Compare commits

..

3 Commits

Author SHA1 Message Date
Daniel Bedeleanu
994920bda2 Build [v1.7.0] - GitGuard - Infrastructure Hardening 2026-04-12 22:45:27 +03:00
Daniel Bedeleanu
0a6368b9f6 modificari de documentatii dupa o implementare noua 2026-04-12 21:43:08 +03:00
Daniel Bedeleanu
26e8f034a2 Update save_version.py: Added automatic master branch synchronization 2026-04-12 21:32:55 +03:00
16 changed files with 334 additions and 150 deletions

View File

@@ -1,81 +1,60 @@
# AI AGENT RULES - MANDATORY ENTRY POINT
# AI AGENT RULES - MANDATORY SSOT ENTRY POINT
**READ THIS ENTIRE FILE BEFORE EXECUTING ANY TASK.**
This is the **Single Source of Truth** for ALL Artificial Intelligence agents (Claude, Gemini, etc.) working on this project.
(Automatic IDE entry points like `GEMINI.md` and `CLAUDE.md` exist solely to redirect agents to this main file).
For technical architecture, data models, and stack details, refer to [PROJECT_ARCHITECTURE.md](PROJECT_ARCHITECTURE.md).
This is the **Single Source of Truth** for ALL AI agents. Refer to [PROJECT_ARCHITECTURE.md](PROJECT_ARCHITECTURE.md) for technical logic.
## 1. Multi-AI Coordination & Memory
- **MANDATORY STARTUP**: Every AI session MUST first read `dev_docs/SESSION_STATE.md` to understand current context.
- **MANDATORY HANDOVER**: At the end of every task/session, update the handover note in `dev_docs/SESSION_STATE.md`. Specify: **Active AI**, **Current Status**, **Technical Context**, and **Next Steps**.
- **ARCHIVAL RULE**: Before writing new state, move all content of the *previous* session handover into the top of `dev_docs/SESSION_HISTORY.md` to prevent bloat.
---
## 1. AI MEMORY, TRACEABILITY & HANDOVER
- **MANDATORY STARTUP**: Read `dev_docs/SESSION_STATE.md` immediately at session start.
- **PLAN RETIREMENT**: Mark a completed "Master Plan" as `[COMPLETED]` in the file itself. Move technical details to `dev_docs/ARCHIVE_LOGS.md` and `PLAN.md` entries to `dev_docs/PLAN_HISTORY.md`.
- **STRICT HANDOVER**: Update `dev_docs/SESSION_STATE.md` at the end of every task with: **Active AI**, **Current Status** (Stable/Broken/In-Progress), **Context**, and **Next Steps**.
- **SESSION ARCHIVE**: Move previous handover content to `dev_docs/SESSION_HISTORY.md` before writing new state.
- **NO INTERACTION OVERLAP**: Never modify a file if another AI session is explicitly working on it.
## 2. Global Operational Laws
- **STRICT ENGLISH POLICY**: All web interfaces, code, variables, scripts, and documentation MUST BE DIRECTLY AND ONLY IN ENGLISH. If you find Romanian text in code, translate it immediately.
- **COMMUNICATION LANGUAGE**: Conversation with the user will be in Romanian or English (preferably Romanian).
- **GIT BINARY PATH**: On this macOS environment, the system `git` is often broken. ALWAYS use the binary path stored in `.git_path` at the project root for all Git operations.
- **COMMIT & PUSH STRICT RULE**:
- Never push to remote (`git push`) or use force flags (`--hard`, `--force`) unless explicitly requested.
- Always update `VERSION.json` on every commit.
- Do NOT add AI co-author signatures (e.g., `Co-Authored-By: AI...`) to commit messages.
- Branching: `master` (stable), `dev` (active development), `vX` (archival releases).
- **PACKAGE MANAGEMENT**: Any new package installed via `pip install` MUST be added to `backend/requirements.txt` with version constraints (e.g., `slowapi>=0.1.9`). Keep requirements.txt synchronized with installed packages.
- **MANDATORY LOGGING**: Document coding/architecture changes in `dev_docs/ARCHIVE_LOGS.md` at the end of the task.
- **TRIPLE CONFIRMATION (Safe Forget)**: You cannot delete a physical location, item, or critical entity without explicit user permission three times.
## 2. ENGINEERING & OPERATIONAL LAWS
- **ENGLISH ONLY**: Interfaces, code, variables, and docs MUST be in English. Translate any Romanian text found in code immediately. (User conversation: Romanian/English).
- **GIT PROTOCOL**: Use the direct binary path in `.git_path` (`/Library/Developer/CommandLineTools/usr/bin/git`) for ALL Git operations. **DO NOT REMOVE OR CHANGE THIS PATH UNDER ANY CIRCUMSTANCES!** Never push or use `--force` unless explicitly asked. Branching: `master` (stable), `dev` (active), `vX` (archive).
- **VERSIONING**: Update `VERSION.json` on every commit. Use `scripts/save_version.py` for automated releases.
- **DEPENDENCIES**: Update `backend/requirements.txt` with version constraints for every new pip package.
- **SSOT INTEGRITY**: Every feature change MUST update: `README.md`, `USER_GUIDE.md`, `PROJECT_ARCHITECTURE.md`, and `export_prod.sh`.
## 3. UI/UX Fidelity Specifications
- **Fidelity First**: Never simplify the UI unless asked. Density and aesthetics must remain "Premium".
- **Styling**: Tailwind CSS. Font weights/spacing must be consistent (Inter/Roboto).
- **Readability**: NO `uppercase` or `tracking-widest` styles. Use standard camel/Title case.
