Products Export
The task writes your products into a file, as CSV, XLS, XML or JSON. It is the route to a partner expecting an article list, and to a price comparison feed somebody collects regularly.
What gets transferred
Whatever you write into the plan, and nothing else. A file brings no fixed fields along: the columns come from the assignments, and a field nobody assigns does not appear in the file.
Which products end up in it you decide in two places inside the Configuration section: through the filter, and through the setting on whether only what has changed is written out.
The task's own configuration
Through the Filter By button you decide which records are eligible at all, with the same conditions you know from the product list: by manufacturer, category, property, price or stock.
Below it sit the settings for a running feed. Incremental writes out only what has changed since the last execution. Only with that tick do Only New Entries appear, narrowing it once more to records that were not included before, and Increment, the point counted from. So without these ticks you get the complete file.
The difference is one between two ways of running: a partner receiving the whole list every night needs the complete file. A system importing changes gets it once and the difference afterwards.
How the transfer plan is built
Eight sections, and the first one describes not the file being read this time but the one being written. In the others the file's columns stand as rows, and each takes its value from a field of the product.

| Section | What sits in it |
|---|---|
| Layout | The record sets of the output: which levels the file gets and which data format writes it |
| Common | The fields on the product itself |
| Descriptions | One record set per language version, with name, description and short text |
| Prices | One record set per price group |
| Stocks | One record set per store |
| Categories | The categories of the product |
| Media | The images, as an address |
| Properties | One tab per property |
As on every task of this channel you create the plan yourself, see Custom.
A record set in the Layout section is a level of the produced file here. For a flat CSV file one is enough: one row per product, and the assignments of the other sections become columns. A second one additionally carries Source and Predecessor and therefore sits below the first, which is what a nested XML or JSON file needs.
The task brings two further plans along as kinds of their own, and Insert offers them as soon as the first plan is there: Product Variants writes the variants of a product, Parent Product its parent article. That way the same article structure that was created on import can be written out again.
A value no field supplies comes from a script. On export that is the more frequent case than on import, because the other side has its own ideas about number formats, yes-no values and composite designations.
Special features and limits
Images go out as an address, not as a file. The row holds the reference the image can be reached at. The file itself is not deposited anywhere by this task.
The execution writes the file at the end, not row by row. Where no endpoint is set up for the task, TRADElube offers it for download afterwards. With an endpoint it goes to that endpoint's data source.
Further reading
- Custom: data formats, endpoints and the other six tasks
- Products Import (Custom): the other way round, from a file into TRADElube
- Products (reference): the fields the columns of the file come from
- Data sources (reference): where an endpoint puts the finished file
- Transfer plan (concepts): layout, record set, grouping and replication