Get Tdocs Batch

Read up to 10 TDocs in one call, for example every TDoc a summary or search result names.

Each TDoc comes back as GET /tdocs/{tdoc_id} returns it (register metadata, sources with roles, and with include its units, text and revision chain), in request order, each with a status: ok, not_indexed (not in the catalog yet: call fetch_tdoc for it), not_found, withdrawn (metadata in tdoc), invalid_tdoc_id, or error. One TDoc that cannot be read does not fail the others. A TDoc in the catalog whose content was not parsed yet has status ok and content.status not_retrieved: call fetch_tdoc for it.

Unit and text budgets are per TDoc. Without max_unit_chars and max_text_chars, 48,000 and 16,000 characters are shared across the TDocs; long unit lists are cut (units_truncated).

Quota: for REST and MCP this call counts as one request per TDoc (requests_counted), the same as calling get_tdoc once per TDoc. If fewer requests remain than TDocs asked for, it fails with 402 before reading anything.

Excerpt limit: each TDoc carries at most 2,000 words of units and text, cut like get_tdoc; excerpt and official_url say where to read the rest.