add example, update README [ci skip]

README.md

@@ -1,3 +1,43 @@
 # crossjoin [![Docker](https://github.com/crossjoin-io/crossjoin/actions/workflows/docker.yml/badge.svg)](https://github.com/crossjoin-io/crossjoin/actions/workflows/docker.yml) [![CLI](https://github.com/crossjoin-io/crossjoin/actions/workflows/go.yml/badge.svg)](https://github.com/crossjoin-io/crossjoin/actions/workflows/go.yml) [![Security scan](https://github.com/crossjoin-io/crossjoin/actions/workflows/shiftleft.yml/badge.svg)](https://github.com/crossjoin-io/crossjoin/blob/main/SECURITY.md)
 
 Crossjoin joins together your data from anywhere.
+
+- Supports PostgreSQL, Redshift, CSV data sources
+- Zero dependency CLI, or a single Docker container
+
+## Example
+
+In the [example](https://github.com/crossjoin-io/crossjoin/tree/main/example) directory, there are two CSVs (adapted from
+this [AWS blog post](https://aws.amazon.com/blogs/big-data/joining-across-data-sources-on-amazon-quicksight/)) representing
+orders and returns data.
+
+The config creates a combined data set in a `joined.db` SQLite3 file.
+
+```yaml
+data_sets:
+  - name: joined
+    data_source:
+      name: orders
+      type: csv
+      path: ./orders.csv
+    joins:
+      - type: JOIN
+        columns:
+          - left_column: Order ID
+            right_column: Order ID
+        data_source:
+          name: returns
+          type: csv
+          path: ./returns.csv
+```
+
+```
+$ crossjoin --config ./config.yaml
+2021/10/14 18:08:06 using config file path config.yaml
+2021/10/14 18:08:06 starting crossjoin
+2021/10/14 18:08:06 creating data set `joined`
+2021/10/14 18:08:06 querying `orders`
+2021/10/14 18:08:06 querying `returns`
+2021/10/14 18:08:06 joining data
+2021/10/14 18:08:06 finished crossjoin
+```

cmd/crossjoin/main.go

@@ -23,6 +23,7 @@ import (
 	"io"
 	"log"
 	"os"
+	"strconv"
 	"strings"
 
 	"github.com/crossjoin-io/crossjoin/config"
@@ -114,13 +115,14 @@ func createDataset(dataset config.DataSet) error {
 	for _, join := range dataset.Joins {
 		joinColumns := []string{}
 		for _, cols := range join.Columns {
-			joinColumns = append(joinColumns, fmt.Sprintf("%s.%s = %s.%s", dataset.DataSource.Name, cols.LeftColumn, join.DataSource.Name, cols.RightColumn))
+			joinColumns = append(joinColumns, fmt.Sprintf(`%s."%s" = %s."%s"`, dataset.DataSource.Name, cols.LeftColumn, join.DataSource.Name, cols.RightColumn))
 		}
 		joinClauses += fmt.Sprintf(" %s %s ON %s", join.Type, join.DataSource.Name, strings.Join(joinColumns, " AND "))
 	}
 
 	log.Println("joining data")
-	_, err = db.Exec(fmt.Sprintf("CREATE TABLE %s AS SELECT * FROM %s %s", dataset.Name, dataset.DataSource.Name, joinClauses))
+	joinQuery := fmt.Sprintf("CREATE TABLE %s AS SELECT * FROM %s %s", dataset.Name, dataset.DataSource.Name, joinClauses)
+	_, err = db.Exec(joinQuery)
 	return err
 }
 
@@ -138,6 +140,9 @@ func fetchSingle(dest *sql.DB, dataSource *config.DataSource) error {
 			return err
 		}
 		columns := firstLine
+		for i := range columns {
+			columns[i] = strconv.Quote(columns[i])
+		}
 
 		_, err = dest.Exec(ngsastOK(fmt.Sprintf("CREATE TABLE %s (%s)", dataSource.Name, strings.Join(columns, ","))))
 		if err != nil {

example/config.yaml

@@ -0,0 +1,15 @@
+data_sets:
+  - name: joined
+    data_source:
+      name: orders
+      type: csv
+      path: ./orders.csv
+    joins:
+      - type: JOIN
+        columns:
+          - left_column: Order ID
+            right_column: Order ID
+        data_source:
+          name: returns
+          type: csv
+          path: ./returns.csv