Elbuz fetches the supplier's price list itself — by link, from a mailbox, from FTP or via API. No more manual uploads after every update.
When you need this:
- The supplier updates prices daily — every import, manual or scheduled, takes the fresh file.
- Suppliers deliver differently — each source has its own method: link, email, FTP or integration.
- No new file arrived — products update from the previous one; the log says if it's overdue.
- The method was set up wrong — the next import removes the wrong file's rows.
Where to start: ask the supplier how they deliver the price list, then open your method's article.
What automatic receiving can do
The supplier decides which method is yours: you collect the price list the way they deliver it. Each method has its own article.
- The supplier gives a link to the fileDirect address, cloud storage or a Google Sheet; date substitution, access key, sign-in to the supplier's account
- The price list arrives by emailMailbox with password and an email selection condition; mailbox connection wizard for 15 providers
- The file sits on an FTP serverServer address, port, login, password and the full path to the file; connection check
- The supplier is on the list of ready-made integrationsTwenty suppliers whose API we support ourselves: columns are mapped in advance, you only provide a login and password
What all methods have in common:
- One method per source — the card shows whose file will be used and warns you if several are set up.
- No new file — the previous one is used, and the program itself tells “not sent yet” from overdue.
- A new file fully replaces the price list rows — the next import fixes a method mistake.
What to get from the supplier
| Method | What to get from the supplier |
|---|---|
| File | the file itself: xls, xlsx, xlsm, xml, yml, csv, txt, tsv, json, a BMEcat catalog, a DATANORM price list or an archive. The program doesn't read PDF: ask the supplier for Excel or CSV |
| URL | the file address; a login and password if it's behind HTTP authentication; an access key if a request header is required |
| the mailbox the supplier sends the price list to and its password; the sender address or the email subject; plus a sample file | |
| FTP | server address, port, user name, password, full path to the file; plus a sample file |
| API | the login and password (token) issued by the supplier; sometimes also the of a pickup point or warehouse |
A price list in a zip archive can be connected by any method — file, link, email, FTP. The program unpacks the archive itself during import; you don't need to extract it manually. This is common with email: suppliers zip a heavy Excel file to fit the attachment size limit.
The sample file for email and FTP isn't a formality: the wizard maps the columns from it. The wizard doesn't ask for mail access or FTP passwords at all — you enter them later in the card. How this works is explained in How to complete the source connection wizard.
For a ready-made API integration, the supplier often also needs our IP address — otherwise they won't let the request through. You don't have to look for it: it's shown right in the access parameters window, in the line “IP address of the Elbuz Jumper server to give to the supplier for access:”.
Why you set up one method, not several
The source type isn't stored as a separate field anywhere. The program works out the method from which fields are filled in.
The import doesn't choose anything. It runs all configured methods in a row, and each next one overwrites the file already downloaded:
| Order | Runs when filled in | Log line |
|---|---|---|
| 1 | file link | “Getting file by link” |
| 2 | mailbox and email selection condition | “Getting file by email” |
| 3 | FTP server | “Getting file via FTP”, and if the file path is empty — next line “No file name specified in FTP settings!” |
| 4 | linked integration and API server address | “Getting file via API” |
The card reads the same fields under the same conditions and highlights the tile of the method whose file will actually be used — the last configured one. If a ready-made integration is linked, it shows “API” and doesn't draw the other tiles at all.
Set up two methods, and you get the file of the one that ran last: API overrides FTP, FTP overrides email, email overrides the link. Two methods don't give you a backup, so fill in the fields of one. With several methods, a warning listing them and naming the winner appears under the tiles: “Several delivery methods are configured: …. On import they run one after another and each overwrites the price file, so the file of the last one is what gets imported — “…”.”
The run log confirms which method actually worked — with the line “Getting file…”.
To drop a method, clear its fields. This doesn't remove a password: an empty password field doesn't erase the saved one — not for the link, FTP or API. This is deliberate, so the password doesn't vanish on every accidental form save, but you need to know it.
1
Current method (URL)2
E-mail3
FTP- The highlighted tile is the method currently used by this source.
