house_activity_reports

Committee activity reports (House and Senate)

One row per captured end-of-Congress committee activity report package, House or Senate (the table's name predates its Senate rows; the package id's hrpt/srpt says which), with what its print adds to its own MODS index. Joins document_citations on package_id = document_key and text_sha256. All columns are stored as VARCHAR.

Coverage. Sampled. House committee activity reports: CRPT packages from the 2023-01-01 issue floor whose titles name activity under the shared title rule. Not an earlier-archive census. (measured 2026-09-28)

Data quality. The MODS is authoritative for bills, laws and Code sections, and this table's own columns say so. Read at full page depth across the eight reports spicy-docs sampled, the print named 0 of 1,406 bills, 0 of 174 public laws, 0 of 37 U.S. Code sections and 0 of 7 Statutes at Large pages that the package MODS does not already state — so distinct_bills_beyond_index, distinct_laws_beyond_index and distinct_usc_sections_beyond_index are near zero, and a nonzero one is worth reading the evidence span for rather than taking as new yield. Measured here 2026-09-20 over 41 reports rather than eight, "zero" is very nearly right and not exactly right: 7 print-only bills of 7,686 distinct, 5 print-only laws of 990, and 6 print-only Code sections. The one apparent survivor, 188-11, was the print's own misprint of 118-11. What the print does add is committees beyond the submitting one (27 resolved codes), 87 RINs, 38 agency dockets and 5 GAO product ids. Except for the Congress of a Senate report's bills. A Senate committee files its report early in the following Congress, and GovInfo stamps the MODS's bills with that filing Congress; printed bills are keyed in covered_congress, the Congress the report itself states, so on a Senate report bills_congress_mismatch and distinct_bills_beyond_index count the MODS's wrong Congress, not new yield (2026-09-26 drift audit: the eight Senate reports on the 117th had published 1,915 citations under 118th-Congress bills before the fix). And for a House report's bills printed under a Congress subheading. CRPT-117hrpt705 and -118hrpt974 print a predecessor bill's history under 116th Congress or 117th Congress inside the current bill's entry, and -118hrpt967 heads each Congress of its committee history, 107th to 118th, the same way; since spicy-docs bill_number 004 those bills are keyed in the subheading's Congress until the next entry, while the MODS stamps them with the filing one, so bills_congress_mismatch and distinct_bills_beyond_index count them (45 on -117hrpt705, 225 on -118hrpt967; replay receipt print-subheading-2026-09-26/). Since bill_number 005 a bill the print sets its own Congress beside (H.R. 6752, 115th Cong., S. Res. 116, 112th Congress) takes that one ahead of any subheading, and those count the same way (38 citations over the 40 reports; receipt print-inline-congress-2026-09-26/). A committee name both chambers hold (Committee on the Judiciary, on the Budget, on Veterans' Affairs, on Armed Services, on Appropriations) resolves to the chamber of the report's own committee, read off the package id; before 2026-09-26 the House won every such name, and the Senate reports published 1,746 House codes for their own committees, 1,718 of them hsju00. A chamber the print names before a committee (Senate Committee on Armed Services) selects that chamber's committees instead (76 rows moved, 24 of them hshm00 to ssga00). Limit: 7 bare committee names in Senate reports reach a name only the House roster holds and keep the House's code, because the print names no chamber: 5 hshm00 (a wrapped Committee on Homeland Security and Governmental Affairs, or its Committee on Home- fragment) and 2 hsso00 (Select Committee on Ethics). Read them as ambiguous, not as House committees. Receipt: fix-print-citations-2026-09-26/replay-qualified.json. committees_unresolved is a floor, not a defect count: the Senate cvc roster states only the committees its listed senators sit on, so a name it does not reach stays unresolved rather than being guessed. stated_page_count is the publisher's own pages field and is never re-derived from the bytes; pages_read is what the extraction saw, and the two are reported separately on purpose (CRPT-118hrpt965 states 282 and a 60-page read saw 60). submitted_by_bioguide_id is NULL where the publisher states the member without an id, which CRPT-118hrpt965 does.

• Parquet file: house_activity_reports.parquet
• MCP query_sql support: Configured; requires an available artifact.
• Publication status: Not established by this schema page or its measurement date.
• Row count: Not stated here; the MCP describe_table reply gives the live count under publication.

