🌱 cron: make CSV header optional (3/n) (#2261)

* Make CSV header optional.

Signed-off-by: Spencer Schrock <[email protected]>

* Appease linter.

Signed-off-by: Spencer Schrock <[email protected]>

* Address PR feedback.

Signed-off-by: Spencer Schrock <[email protected]>

Signed-off-by: Spencer Schrock <[email protected]>

cron/internal/data/iterator.go

@@ -19,6 +19,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"reflect"
 
 	"github.com/jszwec/csvutil"
 
@@ -36,21 +37,45 @@ type Iterator interface {
 func MakeIteratorFrom(reader io.Reader) (Iterator, error) {
 	csvReader := csv.NewReader(reader)
 	csvReader.Comment = '#'
-	dec, err := csvutil.NewDecoder(csvReader)
+	header, err := csvutil.Header(RepoFormat{}, "csv")
+	if err != nil {
+		return nil, fmt.Errorf("error in csvutil.Header: %w", err)
+	}
+	dec, err := csvutil.NewDecoder(csvReader, header...)
 	if err != nil {
 		return nil, fmt.Errorf("error in csvutil.NewDecoder: %w", err)
 	}
 	return &csvIterator{decoder: dec}, nil
 }
 
 type csvIterator struct {
-	decoder *csvutil.Decoder
-	err     error
-	next    RepoFormat
+	decoder     *csvutil.Decoder
+	err         error
+	next        RepoFormat
+	afterHeader bool
+}
+
+// returns true on the first call if the most recently decoded record is a header.
+// always returns false on subsequent calls, as this is only intended to evaluate the first line.
+func (reader *csvIterator) isHeader() bool {
+	if reader.afterHeader {
+		return false
+	}
+	header, err := csvutil.Header(RepoFormat{}, "csv")
+	if err != nil {
+		reader.err = err
+		return false
+	}
+	lastRead := reader.decoder.Record()
+	reader.afterHeader = true
+	return reflect.DeepEqual(header, lastRead)
 }
 
 func (reader *csvIterator) HasNext() bool {
 	reader.err = reader.decoder.Decode(&reader.next)
+	if reader.isHeader() {
+		reader.err = reader.decoder.Decode(&reader.next)
+	}
 	return !errors.Is(reader.err, io.EOF)
 }
 

cron/internal/data/iterator_test.go

@@ -153,6 +153,42 @@ func TestCsvIterator(t *testing.T) {
 				},
 			},
 		},
+		{
+			name:     "Ignore First Header Row",
+			filename: "testdata/ignore_header.csv",
+			outcomes: []outcome{
+				{
+					hasError: false,
+					repo: RepoFormat{
+						Repo: "github.com/owner1/repo1",
+					},
+				},
+				{
+					// will error due to GitHub URL sanity check
+					hasError: true,
+					repo: RepoFormat{
+						Repo: "repo",
+					},
+				},
+			},
+		},
+		{
+			name:     "No Header Row",
+			filename: "testdata/no_header.csv",
+			outcomes: []outcome{
+				{
+					hasError: false,
+					repo: RepoFormat{
+						Repo: "github.com/owner1/repo1",
+					},
+				},
+			},
+		},
+		{
+			name:     "Only Header Row",
+			filename: "testdata/only_header.csv",
+			outcomes: []outcome{},
+		},
 	}
 
 	for _, testcase := range testcases {

cron/internal/data/testdata/ignore_header.csv

@@ -0,0 +1,3 @@
+repo,metadata
+github.com/owner1/repo1,
+repo,metadata

cron/internal/data/testdata/no_header.csv

@@ -0,0 +1 @@
+github.com/owner1/repo1,

cron/internal/data/testdata/only_header.csv

@@ -0,0 +1 @@
+repo,metadata