@erdoai/cli 0.53.0 → 0.54.0

package/README.md

@@ -61,6 +61,15 @@ erdo datasets fetch acme.leads --filter <name>   # opt into a saved filter
 The table is named **`data`** — for file datasets (CSV, Excel, and anything
 written by an event pipeline) that is the name regardless of the dataset's slug
 or resource key, and it is not guessable, so it is the first thing to get right.
+The columns aren't guessable either, so look before you write SQL:
+
+```bash
+erdo datasets schema acme.leads          # column names and types
+```
+
+`datasets schema` reads the stored table itself, so it shows the columns that
+are actually there — the ones `fetch --sql` can reference — rather than a
+declared schema that may lag what recent writes added.
 Database and warehouse datasets are queried through their real table names, which
 come from the dataset's schema. The SQL dialect is DuckDB. A dataset's default
 filters apply on every read; `--filter <name>` adds a saved filter on top, and

package/dist/index.js

@@ -1339,6 +1339,14 @@ function timedOutMessage(threadID) {
 function print(value) {
   console.log(JSON.stringify(value, null, 2));
 }
+function printAlignedTable(columns, rows) {
+  const cell = (v) => v === null || v === void 0 ? "" : typeof v === "object" ? JSON.stringify(v) : String(v);
+  const widths = columns.map((c, i) => Math.max(c.length, ...rows.map((r) => cell(r[i]).length), 0));
+  const line = (cells) => cells.map((c, i) => c.padEnd(widths[i])).join("  ").trimEnd();
+  console.log(line(columns));
+  console.log(widths.map((w) => "-".repeat(w)).join("  "));
+  for (const r of rows) console.log(line(columns.map((_, i) => cell(r[i]))));
+}
 var sleep2 = (ms) => new Promise((r) => setTimeout(r, ms));
 var collect = (v, acc) => {
   acc.push(v);
@@ -3444,7 +3452,7 @@ datasetsCmd.command("fetch <slug>").description(
   "Read rows from a dataset \u2014 the deterministic read, and the one to use for anything mechanical or scripted. Returns {columns, rows, row_count}."
 ).option(
   "-q, --sql <query>",
-  "DuckDB SQL shaping the rows. File datasets (including anything an event pipeline writes) are queried as a table named `data`, whatever the dataset's slug; database and warehouse datasets use their real table names from the schema."
+  "DuckDB SQL shaping the rows. File datasets (including anything an event pipeline writes) are queried as a table named `data`, whatever the dataset's slug; database and warehouse datasets use their real table names from the schema. `erdo datasets schema <slug>` lists the columns you can reference."
 ).option("-l, --limit <n>", "max rows to return", (v) => parseInt(v, 10)).option(
   "-f, --filter <name>",
   "opt into a saved filter by name (repeatable); additive on top of the dataset's default filters (see: erdo datasets filter list <slug>)",
@@ -3463,6 +3471,29 @@ datasetsCmd.command("fetch <slug>").description(
     fail(e);
   }
 });
+datasetsCmd.command("schema <slug>").alias("describe").description(
+  "Show a dataset's physical columns \u2014 the names and types you can reference in `datasets fetch --sql`, read from the stored `data` table itself (DuckDB DESCRIBE) so they are what is actually there, not what was declared. For a database or warehouse dataset, DESCRIBE its real table names instead: `datasets fetch <slug> --sql 'DESCRIBE <table>'`."
+).option("--json", "print the raw JSON result instead of a table").action(async (slug, opts) => {
+  try {
+    const res = await new ErdoClient().fetchDatasetContents(slug, { sql_query: "DESCRIBE data" });
+    if (opts.json) {
+      print(res);
+      return;
+    }
+    const nameIdx = res.columns.indexOf("column_name");
+    const typeIdx = res.columns.indexOf("column_type");
+    if (nameIdx === -1 || typeIdx === -1) {
+      printAlignedTable(res.columns, res.rows);
+      return;
+    }
+    printAlignedTable(
+      ["column", "type"],
+      res.rows.map((r) => [r[nameIdx], r[typeIdx]])
+    );
+  } catch (e) {
+    fail(e);
+  }
+});
 var MAX_UPLOAD_BYTES = 20 * 1024 * 1024;
 datasetsCmd.command("upload <file>").description("Upload a file (CSV, Excel, JSON, ...) and create a dataset from it").option("-n, --name <name>", "display name for the dataset (defaults to the filename)").option("-d, --description <text>", "description shown to agents analyzing the dataset").action(async (file, opts) => {
   try {
@@ -3519,14 +3550,6 @@ datasetsCmd.command("configure-integration <dataset-id>").description("Set an in
     fail(e);
   }
 });
-function printAnalyticsTable(columns, rows) {
-  const cell = (v) => v === null || v === void 0 ? "" : typeof v === "object" ? JSON.stringify(v) : String(v);
-  const widths = columns.map((c, i) => Math.max(c.length, ...rows.map((r) => cell(r[i]).length), 0));
-  const line = (cells) => cells.map((c, i) => c.padEnd(widths[i])).join("  ").trimEnd();
-  console.log(line(columns));
-  console.log(widths.map((w) => "-".repeat(w)).join("  "));
-  for (const r of rows) console.log(line(columns.map((_, i) => cell(r[i]))));
-}
 var analytics = program.command("analytics").description("Page analytics \u2014 what is tracking your published pages, and how they perform with real visitors");
 analytics.command("query <hogql>").description("Run a read-only HogQL query against this org's page-analytics events").option("--json", "print the raw JSON result instead of a table").action(async (hogql, opts) => {
   try {
@@ -3539,7 +3562,7 @@ analytics.command("query <hogql>").description("Run a read-only HogQL query agai
       console.log("Page analytics is not enabled for this organization \u2014 turn it on in the Erdo app, then query real traffic here.");
       return;
     }
-    printAnalyticsTable(res.columns, res.rows);
+    printAlignedTable(res.columns, res.rows);
     if (res.truncated) console.log(`
 (rows truncated at the cap \u2014 aggregate further or add a tighter filter/LIMIT)`);
   } catch (e) {
@@ -3570,7 +3593,7 @@ analytics.command("tracking").description("Show which analytics destinations thi
         rows.push([d.kind, d.enabled ? "on" : "configured, off", d.public_id ?? "", d.provider ?? "", ""]);
       }
     }
-    printAnalyticsTable(["kind", "status", "public id", "provider", "what it does"], rows);
+    printAlignedTable(["kind", "status", "public id", "provider", "what it does"], rows);
     console.log(`
 session replay input masking: ${res.mask_inputs ? "on" : "OFF \u2014 form values are recorded"}`);
     console.log(`consent: ${res.consent}${res.consent === "required" ? " (recording waits on a consent banner, so thin volume may be the gate, not the traffic)" : ""}`);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@erdoai/cli",
-  "version": "0.53.0",
+  "version": "0.54.0",
   "description": "Erdo CLI — drive datasets, pages, and evals from the terminal or CI",
   "type": "module",
   "bin": {