Build Export Sheets — build_export

Converts CONCERT pipeline state into a named list of data frames ready for Excel export via writexl::write_xlsx().

Usage

build_export_sheets(
  raw,
  resolution_state,
  consensus_summary,
  cleaning_audit,
  reference_lists,
  column_tags,
  detection,
  file_info,
  enrichment_cache = NULL,
  detected_data = NULL,
  cleaned_data = NULL,
  toxval_output = NULL,
  harmonize_audit = NULL,
  site_manifest = NULL,
  site_alias_map = NULL,
  script_baseline_state = NULL,
  media_map = NULL,
  media_results = NULL,
  ignored_identifier_cols = character(),
  review_decision_evidence = NULL,
  identity_decisions = NULL,
  candidate_validation = NULL,
  review_reconciliation = NULL,
  candidate_review = NULL,
  source_identifier_evidence = NULL,
  identifier_diagnostics = NULL,
  toxval_identity_mode = c("lookup", "accepted"),
  cleaning_steps = NULL
)

Arguments

raw: Original uploaded data frame
resolution_state: Curated data with consensus_dtxsid, consensus_status
consensus_summary: List with n_agree, n_disagree, etc.
cleaning_audit: Data frame with audit trail (may be NULL)
reference_lists: List with $functional_categories, $stop_words, $block_patterns, and $strip_terms
column_tags: Named list of all applied column tags.
detection: List with $method, $confidence
file_info: List with $name, $size
enrichment_cache: Enrichment cache data frame (may be NULL)
detected_data: Data frame after header detection/extraction. When provided, this is written to Raw Data instead of the headerless ingest snapshot.
cleaned_data: Data frame produced by the Clean Data workflow (may be NULL). When provided, a Cleaned Data sheet is included.
toxval_output: Tibble with 56 ToxVal columns from map_to_toxval_schema(), or NULL (default). When NULL, the ToxVal Output sheet contains a placeholder note row.
harmonize_audit: Tibble with numeric measurement harmonization audit rows, or NULL (default). When provided, an additional Harmonization Audit sheet is appended.
site_manifest: Optional curated site/location manifest from Dataset Context. When provided and non-empty, a Site Manifest sheet is included.
site_alias_map: Optional Dataset Context raw-label alias map. When provided and non-empty, a Site Alias Map sheet is included and the Site Manifest is rebuilt from these mappings.
script_baseline_state: Optional automated resolution state captured before Review Results edits. When provided, cells that differ from resolution_state are persisted as baseline_cell records in Session State so re-imported sessions can regenerate replay review overrides.
media_map: Effective media map, including user overrides.
media_results: Row-level media identity, routing and original-value audit.
ignored_identifier_cols: Deliberately unused identifier metadata columns.
review_decision_evidence: Portable immutable decision evidence.
identity_decisions: Explicit structured source-identity decisions.
candidate_validation: Structured candidate-validation outcomes.
review_reconciliation: Optional reconciliation report.
candidate_review: Optional candidate-validation report.
source_identifier_evidence: Optional source-ID evidence report.
identifier_diagnostics: Optional unused/source-ID diagnostics.
toxval_identity_mode: Portable ToxVal identity policy, lookup or accepted.
cleaning_steps: Optional applied named logical cleaning-step mask.

Value

Named list of data frames with sheet names as keys

Details

The Curated Data sheet marks needs_review = TRUE for error/unresolvable identities, pending FOLLOW-UP flags, incoming needs_review = TRUE values, and unreviewed WQX-only candidates, including exact, alias, and fuzzy matches. To complete WQX identity review, use set_row_flag() with "VERIFIED" (or row_flags in stage_review()), or explicitly accept a WQX identity in Review Results, which records consensus_source = "manual_wqx". These outcomes resolve the WQX requirement, but do not clear a separate incoming review requirement, pending FOLLOW-UP, or error/unresolvable status. BAD alone does not create or clear a review requirement.