NEWS

---

WARDEN 2.0.4

• release() is a new helper that frees a resource_discrete for the current patient and, when resume_event is supplied, automatically schedules that event for the next patient in the resource queue — eliminating the manual if(success & queue_size > 0) new_event(...) pattern.
• release_all() is a new helper that fully purges the current patient from a list of resource_discrete objects (removes from using and from all queue entries) and optionally schedules per-resource resume events. Resume events are only triggered for resources where the patient was actually using (capacity freed).
• release_all_if_using() is a new helper that frees the current patient from a list of resource_discrete objects only if they are currently using them. Does not touch queue entries, preserving queue position for multi-resource workflows.
• resource_discrete() gains discipline ("FIFO" or "LIFO") and max_queue arguments. discipline controls ordering within the same priority level. max_queue caps the waiting list so that attempt_block() returns NA (patient rejected) when the queue is full, enabling M/M/c/k systems.
• resource_discrete() gains new statistics methods: queue_wait_time(), queue_wait_time_current(), had_to_queue(), time_in_use(), utilization(), n_using(), total_patients_blocked(), and total_patients_queued(). queue_wait_time_current() returns 0 at queue entry, grows with elapsed time at each subsequent event while waiting, and returns the total wait on acquisition — replacing the need to manually track time_start_queue. queue_wait_time() returns the final stored wait only after dequeue.
• resource_discrete() gains batch_seize() for seizing a resource for multiple patients in a single C++ call, and attempt_block() / attempt_free() now accept an amount argument for multi-unit acquisitions by a single patient.
• seize() is a new thin wrapper around resource$attempt_block() that reads i and curtime from the calling environment, returning TRUE (acquired), FALSE (queued), or NA (rejected).
• seize_all() is a new helper that atomically seizes a list of resources in C++ under "all_or_none" or "sequential" policy, enabling deadlock-free surgery-scheduling and philosophers-problem patterns. Fixed a bug where "all_or_none" could partially acquire resources when one resource had free capacity but patients already in its queue.
• shared_decr() and shared_incr() are new one-liner helpers for shared_input counters that increment or decrement by delta and return the new value.

WARDEN 2.0.3 (2026-05-27)

• add_item() now works correctly with the native pipe (|>). The .data argument has been moved to the first position (.data = NULL, ..., input), so the LHS of |> is naturally routed to .data without relying on magrittr's . symbol. Existing code using %>%, input=, or named ... arguments is unaffected (#TODO).
• input_block() and run_sim() now correctly handle multiple blocks spanning different simulation levels (e.g., common_all_inputs and common_pt_inputs): n_sensitivity is summed across all blocks, and binary-mode parameter offsets are injected automatically so each block activates at the right DSA iteration.
• input_block() is a new helper that builds a complete pick_val_v() expression from explicit base, psa, sens, and names_out arguments. The binary and dsa_indicators parameters have been renamed to indicator_sens_binary and sens_indicators respectively to align with pick_val_v(). The dsa_names argument now defaults to NULL, meaning all sensitivity_names are treated as scenarios (one iteration per name); supply dsa_names explicitly to designate which names are DSA directions. Setting an entry in sens_indicators to 0 now permanently excludes that parameter from variation in both DSA and scenario analyses, and the engine automatically deduces n_sensitivity from the number of active parameters/groups.
• pick_val_v() now correctly respects indicator_psa in grouped mode (indicator_sens_binary = FALSE). Previously, when sens_bool = TRUE and psa_bool = TRUE, all parameters drew from PSA regardless of indicator_psa; parameters with indicator_psa = 0 now correctly draw from base.

WARDEN 2.0.1 (2025-12-11)

*adj_val now accepts a vectorized_f argument to speed computations in the case of vectorized functions

WARDEN 2.0.0 (2025-10-13)

• Rcpp event based handler has been created using queues for high efficiency. In the new system, a unique event per patient is accepted in the queue, so small changes may be needed in codes where multiple equally named events were set to accommodate the new system. This is to ensure best practices and clarity regarding which events are modified, in the queue, etc.
• Simulation and engines have been redesigned and simplified to more cleanly handle and report errors, debug mode and continue on error functions.
• Constrained based engine has been created, which allows to run resource constrained DES, where resource and/or inputs can be shared across patients within an arm. This can be activated by using run_sim with constrained = TRUE.
• Rcpp based discrete resources with their own queuing system have been implemented (only interacted through R, working similar to R6 objects). This can be created by using resource_discrete()
• Shared inputs have been created to be used in constrained DES (works similar to R6 objects). This can be created by using shared_input().
• New vignette showcasing an example for constrained DES has been created.

