zed-industries/zed · error · Error

Failed to parse UNIT_DATA as JSON: ${e.message}

Error message

Failed to parse UNIT_DATA as JSON: ${e.message}

What it means

Produced at the end of Copilot::sign_in (copilot.rs:810): the whole in-flight sign-in task is remapped with task.map_err(|err| anyhow!("{err:?}")), so this error is the Debug-formatted wrapper around whatever the sign-in flow returned. The underlying task (copilot.rs:758-799) sends the request::SignIn LSP request with the timeout from global_lsp_settings.get_request_timeout(), so typical inner errors are 'copilot sign-in' request failures, timeouts, or server-shutdown errors. On failure the code also resets the sign-in status to NotSignedIn via update_sign_in_status, so the extension returns to a SignedOut state that can retry.

Source

Thrown at script/cargo-timing-info.js:41

    return path.join(localAppData, "Zed");
  } else {
    // Fallback to XDG config dir
    const xdgConfigHome = process.env.XDG_CONFIG_HOME || path.join(os.homedir(), ".config");
    return path.join(xdgConfigHome, "zed");
  }
}

function extractUnitData(htmlContent) {
  // Find the UNIT_DATA array in the file
  const unitDataMatch = htmlContent.match(/const\s+UNIT_DATA\s*=\s*(\[[\s\S]*?\]);/);
  if (!unitDataMatch) {
    throw new Error("Could not find UNIT_DATA in the file");
  }

  try {
    return JSON.parse(unitDataMatch[1]);
  } catch (e) {
    throw new Error(`Failed to parse UNIT_DATA as JSON: ${e.message}`);
  }
}

function formatTime(seconds) {
  if (seconds < 60) {
    return `${seconds.toFixed(2)}s`;
  }
  const minutes = Math.floor(seconds / 60);
  const remainingSeconds = seconds % 60;
  return `${minutes}m ${remainingSeconds.toFixed(2)}s`;
}

function formatUnit(unit) {
  let name = `${unit.name} v${unit.version}`;
  if (unit.target && unit.target.trim()) {
    name += ` (${unit.target.trim()})`;
  }
  return name;

View on GitHub (pinned to bc538def45)

Solutions

1. Retry by calling sign_in again — the error handler already reset status to NotSignedIn, so a fresh attempt is valid.
2. Raise the timeout: set "lsp": { "request_timeout": 300 } in settings.json (read via ProjectSettings global_lsp_settings at copilot.rs:754-756).
3. Check whether the copilot-language-server process is alive (ps / Task Manager) and inspect the Zed log for server stderr/crash output.
4. If the server crashed repeatedly, run copilot: Reinstall to get a fresh binary and retry sign-in.
5. Check network reachability to github.com/GitHub Enterprise URI (copilot_settings.enterprise_uri) if using GHE.

Example fix

// before
let task = copilot.sign_in(cx);
task.await?; // error text is an opaque Debug dump

// after: classify the retryable case
use util::ResultExt;
if let Err(err) = copilot.sign_in(cx).await {
    let msg = format!("{err:#}");
    if msg.contains("timed out") || msg.contains("copilot sign-in") {
        // transient: status was reset to NotSignedIn, safe to re-invoke
        copilot.sign_in(cx).await?;
    } else {
        return Err(anyhow!("copilot sign-in failed: {msg}"));
    }
}

// Check the server is Running before starting sign-in (avoids error 362)
// and confirm the LSP request timeout is generous enough for device flow.
if !matches!(copilot.status(), Status::SignedOut { .. } | Status::Unauthorized) {
    return Ok(()); // nothing to sign in to
}

fn signable(status: &Status) -> bool {
    matches!(
        status,
        Status::SignedOut { awaiting_signing_in: false } | Status::Unauthorized
    )
}

// The status is reset to NotSignedIn on failure, so one bounded retry is safe.
let mut attempt = 0;
while let Err(err) = copilot.sign_in(cx).await {
    attempt += 1;
    if attempt >= 2 || !format!("{err:#}").contains("copilot sign-in") {
        return Err(err.context("copilot sign-in failed twice"));
    }
}

When it happens

Trigger: Calling copilot.sign_in(cx) and awaiting the returned Task while the language server fails to answer the SignIn request: request timeout exceeded (lsp.request_timeout setting, default 120s), language server process crashed or was restarted mid-flow, or the entity was dropped so this.update(cx, ...) inside the task fails.

Common situations: Slow GitHub device-flow authorization exceeding the configured LSP request timeout; Copilot server process crash or OOM kill during sign-in; a very low 'lsp.request_timeout' value in settings.json; antivirus on Windows killing the freshly spawned copilot-language-server process; retrying sign_in after a previous attempt left the server half-initialized.

Understand the failure class

• Parsing and encoding errors: unexpected token, malformed input — why parsers reject input and how to find the real culprit.

Related errors

• HTTP error! status: ${response.status}
• Could not find UNIT_DATA in the file
• GraphQL error: ${JSON.stringify(data.errors)}
• Unexpected response structure: ${JSON.stringify(data)}
• Missing env var `ZED_SERVER_URL`

AI-assisted analysis of zed-industries/zed@bc538def45 (2026-08-16). Data as JSON: /api/errors/b3ae0acee8266c18. Report an issue: GitHub.