FragDB v5.9 — Fragrance Database (multilingual sample)

.gitattributes

@@ -0,0 +1,2 @@
+*.csv filter=lfs diff=lfs merge=lfs -text
+news_sample.parquet filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,32 @@
+# Claude Code / AI tools
+.claude/
+claude.md
+CLAUDE.md
+
+# Internal / credentials
+*.local.json
+*.credentials
+*.secret
+
+# OS
+.DS_Store
+Thumbs.db
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+*~
+
+# Environment
+.env
+.env.local
+.env.*.local
+
+# Logs
+*.log
+
+# Temp
+*.tmp
+*.bak

README.md

@@ -0,0 +1,789 @@
+---
+license: cc-by-nc-4.0
+task_categories:
+  - feature-extraction
+  - text-classification
+language:
+  - en
+  - de
+  - es
+  - fr
+  - cs
+  - it
+  - ru
+  - pl
+  - pt
+  - el
+  - zh
+  - ja
+  - nl
+  - sr
+  - ro
+  - ar
+  - uk
+  - mn
+  - ko
+  - tr
+  - sv
+  - he
+  - hu
+tags:
+  - fragrance
+  - perfume
+  - cosmetics
+  - recommendation-system
+  - e-commerce
+  - retail
+  - fragrantika
+  - multilingual
+size_categories:
+  - n<1K
+configs:
+  - config_name: fragrances
+    data_files: fragrances.csv
+    default: true
+    sep: "|"
+  - config_name: brands
+    data_files: brands.csv
+    sep: "|"
+  - config_name: perfumers
+    data_files: perfumers.csv
+    sep: "|"
+  - config_name: notes
+    data_files: notes.csv
+    sep: "|"
+  - config_name: accords
+    data_files: accords.csv
+    sep: "|"
+  - config_name: translations
+    data_files: translations.csv
+    sep: "|"
+dataset_info:
+  - config_name: fragrances
+    features:
+      - name: pid
+        dtype: int64
+      - name: url
+        dtype: string
+      - name: brand
+        dtype: string
+      - name: name
+        dtype: string
+      - name: year
+        dtype: int64
+      - name: gender
+        dtype: string
+      - name: collection
+        dtype: string
+      - name: main_photo
+        dtype: string
+      - name: info_card
+        dtype: string
+      - name: user_photoes
+        dtype: string
+      - name: video_url
+        dtype: string
+      - name: accords
+        dtype: string
+      - name: notes_pyramid
+        dtype: string
+      - name: perfumers
+        dtype: string
+      - name: description
+        dtype: string
+      - name: rating
+        dtype: string
+      - name: reviews_count
+        dtype: int64
+      - name: appreciation
+        dtype: string
+      - name: price_value
+        dtype: string
+      - name: gender_votes
+        dtype: string
+      - name: longevity
+        dtype: string
+      - name: sillage
+        dtype: string
+      - name: season
+        dtype: string
+      - name: time_of_day
+        dtype: string
+      - name: pros_cons
+        dtype: string
+      - name: by_designer
+        dtype: string
+      - name: in_collection
+        dtype: string
+      - name: reminds_of
+        dtype: string
+      - name: also_like
+        dtype: string
+      - name: news_ids
+        dtype: string
+  - config_name: brands
+    features:
+      - name: id
+        dtype: string
+      - name: name
+        dtype: string
+      - name: url
+        dtype: string
+      - name: logo_url
+        dtype: string
+      - name: country
+        dtype: string
+      - name: main_activity
+        dtype: string
+      - name: website
+        dtype: string
+      - name: parent_company
+        dtype: string
+      - name: description
+        dtype: string
+      - name: brand_count
+        dtype: int64
+      - name: country_de
+        dtype: string
+      - name: country_es
+        dtype: string
+      - name: country_fr
+        dtype: string
+      - name: country_cs
+        dtype: string
+      - name: country_it
+        dtype: string
+      - name: country_ru
+        dtype: string
+      - name: country_pl
+        dtype: string
+      - name: country_pt
+        dtype: string
+      - name: country_el
+        dtype: string
+      - name: country_zh
+        dtype: string
+      - name: country_ja
+        dtype: string
+      - name: country_nl
+        dtype: string
+      - name: country_sr
+        dtype: string
+      - name: country_ro
+        dtype: string
+      - name: country_ar
+        dtype: string
+      - name: country_uk
+        dtype: string
+      - name: country_mn
+        dtype: string
+      - name: country_ko
+        dtype: string
+      - name: country_tr
+        dtype: string
+      - name: country_sv
+        dtype: string
+      - name: country_he
+        dtype: string
+      - name: country_hu
+        dtype: string
+      - name: main_activity_de
+        dtype: string
+      - name: main_activity_es
+        dtype: string
+      - name: main_activity_fr
+        dtype: string
+      - name: main_activity_cs
+        dtype: string
+      - name: main_activity_it
+        dtype: string
+      - name: main_activity_ru
+        dtype: string
+      - name: main_activity_pl
+        dtype: string
+      - name: main_activity_pt
+        dtype: string
+      - name: main_activity_el
+        dtype: string
+      - name: main_activity_zh
+        dtype: string
+      - name: main_activity_ja
+        dtype: string
+      - name: main_activity_nl
+        dtype: string
+      - name: main_activity_sr
+        dtype: string
+      - name: main_activity_ro
+        dtype: string
+      - name: main_activity_ar
+        dtype: string
+      - name: main_activity_uk
+        dtype: string
+      - name: main_activity_mn
+        dtype: string
+      - name: main_activity_ko
+        dtype: string
+      - name: main_activity_tr
+        dtype: string
+      - name: main_activity_sv
+        dtype: string
+      - name: main_activity_he
+        dtype: string
+      - name: main_activity_hu
+        dtype: string
+  - config_name: perfumers
+    features:
+      - name: id
+        dtype: string
+      - name: name
+        dtype: string
+      - name: url
+        dtype: string
+      - name: photo_url
+        dtype: string
+      - name: status
+        dtype: string
+      - name: company
+        dtype: string
+      - name: also_worked
+        dtype: string
+      - name: education
+        dtype: string
+      - name: web
+        dtype: string
+      - name: perfumes_count
+        dtype: int64
+      - name: biography
+        dtype: string
+      - name: status_de
+        dtype: string
+      - name: status_es
+        dtype: string
+      - name: status_fr
+        dtype: string
+      - name: status_cs
+        dtype: string
+      - name: status_it
+        dtype: string
+      - name: status_ru
+        dtype: string
+      - name: status_pl
+        dtype: string
+      - name: status_pt
+        dtype: string
+      - name: status_el
+        dtype: string
+      - name: status_zh
+        dtype: string
+      - name: status_ja
+        dtype: string
+      - name: status_nl
+        dtype: string
+      - name: status_sr
+        dtype: string
+      - name: status_ro
+        dtype: string
+      - name: status_ar
+        dtype: string
+      - name: status_uk
+        dtype: string
+      - name: status_mn
+        dtype: string
+      - name: status_ko
+        dtype: string
+      - name: status_tr
+        dtype: string
+      - name: status_sv
+        dtype: string
+      - name: status_he
+        dtype: string
+      - name: status_hu
+        dtype: string
+      - name: perfumer_name_ru
+        dtype: string
+      - name: perfumer_name_uk
+        dtype: string
+      - name: perfumer_name_ja
+        dtype: string
+      - name: perfumer_name_zh