- **Icons**: Use **Lucide Icons** exclusively. NO emojis.
- **Interactive Affordance**: All select/dropdown boxes MUST have a `ChevronDown` icon. Masked passwords should have reduced opacity (`text-white/50`).
- **Iconography Standardization**:
- **Categories**: ALWAYS use `Layers` (Color: `text-primary`).
- **Item Types**: ALWAYS use `Package` (Color: `text-green-500`).
## 3. UI/UX "PREMIUM" FIDELITY STANDARDS
- **Aesthetics**: Density/aesthetics must remain "Premium". Use Tailwind CSS. NO simplification.
- **Typography Rules**:
- **NO UPPERCASE** or **NO ITALICS** in headers, labels, buttons, or metadata.
- **NO `tracking-widest`**. Use standard camel/Title case.
- Use `font-black` for main headings.
- **Layout**: Main pages MUST use `max-w-7xl`.
- **Unified Headers**: Icon box (`p-4 bg-primary/10 border-primary/20`) + Title (`text-3xl font-black`) + Subtitle (`text-xs text-slate-500`).
- **Iconography**: Use **Lucide Icons** exclusively (NO emojis).
- **Categories**: `Layers` (text-primary).
- **Item Types**: `Package` (text-green-500).
- **Affordance**: Dropdowns MUST have a `ChevronDown`. Passwords: `text-white/50`. Logout MUST be `text-rose-500`.
## 4. Documentation Maintenance
- **SSOT INTEGRITY**: Whenever a feature is added, modified, or removed, you MUST update all corresponding documentation files:
- `README.md` (General usage & technical modes).
- `USER_GUIDE.md` (End-user instructions).
- `PROJECT_ARCHITECTURE.md` (Technical logic & data models).
- `export_prod.sh` (If new scripts or files must be included in the production bundle).
- **REAL-TIME UPDATES**: Documentation updates are NOT optional and must be performed within the same session as the code changes.
## 4. DATA INTEGRITY & AUDIT POLICY
- **RESTRICTED ACTIONS**: `DELETE /items/` and Admin settings require `auth.get_current_admin`.
- **AUDIT IMMUTABILITY**: Deleting an `Item` MUST NOT delete its `AuditLog` entries.
- **TRACEABILITY**: Log deletions to `logs/backend.log` with `USER[id]`, `ITEM[id]`, `Name`, `PN`.
- **CONFIRMATION**:
- **Triple Confirmation**: Deleting critical entities (Locations/Items) requires user confirmation 3 times.
- **Native Alerts**: Use `window.confirm` for all destructive UI actions and Logout.
## 6. AI Command Shortcuts
- **`save-version`**: When the user triggers this command, the AI MUST:
1. Increment the patch version in `VERSION.json`.
2. Stage all current changes (`git add .`).
3. Commit changes with message `Build [vX.Y.Z]`.
4. Create a new branch named `vX.Y.Z` from the current state.
5. Generate a production bundle ZIP (calls `./export_prod.sh`).
6. Stay on the current branch (`dev`).
- *Implementation*: Use `python3 scripts/save_version.py` to ensure consistency.
## 5. AI COMMAND SHORTCUTS
- **`save-version`**:
0. **MANDATORY**: Verify and update ALL documentation (`.md` files: README, USER_GUIDE, ARCHITECTURE, etc.) with explanations of all current changes.
1. Increment `VERSION.json`.
2. Git add/commit (`Build [vX.Y.Z]`).
3. Create branch `vX.Y.Z` (Snapshot).
4. Automatic Sync: Merge changes into `master` branch to keep it up-to-date.
5. Run `./export_prod.sh`.
(Always use `python3 scripts/save_version.py`).
## 7. UI Fidelity Laws (Strict)
- **Consistent Layout**: All main pages (Inventory, Audit, Admin) MUST use `max-w-7xl` for their primary content container.
- **Unified Headers**: Every functional page MUST have an identical header structure:
- Icon inside a styled box (`p-4 bg-primary/10 rounded-[2rem] border border-primary/20`).
- Title: `text-3xl font-black text-white tracking-tight`.
- Subtitle: `text-xs text-slate-500 font-bold mt-1 tracking-wider`.
- **Typographic Integrity**:
- **NO UPPERCASE** (ALL CAPS) allowed anywhere in the UI (headers, labels, buttons, metadata).
- **NO ITALICS** in headers, titles, or active UI elements.
- Consistent font weight (`font-black`) for main headings.
- **Safety Actions**:
- The **Logout** button MUST be clearly differentiated (e.g., `text-rose-500`) and MUST always require a `window.confirm` before execution.
---
## 8. Audit & Deletion Policy
- **Immutability of Logs**: Deleting an `Item` MUST NOT delete its corresponding entries in the `AuditLog` table in the database.
- **Disk Traceability**: Every item deletion MUST be logged to the disk file (`logs/backend.log`) with `USER[id]`, `ITEM[id]`, `Name`, and `PN`.
- **Explicit Confirmation**: Destructive actions (Delete Item from catalog) MUST always trigger a browser-native `window.confirm` before proceeding.
## 5. End of Session Protocol
- Once a phase/task from `PLAN.md` is completed and verified, move that entry into `dev_docs/PLAN_HISTORY.md`.
- After finishing an entire job (including updating session state, architecture logs, and versioning), end your final response on a separate line exactly with:
```
---
✓ Done.
```
- Do not provide unnecessary verbatim summaries of the code.
## END OF SESSION PROTOCOL
End your final response on a separate line exactly with:
```
---
✓ Done.
```

