aboutsummaryrefslogtreecommitdiffstats
path: root/skate
diff options
context:
space:
mode:
authorMartin Czygan <martin.czygan@gmail.com>2021-04-29 15:28:38 +0200
committerMartin Czygan <martin.czygan@gmail.com>2021-04-29 15:28:38 +0200
commit5e7543c3f4612dacd7f1d4e04c75cd48ecb67b0c (patch)
tree699f080cb5b8b8d36b2e1eccce7171a70b75651f /skate
parentd03d398cd703ef399254d3ea8e088c59693ef110 (diff)
downloadrefcat-5e7543c3f4612dacd7f1d4e04c75cd48ecb67b0c.tar.gz
refcat-5e7543c3f4612dacd7f1d4e04c75cd48ecb67b0c.zip
no default prefix
Diffstat (limited to 'skate')
-rw-r--r--skate/cmd/skate-map/main.go14
1 files changed, 9 insertions, 5 deletions
diff --git a/skate/cmd/skate-map/main.go b/skate/cmd/skate-map/main.go
index 259e1aa..ee02875 100644
--- a/skate/cmd/skate-map/main.go
+++ b/skate/cmd/skate-map/main.go
@@ -49,6 +49,7 @@ var (
numWorkers = flag.Int("w", runtime.NumCPU(), "number of workers")
batchSize = flag.Int("b", 50000, "batch size")
verbose = flag.Bool("verbose", false, "show progress")
+ keyPrefix = flag.String("p", "", "a key prefix to use")
extraValue = flag.String("x", "", "extra value to pass to configurable mappers")
)
@@ -60,17 +61,20 @@ func main() {
// [ ] maybe wrap jq and parallel for arbitrary nested keys
availableMappers := map[string]skate.Mapper{
"id": skate.Identity,
- "ff": skate.WithPrefix(skate.CreateFixedMapper(*extraValue), "ff"),
- "ti": skate.WithPrefix(skate.MapperTitle, "ti"),
- "tn": skate.WithPrefix(skate.MapperTitleNormalized, "tn"),
- "ty": skate.WithPrefix(skate.MapperTitleNysiis, "ty"),
- "ts": skate.WithPrefix(skate.MapperTitleSandcrawler, "ts"),
+ "ff": skate.CreateFixedMapper(*extraValue),
+ "ti": skate.MapperTitle,
+ "tn": skate.MapperTitleNormalized,
+ "ty": skate.MapperTitleNysiis,
+ "ts": skate.MapperTitleSandcrawler,
}
switch {
case *mapperName != "":
if f, ok := availableMappers[*mapperName]; !ok {
log.Fatalf("unknown mapper name: %v", *mapperName)
} else {
+ if *keyPrefix != "" {
+ f = skate.WithPrefix(f, *keyPrefix)
+ }
pp := parallel.NewProcessor(os.Stdin, os.Stdout, f.AsTSV)
pp.NumWorkers = *numWorkers
pp.BatchSize = *batchSize