+        dtype: string
+      - name: perfumer_name_ko
+        dtype: string
+      - name: perfumer_name_ar
+        dtype: string
+  - config_name: notes
+    features:
+      - name: id
+        dtype: string
+      - name: name
+        dtype: string
+      - name: url
+        dtype: string
+      - name: latin_name
+        dtype: string
+      - name: other_names
+        dtype: string
+      - name: group
+        dtype: string
+      - name: odor_profile
+        dtype: string
+      - name: main_icon
+        dtype: string
+      - name: alt_icons
+        dtype: string
+      - name: background
+        dtype: string
+      - name: fragrance_count
+        dtype: int64
+      - name: note_name_de
+        dtype: string
+      - name: note_name_es
+        dtype: string
+      - name: note_name_fr
+        dtype: string
+      - name: note_name_cs
+        dtype: string
+      - name: note_name_it
+        dtype: string
+      - name: note_name_ru
+        dtype: string
+      - name: note_name_pl
+        dtype: string
+      - name: note_name_pt
+        dtype: string
+      - name: note_name_el
+        dtype: string
+      - name: note_name_zh
+        dtype: string
+      - name: note_name_ja
+        dtype: string
+      - name: note_name_nl
+        dtype: string
+      - name: note_name_sr
+        dtype: string
+      - name: note_name_ro
+        dtype: string
+      - name: note_name_ar
+        dtype: string
+      - name: note_name_uk
+        dtype: string
+      - name: note_name_mn
+        dtype: string
+      - name: note_name_ko
+        dtype: string
+      - name: note_name_tr
+        dtype: string
+      - name: note_name_sv
+        dtype: string
+      - name: note_name_he
+        dtype: string
+      - name: note_name_hu
+        dtype: string
+      - name: note_group_de
+        dtype: string
+      - name: note_group_es
+        dtype: string
+      - name: note_group_fr
+        dtype: string
+      - name: note_group_cs
+        dtype: string
+      - name: note_group_it
+        dtype: string
+      - name: note_group_ru
+        dtype: string
+      - name: note_group_pl
+        dtype: string
+      - name: note_group_pt
+        dtype: string
+      - name: note_group_el
+        dtype: string
+      - name: note_group_zh
+        dtype: string
+      - name: note_group_ja
+        dtype: string
+      - name: note_group_nl
+        dtype: string
+      - name: note_group_sr
+        dtype: string
+      - name: note_group_ro
+        dtype: string
+      - name: note_group_ar
+        dtype: string
+      - name: note_group_uk
+        dtype: string
+      - name: note_group_mn
+        dtype: string
+      - name: note_group_ko
+        dtype: string
+      - name: note_group_tr
+        dtype: string
+      - name: note_group_sv
+        dtype: string
+      - name: note_group_he
+        dtype: string
+      - name: note_group_hu
+        dtype: string
+  - config_name: accords
+    features:
+      - name: id
+        dtype: string
+      - name: name
+        dtype: string
+      - name: bar_color
+        dtype: string
+      - name: font_color
+        dtype: string
+      - name: fragrance_count
+        dtype: int64
+      - name: name_de
+        dtype: string
+      - name: name_es
+        dtype: string
+      - name: name_fr
+        dtype: string
+      - name: name_cs
+        dtype: string
+      - name: name_it
+        dtype: string
+      - name: name_ru
+        dtype: string
+      - name: name_pl
+        dtype: string
+      - name: name_pt
+        dtype: string
+      - name: name_el
+        dtype: string
+      - name: name_zh
+        dtype: string
+      - name: name_ja
+        dtype: string
+      - name: name_nl
+        dtype: string
+      - name: name_sr
+        dtype: string
+      - name: name_ro
+        dtype: string
+      - name: name_ar
+        dtype: string
+      - name: name_uk
+        dtype: string
+      - name: name_mn
+        dtype: string
+      - name: name_ko
+        dtype: string
+      - name: name_tr
+        dtype: string
+      - name: name_sv
+        dtype: string
+      - name: name_he
+        dtype: string
+      - name: name_hu
+        dtype: string
+  - config_name: translations
+    features:
+      - name: id
+        dtype: string
+      - name: section
+        dtype: string
+      - name: en
+        dtype: string
+      - name: de
+        dtype: string
+      - name: es
+        dtype: string
+      - name: fr
+        dtype: string
+      - name: cs
+        dtype: string
+      - name: it
+        dtype: string
+      - name: ru
+        dtype: string
+      - name: pl
+        dtype: string
+      - name: pt
+        dtype: string
+      - name: el
+        dtype: string
+      - name: zh
+        dtype: string
+      - name: ja
+        dtype: string
+      - name: nl
+        dtype: string
+      - name: sr
+        dtype: string
+      - name: ro
+        dtype: string
+      - name: ar
+        dtype: string
+      - name: uk
+        dtype: string
+      - name: mn
+        dtype: string
+      - name: ko
+        dtype: string
+      - name: tr
+        dtype: string
+      - name: sv
+        dtype: string
+      - name: he
+        dtype: string
+      - name: hu
+        dtype: string
+---
+
+# FragDB v5.9 — Fragrance Database (Multilingual Sample)
+
+The most comprehensive structured fragrance database available. This is a **free sample** of FragDB: **134,577 perfumes, 23 languages** — 10-row CSV samples at root.
+
+Full dataset: [fragdb.net](https://fragdb.net).
+
+## What's New in v5.9
+
+- **Data updated** from v5.8 → v5.9 (parser run 260710):
+  - Fragrances: 134,022 → **134,577** (+555)
+  - Brands: 8,000 → **8,036** (+36) — **154 brand names canonicalized** (restored `Fragrance(s)` suffix; IDs stable)
+  - Perfumers: 3,035 → **3,046** (+11)
+  - Notes: 2,562 → **2,567** (+5)
+  - URL hygiene: pyramid anchors single-domain, photo cache-busters stripped
+- **Free sample refreshed** — 10-record CSV samples rebuilt from v5.9 data
+
+## What's New in v5.8
+
+- **Data updated** from v5.7 → v5.8 (parser run 260701):
+  - Fragrances: 133,392 → **134,022** (+630)
+  - Brands: 7,953 → **8,000** (+47)
+  - Perfumers: 3,020 → **3,035** (+15)
+  - Notes: 2,559 → **2,562** (+3)
+
+## What's New in v5.7
+
+- **Data updated** from v5.6 → v5.7 (parser run 260619):
+  - Fragrances: 132,858 → **133,392** (+534)
+  - Brands: 7,927 → **7,953** (+26)
+  - Perfumers: 3,005 → **3,020** (+15)
+  - Notes: 2,550 → **2,559** (+9)
+
+## What's New in v5.6
+
+- **Data updated** from v5.5 → v5.6 (parser run 260609):
+  - Fragrances: 132,124 → **132,858** (+734)
+  - Brands: 7,881 → **7,927** (+46)
+  - Perfumers: 2,988 → **3,005** (+17)
+  - Notes: 2,533 → **2,550** (+17)
+- **Notes multilingual 100% coverage** (was 99.8%) — 6 previously gap-filled notes now complete across all 22 languages
+- **Photo URL stability** — source cache-buster query params stripped, eliminating phantom diffs across releases
+- Schema unchanged from v5.5 — existing loaders work without modification
+- Free sample files unchanged (10-record structure preserved)
+
+## What's New in v5.5
+
+- **Data updated** from v5.4 → v5.5 (parser run 260601):
+  - Fragrances: 130,949 → 132,124 (+1,175)
+  - Brands: 7,815 → 7,881 (+66)
+  - Perfumers: 2,968 → 2,988 (+20)
+  - Notes: 2,522 → 2,533 (+11)
+
+### Schema unchanged from v5.4
+
+All F column counts identical (30/54/42/55/27/25) — existing scripts work without modification.