View File

@@ -37,6 +37,7 @@ A unified system to maintain an inventory of "items" and their quantities, inclu
### 4.1 AI Usage Policy
- **Routine Operations (Check-in/Out):** Executes entirely on the local device unconditionally using `html5-qrcode` ($0 cost). No AI is allowed here.
- **New Item Onboarding (AI Label OCR):** Uses cloud AI (`gemini-2.0-flash`). The user takes a photo, AI extracts data based on strict templates.
- **AI Box Discovery Mode (v1.6.0):** Supports specialized `mode="box"` prompt that focuses exclusively on prominent container names/hand-written labels, ignoring technical spec noise.
- **Validation Mask:** AI-extracted data is NEVER saved directly. It is presented in a validation UI for human confirmation.
### 4.2 Scanner Technical Specs
@@ -48,6 +49,7 @@ A unified system to maintain an inventory of "items" and their quantities, inclu
- Noise Filtering: Ignores `< 3` chars, decimals, and dates.
- Scoring: Exact S/N (+500), Exact P/N (+200), Token match (+50), Category match (+20).
- Threshold: Minimum **40 points** for auto-match without user intervention.
- **Targeted Field Scanning (v1.6.0):** UI allows "locking" the scanner focus to a specific input field (e.g., `box_label`). The OCR result is then redirected to state without performing regular item lookup.
### 4.3 Box Labeling & Printing System (v1.5.0)
- **Local OCR Priority:** Before checking individual S/Ns, the matching engine searches for `box_label` tokens. If a box is identified:
@@ -83,7 +85,12 @@ To ensure enterprise-grade protection, the following policies are enforced:
### 7.3 Data Privacy
- **Information Scrubbing:** Backend logs are configured to intercept and mask sensitive auth tokens or internal secrets (e.g., `JWT_SECRET_KEY`) during debug output.
- **Direct Bind LDAP:** Authentication uses direct user binding to the LDAP server, avoiding the need for a privileged service account with broad search permissions.
- **Cryptographic Credential Caching:** To support offline operations, the system caches a **PBKDF2-HMAC-SHA256 hash** of the user's Enterprise credentials upon successful online login. Plain text passwords are NEVER stored.
### 7.4 PWA Trust & Security
- **HTTPS Enforcement:** The system requires TLS (Port 3003) for camera access and secure token transmission.
- **Manifest Integrity:** A comprehensive `manifest.json` ensures the app is recognized as a trusted PWA on mobile platforms (iOS/Android).
7.5 Git Infrastructure Hardening (v1.7.0)
To ensure deployment stability on macOS environments with potentially broken developer tool links (`xcode-select` errors):
- **Direct Binary Mapping:** The system bypasses path resolution by using a hardcoded direct link to the Git binary in `.git_path` (`/Library/Developer/CommandLineTools/usr/bin/git`).
- **Persistence Mandate:** This path is protected by mandatory AI rules and must never be removed or modified to ensure `save-version` and automated deployment scripts remain functional.

View File

@@ -34,7 +34,7 @@ Native Linux installation (Alma/Debian/Ubuntu) without Docker dependencies.
To generate a clean production package and snapshot the current state:
1. Use the AI shortcut command: `save-version`.
2. Alternatively, run `./export_prod.sh` manually.
3. A `.zip` archive will be created (e.g., `aInventory-PROD-v1.3.6.zip`).
3. A `.zip` archive will be created (e.g., `aInventory-PROD-v1.7.0.zip`).
4. A backup branch `v.1.3.x` will be created automatically.
---

View File

@@ -19,7 +19,7 @@ The application is a **Progressive Web App**, which means you don't need to down
- **Default User:** On first installation, use `Admin` / `<initial-password>` (check your system administrator for the initial password).
- **Change Password:** We recommend changing your password immediately from the Admin settings.
- **LDAP/Enterprise Login:** If your administrator has configured LDAP integration, you can log in with your company/domain account. The application will cache your password locally to allow offline access (e.g., in areas without signal like basements).
- **LDAP/Enterprise Login:** If your administrator has configured LDAP integration, you can log in with your company/domain account. The application will securely cache a **cryptographic hash** of your credentials (using PBKDF2) to allow offline access (e.g., in areas without signal like basements). **Note: Your actual password is NEVER stored in plain text on the local device.**
- **JWT Tokens:** Your login session is secured with JWT bearer tokens that expire after 8 hours. You will be automatically logged out when your token expires.
---
@@ -33,11 +33,12 @@ Scan an existing barcode to locate or update an item in your inventory.
If no readable text is found, the scanner silently retries on the next cycle.
### Box & Container Scanning (NEW v1.5.0)
You can now scan a generic label on a box (e.g., "SFP Box 1") to identify all its contents at once.
- If the box contains **one type of item**, the application takes you directly to the stock adjustment screen.
- If the box contains **multiple items**, a list will appear for you to select which specific item you are withdrawing or adding.
- This feature works **locally and offline** - no AI costs for scanning boxes.
### Box & Container Scanning (NEW v1.6.0)
You can now manage containers more efficiently with two specialized methods:
- **AI Box Discovery**: When adding a new container through **AI Discovery**, use the **"Box / Container"** toggle. Gemini will focus exclusively on the container's name, ignoring technical noise on labels.
- **Targeted Field Scanning**: In the **Edit Item** modal, tap the small **Camera icon** next to the "Box / Container Label" field. The scanner will capture the next physical label directly into the text field.
- **Automatic Matching**: In the main scanner, scanning a box identifies all its contents. Scanning a box and then an item will suggest linking them together if they aren't already matched.
---
@@ -140,5 +141,5 @@ For detailed technical documentation, see the [Project Architecture](../PROJECT_
---
**Version:** v1.5.0
**Version:** v1.7.0
**Last Updated:** 2026-04-12

