3gpp-tdocs

name	3gpp-tdocs
description	TDoc patterns, filename conventions, metadata structure, HTTP/FTP server access, and TDoc identification. Use when crawling TDocs from FTP directories, parsing TDoc metadata from portal, or validating TDoc numbers.

TDocs (Temporary Documents)

Quick Reference

Pattern	Example	WG	Subgroup
R1-xxxx	R1-2301234	RAN	1
RP-xxxx	RP-230045	RAN	Plenary
S4-xxxx	S4-251209	SA	4
SP-xxxx	SP-240001	SA	Plenary
C1-xxxx	C1-2345678	CT	1
CP-xxxx	CP-123456	CT	Plenary

Regex: [RSC][1-6P].{4,10}\.(zip|txt|pdf)

Overview

TDocs are meeting documents produced by members participating in 3GPP Working Groups and Sub-Working Groups. They include proposals, reports, and other documents related to the development of 3GPP standards.

TDoc Allocation

Every TDoc is always allocated to a specific 3GPP meeting.

TDoc Number Format

Pattern

Regex Pattern:

TDOC_PATTERN = re.compile(r"([RSC][1-6P].{4,10})\.(zip|txt|pdf)", re.IGNORECASE)

Breakdown

Part	Pattern	Description
1st char	[RSC]	Working group: R (RAN), S (SA), C (CT)
2nd char	[1-6P]	Subgroup: 1-6 for WGs, or P for Plenary
3rd part	.{4,10}	4-10 identifier characters
Extension	`.(zip	txt

Examples that Match

Standard format:

• R1-2301234.zip - RAN1 TDoc
• S4-251209.txt - SA4 TDoc
• C1-2345678.pdf - CT1 TDoc

Plenary format:

• RP-230045.txt - RAN Plenary TDoc
• SP-240001.zip - SA Plenary TDoc
• CP-123456.zip - CT Plenary TDoc

Ad-hoc format:

• S4aA220001.zip - SA4 ad-hoc meeting
• R1eE230045.txt - RAN1 ad-hoc meeting

Case variations:

• r1-2301234.ZIP - Case-insensitive matching
• S4-251209.TXT - Case-insensitive matching

Examples that Don't Match

Wrong working group:

• T1-2300456.zip - T is invalid (use C for CT)
• X1-2300456.zip - X is invalid (use R, S, or C)

Wrong subgroup:

• R7-2300456.zip - Subgroup 7 doesn't exist (1-6 only)
• R1-12.zip - Only 2 characters after R1 (need 4-10)
• R1-12345678901.zip - 11 characters, exceeds maximum of 10

Wrong extension:

• R1-2301234.doc - Invalid extension (must be .zip, .txt, or .pdf)
• R1-2301234 - Missing extension

Not a TDoc:

• README.txt - Not a TDoc
• data.csv - Not a TDoc
• agenda.zip - Administrative file, not a TDoc

TDoc FTP/HTTP Server Access

Server Structure

3GPP maintains a file server at FTP-like HTTP URLs:

https://www.3gpp.org/ftp/tsg_<working_group_identifier>/

FTP Root URLs by Working Group

Working Group	Identifier	FTP Root URL
RAN	ran	https://www.3gpp.org/ftp/tsg_ran/
SA	sa	https://www.3gpp.org/ftp/tsg_sa/
CT	ct	https://www.3gpp.org/ftp/tsg_ct/

TDoc URL Pattern

TDocs are available at:

https://www.3gpp.org/ftp/tsg_<wg>/<sub-working_group_identifier>/<meeting_identifier>/Docs/<tdoc_nbr>.zip

URL Components:

• tsg_<wg> - Working group root (e.g., tsg_ran, tsg_sa)
• <sub-working_group_identifier> - Arbitrary path name (NOT official subgroup ID)
• <meeting_identifier> - Arbitrary path name (NOT official meeting ID)
• Docs/ - Typical subdirectory containing TDocs (case-insensitive)
• <tdoc_nbr> - TDoc filename stem
• .zip - File extension (99%+ of TDocs)

Example URLs

RAN WG1 TDoc R1-2301234:
https://www.3gpp.org/ftp/tsg_ran/WG1_RL1/RAN1_98/Docs/R1-2301234.zip

SA4 TDoc S4-251209:
https://www.3gpp.org/ftp/tsg_sa/WG4_S4/SA4_134/Docs/S4-251209.zip

Important Notes:

• Server uses HTTP protocol, not FTP (accessible via standard HTTP requests)
• <sub-working_group_identifier> and <meeting_identifier> are arbitrary path names, do NOT correspond to official IDs
• More than 99% of TDocs are .zip format
• Rare cases use .pdf or .txt extension

TDoc Subdirectory Detection

Problem

TDocs are typically stored in subdirectories like "Docs/" rather than directly in the base meeting directory.