+
+### From v5.4 (unchanged in v5.5)
+- **23 languages** — all labels, note names, accords, countries, statuses translated
+- **9 non-Latin scripts** for perfumer name transliteration
+- **translations.csv** — vocabulary file (34 entries) for gender and voting labels
+- **Compact notes pyramid** — `note_id,opacity,weight` (name/icon via notes.csv JOIN)
+- Each note name variant has its own ID with translations
+- **Gender & voting fields** use translation IDs for multilingual support
+
+## Snapshot freshness
+
+- **Data refreshed**: 2026-07-10 (v5.9)
+
+## Dataset Description
+
+| File | Records | Fields | Description |
+|------|---------|--------|-------------|
+| `fragrances.csv` | 10 | 30 | Iconic fragrances (v5.9) |
+| `brands.csv` | 10 | 54 | Brand profiles + 22 lang translations |
+| `perfumers.csv` | 10 | 42 | Perfumer profiles + 22 lang + 9 name translit |
+| `notes.csv` | 10 | 55 | Fragrance notes + 22 lang translations |
+| `accords.csv` | 10 | 27 | Accords + 22 lang translations |
+| `translations.csv` | 34 | 25 | Gender & voting vocabulary (full) |
+| `comments_sample.parquet` | 25 | 8 | User reviews preview (parquet) |
+| `news_sample.parquet` | 20 | 16 | Editorial articles preview (parquet) |
+| `news_comments_sample.parquet` | 20 | 9 | News comments preview (parquet) |
+| `SPEC.md` | — | — | Parquet schema documentation |
+
+### Loading the data
+
+```python
+from datasets import load_dataset
+
+f = load_dataset("FragDBnet/fragrance-database")           # fragrances (default)
+brands = load_dataset("FragDBnet/fragrance-database", "brands")
+notes  = load_dataset("FragDBnet/fragrance-database", "notes")
+```
+
+## Companion Parquet Datasets — User Reviews, News, and Community Comments
+
+FragDB ships with **three Apache Parquet datasets** containing **4.9 million rows** of user-generated content and editorial coverage — the largest publicly-organized corpus of fragrance reviews and perfumery journalism. Use them for NLP, sentiment analysis, recommendation systems, market research, or training language models on fragrance-specific text.
+
+**Keywords:** fragrance reviews · perfume reviews · multilingual UGC corpus · NLP training data · fragrance sentiment · perfumery journalism · perfume recommendation · scent recommendation · review classification · entity linking · knowledge graph · fragrance industry news · perfume articles
+
+### `comments.parquet` — 4.6 Million User Reviews in 23 Languages
+
+The world's largest collection of structured fragrance reviews. Every entry includes the perfume ID (joinable with `fragrances.csv`), author username, posting date, full review text, avatar URL, and language code.
+
+- **4,643,851 user reviews** covering every major perfume in the database
+- **23 languages** — English (1.69M), Russian, Portuguese, Spanish, Korean, Turkish, Japanese, Polish, Italian, Hungarian, Serbian, Swedish, German, Hebrew, Ukrainian, French, Arabic, Greek, Czech, Chinese, Romanian, Mongolian, Dutch
+- **Coverage:** 70.6% of all fragrances have at least one review (93,305 of 132,160 PIDs)
+- **Deterministic global primary key** — stable comment IDs survive re-scrapes
+- **Zero duplicate rows**, **zero foreign key orphans** against `fragrances.csv.pid`
+- **Independent UGC per language** — genuine localized content, not machine translation
+- **8 fields:** `pid`, `lang`, `comment_id`, `author`, `date`, `text`, `avatar_url`, `gradient_class`
+- **PyArrow large_string format** — combined corpus exceeds 32-bit string offset limit
+
+**Use cases:** sentiment analysis · review classification · recommendation systems · perfume similarity from text · language detection benchmark · multilingual NLP training corpus · fragrance market research · author network analysis · trend detection by language
+
+### `news.parquet` — 24,440 Editorial Articles (2008–2026)
+
+Two decades of professional fragrance journalism. Every article includes title, author, full text (plain + HTML), category, related perfumes/brands/perfumers, publication date, and main image.
+
+- **24,440 editorial articles** from 2008 to 2026 — complete public archive
+- **30+ categories** — New Fragrances (34.9%), Fragrance Reviews (22.8%), Niche Perfumery (10.4%), Designer Brands, Interviews, History, Industry News
+- **Bilingual storage** — `text` (plain) for NLP, `text_html` (markup preserved) for rich display
+- **Linked entities** — `related_pids[]`, `related_brands[]`, `related_perfumers[]` as JSON arrays
+- **0% orphans** over 119,662 PID references
+- **63.1% archived legacy, 36.9% modern** fully-dated articles
+- **16 fields:** `nid`, `title`, `category`, `author`, `url`, `is_archived`, `date_unix`, `description`, `text`, `text_html`, `main_image`, `article_images`, `related_pids`, `related_brands`, `related_perfumers`, `comments_count`
+
+**Use cases:** content recommendation · article search engine · perfume knowledge graph · trend analysis · author influence study · entity linking · timeline analysis · industry research · niche perfumery research
+
+### `news_comments.parquet` — 263,798 Threaded Community Comments
+
+Community discussions attached to editorial articles, with threading support for replies. Joinable with `news.parquet` via `nid`.
+
+- **263,798 threaded comments** across **21,820 articles** (89.3% of news articles have ≥1 comment)
+- **4.9% reply rate** — threaded conversations with reply detection
+- **100% populated timestamps**
+- **9 fields:** `nid`, `comment_id`, `is_reply`, `author`, `date`, `date_unix`, `text`, `avatar_url`, `gradient`
+
+**Use cases:** community engagement analysis · threaded discussion mining · reply network construction · comment sentiment · author activity profiles
+
+### Tier Availability
+
+The parquet datasets ship with **all paid tiers except the $200 Core**:
+
+| Tier | CSV Core | Parquet Datasets |
+|------|----------|------------------|
+| **$200 One-Time Core** | ✅ | ❌ |
+| **$400 One-Time Full Database** | ✅ | ✅ |
+| **Annual Subscription** | ✅ | ✅ (always latest) |
+| **Lifetime Access** | ✅ | ✅ (always latest) |
+
+See https://fragdb.net/#pricing for complete tier comparison.
