Exporting content as CSV, JSON or XML #

This page is for administrators and editors whose data already lives in Exponential as content: a list of service points, a register of decisions, a directory of facilities. Instead of keeping a separate spreadsheet, export the content itself as a dataset file. The tools are the xrowextract extension (xrowextract) and the kernel command exp:ezcsvexport.

Contents #

1. Which tool
2. The Export tab of xrowextract
3. From the command line
4. The kernel command exp:ezcsvexport
5. Publishing the exported file
6. Before you publish an export

Which tool #

xrowextract is installed with Exponential but must be active on your site (ActiveExtensions[]); ask your administrator. Its pages need the extension's policies (xrowextract policies).

The Export tab of xrowextract #

1. Open Export in the top menu (/xrowextract/csv), then One class of content export.
2. Pick the class (for example your service_point class) and the start node.
3. Choose the format: CSV, JSON or XML. Choose the languages if the content is translated: one row per translation.
4. Add filters if needed (date, section, state, visibility, attribute values).
5. Press Preview to see the first rows as a spreadsheet will show them.
6. Download the file, or save the settings as a preset to repeat the export later.

Large exports can run in the background and be downloaded from the Jobs tab when they are done.

From the command line #

# one class below a node, as CSV
php extension/xrowextract/bin/php/csv.php --class=<class identifier> --node=<node id> --output=var/export/service-points.csv

# the same as JSON, visible content only
php extension/xrowextract/bin/php/csv.php --class=<class identifier> --node=<node id> --format=json \
  --visibility=visible --output=var/export/service-points.json

# which columns a class offers
php extension/xrowextract/bin/php/csv.php --list-columns --class=<class identifier>

Schedules (ext:xrowextract:schedule) run exports regularly and can deliver the file to SFTP, FTP/FTPS, a NAS folder, S3, WebDAV or an HTTP endpoint.

The kernel command exp:ezcsvexport #

Without the extension, the kernel exports the subtree below a node to one CSV file per content class:

./console exp:ezcsvexport --storage-dir=var/export/csv <node id>
php bin/php/ezcsvexport.php --storage-dir=var/export/csv <node id>

It has no filters and no column choice; review the columns before publishing.

Publishing the exported file #

The export is a file on the server or in your downloads. To publish it as open data, upload it as a distribution of the dataset (see "Publishing dataset files"), with the date of the export in the dataset's "Last changed" field. For a regular export, replace the file in the same distribution each time so the page address stays the same.

Before you publish an export #

• To do: Remove columns with personal data: owner, creator, modifier, e-mail addresses, internal notes. Exports can include such columns, and user accounts can be exported as special columns by users with the right policy.
• To do: Remove internal identifiers re-users do not need, or keep them deliberately as stable keys.
• To do: Check the character encoding (UTF-8) and the delimiter in a spreadsheet program and a text editor.
• To do: Check one row against the published page.
• To do: Record which export (preset or command) made the file, so the next update is the same.

Related pages #

• xrowextract
• Collected information
• exp:ezcsvexport
• Free documentation: this page is free to copy, translate and improve under the GNU FDL 1.3.