Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Package development depending on RBioFormats: How can I reliably set java.parameters when importing functions from RBioFormats into my package? #38

Open
Gewerd-Strauss opened this issue Mar 12, 2024 · 0 comments

Comments

@Gewerd-Strauss
Copy link

Gewerd-Strauss commented Mar 12, 2024

Hello,

I am a bit at a loss.
I am developing a package that requires RBioFormats::read.image() to lead subsections of images. THus, I need to spin up a java VM prior to loading these functions, and given that the images are 4000x6000 pixels large the default heap size fits not suffice.

In normal scripts, you could ensure

G <- 8
options(java.parameters = paste0("-Xmx",G,"g"))

My thoughts were that using .onLoad() I could successfully apply -Xmx8g.

zzz.R:

..onLoad <- function(lib, pkg) {
    curr_memory <- getOption("java.parameters")
    options(dfl.java_parameters = curr_memory)
    GB <- 8
    if (is.null(curr_memory)) {
        options(java.parameters = paste0("-Xmx",GB,"g"))
        packageStartupMessage(paste0("1Set option <java.parameters> to '",getOption("java.parameters"),"'"))
    } else {
        matches <- regexec("(\\d+)(.*)", curr_memory)
        value <- as.integer(regmatches(curr_memory, matches)[[1]][[2]])
        unit <- regmatches(curr_memory, matches)[[1]][[3]]
        if (isFALSE(unit=="g")) { # we loaded less than a gigabyte - so we must stock up
            options(java.parameters = paste0("-Xmx",GB,"g"))
            packageStartupMessage(paste0("2Set option <java.parameters> to '",getOption("java.parameters"),"'"))
        } else {
            if (value<GB) {
                options(java.parameters = paste0("-Xmx",GB,"g"))
                packageStartupMessage(paste0("3Set option <java.parameters> to '",getOption("java.parameters"),"'"))
            } else {
                packageStartupMessage(paste0("4<java.parameters> is already set to '",getOption("java.parameters"),"'"))
            }
        }
    }
    invisible()
}
.onUnLoad <- function(lib, pkg) {
    ## reset Java parameters
    options(java.parameters = getOption("dfl.java_parameters"))
    options(dfl.java_parameters = NULL)
}

The NAMESPACE:

# Generated by roxygen2: do not edit by hand

export(load_image)
export(prep_loading)
importFrom(RBioFormats,coreMetadata)
importFrom(RBioFormats,read.image)
importFrom(RBioFormats,read.metadata)
importFrom(imager,RGBtoHSV)
importFrom(imager,as.cimg)
importFrom(imager,load.image)
importFrom(imager,sRGBtoRGB)
importFrom(stringr,str_c)

But this approach very reliably does nothing. Executing

RBioFormats::checkJavaMemory()
getOption("java.parameters")

after either calling library(my_package) or executing devtools::load_all(".") however still returns:

> RBioFormats::checkJavaMemory()
[1] 455.5
> getOption("java.parameters")
[1] "-Xmx8g

Functions which require RBioFormats access its functions via

[...]
#'
#' @return hsv-formatted pixel.array, unless HSV==false.
#' @export
#' [...]
#' @importFrom RBioFormats coreMetadata
#' @importFrom RBioFormats read.metadata
#' @importFrom RBioFormats read.image
#'
#' @example man/examples/load_image.R
load_image <- function(image_path, subset_only = FALSE, HSV = TRUE, crop_left=0, crop_right=0, crop_top=0, crop_bottom=0) {
[...]

I want my package to ensure that RBioFormats gets loaded with 8 GB.
How can I do so automatically? What am I not understanding here?


Thank you,
Sincerely,
~Gw

@Gewerd-Strauss Gewerd-Strauss changed the title Package development depending on RBioFormats: How can I reliably set java.parameters when importing functions from RBioFormats? Package development depending on RBioFormats: How can I reliably set java.parameters when importing functions from RBioFormats into my package? Mar 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant