Computed Data and Data Layers¶

Computed.js - the derived-data engine¶

Raw rows read from the spreadsheet are never used directly by reports or checks. Instead, callers request named "compute keys" (for example issues_for_reports_all or top_3_failing_criterias_wcag_21) and Computed.js resolves the dependency graph and calculates only what is needed.

The key data structure is computeStructure_, an 8-level array. Level 1 has no dependencies; each subsequent level may depend on keys from earlier levels. computeData_ walks the levels in order, so dependencies are always satisfied before dependents run.

The file also owns the definitions of the accessibility standards (WCAG 2.0/2.1, Revised Section 508, EN 301 549, EE additions), the issue validation rules, and all per-scenario / per-criteria / per-status aggregations that feed the reports, VPAT/ACR, slides, and quality checks.

Engine¶

Function	Description
`computeData_(currentSheetData, templateSheetData, computeKeys)`	Main entry point. Expands the requested keys into their full dependency set, runs every needed compute function in level order, and returns only the explicitly requested keys.
`resolveComputeKeyDependencies_(key)`	Recursively returns a key plus all of its transitive dependencies; throws on unregistered keys.
`mergeArrayValuesNoDuplicate_(origArray, arr)`	Merges dependency lists without duplicates.
`getWCAGSuffixForComputed_(wcagVerNum)`	Maps a WCAG version to the computed-key suffix (`_wcag_21` / `_wcag_20`). Deprecated.

Standards and criteria¶

Function	Description
`getWCAGCriteriaFromJson_` / `get508CriteriaFromJson_` / `get301CriteriaFromJson_` / `getEEAdditionsCriteriaFromJson_`	Flatten the bundled standards JSON into dictionaries keyed by criteria ref-id (display name, URL, conformance level).
`computeCriteriaWCAG20_` / `computeCriteriaWCAG21_` / `computeCriteria508_` / `computeCriteria301_` / `computeCriteriaEEAdditions_`	Compute functions for the per-standard criteria dictionaries. WCAG 2.1 is built from the Template spreadsheet's criteria tab rather than the bundled JSON.
`computeCriteriaAllPossible_`	Merges all standards into one combined dictionary, used for the Results sheet's data-validation dropdowns.
`computeWCAG21OnlyCriteriaRefIds_`	The criteria new in WCAG 2.1 (present in 2.1 but not 2.0).
`get508CriteriaObjByDisplayName_` / `get301CriteriaObjByDisplayName_` / `getWCAGCriteriaObjByDisplayName_`	Look up a criteria object by the display name stored in the Results sheet.
`addOrEnrichCriteriaInDict_`, `getNumFromEn301RefId_`	Small dictionary and ref-id helpers.

Scenarios¶

Function	Description
`computeScenariosWithStepsNumbered_`	Adds numbered step strings to every scenario.
`computeScenariosNoAdHoc_`	The canonical scenario list: Ad hoc removed, sequential `num` added.
`computeScenariosAllPossible_`	All numbered scenarios plus the Ad hoc scenario (synthesized if missing). Used for the scenario dropdown validation.
`isScenarioWithUnorderedSteps(scenarioName)`	True when a scenario's steps have no meaningful order and should not be numbered.
`isScenarioHasAnyStepsNameValid(scenarioName)`	True when a scenario accepts free-text steps (currently Ad hoc).
`addStepsNumsForScenarioObject_`, `getScenarioStepStringWithNumber_`	Step formatting helpers.

Issue filtering and validation¶

Function	Description
`computeIssuesWithDraftsMarked_` / `computeIssuesNoDraft_`	Tag and filter out draft (effectively empty) issue rows. An issue is a draft when Unique Title, Details, Issue Title, and Image/Video are all blank (`isIssueDraft_`).
`computeIssuesWithValidations_`	Attaches `_missing_required_fields` and `_invalid_fields` to each non-draft issue.
`issueMissingRequiredValues_(issue)`	Lists empty required columns (id, browser, testing tool, scenario, steps, title, issue title, EE status, issue owner, external resource; Steps may be blank for Ad hoc issues).
`issueInvalidValues_(...)`	The single largest validator in the codebase (~320 lines). Validates that the ID is numeric and that Browser / Testing Tool / Scenario / Steps / Success Criteria / Issue Title / Status match the configured lists.
`computeIssuesForReportsAll_`	Non-Ad-hoc issues with a reportable EE Status (blocking/major/minor/advisory; not Fixed/Closed).
`computeIssuesForReportsGenerationWCAG21_` / `...WCAG20_`	Reportable issues logged under a WCAG criteria of the given version.
`computeIssuesForVPATGeneration508_` / `...301_`	Reportable failing issues logged under a 508 / EN 301 criteria, for the ACR.
`computeIssuesForQualityChecks_`	Fully valid non-Ad-hoc issues, the set the quality checks inspect.
`computeIssuesWithErrors_`	Reportable issues with missing or invalid fields; drives the reports pre-flight error list.
`computeIssuesForPreviewResults_`	Non-draft, non-hidden issues enriched for the Preview UI: Drive file id, 300px and 1600px thumbnail URLs, and title/notes rendered from markdown to HTML.
`computeIssueTitlesAllPossible_`	Every issue title across every criteria, for the Issue Title dropdown fallback.
`computeActiveRowIssueId_`	The issue `_id` of the user's currently selected Results row (powers "follow my cursor" in Preview).
`computeAuditIsWebNotSoftware_`	True when the Audit Scope is "Web + mobile browser"; drops 508 Chapter 5 / EN 301 Chapter 11 from the ACR.