View File

@@ -1,12 +1,5 @@
{
"version": "1.5.0",
"last_build": "2026-04-12-2130",
"commit": "HEAD",
"changelog": [
"v1.5.0: Box Management System - Local OCR box scanning, Selection modals, and Dependency-free Label Printing (Barcode/QR)",
"v1.4.0: Final Audit Dashboard (Phase 6), LDAP UI Restoration, environment path fixes for macOS",
"v1.3.9: Maintenance and structural updates",
"v1.3.8: Remove .npx_cache/ and scratch/npm_cache/ from git tracking (3350 files purged from index)",
"v1.3.7: Security hardening — expanded .gitignore, removed ldap_config.json from tracking, added example files"
]
"version": "1.7.0",
"last_build": "2026-04-12-2224",
"codename": "GitGuard"
}

View File

@@ -7,36 +7,55 @@ base_dir = os.path.dirname(os.path.abspath(__file__))
dotenv_path = os.path.join(base_dir, ".env")
load_dotenv(dotenv_path)
def extract_label_info(image_bytes: bytes):
def extract_label_info(image_bytes: bytes, mode: str = "item"):
"""
Orchestrates extraction across multiple AI providers.
Order: Gemini (Flash/Pro) -> Claude (Haiku/Sonnet)
"""
prompt = """
Extract technical inventory information from this label image.
CRITICAL INSTRUCTIONS:
1. Look at the most prominent text (usually top 1-2 rows). This is the product NAME and MODEL.
2. Extract the PART NUMBER (P/N, Model No, Type). If no explicit Part Number is found, synthesize one from the most unique identifier in the header (e.g. 'OM4-MMF-DX').
3. Separate the COLOR (e.g. Turquoise, Yellow, Black).
4. Extract CATEGORY based on the item type (e.g. Patchcord, SFP, Connector).
5. Extract technical SPECS (e.g. '2.0mm', '10G', '850nm').
Return ONLY a valid JSON object:
{
"name": "Full descriptive name from header",
"part_number": "Unique identifier for fast scanning",
"category": "Broad category",
"color": "Color if present",
"specs": "Brief tech specs list",
"barcode": "Barcode value if visible",
"quantity": 1
}
Modes: 'item' (full technical extraction), 'box' (container discovery)
"""
if mode == "box":
prompt = """
Identify the CONTAINER or BOX name from this image.
Look for large, prominent, bold, or hand-written text that identifies a storage unit.
Ignore small technical details, quantities, or fine print.
Return ONLY a valid JSON object:
{
"box_label": "The identified container name",
"name": "Same as box_label",
"category": "Storage",
"specs": "Brief description if useful",
"quantity": 1
}
"""
else:
prompt = """
Extract technical inventory information from this label image.
CRITICAL INSTRUCTIONS:
1. Look at the most prominent text (usually top 1-2 rows). This is the product NAME and MODEL.
2. Extract the PART NUMBER (P/N, Model No, Type). If no explicit Part Number is found, synthesize one from the most unique identifier in the header (e.g. 'OM4-MMF-DX').
3. Separate the COLOR (e.g. Turquoise, Yellow, Black).
4. Extract CATEGORY based on the item type (e.g. Patchcord, SFP, Connector).
5. Extract technical SPECS (e.g. '2.0mm', '10G', '850nm').
Return ONLY a valid JSON object:
{
"name": "Full descriptive name from header",
"part_number": "Unique identifier for fast scanning",
"category": "Broad category",
"color": "Color if present",
"specs": "Brief tech specs list",
"barcode": "Barcode value if visible",
"quantity": 1
}
"""
# 1. Try Gemini
result = gemini.extract(image_bytes, prompt)
if result:
# Maintenance: Ensure fields are mapped if mode was box
if mode == "box" and "box_label" in result and "name" not in result:
result["name"] = result["box_label"]
return result
# 2. Try Claude (Fallback)