+
+### Quick Start — Parquet
+
+```python
+import pyarrow.parquet as pq
+import pandas as pd
+import json
+
+reviews = pq.read_table('comments.parquet').to_pandas()
+fragrances = pd.read_csv('fragrances.csv', sep='|')
+reviews_with_meta = reviews.merge(fragrances, on='pid', how='left')
+
+news = pq.read_table('news.parquet').to_pandas()
+news['related_pids_list'] = news['related_pids'].apply(json.loads)
+
+news_comments = pq.read_table('news_comments.parquet').to_pandas()
+```
+
+Full schema in [`SPEC.md`](SPEC.md).
+
+### Use Cases
+
+**CSV Core (all tiers):**
+- **E-commerce** — Enrich product listings with detailed fragrance data, notes, accords
+- **Mobile Apps** — Build fragrance collection managers, scent discovery apps, perfume catalog apps
+- **Data Analysis** — Analyze fragrance industry trends by brand, country, perfumer, year
+- **Recommendations** — Content-based or collaborative filtering systems using accord/note vectors
+- **Multilingual UIs** — Localized perfume catalogs in 23 languages out of the box
+- **Knowledge Graphs** — Brand → Perfumer → Fragrance → Notes → Accords graph construction
+- **Market Research** — Country-of-origin analysis, parent company portfolios, perfumer productivity stats
+
+**Parquet Datasets ($400+ tiers):**
+- **NLP & Sentiment Analysis** — Train models on 4.6M multilingual fragrance reviews
+- **Recommender Systems** — Hybrid models combining CSV structure with review text similarity
+- **Language Models** — Domain-specific corpus for fragrance/perfumery LLM fine-tuning
+- **Review Classification** — Identify positive/negative reviews, fake review detection
+- **Trend Detection** — News article timeline analysis, emerging fragrance trends
+- **Author Networks** — Identify influential reviewers, perfumery journalists, community leaders
+- **Content-Based Discovery** — "Articles about this perfume" — JOIN news.related_pids with fragrances.pid
+- **Community Analytics** — Reply networks, engagement metrics on editorial content
+- **Cross-Language Studies** — Compare review sentiment across 23 languages for the same fragrance
+- **Search Engines** — Full-text search across reviews, articles, and structured metadata
+- **Knowledge Extraction** — Mine 24K editorial articles for perfume facts, launch dates, perfumer interviews
+
+### Full Database
+
+| | Sample | Full Database |
+|---|--------|---------------|
+| Fragrances | 10 | **134,577** |
+| Brands | 10 | **8,036** |
+| Perfumers | 10 | **3,046** |
+| Notes | 10 | **2,567** |
+| Accords | 10 | **92** |
+| Translations | 34 | **34** |
+| Languages | 23 | **23** |
+| **Total Records** | ~84 | **148,352** |
+
+## Quick Start
+
+```python
+import pandas as pd
+
+fragrances = pd.read_csv('fragrances.csv', sep='|')
+brands = pd.read_csv('brands.csv', sep='|')
+notes = pd.read_csv('notes.csv', sep='|')
+translations = pd.read_csv('translations.csv', sep='|')
+
+# Join and translate
+fragrances['brand_id'] = fragrances['brand'].str.split(';').str[1]
+df = fragrances.merge(brands, left_on='brand_id', right_on='id', suffixes=('', '_brand'))
+trans = translations.set_index('id')
+df['gender_ru'] = df['gender'].map(lambda x: trans.loc[x, 'ru'] if x in trans.index else x)
+print(df[['name', 'name_brand', 'country_ru', 'gender_ru']])
+```
+
+## File Format
+
+- **Format**: CSV (pipe `|` delimited)
+- **Encoding**: UTF-8
+- **Quote Character**: `"` (double quote)
+
+## Links
+
+- **Full Database**: [fragdb.net](https://fragdb.net)
+- **GitHub**: [github.com/FragDB/fragrance-database](https://github.com/FragDB/fragrance-database)
+- **Kaggle**: [kaggle.com/datasets/eriklindqvist/fragdb-fragrance-database](https://www.kaggle.com/datasets/eriklindqvist/fragdb-fragrance-database)
+
+## License
+
+This sample is released under the **CC BY-NC 4.0 License**. Free for non-commercial use with attribution.
+
+## Citation
+
+```bibtex
+@dataset{fragdb2026,
+  title={FragDB Fragrance Database},
+  author={FragDB},
+  year={2026},
+  version={5.9},
+  url={https://fragdb.net},
+  note={Multilingual dataset with 6 files, 23 languages}
+}
+```

SPEC.md

@@ -0,0 +1,635 @@
+# Fragrantika News & Comments Datasets — Technical Specification
+
+**Snapshot date:** 2026-05-05
+**Source:** `fragrantica.com` and 22 localized domains (`fragrantica.de`, `fragrantica.ru`, etc.)
+**Format:** Apache Parquet (Zstandard compression)
+**Total combined size:** ~1.36 GB on disk
+
+This document specifies three datasets that complement the existing
+Fragrantika perfume database (`merged_database`, `brands_database_v2`,
+`perfumers_database_v2`, `notes_database_v2` — already documented and sold
+separately):
+
+1. **`comments.parquet`** — user reviews on individual perfumes
+2. **`news.parquet`** — editorial articles published on Fragrantika
+3. **`news_comments.parquet`** — user comments under those articles
+
+The cross-dataset relationships in §5 explain how to join these with the
+existing perfume/brand/perfumer DBs via primary keys.
+
+---
+
+## 1. Overview
+
+| Dataset | Rows | Compressed Size | Distinct entities |
+|---|---:|---:|---:|
+| `comments.parquet` | 4,643,851 | 1.23 GB | 93,305 perfumes, 23 languages |
+| `news.parquet` | 24,440 articles | 101 MB | NID 1..25000 |
+| `news_comments.parquet` | 263,798 | 33 MB | 21,820 articles with comments |
+
+All three datasets share the same identifier conventions used by the
+existing perfume/brand/perfumer databases:
+
+- **PID** (`int32`) — unique perfume identifier; matches `merged_database.PID`.
+- **NID** (`int32`) — unique news article identifier (1..25000 range).
+- **lang** (`string`) — ISO 639-1 language code identifying which Fragrantika
+  domain the content was sourced from. Each domain hosts independent
+  user-generated content (reviews are not translations of one another).
+
+---
+
+## 2. Dataset 1: `comments.parquet`
+
+User-written reviews of individual perfumes, scraped across 23 localized
+Fragrantika domains. Each comment is a unique user review, not a translation
+or syndication.
