Local source-only mirrors of CRAN and Bioconductor R package repositories

README.md

R Package Repositories - Local Mirrors

by Henrik Bengtsson (2017-11-09 -- 2017-11-17)

Usage

make ## same as 'make sync-all'
make debug
make summary
make sync-all
make sync-cran
make sync-bioconductor

CRAN

Calling

$ make sync-cran

will create/update a local CRAN package repository under cran/ (in the current directory) with the following structure:

$ tree -d cran/
cran/
└── src
    └── contrib
        ├── PACKAGES
        ├── PACKAGES.gz
        ├── PACKAGES.in
        ├── PACKAGES.rds
        ├── A3_1.0.0.tar.gz
        ├── abbyyR_0.5.1.tar.gz
		:
        └── zyp_0.10-1.tar.gz

Bioconductor

Calling

$ make sync-bioconductor

will create/update three local Bioconductor package repositories under bioconductor/$BIOC_VERSION/ (in the current directory), where $BIOC_VERSION is the Bioconductor version (as automatically queried from Bioconductor):

• BioCsoft: Bioconductor Software packages (bioconductor/$BIOC_VERSION/bioc)
• BioCann: Bioconductor Annotation Data packages (bioconductor/$BIOC_VERSION/data/annotation)
• BioCexp: Bioconductor Experimental Data packages (bioconductor/$BIOC_VERSION/data/experiment)

Example of tree structure:

$ tree -d bioconductor/$BIOC_VERSION/bioc
cran/
└── src
    └── contrib
        ├── PACKAGES
        ├── PACKAGES.gz
        ├── PACKAGES.in
        ├── PACKAGES.rds
        ├── a4_1.26.0.tar.gz
        ├── a4Base_1.26.0.tar.gz
		:
        └── zlibbioc_1.24.0.tar.gz

Statistics

As of 2017-11-09, the above package repositories contains:

• CRAN: 11,795 packages, 5.8 GiB disk space
• BioCsoft: 1,476 packages, 4.2 GiB disk space
• BioCann: 910 packages, 61 GiB disk space
• BioCexp: 324 packages, 31 GiB disk space

Makefile

## Query Bioconductor for the current release version
BIOC_VERSION := $(shell curl --silent https://www.bioconductor.org/config.yaml | grep -F "release_version:" | sed -E 's/.*release_version:[ ]*"([0-9.]+)".*/\1/g')

OPTS = --dry-run
OPTS =

all: sync-all

debug:
	@echo BIOC_VERSION=$(BIOC_VERSION)
	du -c -h .

summary:
	@echo "CRAN: $(shell find cran -type f -name *.tar.gz | wc -l) packages ($(shell du -s -h cran | cut -f 1))"
	@echo "BioCsoft $(BIOC_VERSION): $(shell find bioconductor/$(BIOC_VERSION)/bioc -type f -name *.tar.gz | wc -l) packages ($(shell du -s -h bioconductor/$(BIOC_VERSION)/bioc | cut -f 1))"
	@echo "BioCann $(BIOC_VERSION): $(shell find bioconductor/$(BIOC_VERSION)/data/annotation -type f -name *.tar.gz | wc -l) packages ($(shell du -s -h bioconductor/$(BIOC_VERSION)/data/annotation | cut -f 1))"
	@echo "BioCexp $(BIOC_VERSION): $(shell find bioconductor/$(BIOC_VERSION)/data/experiment -type f -name *.tar.gz | wc -l) packages ($(shell du -s -h bioconductor/$(BIOC_VERSION)/data/experiment | cut -f 1))"

sync-all: sync-cran sync-bioconductor

sync-cran:
	mkdir -p cran/; \
	rsync --verbose --human-readable \
	  $(OPTS) \
	  --times \
	  --recursive \
	  --delete \
	  --include 'src/' --include 'src/contrib/' \
          --include 'src/contrib/*.tar.gz' \
          --include 'src/contrib/PACKAGES*' \
	  --exclude '*' \
	  cran.r-project.org::CRAN \
	  cran/

sync-bioconductor:
	mkdir -p bioconductor/$(BIOC_VERSION)/; \
	rsync --verbose --human-readable \
	  $(OPTS) \
	  --times \
	  --recursive \
	  --delete \
	  --include 'src/' \
	  --include 'src/contrib/' \
          --include 'src/contrib/*.tar.gz' \
          --include 'src/contrib/PACKAGES*' \
	  --include 'bioc/' \
	  --include 'data/' \
	  --include 'data/annotation/' \
	  --include 'data/experiment/' \
	  --exclude '*' \
	  master.bioconductor.org::release \
	  bioconductor/$(BIOC_VERSION)/