View File

@@ -66,6 +66,7 @@ _MAX_IMAGE_SIZE = 10 * 1024 * 1024 # 10 MB
async def extract_label(
request: Request,
file: UploadFile = File(...),
mode: str = "item", # 'item' or 'box'
current_user: auth.TokenData = Depends(auth.get_current_user)
):
"""[C-01] Extract label from image — only for authenticated users. [H-02] Rate limit: 10 req/min per IP."""
@@ -86,7 +87,7 @@ async def extract_label(
detail="File exceeds 10MB limit."
)
result = extract_label_info(contents)
result = extract_label_info(contents, mode=mode)
return result
@router.post("/", response_model=schemas.Item, status_code=status.HTTP_201_CREATED)

View File

@@ -1,3 +1,19 @@
### [2026-04-12] v1.5.0: Box Management, Local OCR & Dependency-Free Label Printing
**Purpose:** Implementation of a local-first box scanning workflow, including multi-item container selection and professional label generation without external libraries.
**Actions:**
- `backend/models.py` & `schemas.py` — Added `box_label` field and integrated it into Pydantic models.
- `backend/routers/items.py` — Updated item creation and deletion to include `box_label` in immutable AuditLog snapshots.
- `frontend/lib/db.ts` — Upgraded IndexedDB (Dexie) to v4 with an index on `box_label` for high-speed local searching.
- `frontend/app/page.tsx` — Rewrote `onOCRMatch` to prioritize box matching. Implemented "Box Contents" selection modal and a full Box Inventory management dashboard.
- `frontend/components/AIOnboarding.tsx` — Added Box Label association to the AI-powered onboarding form.
- `frontend/lib/labels.ts` (NEW) — Developed a 100% dependency-free SVG engine for Barcode 128 and QR Code generation.
- `scripts/save_version.py` — Updated script to automatically synchronize `dev` changes into the `master` branch during releases.
- `AI_RULES.md` — Added Rule 1.1 for mandatory "Plan Retirement" and strict traceability.
- `USER_GUIDE.md` & `PROJECT_ARCHITECTURE.md` — Full documentation of the new container-based logic.
**Status:** Stable. Build v1.5.0 release branch created and merged into Master.
---
### [2026-04-12] v1.4.1: Security Hardening, PWA Optimization & Modern CSS Upgrade
**Purpose:** Implementation of security audit recommendations, REST API test suite, PWA asset generation, and visual UI refinements using Modern CSS.
**Actions:**

View File

@@ -1,6 +1,9 @@
# MASTER PLAN: Box/Container Scanning & Printing Architecture
# [COMPLETED] MASTER PLAN: Box/Container Scanning & Printing Architecture
Acest plan detaliat descrie arhitectura tehnică și etapele procedurale necesare pentru a echipa sistemul TFM aInventory cu funcționalități de recunoaștere a cutiilor (Box Scanning). Este structurat astfel încât orice AI să poată prelua dezvoltarea exact din punctul în care a fost oprită.
> [!IMPORTANT]
> **STATUS: FULLY IMPLEMENTED (v1.5.0)**
> Date: 2026-04-12
> This plan is no longer active. All phases (OCR, Smart Routing, Label Printing) have been merged into the main codebase.
---