+
+### 2.1 Schema (8 fields)
+
+| # | Field | Type | NULL? | Description |
+|---:|---|---|---|---|
+| 1 | `pid` | `int32` | NO | Foreign key → `merged_database.PID`. Identifies the perfume being reviewed. |
+| 2 | `lang` | `string` | NO | Domain language (`en`, `ru`, `de`, …). 23 distinct values. |
+| 3 | `comment_id` | `string` | NO | Globally unique deterministic ID, format `{pid}_{lang}_{12hex}` where `12hex` is the first 12 hexadecimal characters of `sha1(author + "|" + date + "|" + text)`. Stable across re-scrapes. |
+| 4 | `author` | `string` | NO | Username as displayed on Fragrantika. May contain Unicode (e.g. `Сергей`, `José`). |
+| 5 | `date` | `string` | NO | Date published as it appears on the page. Most rows: `MM/DD/YY HH:MM` (e.g., `04/11/26 10:05`). Always present. |
+| 6 | `text` | `large_string` (64-bit offsets) | NO | Review body. UTF-8, no HTML tags, paragraph breaks via `\n`. May contain emoji. Length range observed: 1 char (emoji-only) to ~120 KB. |
+| 7 | `avatar_url` | `string` | NO | Full URL to the user's avatar image hosted on Fragrantika's CDN. Verified ~100% reachable. |
+| 8 | `gradient_class` | `string` | NO | Fragrantika's CSS class for the user's color badge (e.g., `tw-gradient-rose`). 7 distinct values observed (`tw-gradient-amber`/`emerald`/`orange`/`rose`/`sky`/`teal`/`violet`). Useful for UI consistency if displaying. |
+
+### 2.2 Volumetrics
+
+- **Total rows:** 4,643,851
+- **Distinct PIDs covered:** 93,305 (70.6% of the 132,160 perfumes in `merged_database`; remaining 38,855 perfumes have zero reviews — verified against Fragrantika's `reviews_count` field)
+- **PID range:** 1 — 130,121
+- **Languages:** 23 (full list in §2.3)
+
+### 2.3 Language distribution
+
+| Lang | Code | Comments | Distinct PIDs |
+|---|---|---:|---:|
+| English | `en` | 1,690,055 | 78,715 (60% of perfumes) |
+| Russian | `ru` | 889,893 | 52,650 |
+| Portuguese (Brazilian) | `pt` | 247,511 | 21,258 |
+| Spanish | `es` | 226,451 | 26,586 |
+| Korean | `ko` | 214,674 | 11,696 |
+| Turkish | `tr` | 191,945 | 11,722 |
+| Japanese | `ja` | 186,836 | 10,767 |
+| Polish | `pl` | 141,242 | 25,350 |
+| Italian | `it` | 115,972 | 22,100 |
+| Hungarian | `hu` | 109,204 | 11,565 |
+| Serbian | `sr` | 108,945 | 15,048 |
+| Swedish | `sv` | 107,666 | 11,155 |
+| German | `de` | 100,218 | 15,505 |
+| Hebrew | `he` | 81,981 | 10,165 |
+| Ukrainian | `uk` | 50,203 | 15,917 |
+| French | `fr` | 42,256 | 13,379 |
+| Arabic | `ar` | 33,077 | 9,071 |
+| Greek | `el` | 28,208 | 8,512 |
+| Czech | `cs` | 24,269 | 3,424 |
+| Chinese (Simplified) | `zh` | 15,843 | 4,466 |
+| Romanian | `ro` | 13,906 | 6,451 |
+| Mongolian | `mn` | 12,100 | 2,056 |
+| Dutch | `nl` | 11,396 | 5,045 |
+
+Each `lang` represents an independent set of user reviews from the
+corresponding national Fragrantika domain. A user reviewing the same
+perfume on `fragrantica.com` (en) and `fragrantica.ru` (ru) would write
+two distinct reviews; both are included.
+
+### 2.4 ID generation
+
+`comment_id` is content-derived (sha1-based), giving three properties:
+- **Globally unique** across all 4.6M rows (verified post-rewrite, 0 collisions).
+- **Stable across re-scrapes**: re-fetching the same comment yields the same
+  ID. Useful for incremental updates and joining with downstream tables.
+- **Idempotent dedup key**: `(nid, comment_id)` and `comment_id` alone are
+  both safe primary keys.
+
+---
+
+## 3. Dataset 2: `news.parquet`
+
+Editorial articles from `fragrantica.com/news/` (English-only). Covers the
+entire archive 2008-2026 (NID 1..25000), including both archived legacy
+articles and modern editorial content.
+
+### 3.1 Schema (16 fields)
+
+| # | Field | Type | NULL? | Description |
+|---:|---|---|---|---|
+| 1 | `nid` | `int32` | NO | News article ID. Range 1..25000. |
+| 2 | `title` | `string` | NO | Article title. Plain text, HTML entities decoded. |
+| 3 | `category` | `string` | NO | Editorial category (e.g., `New Fragrances`, `Fragrance Reviews`, `Interviews`). 30+ distinct values. Top 10 in §3.4. |
+| 4 | `author` | `string` | YES | Comma-separated list of author names (multi-author articles common). |
+| 5 | `description` | `string` | YES | Short summary from `<meta property="og:description">`. Plain text. |
+| 6 | `text` | `string` | NO | Full article body. Plain text, paragraph breaks via `\n`. Footer/byline blocks stripped. UTF-8. |
+| 7 | `text_html` | `string` | NO | Article body **with original HTML preserved** — `<p>`, `<a>`, `<img>`, embedded videos. Useful for rendering or DOM-aware downstream processing. |
+| 8 | `main_image` | `string` | YES | URL to the article's primary image (Fragrantika CDN). Verified reachable. |
+| 9 | `article_images` | `string` | NO | **JSON-encoded array** of all image URLs in the article body. Use `json.loads(field)` → `list[str]`. Empty array `'[]'` if none. |
+| 10 | `url` | `string` | NO | Canonical article URL (e.g., `https://www.fragrantica.com/news/x-12345.html`). |
+| 11 | `is_archived` | `bool` | NO | `True` if article is in Fragrantika's legacy/archived corpus (older HTML template, often missing publication date — see §3.5). |
+| 12 | `related_pids` | `string` | NO | **JSON-encoded array** of PIDs (as decimal strings) referenced in the article. Foreign key → `merged_database.PID`. 0% orphan rate (FK-validated). |
+| 13 | `related_brands` | `string` | NO | **JSON-encoded array** of brand names. Informational metadata; see §5.4 for canonical resolution. |
+| 14 | `related_perfumers` | `string` | NO | **JSON-encoded array** of perfumer names with diacritics preserved (e.g., `François Demachy`, `Carlos Benaïm`). Informational; see §5.5. |
+| 15 | `comments_count` | `int32` | NO | Count of comments in `news_comments.parquet` matching this NID. |
+| 16 | `date_unix` | `int64` | NO | Publication time as Unix timestamp (seconds since epoch). `0` indicates the date is not extractable from the source HTML — see §3.5. |
+
+### 3.2 Volumetrics
+
+- **Total rows:** 24,440 articles
+- **NID range:** 1 — 25,000 (560 NIDs missing — Fragrantika returns HTTP 301 redirect for these — represents truly deleted articles)
+- **Archived:** 15,427 / 24,440 (63.1%)
+- **Non-archived:** 9,013 / 24,440 (36.9%)
+- **With publication date (`date_unix > 0`):** 13,687 (56.0%)
+
+### 3.3 List-fields format note
+
+Four fields hold lists: `article_images`, `related_pids`, `related_brands`,
+`related_perfumers`. **All four are stored as JSON-encoded strings** for
+consistency. Use:
+
+```python
+import json
+images = json.loads(row['article_images'])  # → list[str]
+```
+
+`related_pids` contains PID values as decimal strings (e.g., `'704'`, not
+the int `704`). Cast as needed when joining.
