Feat/check up to date #338
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,146 @@ | ||
| #' Check model and data up to date with outputs | ||
| #' | ||
| #' Functions for checking that the model outputs on disk match the md5 hashes | ||
| #' stored in `bbi_config.json` at run time. In other words, checking that the | ||
| #' model and data files have _not_ changed since the model was last run. See | ||
| #' Details section for what specific files are checked. | ||
| #' | ||
| #' @details | ||
| #' Different files are checked depending on what type of model is being checked. | ||
| #' | ||
| #' **For NONMEM models** | ||
| #' | ||
| #' * The model file (control stream) | ||
| #' | ||
| #' * The data file (referenced in `$DATA` within the control stream) | ||
| #' | ||
| #' **Currently only NONMEM implemented.** | ||
| #' | ||
| #' @param .bbi_object the object to check. Could be | ||
| #' a `bbi_{.model_type}_model` object, | ||
| #' a `bbi_{.model_type}_summary` object, | ||
| #' or a `bbi_log_df` tibble. | ||
| #' @param ... Arguments passed through (currently none). | ||
| #' | ||
| #' @return | ||
| #' | ||
| #' **`bbi_model`** method invisibly returns a logical vector of length 2. The | ||
| #' first element (named `"model"`) refers to the model files mentioned above. | ||
| #' The second element (named `"data"`) refers to the data files mentioned above. | ||
| #' For both elements, they will be `TRUE` if nothing has changed, `FALSE` if | ||
| #' anything has changed. Note: _if no file exists_ at the specified path, | ||
| #' `FALSE` will be returned because that is technically a "change." The file | ||
| #' used to exist and now it does not. | ||
| #' | ||
| #' **`bbi_log_df`** method invisibly returns a named list of lists, with one | ||
| #' element for each row in the input tibble, with the name corresponding to the | ||
| #' value in the `run` column for that row. Each element of the list will contain | ||
| #' the two-element list returned from the `bbi_model` method (described above) | ||
| #' for the relevant model. | ||
| #' | ||
| #' There is no `add_up_to_date()` function because **if you would like to add | ||
| #' these columns to a `bbi_log_df` tibble** you can use [add_config()], which | ||
| #' contains `model_has_changed` and `data_has_changed` columns. Please note: | ||
| #' these contain the opposite boolean values (`check_up_to_date()` returns | ||
| #' `TRUE` if up to date, `*_has_changed` returns `TRUE` if _changed_). | ||
| #' | ||
| #' **The returned value is invisible because a message is printed** alerting the | ||
| #' user of the specific files that have changed, if any. This facilitates | ||
| #' calling the function for this side effect without explicitly handling the | ||
| #' returned value. | ||
| #' | ||
| #' @export | ||
| check_up_to_date <- function(.bbi_object, ...) { | ||
| UseMethod("check_up_to_date") | ||
| } | ||
|
|
||
| #' @export | ||
| check_up_to_date.bbi_nonmem_model <- function(.bbi_object, ...) { | ||
| check_up_to_date_nonmem(.bbi_object) | ||
| } | ||
|
|
||
| #' @export | ||
| check_up_to_date.bbi_nonmem_summary <- function(.bbi_object, ...) { | ||
| check_up_to_date_nonmem(.bbi_object) | ||
| } | ||
|
|
||
| #' @export | ||
| check_up_to_date.bbi_log_df <- function(.bbi_object, ...) { | ||
|
|
||
| check_list <- map(.bbi_object[[ABS_MOD_PATH]], function(.p) { | ||
| tryCatch( | ||
| check_up_to_date(read_model(.p)), | ||
| error = function(.e) { | ||
| .error_msg <- paste(as.character(.e$message), collapse = " -- ") | ||
| if (grepl(CHECK_UP_TO_DATE_ERR_MSG, .error_msg, fixed = TRUE)) { | ||
| message(.error_msg) | ||
| return(as.logical(c(model = NA, data = NA))) | ||
| } else { | ||
| stop(.e) | ||
| } | ||
| } | ||
| ) | ||
| }) | ||
|
|
||
| names(check_list) <- .bbi_object[[RUN_ID_COL]] | ||
| return(invisible(check_list)) | ||
| } | ||
|
|
||
| #################################### | ||
| # PRIVATE implementation functions | ||
| #################################### | ||
|
|
||
|
|
||
| #' Private implementation to check that NONMEM model is up-to-date | ||
| #' | ||
| #' Specifically, check that control stream and data file on disk have not | ||
| #' changed since the model was run. This is accomplished by taking their md5 | ||
| #' hashes and comparing it to the hashes stored in `bbi_config.json`. | ||
| #' | ||
| #' @importFrom jsonlite fromJSON | ||
| #' @importFrom fs file_exists | ||
| #' @importFrom tidyr replace_na | ||
| #' | ||
| #' @inheritParams check_up_to_date | ||
| #' | ||
| #' @keywords internal | ||
| check_up_to_date_nonmem <- function(.mod) { | ||
| config_path <- file.path(get_output_dir(.mod, .check_exists = FALSE), "bbi_config.json") | ||
| if (!fs::file_exists(config_path)) { | ||
| stop(paste(glue("Model {get_model_id(.mod)}:"), CHECK_UP_TO_DATE_ERR_MSG)) | ||
| } | ||
| config <- jsonlite::fromJSON(config_path) | ||
|
|
||
| # check necessary files for changes | ||
| model_file <- get_model_path(.mod) | ||
| data_file <- get_data_path(.mod) | ||
|
|
||
| changed_files <- c( | ||
| config[[CONFIG_MODEL_MD5]] != tools::md5sum(model_file), | ||
| config[[CONFIG_DATA_MD5]] != tools::md5sum(data_file) | ||
| ) | ||
|
|
||
| any_changes <- any(changed_files) | ||
| if(isTRUE(any_changes)) { | ||
| message(paste( | ||
| glue("The following files have changed in {get_model_id(.mod)}"), | ||
| paste("*", names(which(changed_files)), collapse = "\n"), | ||
|
There was a problem hiding this comment. maybe this is just a me thing but it's kinda weird to mix paste works so this isn't wrong I just don't see why we are using it if we have already imported the glue package There was a problem hiding this comment. yea, I see that. I always just use paste unless I explicitly need the |
||
| sep = "\n" | ||
| )) | ||
| } | ||
|
|
||
| na_files <- is.na(changed_files) | ||
| if(isTRUE(any(na_files))) { | ||
| message(paste( | ||
| glue("The following files in {get_model_id(.mod)} ARE NO LONGER PRESENT"), | ||
| paste("*", names(changed_files[na_files]), collapse = "\n"), | ||
| sep = "\n" | ||
| )) | ||
| } | ||
|
|
||
| # build return value | ||
| res <- replace_na(!changed_files, FALSE) | ||
| names(res) <- c("model", "data") | ||
|
|
||
| return(invisible(res)) | ||
| } | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Right now, the bbi_nonmem_summary and bbi_nonmem_model methods are identical so maybe consider just setting this as the default method.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Part of the point of this is prepare for adding Stan stuff so, even though these are the same, they're not gonna be the default methods. I considered adding another parent class that they would share, but that seemed a little excessive to have something like
(
"bbi_model"is already the parent class that they'll share with Stan and other kinds of models)Anyway, that would be a totally valid way to do it, but I opted for just duplicating a couple dispatches instead (and made them use the same private helper to avoid copy/paste).