CSV / XLSX File as a Source

Every operation has a hard-required field: GIP needs Parent SKU and SKU; GIS needs SKU; Get Shipments, Get FR Acknowledgements, and Get Invoices need Fulfillment Request Number; Send FR needs Order ID. Open the mapping template, confirm the required field maps to a valid column, and save before re-running. For header-based feeds, also confirm the header name in the file exactly matches the mapped column name.

Check that (1) the Delimiter setting matches the file's actual delimiter (a pipe-delimited file parsed as comma will break); (2) the decimal format matches your file (period vs. comma); (3) for XLSX, the Excel Sheet Index points to the correct sheet (0 = first); (4) Number of Rows to Skip is set if there are title/comment rows above the header; and (5) for Google Sheets, all numeric columns are formatted as Plain Text.

Check the Download Files Modified in the Last X Days setting: if it is set to 14 but the file was last modified 15 days ago, it is skipped. Increase the value or set it to 0 to always download. For HTTP/HTTPS sources, confirm the URL is not returning a cached copy; for FTP, confirm the file was actually updated on the server after the last run.

Use the File to Extract setting to name the exact file inside the archive (e.g. inventory.csv), matching case and extension exactly. Without it, Flxpoint does not know which file to read and the import fails. If the archive layout changes between deliveries, ask the vendor to keep a consistent filename.

Use a wildcard * in the Remote File Name Pattern setting to match every file that includes a dynamic or numeric value. For example, if the vendor delivers ABC080726.csv and ABC080826.csv, set the pattern to ABC*.csv and Flxpoint reads all matching files.

Set the ZIP file's name in the Remote File Name Pattern setting, then name the specific file to read from inside the archive in the File to Extract box under Advanced settings.

This often occurs when a SEMA CSV feed has an inconsistent number of columns per row due to a variable count of asset columns (images or videos). This causes column shifting where numeric fields like Weight or Length map to asset filenames instead of numbers. If the product save fails due to invalid numeric data and Archive Missing Products is enabled, Flxpoint may archive the SKU. To resolve this, request a feed from SEMA with consistent delimiters (such as double-quote string delimiters) or omit mapping the Weight, Height, Width, and Length fields in your template to prevent save errors.

Five: Manual File (upload a CSV/XLSX from your computer, one-time or ad hoc); HTTP/HTTPS (paste a direct file URL, optional Basic Auth); FTP/SFTP (host, port, username, and password or key); Email (Inbound) (an IMAP/POP server that retrieves file attachments from an inbox); and Google Sheets (a CSV export URL in the form https://docs.google.com/spreadsheets/d/YOUR_DOC_ID/export?format=csv). For protected sheets, the link must be accessible to Flxpoint.

File Has Headers (header-based vs. index-based mapping), File Format (CSV, TSV, XLSX), Delimiter (comma, pipe, tab, etc.), Number of Rows to Skip (skip title comments or blank rows at the top), Excel Sheet Index (which sheet tab in an XLSX file, 0-indexed), Download Files Modified in the Last X Days, Archive/Un-Archive Inventory, Save Products in Small Batch (reduces memory pressure on large files), and File to Extract (filename to use when the file arrives in a ZIP/RAR archive).

GIS variant fields: SKU (required), Quantity, Cost, MSRP, MAP, Source List Price, Dropship Fee, Shipping Cost, Weight, Weight Unit, Dimension Unit, Height, Length, Width, and Allow Backorders. GIS is built for frequent price/quantity updates from a vendor's separate PQS file: map SKU as the matching key plus whichever pricing and quantity fields the vendor includes, and keep product-level fields like title and description in Primary.

In header-based mode you map by column name (e.g. map the "Parent SKU" header to the Parent SKU field), and the "Process Feed Even If Some Headers Are Missing" setting applies. In index-based mode you map by column position (Column A = first, Column B = second, etc.); all columns must be present in the correct order regardless of settings, so removing or reordering a column breaks the mapping. Use header-based for any file that includes a header row, and index-based only for headerless files with a guaranteed-stable column order.

It reads a vendor-provided acknowledgement file and maps Fulfillment Request Number (required: links the ack to its FR), Acknowledgment Status (e.g. Accepted, Rejected), Acknowledgment Date, and Note at the header level, plus SKU, Quantity, and Item Status at the line-item level. Use it when your vendor returns order acceptance/rejection as a file rather than in real time via API; sample acknowledgement files are available in the setup for format reference.