Hi Colin. CSVW does support multiple values per column, either ordered or unordered using the “separator” property [1]. For example see examples 11-13 in the data-model document [2]. You can also find more examples here [3]. Look at tree-ops-ext.csv and tree-ops-ext.csv-metadata.json Gregg Kellogg gregg@greggkellogg.net [1] http://w3c.github.io/csvw/metadata/#cell-separator [2] http://w3c.github.io/csvw/syntax/#parsing-examples [3] https://github.com/w3c/csvw/tree/gh-pages/examples > On Oct 8, 2015, at 3:26 AM, Colin Maudry <colin@maudry.com> wrote: > > Hello, > > I work with the French government on a project that aims at generating metadata for the tabular data published on http://data.gouv.fr <http://data.gouv.fr/>. CSVW seems to be the right vocabulary for this purpose. > > It's quite common in data.gouv.fr data to see columns that contain lists of values separated by a different delimiter, or quoted. > > color,size,cities > red,12,London;Madrid;San Diego > > OR > > red,12,"London,Madrid,San Diego" > > In the CSVW vocabulary, I spotted a csvw:ordered boolean property that defines whether the order of the values in a column should be preserved, which makes me think CSVW supports this use case, somehow. > > Are there examples of CSVW schemas that cover this situation and define: > > - which column potentially contains lists of values > - what delimiter is used to separate the items of the lists > > Thanks, and good job with the spec! > > Colin Maudry > http://colin.maudry.com <http://colin.maudry.com/>
Received on Thursday, 8 October 2015 20:30:39 UTC