EA·Lattice

SEC Filings

Every filing a tracked registrant has made to EDGAR, with its form type, filing date, period of report, acceptance timestamp and document references. Amendments are linked to what they amend, so a restatement is visible as a chain rather than as a replacement.

degradedfresh25h agopit: publishedparquetU.S. Securities and Exchange Commission

Collection

Expected frequency
1 day
Stale after
3 days
Last successful run
2026-09-15 00:27:17 UTC
Latest source period
2026-09-14 00:00:00 UTC
Earliest coverage
1994-01-04 00:00:00 UTC
Latest coverage
2026-09-14 00:00:00 UTC
Rows
728,801
Partitions
1
Owner

Point in time

What a historical query can reconstruct

Level
published
On revision
immutable
Effective field
filing_date
Published field
acceptance_datetime
Primary key
cik, accession_number

Quality

CheckResultSeverityDetail
row count anomalypasswarnonly 3 prior run(s); need 5 to judge
schema driftpasswarnobserved columns match the manifest
freshnesspasswarnlast update 24.5h ago, within the 72.0h SLA
timestamp orderingwarnwarn36 row(s) have out-of-order timestamps
primary key uniquepassfail1000 row(s), all keys distinct
required non nullpassfailall 4 required column(s) populated
provenance resolvespassfailall 595 cited raw object(s) are catalogued

Known limitations

What this dataset does not cover

  • Coverage is the tracked universe of registrants, not all of EDGAR. A company absent from the tracked list has no filings here, which is a gap rather than an absence of filings.
  • Amendment linkage is derived from the form suffix and the reported period, because EDGAR does not state which specific filing an amendment supersedes. Where that is ambiguous the link is left empty rather than guessed.
  • Acceptance timestamps are UTC, but EDGAR's business day runs on US Eastern time. A filing accepted late in the Eastern evening is dated the following business day, so acceptance time and filing date can disagree by more than they appear to.
  • The submissions API returns recent filings inline and older ones in separate index files.