====== Reading CSV Files ======
  - Adding Apache commons csv library as dependencies
  - Putting your csv file in the project
  - Read the CSV file in your code

===== Adding apache commons text library as dependencies =====
Put the following in your build.gradle dependencies{} section. The current version of commons cvs is 1.6. You might want to use a newer version upon its release.
<code>
compile group: 'org.apache.commons', name: 'commons-csv', version: '1.6'
</code>

===== Putting your csv file in the project =====
For Grails 3 project, one of the good place to store your file is the ''/grails-app/conf'' directory. You can even create your own directory within it. We are going to use **MTR Lines (except Airport Express & Light Rail) Fares** ''mtr_lines_fares.csv'' as an example, and it can be downloaded at [[https://data.gov.hk/en-data/dataset/mtr-data-routes-fares-barrier-free-facilities|here]].
{{ :reading_csv_files1.png?direct&200 |}}

The data look like the follow. Play attention to the header.
^ SRC_STATION_NAME	^ SRC_STATION_ID	^ DEST_STATION_NAME	^ DEST_STATION_ID	^ OCT_ADT_FARE	^ OCT_STD_FARE	^ SINGLE_ADT_FARE	^ OCT_CON_CHILD_FARE	^ OCT_CON_ELDERLY_FARE	^ OCT_CON_PWD_FARE	^ SINGLE_CON_CHILD_FARE	^ SINGLE_CON_ELDERLY_FARE
| Central	| 1	| Central	| 1	| 0	| 0	| 0	| 0	| 0	| 0	| 0	| 0
| Central	| 1	| Admiralty	| 2	| 4.6	| 3	| 5	| 3	| 2	| 2	| 3	| 3
| ... ||


===== Read the CSV file in your code =====
We can read the csv in a Grails controller or service. We are going to do this in a controller.
<code>
class PriceController {
    def grailsResourceLocator

    def index() {
        String strings = grailsResourceLocator.findResourceForURI('classpath:/mtr_lines_fares.csv').inputStream.getText(StandardCharsets.UTF_8.name())
        Iterable<CSVRecord> records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(new StringReader(strings))
        StringBuilder sb = new StringBuilder()
        for (CSVRecord cvsRecord : records) {
            try {
                sb.append(cvsRecord.get("SRC_STATION_NAME")).append(", ")
                sb.append(cvsRecord.get("SRC_STATION_ID")).append(", ")
                sb.append(cvsRecord.get("DEST_STATION_NAME")).append(", ")
                sb.append(cvsRecord.get("DEST_STATION_ID")).append(", ")
                sb.append(cvsRecord.get("OCT_ADT_FARE")).append(", ")
                sb.append(cvsRecord.get("OCT_STD_FARE")).append(", ")
                sb.append(cvsRecord.get("SINGLE_ADT_FARE")).append(", ")
                sb.append(cvsRecord.get("OCT_CON_CHILD_FARE")).append(", ")
                sb.append(cvsRecord.get("OCT_CON_ELDERLY_FARE")).append(", ")
                sb.append(cvsRecord.get("OCT_CON_PWD_FARE")).append(", ")
                sb.append(cvsRecord.get("SINGLE_CON_CHILD_FARE")).append(", ")
                sb.append(cvsRecord.get("SINGLE_CON_ELDERLY_FARE"))
                sb.append("<br>")
            } catch (Exception ignored) {
                log.error("Unable to convert ${cvsRecord.toString()}")
            }
        }
        render(sb.toString())
    }
}
</code>

First we load our csv file with grailsResourceLocator. 
<code>
String strings = grailsResourceLocator.findResourceForURI('classpath:/mtr_lines_fares.csv').inputStream.getText(StandardCharsets.UTF_8.name())
</code>

Than parse it with ''withFirstRecordAsHeader''. It will return an iterable CSVRecord, which is basically a map with headers as keys.
<code>
Iterable<CSVRecord> records = CSVFormat.RFC4180.withFirstRecordAsHeader().parse(new StringReader(strings))
</code>

With the iterable records, we can use a loop to loop though them to extract the data we want. Here we use the csv header as the key to obtains data in each column.
<code>
for (CSVRecord cvsRecord : records) {
   try {
      sb.append(cvsRecord.get("SRC_STATION_NAME")).append(", ")
      .
      .
      .
   } catch (Exception ignored) {
      log.error("Unable to convert ${cvsRecord.toString()}")
}
</code>
{{ :reading_csv_files2.png?direct&400 |}}

===== CSV without header =====
You can parse the string with 
<code>
Iterable<CSVRecord> records = CSVFormat.RFC4180.parse(new StringReader(strings))
</code>
And loop though the records with
<code>
for (CSVRecord cvsRecord : records) {
   try {
      sb.append(cvsRecord.get(0).append(", ")
      sb.append(cvsRecord.get(1).append(", ")
      sb.append(cvsRecord.get(2).append(", ")
      .
      .
      .
   } catch (Exception ignored) {
      log.error("Unable to convert ${cvsRecord.toString()}")
}
</code>