Skip to content

Tags

Tags are the lightweight grouping mechanism for TDocs, specs, and other items. A tag row points at an item via its kind and ID — no workspace directories, no copying.

Commands

tag add

3gpp-crawler tag add tdoc S4-250638 --tag codec
3gpp-crawler tag add tdoc S4-250638 --tag codec,rel-18   # multiple tags
3gpp-crawler tag add spec 26260 --tag rel-18
3gpp-crawler tag add other "ETSI TS 103 224" --tag reference
Argument Description
KIND tdoc, spec, or other.
ITEMS… Item identifiers: TDoc IDs (S4-250638), spec numbers (26260), or freeform IDs.
-t, --tag TAGS Tags to apply; comma-separate multiple values.

tag remove

3gpp-crawler tag remove tdoc S4-250638 --tag codec   # remove specific tags
3gpp-crawler tag remove tdoc S4-250638               # remove all tags

tag list

3gpp-crawler tag list                     # all tags
3gpp-crawler tag list --kind tdoc         # only TDoc tags
3gpp-crawler tag list --item S4-250638    # tags on one item
3gpp-crawler tag list --tag codec         # items carrying a tag
3gpp-crawler tag list -o json             # machine-readable (table, json, yaml, toon, ison)

database export-tags and database import-tags

Tags are exported with their natural keys (kind, item_id, tag), never the internal row ID, so a dump stays valid after the database is rebuilt.

3gpp-crawler database export-tags                          # JSON to stdout
3gpp-crawler database export-tags --kind spec -o yaml      # YAML to stdout
3gpp-crawler database export-tags -f tags.json             # write a backup file
3gpp-crawler database import-tags tags.json                # batch import (existing tags skipped)

database import-tags accepts .json, .yaml and .yml files. Re-importing a dump is safe: rows that already exist are skipped.

Rebuilding the database keeps tags

3gpp-crawler database clear-all deletes the database, so it first saves all tags to tags-backup.json next to the database, then restores them into the new database. The backup file is deleted after a successful restore; if the restore fails, it is kept so the tags can be imported manually with database import-tags.

Tag filters on queries

Both tdoc query and spec query accept --tag filters:

# OR within one option: items matching codec OR rel-18
3gpp-crawler tdoc query --tag codec,rel-18

# AND across repeated options: items matching codec AND rel-18
3gpp-crawler tdoc query --tag codec --tag rel-18

# Combine: (codec OR codec2) AND rel-18
3gpp-crawler tdoc query --tag codec,codec2 --tag rel-18

Tag filtering runs before --limit, so --tag atias --limit 300 returns tagged rows even when they sort beyond the first 300.

Kinds are independent

Tags on TDocs do not affect spec queries and vice versa — each kind has its own tag namespace:

3gpp-crawler tag add tdoc S4-250638 --tag codec
3gpp-crawler tag add spec 26260 --tag codec
# Both rows exist; queries filter within their own kind.

MCP

Tags are also exposed via the MCP server: tag_add, tag_remove, tag_list, plus database_export_tags / database_import_tags for backup dumps — see MCP Server.