Inventory Column Analysis

01 Raw Data Preview

First entries from the Excel file showing all 6 product types across the 7 columns.

02 Column Definitions

What each column means in the containerboard industry, verified against CEPI and FEFCO standards.

03 Product Classification

How each raw product name maps to B2BPaper's paper_type, fiber_type, and other model fields.

04 GSM Range Analysis

Where each product's basis weight falls in the containerboard GSM spectrum.

05 FSC Certification Types

Forest Stewardship Council certifications found in this inventory, and how they should be stored.

06 Column → Field Mapping

Complete mapping from Excel columns to B2BPaper Product model fields with transformation rules.

07 Parser Gap Analysis

What the current paper_type_aliases.py handles vs. what's missing.

08 Recommendations

Proposed rules to close the parser gaps and enrich imported products automatically.

# Step 1: keyword match on raw product name (before suffix stripping)
if "recycled" in raw_name.lower():
    fiber_type = "recycled"

# Step 2: default by paper_type (if no keyword match)
FIBER_DEFAULTS = {
    "kraftliner": "virgin",      # >80% virgin kraft pulp by definition
    "testliner": "recycled",     # recycled fiber by definition
    "fluting": "virgin",         # unqualified "medium" = semi-chemical (virgin)
    "white_top_testliner": "recycled",
    "sack_kraft": "virgin",
    "mg_kraft": "virgin",
}
fiber_type = fiber_type or FIBER_DEFAULTS.get(paper_type, None)

import re

CERT_PATTERNS = [
    (re.compile(r'FSC\s+MIX\s+CREDIT(?:\s+CERTIFIED)?', re.I), "FSC Mix Credit"),
    (re.compile(r'FSC\s+MIX(?:\s+CERTIFIED)?', re.I), "FSC Mix"),
    (re.compile(r'FSC\s+RECYCLED', re.I), "FSC Recycled"),
    (re.compile(r'FSC\s+100%?', re.I), "FSC 100%"),
    (re.compile(r'FSC', re.I), "FSC"),         # bare FSC = generic
    (re.compile(r'PEFC', re.I), "PEFC"),
]

def extract_certifications(raw_name: str) -> list[str]:
    certs = []
    for pattern, label in CERT_PATTERNS:
        if pattern.search(raw_name):
            certs.append(label)
            break  # most specific match wins
    return certs

HS_CODES = {
    ("kraftliner", "virgin"):  "4804.11",  # >80% virgin sulfate, >115 g/m2
    ("testliner", "recycled"):  lambda gsm: "4805.24" if gsm <= 150 else "4805.25",
    ("fluting", "virgin"):      "4805.11",  # semi-chemical fluting
    ("fluting", "recycled"):    "4805.24",  # recycled fluting
    ("white_top_testliner", "recycled"): lambda gsm: "4805.24" if gsm <= 150 else "4805.25",
}

BRAND_MAP = {
    "liner xp": "Cascades XP",
    "top kl": None,  # placeholder for future brands
}

# Apply after paper_type resolution:
brand = BRAND_MAP.get(cleaned_name.lower(), None)

@dataclass
class PaperClassification:
    paper_type: str | None       # "fluting"
    fiber_type: str | None       # "recycled"
    certifications: list[str]    # ["FSC Mix Credit"]
    brand: str | None            # "Cascades XP"
    hs_code: str | None          # "4805.24"

def classify_paper(raw_value: str) -> PaperClassification:
    """Full classification from a raw product name string."""
    certs = extract_certifications(raw_value)
    fiber = infer_fiber_type(raw_value)
    paper_type = normalize_paper_type(raw_value)
    brand = BRAND_MAP.get(clean(raw_value), None)
    hs = derive_hs_code(paper_type, fiber, gsm=None)
    return PaperClassification(paper_type, fiber, certs, brand, hs)

09 HS Code Reference

Harmonized System codes for containerboard products in this inventory.

10 Inventory Summary

Aggregated view of the full inventory file by product type.

Sources & References