- E-mail — the supplier sends the price list by email.
- FTP — the price list is downloaded from the supplier's FTP server.
What happens when a new file doesn't arrive
A failed download doesn't stop the run. The reason doesn't matter: the server didn't respond, the link returned 404, FTP refused, a private Google file wasn't shared, the mailbox has no emails. The program logs the reason, goes through the rest of the methods and then acts as if there's simply no new file.
And with no new file, the old one is used, and products still get updated from yesterday's data. Whether that counts as an error depends on the reason:
- Failure on the supplier's side — the site returned an error code, showed a Cloudflare check or a page instead of the file, returned an empty file, or the mailbox refused the program. The reason appears as a red line, and the run is marked as an error. These lines are explained in the article about the import log.
- The supplier just hasn't sent an update yet — there's no new email in the mailbox. The log shows the warning “No new file received from the supplier — products from the previous file are in use. This is not an error if the supplier has not sent an update yet.”, followed by the note “Last new price list: ‹date› (N days ago).” The run ends with the “Done” status.
- No new file for longer than usual — then the line “No new price list for N days (last one: ‹date›)…” and the “Error” status. What counts as “usual” is set by “How often the supplier sends a new price list”.
The run stops in only one case — when there's no file at all, neither fresh nor previous. Then the last log line is “No file found to import: ‹name›”.
- Action or result
- Decision: the path depends on the answer
So after setting up a new method, run the first import manually and find the “Getting file…” line in the log with the file name after it. If instead there's a red line with the reason or “No new file received from the supplier”, the method didn't bring a file, even though the run finished. All messages and what to do about them are in the table below.
How often the supplier sends a new price list
To tell “the supplier hasn't sent it yet” from “the supplier has gone quiet”, the program needs to know how often a new file arrives. That's the “How often the supplier sends a new price list” field in the source card, “Schedule” section, below the auto-import settings button. Options: “Detect automatically” (default), “Every day”, “Once a week”, “Every two weeks”, “Once a month”, “Do not monitor”. The value is saved as soon as you pick it. Site parser sources don't have this field.
1
How often the supplier sends a new price list2
Hint about the expected new file- Select how often the supplier sends a new price list; the default is automatic detection.
- The hint shows how often the program expects a new file and when the last new one arrived.
“Detect automatically” learns from the source's own history. The program takes up to eleven recent runs that brought a genuinely new file and uses the median interval between them — a typical gap that one random long pause doesn't skew. Until there are three such runs, it expects a new file at least every 7 days. A run where the file didn't arrive or was byte-for-byte the same isn't added to the history.
Set the frequency manually when the history misleads the program. Say the supplier sends a price list once a month, but automatic detection doesn't have three new files yet and starts flagging overdue after 11 days. Or they send irregularly — twice a week, then once in three weeks — and the median comes out shorter than the real pauses. “Do not monitor” turns the check off: the log keeps only the no-new-file warning, without the note or overdue.
A file becomes overdue when more days have passed since the last new file than the larger of two numbers: the interval plus one day, or one and a half intervals.
| Frequency | The run becomes an error if there's no new file for |
|---|---|
| Every day | more than 2 days |
| Once a week, and while history is short | more than 11 days |
| Every two weeks | more than 21 days |
| Once a month | more than 45 days |
Below the field, the program shows what it expects now: “The program currently expects a new file every N days.”, “A new file is expected every N days.” or “There is little import history yet — the program expects a new file at least every 7 days.”, then “Last new one: ‹date› (N days ago).” If the deadline has passed, it adds “The new file is overdue.” A price list uploaded manually as a file isn't checked during import.