+
+### 3.4 Top categories
+
+| Category | Articles | % |
+|---|---:|---:|
+| New Fragrances | 8,534 | 34.9% |
+| Fragrance Reviews | 5,583 | 22.8% |
+| Niche Perfumery | 2,537 | 10.4% |
+| Art Books Events | 1,570 | 6.4% |
+| Columns | 1,346 | 5.5% |
+| Fragrant Horoscope | 851 | 3.5% |
+| Fragrance News | 760 | 3.1% |
+| Interviews | 682 | 2.8% |
+| Vintages | 437 | 1.8% |
+| Raw Materials | 384 | 1.6% |
+
+Remaining 1,756 articles span 33 smaller categories.
+
+### 3.5 Date coverage
+
+`date_unix == 0` for 10,753 articles (44%). Distribution by NID range:
+
+| NID range | Articles | with date | % | archived % |
+|---|---:|---:|---:|---:|
+| 1 — 5,000 | 4,888 | 2 | 0.0% | 100% |
+| 5,001 — 10,000 | 4,799 | 0 | 0.0% | 100% |
+| 10,001 — 15,000 | 4,890 | 3,822 | 78.2% | 100% |
+| 15,001 — 20,000 | 4,900 | 4,900 | 100% | 17.3% |
+| 20,001 — 25,000 | 4,963 | 4,963 | 100% | 0% |
+
+The 10,753 articles without date come entirely from the archived legacy
+template: 9,685 in NIDs 1–10,000 (effectively all of them) and 1,068 in
+NIDs 10,001–15,000 (the remaining 22% of that range — the other 78% have
+recoverable visible-text dates). Fragrantika's old archived HTML does not
+embed any date marker (no `<time>`, no `<meta>`, no visible date string)
+for these. This is an upstream HTML limitation, not a parser issue.
+**For all non-archived articles (9,013 of them), `date_unix` is 100%
+populated.**
+
+---
+
+## 4. Dataset 3: `news_comments.parquet`
+
+User comments under news articles. Pure UGC; same comment-card schema
+Fragrantika uses on perfume pages.
+
+### 4.1 Schema (9 fields)
+
+| # | Field | Type | NULL? | Description |
+|---:|---|---|---|---|
+| 1 | `nid` | `int32` | NO | Foreign key → `news.parquet.nid` (0% orphan rate). |
+| 2 | `comment_id` | `string` | NO | Comment ID as Fragrantika assigns it (page-anchor format). Globally unique within parquet. |
+| 3 | `author` | `string` | NO | Commenter username. |
+| 4 | `date` | `string` | NO | Date as displayed on page (e.g., `04/11/26 10:05`). |
+| 5 | `date_unix` | `int64` | NO | Parsed Unix timestamp. **100% populated** (all rows have `date_unix > 0`). |
+| 6 | `text` | `string` | NO | Comment body, plain text. |
+| 7 | `avatar_url` | `string` | NO | Full URL to author avatar (Fragrantika CDN). |
+| 8 | `gradient` | `string` | NO | CSS class for color badge (same scheme as `comments.parquet.gradient_class`). |
+| 9 | `is_reply` | `bool` | NO | `True` if this comment is a reply to another comment (threaded discussion); `False` for root comments. 4.9% of rows are replies (12,898 / 263,798). |
+
+### 4.2 Volumetrics
+
+- **Total rows:** 263,798
+- **Distinct NIDs:** 21,820 (89.3% of articles have at least one comment)
+- **Replies:** 12,898 (4.9%)
+- **Average comments per article (in articles with comments):** 12.1
+
+---
+
+## 5. Cross-dataset relationships
+
+This section is the load-bearing piece for buyers integrating with the
+existing perfume/brand/perfumer DBs.
+
+### 5.1 Diagram
+
+```
+┌────────────────────────────────────────────────────────┐
+│   merged_database.csv  (existing, 132,160 perfumes)   │
+│   ─────────────────────                                │
+│   PID   ←──────────────────────┐                       │
+│   Brand                        │                       │
+│   Name                         │                       │
+│   noses_f  (perfumers)         │                       │
+│   ...30 fields total           │                       │
+└────────────────────────────────┼───────────────────────┘
+                                 │
+                ┌────────────────┴───────────────┐
+                │                                │
+                ▼                                ▼
+   ┌──────────────────────┐         ┌──────────────────────┐
+   │ comments.parquet     │         │ news.parquet         │
+   │ ──────────────────── │         │ ──────────────────── │
+   │ pid    ─────────────►│         │ related_pids ───────►│
+   │ lang   (23 langs)    │         │ related_brands       │
+   │ comment_id (sha1)    │         │ related_perfumers    │
+   │ author/date/text     │         │ nid ◄────┐           │
+   │ +avatar/gradient     │         │ ...16 fields total   │
+   └──────────────────────┘         └──────────┼───────────┘
+                                               │
+                                               │
+                                    ┌──────────┴───────────┐
+                                    │ news_comments.parquet│
+                                    │ ──────────────────── │
+                                    │ nid ─────────────────│ FK → news.nid
+                                    │ comment_id           │
+                                    │ author/date/text     │
+                                    │ is_reply             │
+                                    └──────────────────────┘
+```
+
+### 5.2 Linking comments to perfumes (PID)
+
+Foreign key: **`comments.pid → merged_database.PID`** — verified 0% orphan rate.
+
+```sql
+-- All English reviews for the perfume "Mugler Angel" (PID 704):
+SELECT c.author, c.date, c.text
+FROM read_parquet('comments.parquet') c
+JOIN read_csv('merged_database.csv', delim='⏸') m ON c.pid = m.PID
+WHERE m.Brand = 'Mugler' AND m.Name = 'Angel' AND c.lang = 'en'
+ORDER BY c.date DESC LIMIT 100;
+```
+
+```python
+import pyarrow.parquet as pq
+import pyarrow.compute as pc
+
+c = pq.read_table('comments.parquet')
+angel_reviews = c.filter(pc.and_(pc.equal(c['pid'], 704),
+                                  pc.equal(c['lang'], 'en')))
+```
+
+### 5.3 Linking news to perfumes (related_pids)
+
+Foreign key: **`news.related_pids → merged_database.PID`** — JSON-array
+field, 0% orphan rate over 119,662 references.
+
+```sql
+-- All news mentioning Aventus (PID 9828):
+SELECT nid, title, category, date_unix
+FROM read_parquet('news.parquet')
+WHERE list_contains(json_extract(related_pids, '$'), '9828');
+```
+
+```python
+import json
+n = pq.read_table('news.parquet')
+mentions_aventus = []
+for i, rp in enumerate(n.column('related_pids').to_pylist()):
+    if rp and '9828' in json.loads(rp):
+        mentions_aventus.append(i)
+```
+
+### 5.4 Linking news to brands
+
+`news.related_brands` is a **JSON array of brand-name strings** (informational).