Column	Type	Description
package_id	VARCHAR	The GovInfo package id, which is this row's identity.
congress	VARCHAR	The numbered Congress, as the keyed summary states it: the Congress the report was filed in. For a Senate report that is the Congress after the one it covers; covered_congress is the one its printed bills belong to.
session	VARCHAR	The session of Congress the summary states.
title	VARCHAR	The package title as the keyed summary states it.
date_issued	VARCHAR	The date the package was issued.
last_modified	VARCHAR	When the publisher last modified the package; the merge prefers the larger value.
committee_system_code	VARCHAR	The authorityId of the first congCommittee the MODS names: the committee whose activity this report is, already keyed the way committees.system_code is. Never read off the printed cover, which the citation rules read for a different purpose.
committee_name	VARCHAR	That committee's authority-standard name, as the MODS spells it.
submitted_by_bioguide_id	VARCHAR	The bioGuideId of the member the MODS names with role="SUBMITTEDBY". The one bioguide id any of these documents states -- the citation rules measured zero printed ones across ten families -- and NULL where the publisher states the member without an id, which CRPT-118hrpt965 does.
committee_count	VARCHAR	How many committees the MODS names; every one is in committees_json.
committees_json	VARCHAR	Every congCommittee the MODS names, as a JSON array of objects carrying the authority id, chamber, name and the Congress the authority record itself states -- which is the roster's Congress and not necessarily this report's.
associated_bill_count	VARCHAR	How many bills the MODS names; every one is in associated_bills_json.
associated_bills_json	VARCHAR	Every root-level <bill> the MODS names, as a JSON array of objects carrying the bill's natural key and the publisher's own context marker, in document order. The context is kept because document order is not priority order.
associated_law_count	VARCHAR	How many laws the MODS names; every one is in associated_laws_json.
associated_laws_json	VARCHAR	Every root-level <law> the MODS names, as a JSON array of joined laws identities, in document order. Carried because it is what the beyond-index bill and law counts are measured against.
associated_usc_section_count	VARCHAR	How many U.S. Code sections the MODS names; every one is in associated_usc_sections_json.
associated_usc_sections_json	VARCHAR	Every <USCode> section the MODS names, as a JSON array of objects carrying the {title}-{section} key and the publisher's own subsection detail, in document order. A chapter-only <USCode> block contributes nothing: a chapter is not a section and has no hosted key.
related_report_count	VARCHAR	How many sibling reports the MODS names; every one is in related_reports_json.
related_reports_json	VARCHAR	Every <congReport> the MODS names, as a JSON array of CRPT package ids, in document order: the report-to-report edge, stated by the publisher rather than read off the print.
distinct_bills	VARCHAR	How many distinct bill keys the print names, by the shared citation rules. A floor, not a total: it counts what pages_read reached, and the MODS is the authoritative list. A key carries covered_congress, because a print mostly writes H.R. 7806 with no Congress beside it, unless the print states the bill's own: set right after it (H.R. 6752, 115th Cong.), else in a Congress subheading over it (116th Congress), which it takes instead; where the report states none, these are the printed forms, unresolved.
distinct_bills_beyond_index	VARCHAR	How many of those the MODS does not already state, which is a measured floor and not a discovery rate: 0 of 1,406 across all eight sampled reports at full page depth. The per-row document_citations.stated_by_index carries the same comparison exactly; this column is its summary for one document. NULL where covered_congress is, because a printed bill without a Congress cannot be compared with the MODS's keys.
bills_congress_mismatch	VARCHAR	How many printed bills the MODS states under a different Congress than covered_congress: the same comparison run a second time on (bill_type, number) alone. Nonzero on a Senate report is the publisher's index disagreeing with the report -- GovInfo stamps a Senate report's <bill> list with the filing Congress, so CRPT-118srpt99's H.R. 5376 is 118 in its MODS and 117 in its print, which reports on the 117th. Nonzero on a House report counts the bills the print keys in an earlier Congress -- beside the bill or in a subheading over it -- which the MODS stamps with the filing one, and any measure of another Congress the print names without saying so, whose row's bill_id is then wrong. NULL where covered_congress is.
distinct_laws	VARCHAR	How many distinct public laws the print names; a floor bounded by pages_read.
distinct_laws_beyond_index	VARCHAR	How many of those the MODS does not already state: 0 of 174 across all eight sampled reports at full page depth. A floor with stated_by_index semantics, not a yield estimate.
distinct_usc_sections	VARCHAR	How many distinct U.S. Code sections the print names; a floor bounded by pages_read.
distinct_usc_sections_beyond_index	VARCHAR	How many of those the MODS does not already state: 0 of 37 across all eight sampled reports at full page depth. This is the correction that removed U.S. Code sections from the family's claimed new yield; the kinds that survive are committees, RINs, agency dockets and GAO ids.
committees_resolved	VARCHAR	How many distinct committee system_codes the printed committee names settled to against the rosters the caller supplied.
committees_unresolved	VARCHAR	How many distinct printed committee candidates no supplied roster settled. Not a defect count: the pinned Senate roster excerpt reaches only the committees its listed senators sit on, so this is a floor on what a full roster would settle.
citation_rows	VARCHAR	How many citation rows this document produced, across every kind.
pages_read	VARCHAR	How many pages the extraction actually read, which a capped read makes smaller than page_count.
stated_page_count	VARCHAR	How many pages the document has, as the keyed summary's own pages field states it -- not re-derived from the bytes. Named apart from committee_reports.page_count, which is a package-keyed column holding how many pages that extraction read: the two would otherwise collide on one package_id with opposite meanings (282 against 60 on CRPT-118hrpt965).
pages_capped	VARCHAR	Whether the read stopped short of the document, so every count above is a floor.
body_rendition	VARCHAR	Which rendition the text was derived from. pdf here because the acquirer is asked for sources.govinfo.bodies.PRINT_BODY_PREFERENCE -- the sealed order with PDF first -- and not because it is all an activity report offers: these packages state htm too, and under the sealed default 10 of 41 refused on HTML nesting depth while the 31 that were read stated no page at all, which is what the four page-stating columns here need.
body_derivation	VARCHAR	How that rendition became text.
text_sha256	VARCHAR	Digest of the normalized text the citation spans index into.
rule_set_version	VARCHAR	Digest over every citation rule's name, version, pattern and rejects, so these counts name the rules that produced them.
covered_congress	VARCHAR	The Congress the report says it reports on, read from its own words -- the index title, else the print's cover, else its first five pages, and covered_congress_source says which -- and the Congress a printed bill key is built in where the print states no other for it. Beside congress rather than replacing it: the two differ on every Senate report, which is filed early in the following Congress (CRPT-118srpt99 reports on the 117th). NULL where the report states no Congress, or where the first source stating one states several; its printed bills the print states no Congress for then stay unresolved rather than being stamped with a guess.
covered_congress_source	VARCHAR	Which of the report's own statements covered_congress was read from: title, cover or front_matter, strongest first, the first to state any Congress deciding. Set beside a NULL covered_congress where that source states several; NULL where none states one.