WARDEN 1.3.5

• Accumulation backwards now uses active bindings to recognize which ongoing outputs are being modified.
• Deprecated modify_item_seq, modify_item as they are no longer needed to run even if backwards = TRUE. Removed from all examples.

WARDEN 1.3.4

• Unit tests for shared_inputs and run_sim_parallel added.
• add_item and add_item2 have now been integrated into add_item, with both behaviors being accepted. This means load_inputs has been overwritten with the load_inputs2 form.

WARDEN 1.3.0

• Rcpp versions of key functions have been implemented for speed improvements: disc_cycle_v, disc_ongoing_v, disc_instant_v, qcond_*, qtimecov, luck_adj

WARDEN 1.2.4 (2025-07-04)

• Added "adj_val" function, to reduce the need to call cycle events which can hinder model speed.
• Added "qtimecov" function, to predict time to events with time changing covariates, to reduce the need to call cycle events which can hinder model speed.

WARDEN 1.2.3

• Added model run unit tests
• Now the model can be run without specifying utilities, costs or other outputs (only LYs will be accounted for)
• Fixed an issue when timed_freq was active (now outputs are correct)

WARDEN 1.2.2 (2025-06-25)

• Per cycle outcomes now correctly work, and a maximum number of cycles argument has been implemented.

WARDEN 1.2.1 (2025-06-19)

• Discounting now correctly allocates to drq or drc depending on output. Other inputs use drc.

WARDEN 1.2.0 (2025-06-11)

• Speed gains from rewritten internal compute_outputs function.
• Implemented "random_stream" function (using method similar to R6) to facilitate careful handling of random numbers.

WARDEN 1.1.0

• Added "add_item2", which allows to incorporate expressions directly instead of a list, being faster and more consistent with "add_tte" and "add_react_evt".
• Engine now fully utilizes environment for slightly faster analysis when loading inputs.

WARDEN 1.0 (2025-05-28)

• Major update: now the engine uses environments instead of lists, which allows user to remove "modify_item" and "modify_item_seq" from their code, improving running speed by 20-40%
• Secondary changes to accommodate this update applied throughout (extract reactions, debug mode).
• Debug mode now uses abstract syntax tree to capture assignments, which can be limited in the presence of dynamic code assignments.
• sens_iterator function added to facilitate looping through DSA or multiple scenarios in a single model run. See the input selectors vignette in the website for an example.

WARDEN 0.99.3 (2025-04-04)

• Minor fix in run_sim_parallel to ensure compatibility with future package (had some "=T" instead of "=TRUE")
• Added BugsReport link in Description

WARDEN 0.99.2

• Added qgamma_mse function
• Added two articles for the website explaining more in detail WARDEN and the use of sobol sequences

WARDEN 0.99.1 (2024-12-13)

• CRAN feedback implemented, including changes in documentation:
• runif_stream function has been removed due to violation of CRAN policy of global environment modification. It is suggested that the user employs different methods (e.g., pre-drawing random numbers)
• Now the user is allowed to select the starting seed for the analysis

WARDEN 0.99

• Now debug and continue_on_error will work on all stages, not only for simulations
• CRAN preparation changes

WARDEN 0.98

• Repository is now public, Github Website has been set up
• Website references are now split by topic
• Added auxiliary functions to extract items and events from reactions to easily see interconections in models

WARDEN 0.97

• Update based on validation comments from Gabriel
• Modified conditional quantile functions weibull and llogistic to better match default R stats behavior
• Set License to be GPL >=3

WARDEN 0.96

• Update based on validation comments from Gabriel.
• Renamed conditional quantile functions for consistency.

WARDEN 0.95

• Seeds used by default have been changed to guarantee uniqueness
• Added possibility of continuing to the next simulation on error (if it occurs at the patient/arm level, not at the statics/structural loading level)
• Debug mode now exports log even if the simulation stops due to error. If combined with continue on error, it will continue to export log with the timestamp

WARDEN 0.94

• Added possibility of accumulating outputs continuously backward or forward using the accum_backward option in run_sim and run_sim_parallel

WARDEN 0.93

• Conditional quantile functions added and adjusted
• Luck adjustment function added with instructions for user

WARDEN 0.92

• Progress bar added for both parallel and standard computing model
• To use progress bar while in batch mode for a quarto document, make sure to add in the knitr options knitr: opts_chunk: R.options: progressr.enable: true

WARDEN 0.91

• Debug mode exports a txt file

WARDEN 0.9

• Warning, this commit will change previous results.
• Sensitivity-level and simulaton-level seeds have been moved outside the input loading loop as this caused correlation between inputs loaded at those stages.

WARDEN 0.5

• Initial set-up of news file
• Summary of inputs has been overhauled to provide INMB through a WTP argument
• Summary now can also be provided across analyses to quickly obtain DSA/scenario analysis results summarized