73 lines
2.8 KiB
YAML
73 lines
2.8 KiB
YAML
version: '3'
|
|
|
|
tasks:
|
|
main:
|
|
desc: Powerhouse Museum Tutorial
|
|
vars:
|
|
DIR: '{{splitList ":" .TASK | first}}' # results in the task namespace, which is identical to the directory name
|
|
cmds:
|
|
- task: refine
|
|
- task: :check # check OpenRefine log for any warnings and exit on error
|
|
vars: {DIR: '{{.DIR}}'}
|
|
|
|
refine:
|
|
dir: ./{{.DIR}}
|
|
vars:
|
|
DIR: '{{splitList ":" .TASK | first}}'
|
|
PROJECT: phm
|
|
PORT: 3336 # assign a different port for each project
|
|
RAM: 2048M # maximum RAM for OpenRefine java heap space
|
|
LOG: '>(tee -a "{{.PROJECT}}.log") 2>&1' # be careful when making changes here, as the path to the log file should match the server log (see main task "start")
|
|
deps:
|
|
- task: download # will be executed each run independent of up-to-date check
|
|
cmds:
|
|
- task: :start # launch OpenRefine
|
|
vars: {DIR: '{{.DIR}}', PROJECT: '{{.PROJECT}}', PORT: '{{.PORT}}', RAM: '{{.RAM}}'}
|
|
- > # import file
|
|
"$CLIENT" -P {{.PORT}}
|
|
--create "$(readlink -m input/phm-collection.tsv)"
|
|
--processQuotes false
|
|
--guessCellValueTypes true
|
|
--projectName "{{.PROJECT}}"
|
|
> {{.LOG}}
|
|
- > # apply transformation rules
|
|
"$CLIENT" -P {{.PORT}} "{{.PROJECT}}"
|
|
--apply config/phm-transform.json
|
|
> {{.LOG}}
|
|
- mkdir -p output
|
|
- > # export to file
|
|
"$CLIENT" -P {{.PORT}} "{{.PROJECT}}"
|
|
--output "$(readlink -m output/phm-results.tsv)"
|
|
> {{.LOG}}
|
|
- | # print allocated system resources
|
|
PID="$(lsof -t -i:{{.PORT}})"
|
|
echo "used $(($(ps --no-headers -o rss -p "$PID") / 1024)) MB RAM" > {{.LOG}}
|
|
echo "used $(ps --no-headers -o cputime -p "$PID") CPU time" > {{.LOG}}
|
|
- task: :stop # shut down OpenRefine and archive the OpenRefine project
|
|
vars: {DIR: '{{.DIR}}', PORT: '{{.PORT}}', PROJECT: '{{.PROJECT}}'}
|
|
sources:
|
|
- Taskfile.yml
|
|
- input/**
|
|
- config/**
|
|
generates:
|
|
- ./{{.PROJECT}}.openrefine.tar.gz
|
|
- output/**
|
|
ignore_error: true # workaround to avoid an orphaned Java process on error https://github.com/go-task/task/issues/141
|
|
|
|
download:
|
|
dir: ./{{.DIR}}
|
|
vars:
|
|
DIR: '{{splitList ":" .TASK | first}}'
|
|
cmds:
|
|
- mkdir -p input config
|
|
- > # Download input
|
|
wget --no-verbose -O input/phm-collection.tsv
|
|
https://github.com/opencultureconsulting/openrefine-batch/raw/master/examples/powerhouse-museum/input/phm-collection.tsv
|
|
- > # Download config
|
|
wget --no-verbose -O config/phm-transform.json
|
|
https://github.com/opencultureconsulting/openrefine-batch/raw/master/examples/powerhouse-museum/config/phm-transform.json
|
|
|
|
default: # enable standalone execution (running `task` in project directory)
|
|
cmds:
|
|
- DIR="${PWD##*/}:main" && cd .. && task "$DIR"
|