Merge pull request #67 from stitam/docker

Dockerfile, NAMESPACE, examples

.Rbuildignore

@@ -14,4 +14,5 @@ README.Rmd
 /cache/
 
 # scripts that are not part of the package
-scripts
+^docker$
+^scripts$

NAMESPACE

@@ -8,6 +8,7 @@ export(mgnify_instance)
 export(mgnify_list)
 export(ncbi_download_genome)
 export(ncbi_get_meta)
+export(ncbi_get_uid)
 export(ncbi_link_uid)
 export(ncbi_parse)
 export(ncbi_parse_assembly_xml)

R/ncbi_db_links.R

@@ -1,14 +1,12 @@
-#' Which NCBI databases can an database be linked to?
+#' Which NCBI databases can a database be linked to?
 #'
 #' This function is a wrapper around \code{rentrez::entrez_db_links()}. Results
 #' are cached and reused to validate input for \code{ncbi_link_uid()}.
 #' @param db character; the name of an NCBI database
 #' @return a character vector of NCBI database names the original database can
 #' be linked to.
 #' @examples
-#' \dontrun{
 #' ncbi_db_links("assembly")
-#' }
 #' @noRd
 ncbi_db_links <- function(db) {
   if (!dir.exists(tempdir())) dir.create(tempdir())

R/ncbi_get_uid.R

@@ -25,11 +25,10 @@
 #' However, if the search terms are very long, the function may fail with an
 #' error message. In this case, try reducing the \code{batch_size} value.
 #' @examples
-#' \dontrun{
 #' ncbi_get_uid("GCA_003012895.2", db = "assembly")
 #' ncbi_get_uid("Autographiviridae OR Podoviridae", db = "biosample")
 #' ncbi_get_uid(c("WP_093980916.1", "WP_181249115.1"), db = "protein")
-#' }
+#' @export
 ncbi_get_uid <- function(
     term,
     db,

R/ncbi_link_uid.R

@@ -42,10 +42,8 @@
 #' behaviour of the function with web histories is unreliable. The option is
 #' there but it is recommended NOT to use web histories with this function.
 #' @examples
-#' \dontrun{
 #' ncbi_link_uid("4253631", "assembly", "biosample")
 #' ncbi_link_uid(c("1226742659", "1883410844"), "protein", "nuccore")
-#' }
 #' @export
 ncbi_link_uid <- function(
     query,

docker/Dockerfile

@@ -0,0 +1,33 @@
+FROM debian:bookworm-20240211
+
+ARG DEBIAN_FRONTEND=noninteractive
+
+RUN apt-get update
+
+# Install system dependencies
+RUN apt-get install -y \
+  libcurl4-openssl-dev
+
+# Install R and R packages from apt
+RUN apt-get install -y \
+  r-base \
+  r-cran-curl \
+  r-cran-devtools \
+  r-cran-dplyr \
+  r-cran-httr \
+  r-cran-jsonlite \
+  r-cran-rentrez \
+  r-cran-rmarkdown \
+  r-cran-stringi \
+  r-cran-stringr \
+  r-cran-testthat \
+  r-cran-tibble \
+  r-cran-tidyr \
+  r-cran-xml \
+  r-cran-xml2
+
+# Install webseq
+RUN R -e 'options(warn = 2); devtools::install_github("stitam/webseq", upgrade = "never")'
+
+# Set R library path to avoid binding from host system
+ENV R_LIBS_USER /usr/local/lib/R/site-library

docker/run.sh

@@ -0,0 +1,16 @@
+account="stitam"
+container="webseq"
+version="0.1.0.dev"
+
+# build docker image
+sudo docker build -f Dockerfile -t ${account}/${container}:${version} .
+
+# OPTION 1 convert docker to singularity in one step
+# sudo singularity build ${account}-${container}-${version}.img docker-daemon://${account}/${container}:${version}
+
+# OPTION 2 export docker as tar and then convert to singularity
+sudo docker image save ${account}/${container}:${version} -o ${account}-${container}-${version}.tar
+sudo singularity build ${account}-${container}-${version}.img docker-archive:${account}-${container}-${version}.tar
+
+# push to Docker Hub
+# sudo docker push ${account}/${container}:${version}