How long the program waits for the file
No method has a file size cap: the program doesn't check how big the supplier's file is. The wait time depends on the method, and you can't change it in the settings.
| Method | How long it waits |
|---|---|
| URL | the download itself has no overall limit — a large price list can take a long time. But connecting to the supplier's server gets 60 seconds, and if the server “hangs” mid-transfer and sends almost nothing for more than 5 minutes, the program aborts the download. Auxiliary requests — sign-in to the supplier's account and the page where the program looks for the file link — wait up to 60 seconds each |
| FTP | 400 seconds for the download. If it doesn't finish, the partial file is deleted, the error text goes to the log, and the run continues on the previous file |
| API | each integration has its own response timeout. For example, a request to BRAIN waits 120 seconds, to ERC — 300 seconds. The BRAIN price file itself is downloaded separately: if the BRAIN server sends no data for 10 minutes, the program aborts the download, and the run continues on the previous file |
What happens to the price list when the file arrives
A file that arrived on its own — by link, email, FTP or API — is processed by the import the same way as one uploaded manually: first it deletes all previous rows of this source, then writes the rows of the new file. Other sources aren't touched.
Manual edits in price list rows won't survive the next fetch. A name, price or quantity corrected on the source's “Products” tab sits in the same row the next run deletes and rewrites from the file.
You can lock your own value only in the base catalog product card. The “Do not update” checkbox excludes the card from base catalog updates from price lists. “Fixed Price” keeps your price, while the update changes other fields as usual. For a fixed price you can set “Date to remove the “Fixed price” flag”: once the date passes, the checkbox is cleared during the base catalog update, and the price comes from the price list again.
Items missing from the new file disappear from the price list. They aren't deactivated and their stock isn't zeroed — the row is simply gone. What happens to the matching card in the base catalog is decided by the “Product activity” section in the base catalog update — see How to import a price list and get prices into your catalog.
Old rows stay in place in three cases:
- if the new file didn't arrive — the run uses the previous one (see above);
- if the file arrived but not a single row could be read — deletion doesn't start, and the log says “No data received, previous product list will be displayed”;
- if a stop rule with the “Block” action fired — “Import canceled, old data kept”.
If the method was set up wrong
A method mistake — wrong link, wrong FTP path, an extra request parameter — is fixed by the next import. Correct the field and run the source import again: the new run deletes all rows that came from the wrong file and writes the rows of the right one. No need to clean up items one by one. The case with “Price list scope” in BRAIN is the same: set the parameter back to 0 (the value is saved immediately) and rerun the import.
The “Sharp drop in item count” stop rule can get in the way. By default it only warns when the item count falls by more than 50%. But if you switched it to “Block”, a corrective import that cuts the item count several times over is canceled with the “Import blocked by a stop-rule” window, and the extra rows stay. Set this rule to “Warn” while you fix things. A source's own stop rules are set on the “Log” tab of its card.
To keep such a mistake out next time, set the “Sharp rise in item count” rule to “Block”. By default it only warns — when the item count grows by more than 200%, that is, over three times. Then a file with a suddenly bloated catalog is rejected, and the previous data stays in the database. On a source's first import this rule doesn't fire at all: there's nothing to compare with. Details: Stop rules and supplier price control.
If the wrong items have already reached the base catalog through a catalog update, you need to roll that back too — the steps are in How to fix a price list after a bad file was imported.
To have the file arrive on its own
Scenarios
The supplier rarely updates the price list, but imports run daily on a schedule
What to do. In the source card, “Source” section, “Import filters” block, turn on “Process the price list only if the file has changed”. If the supplier updates the price list once a month, also set “How often the supplier sends a new price list” to “Once a month” in the “Schedule” section, so weeks without a new file aren't flagged as overdue.
1
Process only if the file changed2
Show products in Trash- Process the price list only if the file has changed — skips the import when the file is identical to the last one.
- Show products in Trash that are not assigned to catalog categories — keeps unmatched items visible in the Trash.
What you get. The file is still downloaded, but if its content matches the previous one byte for byte and the price list already has products, the run stops with the line “Price list data has not changed, skipping import”. The price list rows stay as they were.
What won't happen. The checkbox works only when all active price lists are processed together and on scheduled runs. A manual import of one source always processes the file. It also won't help if the supplier changes anything in the file each time — the generation date, row order: for the program, that's already a different file.
The supplier switched from email to a link
What to do. In the “Source” section, clear the email fields — the mailbox and the email selection condition — and fill in the “URL” method fields. Run the source import manually and find the “Getting file by link” line in the log.