+
+**Authoritative brand resolution** for any news article uses the PID:
+```
+news.related_pids → merged_database.PID → merged_database.Brand
+```
+
+The `related_brands` field is provided as supplementary metadata. Approx.
+12% of strings in `related_brands` will not match `brands_database_v2.name`
+exactly because Fragrantika has renamed some brands over time
+(e.g., `Christian Dior` → `Dior`, `Annick Goutal` → `Goutal`,
+`Paco Rabanne` → `Rabanne`). The original captured name is preserved.
+
+For canonical brand lookup, always join via `related_pids`:
+
+```sql
+SELECT DISTINCT m.Brand
+FROM read_parquet('news.parquet') n
+JOIN read_csv('merged_database.csv', delim='⏸') m
+  ON list_contains(json_extract(n.related_pids, '$'), CAST(m.PID AS VARCHAR))
+WHERE n.nid = 17644;
+```
+
+### 5.5 Linking news to perfumers
+
+Same pattern as brands. `news.related_perfumers` is a **JSON array of perfumer-name
+strings** with diacritics preserved (e.g., `François Demachy`, `Carlos Benaïm`,
+`Cécile Zarokian` — verified canonical post-fix).
+
+**Authoritative perfumer resolution** uses PID:
+```
+news.related_pids → merged_database.PID → merged_database.noses_f
+```
+
+`merged_database.noses_f` contains the full list of perfumers per perfume in
+the project's standard format. Approximately 11% of `related_perfumers`
+strings will not exactly match `perfumers_database_v2.name` due to the
+upstream perfumer DB not yet covering all names mentioned in news (e.g.,
+`Jean Claude Ellena` ×191 mentions). Use the PID-join for guaranteed
+matches.
+
+### 5.6 Linking news_comments to articles (NID)
+
+Foreign key: **`news_comments.nid → news.nid`** — 0% orphan rate.
+
+```sql
+-- All comments under article NID 17644:
+SELECT author, date, text, is_reply
+FROM read_parquet('news_comments.parquet')
+WHERE nid = 17644
+ORDER BY date_unix ASC;
+```
+
+### 5.7 Combined query examples
+
+**Q1.** All news articles + their root comments mentioning Creed Aventus
+(PID 9828) in 2024:
+
+```sql
+SELECT n.nid, n.title, n.date_unix, nc.author, nc.text
+FROM read_parquet('news.parquet') n
+LEFT JOIN read_parquet('news_comments.parquet') nc ON n.nid = nc.nid
+WHERE list_contains(json_extract(n.related_pids, '$'), '9828')
+  AND n.date_unix BETWEEN 1704067200 AND 1735689599  -- 2024
+  AND nc.is_reply = false
+ORDER BY n.date_unix DESC, nc.date_unix ASC;
+```
+
+**Q2.** Top-10 most-reviewed perfumes by language:
+
+```sql
+SELECT lang, pid, COUNT(*) AS review_count
+FROM read_parquet('comments.parquet')
+GROUP BY lang, pid
+QUALIFY ROW_NUMBER() OVER (PARTITION BY lang ORDER BY COUNT(*) DESC) <= 10;
+```
+
+**Q3.** Sentiment correlation between news articles and reviews — pairs of
+(news article, perfume reviewed in same time window):
+
+```sql
+SELECT n.nid, n.title, m.Brand, m.Name, COUNT(c.comment_id) AS reviews_in_30d
+FROM read_parquet('news.parquet') n
+JOIN read_csv('merged_database.csv', delim='⏸') m
+  ON list_contains(json_extract(n.related_pids, '$'), CAST(m.PID AS VARCHAR))
+JOIN read_parquet('comments.parquet') c
+  ON c.pid = m.PID
+WHERE n.date_unix > 0
+  AND c.date_unix BETWEEN n.date_unix AND n.date_unix + 2592000  -- 30 days
+GROUP BY n.nid, n.title, m.Brand, m.Name;
+```
+
+---
+
+## 6. Format specification
+
+### 6.1 Storage
+
+- **Container:** Apache Parquet 2.x
+- **Compression:** Zstandard (`zstd`)
+- **Row groups:**
+  - `comments.parquet` — 5 row groups
+  - `news.parquet` — 1 row group
+  - `news_comments.parquet` — 1 row group
+
+### 6.2 String types
+
+PyArrow uses two string types:
+- `string` — backed by 32-bit offsets (per-array limit ~2 GB total).
+- `large_string` — backed by 64-bit offsets (no practical size limit).
+
+`comments.parquet` uses `large_string` for the `text` column (because the
+combined corpus of 4.6M reviews exceeds the 32-bit offset limit). All
+other string columns in all three datasets use `string`. Pandas
+automatically converts both to its native `object` dtype on load — buyers
+generally do not need to distinguish between the two types unless writing
+custom Arrow code.
+
+Exact field-level types are listed in the schema tables of §2.1, §3.1, §4.1.
+
+### 6.3 Encoding
+
+- **Character encoding:** UTF-8 throughout.
+- **Normalization:** NFC where applicable; original Unicode codepoints
+  preserved (including emoji).
+- **Newlines:** `\n` only (Unix-style). No `\r`.
+- **HTML entities:** decoded (no `&amp;`, `&lt;`, etc. in text fields).
+- **Replacement chars (U+FFFD):** zero — rare upstream-broken bytes were
+  stripped, with the surrounding text preserved.
+
+### 6.4 List fields (news.parquet)
+
+Four columns store JSON-encoded arrays-as-strings:
+- `article_images`: `list[str]` of image URLs
+- `related_pids`: `list[str]` of decimal-string PIDs
+- `related_brands`: `list[str]` of brand names
+- `related_perfumers`: `list[str]` of perfumer names
+
+Empty list is stored as the literal string `'[]'` (not NULL, not empty
+string). This makes parsing branchless:
+
+```python
+items = json.loads(row[field])  # always returns a list
+```
+
+---
+
+## 7. Data quality
+
+The datasets passed a multi-track validation audit (full report:
+`reports/data_quality/POST_FIX_VALIDATION.md`). Summary:
+
+| Check | Result |
+|---|:---:|
+| Duplicate rows by primary key (all 3 datasets) | 0 |
+| FK integrity (`comments.pid → PID`) | 0 orphans |
+| FK integrity (`news.related_pids → PID`) | 0 orphans (over 119,662 refs) |
+| FK integrity (`news_comments.nid → nid`) | 0 orphans |
+| `comment_id` global uniqueness (`comments.parquet`) | 0 collisions over 4.6M rows |
+| HTML/CSS pollution in text fields | 0 |
+| Cloudflare challenge pages in data | 0 |
+| Mojibake / replacement char (U+FFFD) | 0 |
+| Language attribution accuracy (langdetect on sample) | >95% per language |
+| `news_comments.date_unix > 0` populated | 100% |
+| `news.date_unix > 0` populated (non-archived) | 100% |
+
+### 7.1 Provenance
+
+Data scraped via headless browser fingerprinting (chrome120 impersonation
+through curl_cffi) over a residential proxy network. Each page was
+individually fetched, parsed by a custom HTML parser, and post-processed
+through a content sanitization pipeline. Re-scrapes are idempotent thanks
+to deterministic comment IDs (§2.4).
