What an empty result means
Outcome: you will be able to read any empty page or missing record correctly, and you will know why minting and inscribing keep working while the record is still incomplete.
Why this needs a page
Section titled “Why this needs a page”The indexer reads the Zcash chain from the beginning before it can list what is on it, and that takes time. While it does, a section with nothing in it is not a statement that nothing exists: the records may sit in blocks it has not reached yet. Most explorers blur this into a generic empty state. This one never does.
The four situations
Section titled “The four situations”- The chain is still being read. The result says so where it appears, instead of reporting a count. Absence here means nothing at all, and no page turns it into a zero: a count of Zordinals, ZRunes, collections or activity is shown only when it can be right.
- No records yet. Shown only once the whole chain has been read and replayed under the current rules (what that takes). This is a real, checkable statement that nothing of that kind exists.
- Indexer unreachable. The service could not be reached, so nothing can be said either way. Nothing is hidden or lost. The outage model.
- Data is stale. The last known values are still shown, labeled, with the time they were last confirmed.
There is no site-wide reading badge in the header or the account panel. The answer sits on the result it applies to, and the network selector and wallet controls stay where they always were.
Read is not the same as qualified
Section titled “Read is not the same as qualified”Two different facts sit behind every answer, and an empty result needs both:
- Scanned: the indexer has stored every block from the first one it is required to read up to its current checkpoint.
- Qualified: the exact indexer release that is serving you replayed those blocks itself, without a gap, under the exact rules and settings it runs now.
On mainnet the required start is block 0, the first block of the chain. The record is qualified only when all seven readings below were replayed from block 0 to the exact current checkpoint, block hash included:
| Reading | Rule set |
|---|---|
| Zordinals | universe-zerdinals-v1 |
| Collections | collections-v1 |
| ZRunes | zrunes-v1 (no ZRune can exist before block 3,470,000) |
| ZRC-721 | zrc721-v1 |
| ZRC-20, zord reading | zord |
| ZRC-20, Zecscriptions reading | zecscriptions |
| ZkMap | zkmap-v1 |
A replay that started late stays unqualified, however far it gets. Catching up to the newest block extends it forward; it never fills in the blocks it skipped at the start. A new indexer release whose reading code or settings changed also starts unqualified, and has to replay from block 0 again.
Strict and draft answers
Section titled “Strict and draft answers”Most answers are strict: they are given only from qualified, current
history. When that history is missing, a strict answer does not guess. It
returns a typed unavailable answer that names the reason, for example
replay_incomplete, and the product shows that the record cannot answer yet.
An unavailable answer is never an empty list and never “nothing here”.
A few views, such as the ZkMap block map, can also be read as a draft. A draft shows what the service can already see: a block claimed under a checkpoint the node confirms, a claim waiting in the mempool, and otherwise unknown. A draft never marks a block free to claim and never stands in for an accepted claim receipt. Only a strict answer can do that.
Qualification does not decide whether you can create. Inscribing, etching, minting and ZkMap claims keep the rules in the next section; a receipt that needs qualified history arrives once the record can vouch for it.
Creating does not wait for the reading
Section titled “Creating does not wait for the reading”Reading the record and writing to the chain are separate jobs. An inscription, a batch, a ZRune etch or mint and a ZkMap claim are built from the service’s own Zcash node, not from how far the record has read. Each one checks only what that transaction needs:
- the node is on the right network and reports a usable chain height;
- the node publishes the rules the next block is signed under;
- for ZRunes, the node’s height has reached the activation block;
- the path you chose is open: your connected wallet release, or the service’s own signer and payment watcher;
- every coin it spends is proven free of assets, one output at a time.
While the record is incomplete, a result you create may take a while to appear on its pages. That is the reading, not the order. Sending, listing and buying still wait for the full record, because they move assets the record has to see first.
Every output is checked on its own
Section titled “Every output is checked on its own”Spending an output that carries an inscription or a ZRune as a fee destroys it, so nothing is spent until that output is proven clean. The proof is per output, not a wait for the whole chain:
- the record has read the block that created it and found nothing on it; or
- the service’s node traces where the coins came from and every step is plain ZEC, back to newly mined coins or a shielded balance.
An output that cannot be proven either way is unchecked, never clear, and is left untouched. An output the record lists an asset on is never spent as a fee. Protect asset-bearing outputs.
If your payment waits for proof
Section titled “If your payment waits for proof”Coins sent from a shielded balance are proven as soon as they confirm. Coins that went through many transparent hops may have to wait for the record.
- Paying an invoice: the order stays at its confirming step, and its status carries a note that the deposit is waiting for proof. Nothing is spent or refunded while it waits, and the order continues by itself once the deposit is proven.
- Connected wallet: if the address holds enough ZEC but not enough of it
is proven yet, the answer says exactly that (
FUNDING_UNPROVEN) rather than calling it a shortfall.
Either way, the quickest fix is to send the ZEC from a shielded balance.
How to check coverage yourself
Section titled “How to check coverage yourself”How far the record has read is public:
https://zordinals.run/idx/zcash-metaprotocols/status| Field | What it tells you |
|---|---|
coverage.scannedHeight, coverage.networkHeight | How far the scan has reached, and the network’s height |
coverage.scanComplete | Every block from the required start to the checkpoint is stored |
coverage.protocolQualified | All seven readings were replayed by this release from the required start to this checkpoint |
coverage.freshness | The verdict in one word: ok, or the reason it is not, such as replay_incomplete or indexer_lag |
coverage.chainComplete | true only when the scan is complete, the replay is qualified and the checkpoint is current |
An empty result is a fact only while chainComplete is true.
The status page records the
last verified values.
How this behavior is tested
Section titled “How this behavior is tested”Functional tests run on the public Zcash Testnet, with real wallets, real transactions and the same indexer rules. Bitcoin Signet cannot be used for these flows: it runs Bitcoin’s consensus rules, not Zcash’s, so it cannot carry a Zcash transaction or the readings built on one. Signet is used only for flows that are actually Bitcoin flows. Crash and chain reorganization cases are rehearsed separately on a private test chain and reported separately.