Files
kemal b63852e775 SimurgArray: master plan, repo scaffolding and data/ref fetch scripts
Phase 0 deliverables: full technical specification and roadmap
(SIMURGARRAY_MASTER_PLAN.md), bootstrap script for reference
open-source repos and tiered dataset fetch script with checksums.
2026-09-27 15:20:06 +01:00

25 lines
973 B
Bash
Executable File

#!/usr/bin/env bash
# Shallow-clone reference open-source repositories into refs/ (git-ignored).
# GPL repos are for use as external oracles only — do not copy their code into SimurgArray.
set -euo pipefail
cd "$(dirname "$0")/.."
mkdir -p refs
repos=(
"Illumina/BeadArrayFiles" # BSD-2
"Illumina/GTCtoVCF" # Apache-2.0
"freeseek/gtc2vcf" # MIT
"freeseek/mocha" # MIT
"broadinstitute/picard" # MIT (see src/main/java/picard/arrays)
"WGLab/PennCNV" # Unlicense
"sbslee/pypgx" # MIT
"PharmGKB/PharmCAT" # MPL-2.0 (oracle)
"HenrikBengtsson/illuminaio" # GPL-2 (oracle only)
"gjun/verifyIDintensity" # GPL-3 (oracle only)
)
for r in "${repos[@]}"; do
d="refs/${r#*/}"
if [[ -d "$d/.git" ]]; then echo "skip $d"; continue; fi
git clone --depth 1 "https://github.com/$r.git" "$d"
done
echo "done: $(ls refs | wc -l) repos in refs/"