verse owns parallel text, references, commentary, and verse Study Marks;
• a chapter owns chapter navigation and chapter-level Language Study entry;
• a book owns outline and book-level study context;
• global/user tools own personal data, package, and diagnostic functions.

On desktop, the study workspace offers Compact, Standard, and Expanded widths, with Standard as the default. The reader and study workspace scroll independently, so longer study material does not displace the reading location. At 768px and below, the workspace continues to use the full-screen mobile drawer.

The contextual workspace hierarchy is Word → Verse. Word is present only for exact canonical word or source-token context; Verse retains its parallel, reference, commentary, Language Study, and Study Marks actions. Chapter Language Study and Book Outline remain reader-header actions rather than persistent workspace groups. The shared detail pane preserves panel lock, highlight, and history behavior. Exact-token Meaning and Study Marks open in contained surfaces inside the workspace while the underlying work area is inert.

Screenshots

These 19 captures are the current, manually reviewed, accepted public-preview evidence for the reader and personal-study experience. They are generated by the maintained capture workflow from deterministic UI state and reviewed at actual size. Standard is the normal public desktop width; the two mobile images show the existing full-screen drawer. The interlinear*.png filenames remain technical, while the product surface they show is Language Study.

Reader and navigation

View	Light	Dark
Reader
Standard Word → Verse workspace
Full-screen mobile drawer

• Book picker
• Verse context controls

Study surfaces

View	Light	Dark
Language Study	John 11:35 Language Study	Dark John 11:35 Language Study
Hebrew and Strong's detail	Exact Hebrew token and Strong's detail	Dark exact Hebrew token and Strong's detail
Study Marks	Contained exact-token Favorite workflow	Dark Study Marks index with the seeded exact-token Favorite

• Populated search results
• Contained exact-source-token Meaning surface with saved origin

My Data

• My study data summary
• Backup and restore controls
• Completed local maintenance

Run Locally

Prerequisites

• Node.js 20 or newer.
• A modern browser.
• Microsoft Edge on Windows when running the complete automated browser suite.

npm ci
npm run serve

npm run serve is the deterministic development/test server and sends Cache-Control: no-store. For local distribution or publish-like validation, use npm run serve:publish; it revalidates mutable files with ETag and Last-Modified validators so unchanged responses can return 304 Not Modified while changed files become visible without clearing browser storage.

Open:

http://127.0.0.1:8000/#/read/bsb/psalms/23

Routes are hash-based, so the app can run from the included Node static server without a framework-specific deployment runtime.

Verification

npm run inventory:check
npm run test:static
npm run test:browser
npm run test:browser:mobile
npm run verify
npm audit --audit-level=low
gitleaks detect --source . --no-git=false
git diff --check

npm run verify runs the static, domain, accessibility-source, desktop-browser, mobile-browser, inventory, and publish-audit suites. The complete automated browser suite uses Microsoft Edge on Windows, and maintained focused suites also support Chrome where specified. Broader Edge and Chrome QA is complete under issue #7; unavailable Firefox, Safari, Android Chrome, screen-reader, and real browser-UI zoom evidence is explicitly recorded there.

See the test inventory for the executable coverage map.

Architecture

The application is intentionally deployable as static files:

• app/index.html, app/app.js, and the app stylesheets provide the shell.
• Focused ES modules under app/src/ implement routing, rendering, panel state, study tools, semantic targets, persistence, and package state.
• Deterministic runtime datasets live under app/data/.
• Schemas and data-generation tools live under app/schemas/ and app/tools/.
• Repository-level integrity and regression tests live under tests/.

Further documentation:

• Architecture
• Data model
• Security posture
• UI functionality contract
• Test inventory

Repository-wide documentation and loose-file reconciliation is completed through issue #15. The dependency-ordered program roadmap is issue #22.

Package Inventory and Repository Size

The current full-study package contains:

• 10 reader translations;
• 29 feature packs;
• 2,804 packaged files;
• 954,311,610 aggregate bytes;
• 180,460,807 aggregate gzip bytes.

The repository is much larger than a typical static web project. Keeping the data together allows the preview to run without a hosted data service. On the exact performance candidate measured under issue #6, the source archive was 189,852,720 bytes, the extracted tree was approximately 979.4 MB, and a full clone occupied approximately 1.16 GB including Git metadata. These are single-environment measurements rather than guaranteed download sizes.

Issue #6 found no measured performance release blocker and recommends retaining the complete bundled-data model for the intended public preview. Non-blocking post-release optimization work is tracked in issues #44, #45, and #46.

Data Rights

Application code, tests, scripts, schemas, and tooling are available under the MIT License. Bundled Bible and study data retains its source rights and notices and is not described as MIT-licensed.

Before redistributing bundled content, review:

• NOTICE.md
• app/data/source-manifest.json

Some retained source notices contain both permission or copyright language and later public-domain wording. The repository preserves those notices and the recorded transformations so downstream users can inspect provenance rather than rely on an oversimplified license summary. Publication of this repository does not create a blanket relicensing conclusion for bundled data.

Security and Privacy Model

Bible App Reader has no server-side account system, analytics service, payment flow, remote write API, or application backend. Personal study state remains in the current browser profile unless the user exports it.

The static application includes a Content Security Policy and sanitizes commentary HTML, but changes involving HTML rendering, imported data, browser persistence, or bundled third-party content still require review.

See SECURITY.md for vulnerability reporting and the current repository-security posture.

Current Boundaries

• Browser-local study data does not automatically synchronize across devices or browser profiles.
• There is no collaborative account system or cloud backup.
• The complete automated browser suite is Edge-focused; focused Chrome and broader manual Edge/Chrome evidence are maintained separately.
• The bundled package increases clone and checkout size.
• The flexible study workspace, Meaning, Study Marks, and My Data interfaces are active-development surfaces rather than stable APIs.
• Bundled data should be redistributed only after reviewing the included source notices and manifest.

Project Status

The repository is PUBLIC PREVIEW — ACTIVE DEVELOPMENT. Public visibility is separate from a stable release, release tag, API promise, or blanket relicensing of bundled data.

Flexible Word → Verse context, workspace width and independent-scrolling behavior, unified target-aware Study Marks, exact source-token Meaning, consolidated My Data, documentation reconciliation, maintained screenshots, broader browser QA, hover-first evidence, and the package/runtime performance classification are complete.

The remaining release decision work is tracked under issue #5: final rights, security, metadata, clean-checkout, required-check, unavailable-evidence, and owner authorization gates. Issues #44, #45, and #46 are non-blocking post-release optimization work.

No release or tag is authorized by this status or by automated checks alone.

Contributing

Focused bug reports, documentation corrections, accessibility findings, data-rights questions, and reproducible browser issues are welcome. See CONTRIBUTING.md before opening a pull request.