Bigquery Query

Switch project if needed:

gcloud config set project skilled-fulcrum-90207   # production
gcloud config set project holistics-data-294707   # development

Workflow

1. Check project — gcloud config get-value project to confirm environment.
2. Discover — List datasets and tables, inspect schemas before writing queries.
3. Query — Run the query.

Core Commands

Run a Query

bq query --use_legacy_sql=false --project_id=PROJECT_ID \
  "SELECT * FROM dataset.table LIMIT 10"

Dry Run (Cost Estimation)

Always run a dry run first on large or unfamiliar tables to check bytes processed:

bq query --use_legacy_sql=false --dry_run --project_id=PROJECT_ID \
  "SELECT * FROM dataset.table"

List Datasets

bq ls --project_id=PROJECT_ID

List Tables in a Dataset

bq ls --project_id=PROJECT_ID dataset_name

Get Table Schema

bq show --format=json PROJECT_ID:dataset.table | jq '.schema.fields'

Get Table Info (Row Count, Size)

bq show --format=json PROJECT_ID:dataset.table | jq '{rows: .numRows, bytes: .numBytes, type: .type}'

Tips

• Check your active project first — run gcloud config get-value project before querying.
• Always filter on partition columns to avoid full table scans and reduce costs.
• Use --dry_run before running queries on large tables to check estimated bytes.
• Use LIMIT 10 during exploration; remove it only for final queries.
• Fully qualify table names as project.dataset.table to avoid ambiguity.
• Use SAFE_ prefix functions (e.g., SAFE_DIVIDE, SAFE_CAST) to avoid query failures on bad data.
• BigQuery charges by bytes scanned; SELECT * on wide tables is expensive. Select only needed columns.
• Use bq show --format=json to inspect table schemas before writing queries.