-
Notifications
You must be signed in to change notification settings - Fork 36
A quick and easy conversion
- Installing csv2rdf4lod automation.
- It would help to be familiar with the material in Conversion process phase: name and Conversion process phase: retrieve, but you can charge ahead and come back to it if things get hairy.
Let's say we are interested in some oil well data shown at ScraperWiki, which offers a URL for the CSV. As data curators, we'll need to choose identifiers for our source, dataset, and version, so we choose scraperwiki-com, uk-offshore-oil-wells, and 2011-Jan-24, respectively. Knowing these values, we can make the directory:
bash-3.2$ mkdir ~/Desktop/source # Creates the directory for all data that you collect and convert.
bash-3.2$ cd ~/Desktop/source
bash-3.2$ mkdir -p scraperwiki-com/uk-offshore-oil-wells/version/2011-Jan-24
Next, we'll hop into our conversion cockpit and set up shop:
~/Desktop/source
bash-3.2$ cd scraperwiki-com/uk-offshore-oil-wells/version/2011-Jan-24
bash-3.2$ mkdir source
bash-3.2$ mkdir manual
Hop into source/ and grab the data (with pcurl.sh!) and get back into our conversion cockpit:
~/Desktop/source/scraperwiki-com/uk-offshore-oil-wells/version/2011-Jan-24
bash-3.2$ cd source/
bash-3.2$ pcurl.sh http://purl.org/twc/query/scraperwiki/uk-offshore-oil-wells -e csv
bash-3.2$ cd ..
(If scraperwiki doesn't return the CSV, you can point pcurl.sh to our cached copy of uk-offshore-oil-wells.csv instead.)
[Make the conversion trigger](Conversion process phase: create conversion trigger), [pull it](Conversion process phase: pull conversion trigger) (if you hit a memory error, see FAQ), and see [what it did](A guided tour of csv2rdf4lod's Turtle dump file):
~/Desktop/source/scraperwiki-com/uk-offshore-oil-wells/version/2011-Jan-24
bash-3.2$ cr-create-convert-sh.sh -w source/uk-offshore-oil-wells.csv
bash-3.2$ ./convert-uk-offshore-oil-wells.sh
bash-3.2$ vi automatic/uk-offshore-oil-wells.csv.raw.ttl
There's your RDF --------^^ as a verbatim interpretation of the tabular literals (if you hit a memory error, see FAQ). But having an enhanced version is better! Start by reviewing the Conversion process phases, one of which will show you how to make an enhancement to add to your initial conversion. conversion:Enhancement shows a good set of things you can tell the converter to make nicely-structured RDF from a relatively uninformative bucket of literals. But if you want to cheat, grab my enhancement parameters for this dataset and plop them into the manual/ directory of your conversion cockpit; then run ./convert-uk-offshore-oil-wells.sh again (it'll realize that you already ran the raw conversion and move on with enhancing).
- Step through the Conversion process phases
- Real world examples