Detection Process

1. Fetch base meeting directory from files_url
2. Parse HTML to extract directory links
3. Check for TDoc subdirectories (case-insensitive matching)
4. If subdirectories found: Crawl each subdirectory
5. If no subdirectories: Crawl base directory directly

TDoc Subdirectories

Common subdirectory names (case-insensitive matching):

• Docs/
• Documents/
• Tdocs/
• TDocs/
• DOCS/

Excluded Directories

Non-TDoc directories to skip during crawling:

EXCLUDED_DIRS = {"Inbox", "Draft", "Drafts", "Agenda", "Invitation", "Report"}

TDoc Direct Links

Portal TDoc View URL

When TDoc number is known, use 3GPP portal to query metadata:

https://portal.3gpp.org/ngppapp/CreateTdoc.Aspx?mode=view&contributionUid=<tdoc_nbr>

Example:

https://portal.3gpp.org/ngppapp/CreateTdoc.Aspx?mode=view&contributionUid=R1-2301234

TDoc Metadata Fields

When validating TDocs via portal page, parse these fields:

Required Fields

• title: Document title
• meeting: The meeting identifier (e.g., "SA4#133")
• contact: Contact person
• source: Responsible organization
• tdoc_type: Document type classification
• for: Purpose (agreement, discussion, information, etc.)
• agenda_item: Associated agenda item (split into agenda_item_nbr and agenda_item_title)
• status: Document status

Optional Fields

• is_revision_of: Reference to previous TDoc version (self-referencing FK)

Metadata Parsing

1. Ensure authenticated with 3GPP portal before fetching
2. Fetch portal page with mode=view parameter
3. Parse HTML using BeautifulSoup
4. Extract form fields (labels and values)
5. Handle agenda_item split: Separate number and title
6. Return dictionary of key-value pairs

Code Pattern

from bs4 import BeautifulSoup


def parse_tdoc_metadata(html_content: str, tdoc_id: str) -> dict[str, str]:
    """Parse TDoc metadata from portal HTML page."""
    soup = BeautifulSoup(html_content, "html.parser")

    metadata = {}

    # Parse form fields (simplified example)
    for label in soup.find_all("label"):
        value_element = label.find_next_sibling("input")
        if value_element and value_element.get("value"):
            field_name = label.get_text().strip(": ").lower()
            metadata[field_name] = value_element.get("value")

    return metadata

TDoc ID Normalization

Always normalize TDoc IDs to uppercase for case-insensitive matching and database lookups:

def normalize_tdoc_id(tdoc_id: str) -> str:
    """Normalize TDoc ID to uppercase for case-insensitive lookup."""
    return tdoc_id.upper().strip()

Example

# User input (any case)
tdoc_id = "r1-2301234"

# Normalized for database/storage
normalized_id = normalize_tdoc_id("r1-2301234")  # Returns "R1-2301234"

Usage in Code

When Crawling TDocs

1. Use TDOC_PATTERN to match files in HTTP directory listings
2. Extract TDoc ID from filename stem using group(1) of regex match
3. Store normalized ID (uppercase) in database
4. Build full HTTP URL to TDoc file
5. Validate via portal if needed (uses 3gpp-portal-authentication skill)

Example Implementation

import re
from bs4 import BeautifulSoup
import requests

# Match TDoc file
match = TDOC_PATTERN.search(href)
if match:
    tdoc_id = match.group(1).upper()
    full_url = base_url + match.group(0)
    
    # Later validate via portal
    metadata = fetch_tdoc_metadata(tdoc_id, credentials)

Working Group Inference

Infer working group from TDoc ID first character:

First Char	Working Group
R	RAN
S	SA
C	CT

def infer_working_group_from_tdoc(tdoc_id: str) -> str | None:
    """Infer working group from TDoc ID."""
    mapping = {"R": "RAN", "S": "SA", "C": "CT"}
    return mapping.get(tdoc_id[0].upper(), None)

Cross-References

• 3gpp-working-groups - For understanding working group and subgroup structure
• 3gpp-meetings - For understanding how TDocs are associated with meetings
• 3gpp-portal-authentication - For accessing TDoc metadata from portal

Key Points

• Case-insensitive: TDoc IDs should be normalized to uppercase for database operations
• 99%+ .zip files: Most TDocs use .zip format
• Public FTP access: Files are publicly accessible without authentication
• Portal metadata: Requires EOL account for detailed metadata
• Subdirectory handling: Always check for Docs/, Documents/, etc. before crawling base directory
• Excluded directories: Skip Inbox/, Draft/, Agenda/, Invitation/, Report/ as they are not TDocs

Resources

• 3GPP Official: https://www.3gpp.org/
• 3GPP Portal: https://portal.3gpp.org/
• 3GPP FTP: https://www.3gpp.org/ftp/