# Aumentum Storage Structure - Visual Guide ## 🎯 The Simple Truth **When you upload multiple files for the same document number at different times, they go into different folders based on WHEN they were uploaded, not WHAT document they belong to.** --- ## 📁 Real Example: Your PL21825 ### What You Did ``` 09:18 AM - Uploaded Type 103 (50 pages) 09:25 AM - Uploaded Type 127 (2 pages) 09:29 AM - Uploaded Type 126 (2 pages) ``` ### Where They Went ``` /contentstore/ └── 2025/ └── 11/ └── 4/ ├── 9/ ← Hour 9 (9:00 AM) │ └── 15/ ← Minute 15 (9:15 AM) │ └── [9 .bin files for Type 103] │ └── 10/ ← Hour 10 (10:00 AM) ├── 1/ ← Minute 1 (10:01 AM) │ └── [39 .bin files for Type 127] │ └── 4/ ← Minute 4 (10:04 AM) └── [8 .bin files for Type 126] ``` **Total: 56 files across 3 different directories for ONE document number!** --- ## 🔄 How The System Processes This ### Step 1: Upload ``` User Action: "Upload PL21825, Type 103" ↓ System: "What time is it? 09:15" ↓ System: "Create folder 2025/11/4/9/15/" ↓ System: "Save 50 pages as UUID.bin files in that folder" ``` ### Step 2: Database Entry ``` Create record in lr_source_document: document_number: PL21825 document_type: 103 page_count: 50 create_date: 2025-11-04 09:18:30 ``` ### Step 3: Linking (Should happen, but delayed for your case) ``` Create Alfresco node: node_id: 2443208 uuid: 46974fd7-af5d-4e1d-9719-3b63d0a2542b Link to document: property: targetRids value: "PL21825" Link to files: content_url: store://2025/11/4/9/15/file1.bin content_url: store://2025/11/4/9/15/file2.bin ... (50 files total) ``` --- ## 🔍 How Searching Works ### When Someone Searches "PL21825" ``` Step 1: Database Query "Find all nodes where targetRids = 'PL21825'" Result: Node 2443208 Step 2: Get File Locations "What files does Node 2443208 point to?" Result: store://2025/11/4/9/15/uuid1.bin store://2025/11/4/9/15/uuid2.bin ... store://2025/11/4/10/1/uuid50.bin ... store://2025/11/4/10/4/uuid54.bin Step 3: Convert to Filesystem Paths store://2025/11/4/9/15/uuid1.bin ↓ /contentstore/2025/11/4/9/15/uuid1.bin Step 4: Convert Each .bin to PDF Read JPEG from .bin file Convert to PDF Combine all PDFs into one document Step 5: Display to User "Here's your PL21825 document (54 pages)" ``` **The user never knows the files are in different folders!** --- ## 💡 Key Concepts ### 1. Directory = Time, Not Document ``` ❌ WRONG: Document PL21825 goes in folder "PL21825/" ✅ RIGHT: Document PL21825 uploaded at 9:15 goes in folder "2025/11/4/9/15/" ``` ### 2. One Document Number → Many Directories ``` Document Number: PL21825 ↓ Multiple Uploads (at different times) ↓ Multiple Directories (one per upload time) ↓ System combines them when you search ``` ### 3. Directory Structure is Always: YYYY/MM/DD/HH/MM ``` Year = 2025 Month = 11 Day = 4 Hour = 9 (or 10) Minute = 15 (or 1, or 4) Path = 2025/11/4/9/15/ ``` ### 4. The Database Links Everything Together ``` Document Number "PL21825" ↓ (stored in alf_node_properties) Node ID 2443208 ↓ (stored in alf_content_url) Store URLs: store://YYYY/MM/DD/HH/MM/UUID.bin ↓ (map to filesystem) Physical Files: /contentstore/YYYY/MM/DD/HH/MM/UUID.bin ``` --- ## 📊 Comparison: Multiple Document Numbers vs Same Document Number ### Scenario A: Different Documents Uploaded at Same Time ``` 9:15 AM - Upload PL21825 (50 pages) 9:15 AM - Upload PL21826 (30 pages) 9:15 AM - Upload PL21827 (20 pages) Storage: 2025/11/4/9/15/ ├── file1.bin (PL21825) ├── file2.bin (PL21825) ├── ... ├── file50.bin (PL21825) ├── file51.bin (PL21826) ├── ... ├── file80.bin (PL21826) ├── file81.bin (PL21827) └── file100.bin (PL21827) ALL 100 FILES IN SAME DIRECTORY! ``` ### Scenario B: Same Document Uploaded at Different Times (YOUR CASE) ``` 9:15 AM - Upload PL21825 Type 103 (50 pages) 10:01 AM - Upload PL21825 Type 127 (2 pages) 10:04 AM - Upload PL21825 Type 126 (2 pages) Storage: 2025/11/4/9/15/ ├── file1.bin (PL21825 Type 103) ├── ... └── file50.bin (PL21825 Type 103) 2025/11/4/10/1/ ├── file51.bin (PL21825 Type 127) └── file52.bin (PL21825 Type 127) 2025/11/4/10/4/ ├── file53.bin (PL21825 Type 126) └── file54.bin (PL21825 Type 126) SAME DOCUMENT, 3 DIFFERENT DIRECTORIES! ``` --- ## 🎯 Why This Design? ### Advantages 1. **Temporal Organization** - Easy to find files by date/time - Helps with backup and archival - Performance: prevents huge directories 2. **Scalability** - No single directory gets too large - Distributed storage is possible - Easy to partition by time 3. **Debugging** - Can trace uploads by time - Upload problems isolated by timeframe - Audit trail built into structure ### Trade-offs 1. **One document can span multiple directories** - System must query database to find all pieces - Can't simply look in one folder 2. **Requires robust database linking** - If database link breaks, files become orphaned - Must maintain database integrity --- ## 🚨 Current Situation with PL21825 ### What's Working ``` ✅ Files uploaded to filesystem (56 files in 3 directories) ✅ lr_source_document records created (3 records) ✅ alf_node created (Node 2443208) ✅ alf_node_properties created (PL21825 → Node 2443208) ``` ### What's Missing ``` ❌ alf_content_data not populated ❌ alf_content_url entries not created ❌ Node 2443208 doesn't point to any files yet ``` ### Result ``` Files exist: ✅ Database knows about PL21825: ✅ Database knows which files belong to PL21825: ❌ Status: INDEXING INCOMPLETE ``` **This is why you can't access the document yet in WebAccess!** --- ## 📋 Summary ### Your Theory: ✅ CONFIRMED ``` "Multiple files for same document go to different directories based on upload time" → 100% CORRECT! ``` ### How It Works 1. **Directory = Upload Time** (YYYY/MM/DD/HH/MM) 2. **Same document at different times = Different directories** 3. **Database links them all together via document_number** 4. **Search queries database, not filesystem** 5. **System transparently combines files from all directories** ### Your PL21825 Specifically - ✅ Correctly stored in 3 time-based directories - ❌ Database linking not complete yet - ⏳ Waiting for indexing/linking process to finish --- ## 🎓 Teaching Example If you were explaining this to someone new: ``` "Think of the contentstore as a library organized by TIME, not by BOOK TITLE. Books arrive throughout the day and get shelved based on when they arrived. To find all chapters of a book, you don't look in one place. You look in the card catalog (database), which tells you: 'Chapter 1-50: Shelf 9:15' 'Chapter 51-52: Shelf 10:01' 'Chapter 53-54: Shelf 10:04' Then you collect all chapters and bind them together." ``` That's exactly how Aumentum works! 📚