tdocs

Search Tdocs

Search the text of 3GPP TDocs (meeting contributions, CRs, LSs, summaries, chair notes, reports) by meaning and by exact term.

Use for "are there any TDocs about X", "who is proposing Y", or finding where an IE, parameter, spec, or TDoc ID is discussed. Merges semantic (vector) and keyword (lexical) matches, then reranks. Exact identifiers in the query (TDoc IDs such as R2-2008188, IE names such as ta-CommonDrift, spec numbers such as 38.213, quoted phrases) are matched literally, and passages that contain them come first.

filters takes the same fields as GET /tdocs (source company with alias matching, meeting, group, agenda_item, doc_kind, status, work_item, spec, release, since, until). unit_types limits results to passages holding numbered proposals or observations.

Each result is one passage with its TDoc's ID, title, doc_kind, register status and status_label, meeting, agenda item, source companies with roles, official_url, unit type and labels (for example "Proposal 3"), page and paragraph anchor, and text. anchor is where the passage starts; units gives each numbered proposal or observation in the passage with its own anchor, which is the one to cite for that unit. coverage lists the meetings whose TDoc text is indexed, so an empty result is a real "nothing found in these meetings".

Evidence rules: a passage from a contribution is its source companies' proposal, not an agreement. Use list_meeting_decisions for what was agreed, and get_tdoc to read a whole TDoc.

Errors: 400 invalid_parameter, 409 organization_ambiguous (candidates returned), 502 search_unavailable.

post/tdocs/search

Request

  • Base URL: https://api.3gppscout.com (the document declares no server; this is the origin it was published from)
  • URL: https://api.3gppscout.com/tdocs/search
  • Auth: none declared

Request body

querystring required

What the TDocs should discuss, in plain words or exact terms, e.g. 'timing advance pre-compensation for NTN', 'ta-CommonDrift', 'R2-2008188'.

unit_typesstring[] nullable

Only chunks holding these units: 'proposal', 'observation', or 'body' (text outside numbered proposals and observations).

rerank_top_kinteger

Results to return after reranking.

max_per_tdocinteger

At most this many passages from one TDoc. Use 1 for a list of distinct TDocs.

include_textboolean

Include each passage's text.

rerankboolean

Rerank candidates for precision. Leave true.

Response

Successful Response

unknown required

Changes

No recorded changes to this operation across all 1 revision of this API.