Aggregations¶

Each aggregation follows the same pattern: a generic function plus thin ...WCAG21_ / ...WCAG20_ compute-key wrappers.

Function	Description
`computeIssuesByScenarioByStatus_` (+ `NoAdvisory_` variant)	Scenario → status → issues map (optionally excluding Advisory).
`computeIssuesByScenarioByStepByStatus_`	Scenario → step → status → issues, for step-level tables.
`computeFeaturedIssuesByScenario_`	Per scenario, the issues flagged "Featured".
`computeIssuesByCriteriasByStatus_` / `computeIssuesByCriterias_` / `computeIssuesByCriteriaByStatusByTitle_`	Criteria-keyed groupings.
`computeIssuesByIssueTitleByStatus_` / `computeIssuesByStatus_`	Title-keyed and flat status groupings.
`computeImportantIssueByScenario_`	The single most important issue per scenario (worst status first) for summary slides.
`computeCriteriasSortedByFailingIssuesCount_` / `computeIssueTitlesSortedByIssuesCount_`	Sorted failure rankings.
`computeTop3FailingCriterias_` / `computeTop3FailingIssues_`	The top-3 lists used in reports.
`computeVPATIssuesCriteriaInfo_` (+ `508_`, `En301_`, `Combined_` variants)	Per-criteria conformance level and remarks for the ACR, including 508 and EN 301 criteria calculated as combinations of WCAG criteria.
`computeNaOrSupportInfo_`	Resolves each criteria's supported / NA remark template from the "Report: NA or Support - v3" sheet and substitutes variables. Emits a sentinel error value when the selected scenario is wrong or missing, which the reports pre-flight check looks for.

Sheets_data.js - the sheet read layer¶

Every feature starts by calling sheetsData(ssId, extractionConfig) to pull only the tabs it needs. The file also defines dataDefaultOptions (per-tab extraction defaults) and sheetsConfig (physical layout: header row counts, start columns, frozen columns, and the canonical Results column order). Nothing else in the codebase should read raw ranges from the audit tabs directly.

Function	Description
`sheetsData(ssId, extractionConfig)`	Entry point. Opens the spreadsheet (or reuses the active one), dispatches each requested key to its extractor, and returns `{key: extractedData}`.
`getAuditUser_`	Returns the current Google account (feeds role info into the HTML app).
`getAuditSs_`	The spreadsheet object plus id, URL, and name.
`getAuditSheets_`	All tabs, plus an `audit_sheets` map indexing the known audit tabs by logical key.
`getIssues_`	The Results tab as issue objects via `getResultsTabEntries_`, honoring the `mark_hidden_rows` option.
`getReportNaOrSupportV3_`	The "Report: NA or Support - v3" tab as a criteria-keyed object of `{is_supported, details}`.
`getReportNotes_`	Legacy report-notes reader (the notes now live on the Configuration tab).
`getScenarios_`	The Scenarios tab read column-wise (each scenario is a column: name, overview, then steps).
`getConfiguration_`	The Configuration tab's key/value pairs, applying per-key options (string vs array, EE-formatted dates).
`getPlatformBrowserScope_`	The "Platform / Browser by scope" tab, notably the third-party company and product lists with URLs.
`getReviewReport_`	The Review Report tab: the current home of the Platform/Browser, Testing Tools, and Technologies lists used to validate issue rows.
`issueSpecialField_(field, suffix)`	Builds synthetic field names like `title__to_html`, the convention for computed fields on issue objects.

Google_drive_data.js - the Drive data layer¶

The Drive-side counterpart of sheetsData. Currently the only key is audit_reports_folder, the folder where generated reports are written.

Function	Description
`googleDriveData_(data, config)`	Key-dispatch entry point mirroring `sheetsData`.
`getAuditReportsFolder_(data)`	Returns `{folder, id, url}` for the audit's reports folder.
`_getCurSSReportsFolder_NEW(data)`	Finds the spreadsheet's parent Drive folder, then finds or creates the `<Product Name> Report` subfolder. Throws user-facing messages when the spreadsheet sits in Drive root or Product Name is unset.

Results tab parser (src/resultsSheet/getResultsTabEntries.ts)¶

A single TypeScript parser that turns the raw Results tab into structured objects used by reports, the preview UI, and the AI review feature.

Function Description

getResultsTabEntries_(resultsSheet, isToMarkHiddenRows) Reads every row and column (display values plus formulas) and returns {data: ResultsTabEntry[], columns_indexes}. Headers are translated through the colNames map into system names such as title and ee_status. Derived fields are added: ...__url from HYPERLINK formulas, steps__order_num plus a stripped step name, a boolean featured from the checkbox column, source_code wrapped in a markdown code fence and converted to HTML, and a numeric _id. When isToMarkHiddenRows is true it calls the advanced Sheets API and sets _filtered_from_view per row.