Free · No sign-up · Nothing installed

BigQuery to ER Diagram

Paste the DDL from INFORMATION_SCHEMA or the console — backticked project.dataset.table, INT64/STRING, ARRAY<STRUCT<…>> and NOT ENFORCED keys included.

SQL input — CREATE TABLE · DDL · dumpbigquery

CREATE TABLE `analytics.blog.users` (
  id INT64 NOT NULL,
  email STRING NOT NULL,
  display_name STRING,
  created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP()
)
CLUSTER BY id
OPTIONS(description="blog user accounts");

CREATE TABLE `analytics.blog.posts` (
  id INT64 NOT NULL,
  author_id INT64 NOT NULL,
  title STRING NOT NULL,
  body STRING,
  labels ARRAY<STRING>,
  attrs JSON,
  published_at TIMESTAMP
)
OPTIONS(description="blog posts");

CREATE TABLE `analytics.blog.comments` (
  id INT64 NOT NULL,
  post_id INT64 NOT NULL,
  commenter_id INT64 NOT NULL,
  body STRING NOT NULL,
  created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP()
)
OPTIONS(description="post comments");

CREATE TABLE `analytics.blog.tags` (
  id INT64 NOT NULL,
  slug STRING NOT NULL
)
OPTIONS(description="tag dictionary");

CREATE TABLE `analytics.blog.post_tags` (
  post_id INT64 NOT NULL,
  tag_id INT64 NOT NULL
)
OPTIONS(description="post to tag join");

-- BigQuery never enforces keys, but the constraint statements still
-- describe the join structure — drawn as relationship edges:
ALTER TABLE `analytics.blog.post_tags` ADD PRIMARY KEY (post_id, tag_id) NOT ENFORCED;
ALTER TABLE `analytics.blog.post_tags` ADD FOREIGN KEY (post_id) REFERENCES analytics.blog.posts (id) NOT ENFORCED;
ALTER TABLE `analytics.blog.post_tags` ADD FOREIGN KEY (tag_id) REFERENCES analytics.blog.tags (id) NOT ENFORCED;
ALTER TABLE `analytics.blog.comments` ADD FOREIGN KEY (post_id) REFERENCES analytics.blog.posts (id) NOT ENFORCED;
ALTER TABLE `analytics.blog.comments` ADD FOREIGN KEY (commenter_id) REFERENCES analytics.blog.users (id) NOT ENFORCED;
ALTER TABLE `analytics.blog.posts` ADD FOREIGN KEY (author_id) REFERENCES analytics.blog.users (id) NOT ENFORCED;

bigquerySample: · · · · ·

Diagram—

Your diagram renders here.

Parser coverage

What we parse in bigquery

Nested types — ARRAY<STRING> and JSON render verbatim in the column row
Backticked paths — `project.dataset.table` unwrapped, dataset shown as the group
Column OPTIONS(description=…) — stripped as noise, never rendered as columns
NOT ENFORCED keys — ALTER TABLE … ADD FOREIGN KEY statements become relationship edges

From your database

Get just the schema

SELECT ddl FROM `my-project.mydataset`.INFORMATION_SCHEMA.TABLES;

Run it in the BigQuery console (or bq query) — INFORMATION_SCHEMA is free metadata, no bytes scanned. Paste the whole ddl column above.

FAQ

Questions, answered

How do I get the DDL out of BigQuery?

Query your dataset's INFORMATION_SCHEMA: SELECT ddl FROM `project.dataset`.INFORMATION_SCHEMA.TABLES — or open a table's Details tab in the console and copy the DDL.

Does it understand nested types?

Yes — ARRAY<STRING>, STRUCT<…> and JSON render verbatim in the column's type slot, exactly as BigQuery writes them.

BigQuery doesn't enforce foreign keys — what do I get?

NOT ENFORCED PRIMARY KEY / FOREIGN KEY statements are drawn as relationship edges anyway, so the join structure still shows.

Will the INFORMATION_SCHEMA query cost me anything?

No — INFORMATION_SCHEMA queries scan metadata only and are free. Pasting here is free too, and nothing is stored.

More converters

BigQuery to ER Diagram

What we parse in bigquery

Get just the schema

Questions, answered

Also available