Handy scripts for subject set management

download_subject_set.sh

#!/bin/bash

set -e

# Downloads a subject set and writes it to a text file.
#
# Takes a subject set ID as an argument.
#
# Example usage:
# $ ./download_subject_set.sh 1234

SUBJECT_SET_ID=$1
OUTPUT_FILE="subject_set_${SUBJECT_SET_ID}.txt"

echo "Fetching subjects from set \`$SUBJECT_SET_ID\`"
IDS="$(panoptes subject ls -q -s $1)"

echo "Formatting"
FORMATTED_IDS="${IDS// /$'\n'}"

echo "Writing subjects to \`$OUTPUT_FILE\`"
cat <<< "$FORMATTED_IDS" > "$OUTPUT_FILE"

split_file_by_lines.sh

#!/bin/bash

set -e

# Splits a file into smaller files, of n lines each.
#
# Takes a source file and number of lines per file as arguments.
#
# Example usage:
# $ ./split_file_by_lines.sh example.txt 1000

FILE=$1
NUMBER_OF_LINES=$2

split -l $NUMBER_OF_LINES $FILE

upload_sets.sh

#!/bin/bash

set -e

# Uploads a directory of `.txt` files as new subject sets.
#
# Takes a project ID as an argument.
#
# Text files should consist of one subject ID per line. Subject set names are
# taken from the text file's filename.
#
# Example usage:
# $ ./upload_sets.sh 1234

FILES=*.txt
PROJECT_ID=$1

for f in $FILES
do
  # Get the filename without the extension
  NEW_SET_NAME="${f%.*}"

  echo "Processing file \`$f\`..."

  echo "Creating new set \`$NEW_SET_NAME\`"
  create_set_output=$(panoptes subject-set create $PROJECT_ID $NEW_SET_NAME)

  # Get the new set ID from the panoptes output
  NEW_SET_ID="$( cut -d ' ' -f 1 <<< $create_set_output )"

  echo "Uploading subjects from $f to set #$NEW_SET_ID"
  panoptes subject-set add-subjects -f $f $NEW_SET_ID

  echo "Done!"
done