+
+---
+
+## 8. Known limitations (transparent disclosure)
+
+1. **Archived news articles without dates (10,753 / 24,440 = 44%).**
+   Fragrantika's archived HTML template does not embed any date marker
+   (`<time>`, meta tag, or visible string). For these articles, `date_unix`
+   is `0`. The no-date subset is concentrated in NIDs 1–15,000: 9,685 of
+   9,687 archived articles in NIDs 1–10,000 and 1,068 of 4,890 in NIDs
+   10,001–15,000. (850 archived articles in NIDs 15,001–20,000 do have
+   dates.) **All non-archived articles (9,013 of them) have 100% date
+   coverage.**
+
+2. **`related_brands` matching gap (~12%).** Fragrantika has renamed
+   ~50 brands over the years (e.g., `Christian Dior → Dior`); the captured
+   name in news may be the older form, while `brands_database_v2.csv`
+   carries only the canonical (post-rename) name. Use `related_pids` for
+   authoritative brand resolution (§5.4).
+
+3. **`related_perfumers` matching gap (~11%).** Some perfumer names
+   referenced in news (e.g., `Jean Claude Ellena` ×191) are not yet in the
+   companion `perfumers_database_v2.csv`. The names in news are
+   diacritic-correct; the gap is in the reference DB coverage. Use
+   `related_pids → merged_database.noses_f` for authoritative resolution
+   (§5.5).
+
+4. **Cross-language mirrored short comments (~0.014%).** Approximately
+   650 short, identical-text comments appear in multiple languages
+   (e.g., a "5/10" rating posted by the same author on multiple subdomains).
+   These are real (Fragrantika's display behavior), not parser duplicates.
+
+5. **Comments PID coverage gap.** 38,855 perfumes (29%) have zero comments
+   in any language. Verified against Fragrantika's `reviews_count` field:
+   99.8% of these perfumes truly have zero reviews on the platform. A
+   targeted re-collection of the 73 PIDs flagged as `reviews_count > 0`
+   yet missing comments recovered 68 of them (305 new comments added);
+   the remaining 5 PIDs have `reviews_count > 0` in Fragrantika's metadata
+   but the page returns no review HTML — treated as upstream
+   metadata/page mismatch.
+
+---
+
+## 9. Loading examples
+
+### 9.1 Python (PyArrow + Pandas)
+
+```python
+import pyarrow.parquet as pq
+import json
+
+# Comments
+comments = pq.read_table('comments.parquet')
+df = comments.to_pandas()
+print(df.head())
+
+# News list-field unpack
+news = pq.read_table('news.parquet').to_pandas()
+news['related_pids'] = news['related_pids'].apply(json.loads)
+news['related_brands'] = news['related_brands'].apply(json.loads)
+```
+
+### 9.2 DuckDB (zero-copy SQL on Parquet)
+
+```sql
+-- Direct query, no import needed
+SELECT lang, COUNT(*) FROM 'comments.parquet' GROUP BY lang ORDER BY 2 DESC;
+
+-- Multi-file join
+SELECT n.title, COUNT(nc.comment_id) AS replies
+FROM 'news.parquet' n
+LEFT JOIN 'news_comments.parquet' nc ON n.nid = nc.nid AND nc.is_reply = true
+GROUP BY n.nid, n.title HAVING replies > 10;
+```
+
+### 9.3 Polars
+
+```python
+import polars as pl
+df = pl.read_parquet('comments.parquet')
+df.filter(pl.col('lang') == 'en').head()
+```
+
+---
+
+## 10. Sample files
+
+The `samples/` directory contains three small Parquet files that
+demonstrate the cross-dataset relationships using real production data:
+
+| File | Rows | Content |
+|---|---:|---|
+| `comments_sample.parquet` | 25 | Reviews of 5 popular perfumes (Mugler Angel, Guerlain Shalimar, Creed Aventus, JPG Le Male, Dior Poison), each in 5 languages (en/ru/de/es/fr). |
+| `news_sample.parquet` | 20 | News articles that reference these 5 perfumes via `related_pids`. Diverse categories and archived/recent mix. |
+| `news_comments_sample.parquet` | 20 | Comments from 7 of the news articles in `news_sample.parquet`, mix of root and replies. |
+
+The samples are produced by `scripts/build_sales_samples.py` and use the
+**identical Parquet schema and Zstandard compression as the production
+files** — buyers can verify their loading code against the samples before
+committing to the full datasets.
+
+### 10.1 Cross-link verification (using the samples)
+
+```python
+import pyarrow.parquet as pq
+import json
+
+c = pq.read_table('samples/comments_sample.parquet')
+n = pq.read_table('samples/news_sample.parquet')
+nc = pq.read_table('samples/news_comments_sample.parquet')
+
+sample_pids = set(c.column('pid').to_pylist())
+print(f"Comment-sample PIDs: {sorted(sample_pids)}")
+# {53, 218, 430, 704, 9828}
+
+# Find news rows that reference these PIDs:
+for i, rp in enumerate(n.column('related_pids').to_pylist()):
+    pids = set(int(p) for p in json.loads(rp) if p.isdigit())
+    if pids & sample_pids:
+        nid = n.column('nid').to_pylist()[i]
+        title = n.column('title').to_pylist()[i]
+        print(f"  NID {nid} ({title[:60]}...) references PIDs {pids & sample_pids}")
+
+# All news_comments NIDs are subset of news_sample NIDs:
+sample_nids = set(n.column('nid').to_pylist())
+nc_nids = set(nc.column('nid').to_pylist())
+assert nc_nids <= sample_nids, "FK integrity holds in samples"
+print(f"\nAll news_comments_sample NIDs link back to news_sample: {nc_nids <= sample_nids}")
+```
+
+Running this script on the supplied samples will print a verification
+trace and confirm relational integrity.
+
+---
+
+## 11. Versioning and updates
+
+| Field | Value |
+|---|---|
+| Snapshot date | 2026-05-05 |
+| Comments PID range | 1 — 130,121 |
+| News NID range | 1 — 25,000 |
+| Schema version | v1 (8 / 16 / 9 fields respectively) |
+| Compression | zstd |
+| Format | Parquet 2.x |
+
+Updates to these datasets are produced as full snapshots; incremental
+diffs (per-PID / per-NID delta exports) are available on request.

accords.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a3991b11c3f0c226f37701f69634c1a7eb137feb531282610db4936c01728d9
+size 3307

brands.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b62069ad3fa04f7d74851b3cc36d00947b9b3751bef76dcd89d9122fd42a953f
+size 33955

fragrances.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddca201eb42afb1915706356dabcb68a8072a63dfafa7487ee904a3d754b6bfc
+size 67536

news_sample.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b845c176dcf4ca75e23041661f0443d4134308eee7f4ed463d62c1847418d5af
+size 124859

notes.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a70ba110c929c9eb189d2209dd54cc9dabdc8c05ba7821be7745a15d1514f4
+size 12565

perfumers.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2789476c73b6a30cb506871288023b1a1478a4200133ca38b4d986b271479d8e
+size 30431

translations.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00529c19e4b6042589fab28039c8a930f7d624aae693ed274201e8d63f4a930e
+size 10195