View File

@@ -2,14 +2,14 @@
**Active AI:** Gemini (Antigravity)
**Last Updated:** 2026-04-12
**Current Version:** v1.4.0
**Current Version:** v1.6.0 (BoxMaster)
**Branch:** dev
---
## STATUS: 🟢 STABLE — BOX MANAGEMENT & LABEL PRINTING COMPLETE (v1.5.0)
## STATUS: 🟢 STABLE — ADVANCED BOX MANAGEMENT & AI MODES COMPLETE (v1.6.0)
The TFM aInventory v1.5.0 has been upgraded with a powerful local-first Box/Container management system. Users can now group items into boxes, identify them instantly via local OCR, and print physical Barcode/QR labels directly from the PWA.
**CRITICAL FOR NEXT AI:** The "Box/Container Management" feature is **FINISHED**. Do NOT attempt to re-implement or look for a plan. The core logic is already in `frontend/app/page.tsx` (`onOCRMatch` and `BoxManager`), `backend/models.py`, and `frontend/lib/labels.ts`.
---
@@ -30,10 +30,16 @@ The TFM aInventory v1.5.0 has been upgraded with a powerful local-first Box/Cont
- **Box Manager Dashboard** — Added a dedicated UI to view all existing boxes and trigger label generation.
- **Hybrid Printing** — Implemented CSS `@media print` for professional desktop printers and "Save as PNG" rasterization for portable Bluetooth printers on mobile.
### 4. UI/UX Excellence
- **Search Integration** — Integrated `box_label` into the global search filter.
- **Onboarding Support** — Added Box Label association to the AI-powered onboarding workflow with smart `datalist` suggestions.
- **Visual Polish** — Applied Lucide `Package` iconography and consistent branding to the new modules.
### 4. UI/UX: Targeted Field Scanning
- **Camera Capture** — Added a dedicated scan button in Edit modals that redirects OCR results directly to the "Box Label" field without performing general item matches.
### 5. Multi-Mode AI Discovery
- **Contextual Prompts** — Implemented a dual-mode toggle (Item/Box) in the AI Onboarding screen.
- **Box Extraction** — Created a specialized prompt for Gemini 2.0 Flash to extract container names while filtering out technical noise from product labels.
### 6. Operational Rigor: Step 0 Rule
- **Mandatory Documentation** — Updated `AI_RULES.md` to force documentation verification before any `save-version` (git commit) operation.
- **Master Branch Sync** — Confirmed `scripts/save_version.py` logic to keep `master` branch in sync with the latest releases automatically.
---
@@ -51,7 +57,10 @@ The TFM aInventory v1.5.0 has been upgraded with a powerful local-first Box/Cont
**Active database:** `<project_root>/data/inventory.db`
**LDAP config:** `backend/config/ldap_config.json`
**Production Bundle:** `aInventory-PROD-v1.4.0.zip` (Clean)
**Production Bundle:** `aInventory-PROD-v1.6.0.zip` (BoxMaster Final)
> [!IMPORTANT]
> **Git Access Fix**: The `xcode-select` breakage is bypassed by using the direct binary path: `/Library/Developer/CommandLineTools/usr/bin/git` (stored in `.git_path`). **DO NOT change this path.** Operations now work correctly via this direct link.
**How to start:**
```bash

View File

@@ -1,9 +1,10 @@
'use client';
import { useState, useEffect } from 'react';
import { useState, useEffect, useCallback } from 'react';
import { db, Item } from '@/lib/db';
import { inventoryApi } from '@/lib/api';
import PageShell from '@/components/PageShell';
import Scanner from '@/components/Scanner';
import { toast } from 'react-hot-toast';
import {
Package,
@@ -17,7 +18,8 @@ import {
X,
AlertTriangle,
Tag,
Edit2
Edit2,
Camera
} from 'lucide-react';
import { clsx, type ClassValue } from 'clsx';
import { twMerge } from 'tailwind-merge';
@@ -50,6 +52,10 @@ export default function InventoryPage() {
const [catEditedName, setCatEditedName] = useState('');
const [catEditedDesc, setCatEditedDesc] = useState('');
// Scanner state
const [showScanner, setShowScanner] = useState(false);
const [fieldScanning, setFieldScanning] = useState<{ active: boolean, field: string } | null>(null);
useEffect(() => {
setMounted(true);
const savedUser = localStorage.getItem('inventory_user');
@@ -182,6 +188,31 @@ export default function InventoryPage() {
}
};
const onOCRMatch = useCallback(async (text: string) => {
const cleanText = text.toUpperCase().replace(/[^A-Z0-9\s/+-]/g, ' ');
const tokens = cleanText.split(/[\s\n,]+/).filter(t => t.length >= 3);
if (fieldScanning?.active && fieldScanning.field === 'box_label') {
const label = tokens[0] || cleanText;
setEditedItem(prev => ({ ...prev, box_label: label }));
setFieldScanning(null);
setShowScanner(false);
toast.success(`Captured: ${label}`);
return;
}
}, [fieldScanning]);
const onScanSuccess = useCallback((barcode: string) => {
// Inventory page doesn't do check-in via scanner, it just finds the item
const item = inventory.find(i => i.barcode === barcode);
if (item) {
setSelectedItem(item);
setShowScanner(false);
} else {
toast.error(`Item with barcode ${barcode} not found in catalog`);
}
}, [inventory]);
// Group items by category
const categories = Array.from(new Set(inventory.map(i => i.category)));
const filteredCategories = categories.filter(c =>
@@ -189,8 +220,9 @@ export default function InventoryPage() {
inventory.some(i => i.category === c && i.name.toLowerCase().includes(searchQuery.toLowerCase()))
);
// Extract unique item types for suggestions
// Extract unique item types and box labels for suggestions
const existingTypes = Array.from(new Set(inventory.map(i => i.type).filter(Boolean))).sort() as string[];
const existingBoxes = Array.from(new Set(inventory.map(i => i.box_label).filter(Boolean))).sort() as string[];
if (!mounted) return null;
