mapper: add cdxu

2 files changed, 22 insertions, 0 deletions

diff --git a/skate/cmd/skate-map/main.go b/skate/cmd/skate-map/main.go
index 5b86250..f87c02f 100644
--- a/skate/cmd/skate-map/main.go
+++ b/skate/cmd/skate-map/main.go
@@ -80,6 +80,7 @@ func main() {
 		"rcns": skate.MapperReleaseContainerName,
 		"vcns": skate.MapperReleaseResolvedContainerName,
 		"isbn": skate.MapperOpenLibraryReleaseNormalizedISBN,
+		"cdxu": skate.MapperCdxSummary,
 	}
 	if *logFile != "" {
 		f, err := os.OpenFile(*logFile, os.O_CREATE|os.O_APPEND, 0644)
diff --git a/skate/map.go b/skate/map.go
index 1a8feef..ef9c018 100644
--- a/skate/map.go
+++ b/skate/map.go
@@ -44,6 +44,17 @@ type PartialRef struct {
 	Year         string `json:"release_year"`
 }
 
+// cdxSummary is an ad-hoc CDX summary format, created currently by skate-cdx-lookup.
+type cdxSummary struct {
+	Line    string `json:"line"`
+	NumRows int64  `json:"numRows"`
+	Summary struct {
+		Delta int64  `json:"delta"`
+		Last  string `json:"last"`
+		Ok    string `json:"ok"`
+	} `json:"summary"`
+}
+
 // Mapper maps a blob to an arbitrary number of fields, e.g. for (key,
 // doc). We want fields, but we do not want to bake in TSV into each function.
 type Mapper func([]byte) ([][]byte, error)
@@ -316,6 +327,16 @@ func MapperOpenLibraryReleaseNormalizedISBN(p []byte) (fields [][]byte, err erro
 	return [][]byte{key, p}, nil
 }
 
+func MapperCdxSummary(p []byte) (fields [][]byte, err error) {
+	var (
+		cdx cdxSummary
+	)
+	if err := json.Unmarshal(p, &cdx); err != nil {
+		return nil, err
+	}
+	return [][]byte{[]byte(cdx.Line), p}, nil
+}
+
 // MapperPartial works on partial documents.
 func MapperPartial(p []byte) (fields [][]byte, err error) {
 	// TODO: Group by some normlized container name or identifier.