What you get. There's no multiple-methods warning under the tiles, the “URL” tile is highlighted, and the file arrives from the new address.
What won't happen. An empty password field doesn't erase the saved password — not for the link, FTP or API. If the old email fields are still filled in, email will override the link: it runs later.
What to do when the file doesn't arrive
| What you see | Why | What to do |
|---|---|---|
| “No file found to import: ‹name›” | There's neither a fresh file nor a previous one | Check that the method is fully set up: email needs a mailbox and a condition, FTP — a server and a path, API — an address and a login (about methods) |
| “No new file received from the supplier — products from the previous file are in use. This is not an error if the supplier has not sent an update yet.” | No configured method brought a new file, but there was no failure either — after a red line with a reason this warning isn't written; products updated from yesterday's data, the run ended with the “Done” status. Runs before 18.10.2026 show “No new file received, using the previous one” with the “Error” status here | Nothing, if the next line is the note “Last new price list: …”. If the supplier should have sent a file, check the email selection condition (when the file didn't arrive) |
| “No new price list for N days (last one: ‹date›)…” | No new file for longer than the threshold — the larger of two: the interval plus one day, or one and a half intervals | Ask the supplier whether they're still sending the price list; if they send less often than the program thinks, set the frequency manually (how often the supplier sends a price list) |
Messages specific to each method are covered in their articles: link, email, FTP, API.
FAQ
Can I set up several methods as a backup?
No. The import runs all configured methods in a row, and the file of the one that ran last is kept. The card highlights the winner's tile and warns you if there are several methods. Details: why one method.
Does the program fetch the price list on its own, without me?
Yes, on every import of the source. To have the import start without you too, set up a schedule — schedule and task queue. Run the first import after setup manually and check the log.
Why do email and FTP need a sample file?
The wizard maps the columns from it: the program needs to see once where the name, price and SKU are in the price list. The mailbox or server is entered later, in the source card.
Can I import a price list in PDF?
No. The program doesn't read PDF, so ask the supplier for the same price list in Excel or CSV. If an Excel file won't open because it's password-protected, open it in Excel and save it as .xlsx without a password; the message in the import log says the same.
The supplier sends a price list once a month, but the log says “No new price list”. What should I change?
In the “Schedule” section of the source card, set “How often the supplier sends a new price list” to “Once a month”. Then it will be overdue only after 45 days without a new file. Details: how often the supplier sends a price list.
Where it is and who has access
- Menu
- Directories → Processing price lists → List of price lists
- Source card
- the “Source” section — method tiles and all connection fields
- New source wizard
- step 2 “From where”; ready-made integrations are chosen only here
- Access right
- the “List of price lists” window plus “Import settings” and “…price list file”
- Prepare
- access details from the supplier; for email and FTP also a sample file
All connection settings live in one place — the source card, “Source” section. It has a row of four tiles: “File”, “URL”, “E-mail”, “FTP”; the selected tile shows its fields below. You can switch tiles at any time: a tile only shows its set of fields, without erasing or moving anything.
Two methods aren't in this row. A ready-made API integration is linked once in the wizard — after that the section shows a single “API” tile instead of four. A site isn't chosen in the card at all; that's also a wizard decision.
1
List of price lists- List of price lists — opens the window with all price list sources.
The “List of price lists” window has a desktop shortcut. The other windows open only from the source card or the wizard.
Without both “Import settings” rights, the card simply has no “Source” section. This split is deliberate: the section holds the supplier's email and FTP passwords. A manager who works with imported products doesn't need these rights — the “Products” tab has its own right.
The mail server form and the supplier account access window check their own right. The mailbox connection wizard and the API parameters window don't: anyone who can open the “Source” section can open them too.
Related topics
- If the source isn't created yet — How to connect a price list source with the wizard.
- If the file arrives, but the import needs to run on a schedule — Schedule and task queue.
- If the log has a red line and it's unclear what it means — Errors in the import log.
- If you worry that a broken supplier file will reach your website — Price list stop rules.