@@ -200,6 +232,9 @@ export default function InventoryPage() {
<datalist id="existing-types">
{existingTypes.map(t => <option key={t} value={t} />)}
</datalist>
<datalist id="existing-boxes">
{existingBoxes.map(b => <option key={b} value={b} />)}
</datalist>
<header className="flex items-center gap-5 mb-10">
<div className="p-4 bg-primary/10 rounded-[2rem] text-primary border border-primary/20 shadow-xl shadow-primary/5">
@@ -412,6 +447,33 @@ export default function InventoryPage() {
className="w-full bg-slate-950 border border-slate-800 rounded-xl py-3 px-4 text-sm outline-none text-slate-100"
/>
</div>
<div className="col-span-2">
<label className="text-xs font-black text-slate-500 ml-1">Box / Container Label</label>
<div className="relative flex items-center">
<input
type="text"
list="existing-boxes"
value={editedItem.box_label || ''}
onChange={e => setEditedItem({...editedItem, box_label: e.target.value})}
className="w-full bg-slate-950 border border-slate-800 rounded-xl py-3 pl-4 pr-12 text-sm outline-none text-slate-100 placeholder:text-slate-700 focus:border-primary transition-colors"
placeholder="e.g. SFPs 40G Cisco"
/>
<button
type="button"
onClick={() => {
setFieldScanning({ active: true, field: 'box_label' });
setShowScanner(true);
toast.success("Ready to scan Box label...");
}}
className={cn(
"absolute right-2 p-2 rounded-lg transition-all",
fieldScanning?.active ? "bg-primary text-white animate-pulse" : "text-slate-500 hover:bg-slate-800"
)}
>
<Camera size={18} />
</button>
</div>
</div>
</div>
<div>
<label className="text-xs font-black text-slate-500 ml-1">Specs / Comments</label>
@@ -542,6 +604,13 @@ export default function InventoryPage() {
</div>
)}
{showScanner && (
<Scanner
onScanSuccess={onScanSuccess}
onOCRMatch={onOCRMatch}
/>
)}
</div>
</PageShell>
);

View File

@@ -10,6 +10,7 @@ import PageShell from '@/components/PageShell';
import { toast } from 'react-hot-toast';
import {
Package,
Camera,
Plus,
Minus,
Trash2,
@@ -80,6 +81,7 @@ export default function Home() {
const [syncing, setSyncing] = useState(false);
const [currentUser, setCurrentUser] = useState<any | null>(null);
const [categories, setCategories] = useState<any[]>([]);
const [fieldScanning, setFieldScanning] = useState<{ active: boolean, field: string } | null>(null);
useEffect(() => {
if (!localStorage.getItem('inventory_token')) {
@@ -240,6 +242,17 @@ export default function Home() {
if (tokens.length === 0) return;
// [NEW] Targeted Field Scan Logic
if (fieldScanning?.active) {
if (fieldScanning.field === 'box_label') {
const potentialLabel = tokens[0] || cleanText;
setEditedItem(prev => ({ ...prev, box_label: potentialLabel }));
setFieldScanning(null);
toast.success(`Captured: ${potentialLabel}`);
return;
}
}
// Toast only potentially useful scans
toast(`Scanning: ${cleanText.substring(0, 30)}...`, { icon: '🔍', duration: 1500, id: 'ocr-scan' });
@@ -662,14 +675,30 @@ export default function Home() {
</div>
<div className="col-span-2">
<label className="text-xs font-black text-slate-500 ml-1">Box / Container Label</label>
<input
type="text"
list="existing-boxes"
value={editedItem.box_label || ''}
onChange={e => setEditedItem({...editedItem, box_label: e.target.value})}
className="w-full bg-slate-950 border border-slate-800 rounded-xl py-3 px-4 text-sm outline-none text-slate-100"
placeholder="e.g. SFPs 40G Cisco"
/>
<div className="relative flex items-center">
<input
type="text"
list="existing-boxes"
value={editedItem.box_label || ''}
onChange={e => setEditedItem({...editedItem, box_label: e.target.value})}
className="w-full bg-slate-950 border border-slate-800 rounded-xl py-3 pl-4 pr-12 text-sm outline-none text-slate-100 placeholder:text-slate-700 focus:border-primary transition-colors"
placeholder="e.g. SFPs 40G Cisco"
/>
<button
type="button"
onClick={() => {
setFieldScanning({ active: true, field: 'box_label' });
setShowScanner(true);
toast.success("Scanning for BOX label...");
}}
className={cn(
"absolute right-2 p-2 rounded-lg transition-all",
fieldScanning?.active ? "bg-primary text-white animate-pulse" : "text-slate-500 hover:bg-slate-800"
)}
>
<Camera size={18} />
</button>
</div>
</div>
<div className="col-span-2">
<label className="text-xs font-black text-slate-500 ml-1">Specs</label>

View File

@@ -16,6 +16,7 @@ export default function AIOnboarding({ onCancel, onComplete, categories, invento
const [image, setImage] = useState<string | null>(null);
const [uploading, setUploading] = useState(false);
const [extractedData, setExtractedData] = useState<any>(null);
const [mode, setMode] = useState<'item' | 'box'>('item');
// Extract unique item types for suggestions
const existingTypes = Array.from(new Set(inventory.map(i => i.type).filter(Boolean))).sort() as string[];
@@ -42,7 +43,7 @@ export default function AIOnboarding({ onCancel, onComplete, categories, invento
const formData = new FormData();
formData.append('file', blob, 'label.jpg');
const data = await inventoryApi.analyzeLabel(formData);
const data = await inventoryApi.analyzeLabel(formData, mode);
if (data.error) {
toast.error(`AI Error: ${data.error}`);
@@ -51,7 +52,11 @@ export default function AIOnboarding({ onCancel, onComplete, categories, invento
}
setExtractedData(data);
toast.success("AI extraction complete!");
if (mode === 'box') {
toast.success(`Box identified: ${data.box_label || data.name}`);
} else {
toast.success("AI extraction complete!");
}
} catch (error) {
toast.error("Failed to process image with AI");
console.error(error);
