NEWS

omock 0.7.0.9000

Added subsetVocabularyTables() to subset vocabulary tables to selected concept IDs while keeping related vocabulary tables consistent.
Added conceptSet, includeRelated, and keepDomains arguments to mockCdmReference(), mockVocabularySet(), and mockVocabularyTables().
Added a new vignette showing how to subset vocabulary concepts.
mockMeasurement() now creates more realistic mock measurements by filling value_as_number, unit_concept_id, and value_as_concept_id.
mockCohort() now creates an omopgenerics-style cohort attrition table with an initial qualifying events row.

changeCdmVersion() now uses omopgenerics::compareOmopTableFields() to identify added, removed, and renamed fields between OMOP CDM versions.
changeCdmVersion() now aligns local OMOP tables to the target CDM version, including removal of fields that do not exist in the target version.
mockCdmFromDataset() now validates cdmVersion and converts loaded datasets to the requested CDM version when needed.
Mock dataset CDM names now resolve to a downloadable dataset version before download, avoiding invalid URL errors when a requested target CDM version is not directly hosted.

mockCdmFromTables() now gives a more informative error when input tables contain missing dates in columns used to derive observation periods.
mockObservationPeriod() now returns an empty observation period with an informative message when the person table is empty.
mockConditionOccurrence() now generates round(recordPerson * nrow(person)) records in total and samples condition concepts across those records.
Downloaded mock datasets within 99.99% of their expected size are now reused; users are only asked to delete a dataset when it appears incomplete.
Improved the prompt shown when a downloaded dataset appears incomplete.
Updated documentation with shared roxygen templates and clearer parameter descriptions.

Deprecated mockConcepts() because it creates placeholder concept rows that may be mistaken for real OMOP vocabulary content. Use mockCdmReference(vocabularySet = "eunomia"), mockVocabularyTables(), or subsetVocabularyTables() instead.

Improved dataset downloads with retry logic and timeout handling for more reliable fetching.
Updated mock dataset URLs to a new storage location and improved join logic.
Enhanced mock CDM examples, including clearer error handling and more robust behaviour.
Added an internal GI Bleed dataset, along with the associated loading logic, for expanded example coverage.

Removed unused input validation and table-checking functions to simplify the codebase.
Fixed the placement of rows_to_keep calculation in mockCohort to ensure correct behaviour.