@@ -97,13 +102,34 @@ export default function AIOnboarding({ onCancel, onComplete, categories, invento
{!image ? (
<div className="flex-1 flex flex-col gap-6 min-h-0">
<div className="flex bg-slate-900/50 p-1.5 rounded-2xl border border-slate-800/50 shrink-0">
<button
onClick={() => setMode('item')}
className={`flex-1 flex items-center justify-center gap-2 py-3 rounded-xl font-bold transition-all ${mode === 'item' ? 'bg-primary text-white shadow-lg' : 'text-slate-500 hover:text-slate-300'}`}
>
<Package size={18} />
<span className="text-xs">Item Label</span>
</button>
<button
onClick={() => setMode('box')}
className={`flex-1 flex items-center justify-center gap-2 py-3 rounded-xl font-bold transition-all ${mode === 'box' ? 'bg-primary text-white shadow-lg' : 'text-slate-500 hover:text-slate-300'}`}
>
<Layers size={18} />
<span className="text-xs">Box / Container</span>
</button>
</div>
<div className="flex-1 flex flex-col items-center justify-center border-2 border-dashed border-slate-800 rounded-[2.5rem] bg-slate-900/30 overflow-hidden px-4">
<div className="w-20 h-20 bg-slate-900 rounded-3xl flex items-center justify-center mb-6 shadow-inner">
<Camera size={32} className="text-slate-500" />
<Camera size={32} className={mode === 'box' ? 'text-primary' : 'text-slate-500'} />
</div>
<p className="text-slate-300 mb-2 text-center font-bold">Label Insight Mode</p>
<p className="text-slate-300 mb-2 text-center font-bold">
{mode === 'box' ? 'Container Discovery Mode' : 'Label Insight Mode'}
</p>
<p className="text-xs text-slate-500 px-8 text-center leading-relaxed font-bold">
Scan or upload a sharp photo of the item specifications
{mode === 'box'
? 'Scan the large, prominent label or hand-written name on the container'
: 'Scan or upload a sharp photo of the item specifications'}
</p>
</div>

View File

@@ -25,6 +25,7 @@ export default function Scanner({ onScanSuccess, onOCRMatch, paused }: ScannerPr
const [isSelecting, setIsSelecting] = useState(false);
const [capturedImage, setCapturedImage] = useState<string | null>(null);
const isBusy = useRef(false);
const isTransitioning = useRef(false); // Flag to track start/stop transitions
const scannerId = "reader-container-unique";
useEffect(() => {
@@ -33,11 +34,20 @@ export default function Scanner({ onScanSuccess, onOCRMatch, paused }: ScannerPr
}
const startScanner = async () => {
if (isBusy.current) return;
if (isBusy.current || isTransitioning.current) return;
isBusy.current = true;
isTransitioning.current = true;
try {
// If already scanning, we MUST stop it first
if (html5QrCodeRef.current?.isScanning) {
await html5QrCodeRef.current.stop();
try {
await html5QrCodeRef.current.stop();
} catch (e) {
console.warn("Graceful stop failed, might be in transition:", e);
// If it's already in transition, we just return and wait for the next effect trigger
return;
}
}
const config = {
@@ -85,6 +95,7 @@ export default function Scanner({ onScanSuccess, onOCRMatch, paused }: ScannerPr
}
} finally {
isBusy.current = false;
isTransitioning.current = false;
}
};
@@ -94,11 +105,15 @@ export default function Scanner({ onScanSuccess, onOCRMatch, paused }: ScannerPr
return () => {
if (html5QrCodeRef.current?.isScanning) {
isTransitioning.current = true;
html5QrCodeRef.current.stop()
.then(() => {
setIsStarted(false);
})
.catch(e => console.error("Stop failed", e));
.catch(e => console.error("Unmount stop failed", e))
.finally(() => {
isTransitioning.current = false;
});
}
};
}, [paused, onScanSuccess]);

View File

@@ -75,8 +75,8 @@ export const inventoryApi = {
}
},
analyzeLabel: async (formData: FormData) => {
const res = await axiosInstance.post('/items/extract-label', formData, {
analyzeLabel: async (formData: FormData, mode: string = "item") => {
const res = await axiosInstance.post(`/items/extract-label?mode=${mode}`, formData, {
headers: { 'Content-Type': 'multipart/form-data' }
});
return res.data;

View File

@@ -24,7 +24,7 @@ def run_command(cmd):
def main():
git = get_git_path()
# 1. Read and Increment Version
# 1. Read and Parse Version
if not os.path.exists(VERSION_FILE):
print(f"Error: {VERSION_FILE} not found.")
sys.exit(1)
@@ -34,10 +34,20 @@ def main():
old_version = data.get('version', '1.0.0')
parts = old_version.split('.')
if len(parts) == 3:
parts[2] = str(int(parts[2]) + 1)
if len(parts) < 3:
parts.extend(['0'] * (3 - len(parts)))
# Check for --minor or --major flags
if '--minor' in sys.argv:
parts[1] = str(int(parts[1]) + 1)
parts[2] = '0'
elif '--major' in sys.argv:
parts[0] = str(int(parts[0]) + 1)
parts[1] = '0'
parts[2] = '0'
else:
parts.append('1')
# Default: patch increment
parts[2] = str(int(parts[2]) + 1)
new_version = '.'.join(parts)
data['version'] = new_version
@@ -57,11 +67,18 @@ def main():
branch_name = f"v{new_version}"
run_command([git, 'branch', branch_name])
# 4. Create Production Bundle
# 4. Update master branch
print(f"🔄 Updating 'master' branch to v{new_version}...")
current_branch = run_command([git, 'rev-parse', '--abbrev-ref', 'HEAD'])
run_command([git, 'checkout', 'master'])
run_command([git, 'merge', current_branch])
run_command([git, 'checkout', current_branch])
# 5. Create Production Bundle
print("📦 Generating production bundle...")
run_command(['./export_prod.sh'])
print(f"Successfully saved version {new_version}, created branch {branch_name}, and generated production ZIP.")
print(f"Successfully saved version {new_version}, created branch {branch_name}, updated 'master', and generated production ZIP.")
print("Verification: check for the .zip file in the root.")
if __name__ == "__main__":