Refactor: move io funcitons in separate file

3 files changed, 85 insertions, 79 deletions

diff --git a/io_helpers.go b/io_helpers.go
new file mode 100644
index 0000000..3bf9600
--- /dev/null
+++ b/io_helpers.go
@@ -0,0 +1,82 @@
+package main
+
+import (
+	"bufio"
+	"encoding/csv"
+	"encoding/json"
+	"fmt"
+	"io/ioutil"
+	"os"
+)
+
+func readLines(filepath string) []string {
+	file, err := os.Open(filepath)
+	if err != nil {
+		panic(err)
+	}
+	defer file.Close()
+
+	resp := []string{}
+	scanner := bufio.NewScanner(file)
+	for scanner.Scan() {
+		resp = append(resp, scanner.Text())
+	}
+
+	if err := scanner.Err(); err != nil {
+		panic(err)
+	}
+	return resp
+}
+
+// writeLines writes the lines to the given file.
+func writeLines(lines []string, path string) error {
+	file, err := os.Create(path)
+	if err != nil {
+		return err
+	}
+	defer file.Close()
+
+	w := bufio.NewWriter(file)
+	for _, line := range lines {
+		fmt.Fprintln(w, line)
+	}
+	return w.Flush()
+}
+
+func readJson(filepath string) map[string]string {
+	data := make(map[string]string)
+	plan, err := ioutil.ReadFile(filepath)
+	if err != nil {
+		return data
+	}
+	err = json.Unmarshal(plan, &data)
+	if err != nil {
+		panic(err)
+	}
+	return data
+}
+
+func writeJson(data map[string]string) {
+	metadataJson, _ := json.MarshalIndent(data, "", "  ")
+	err := ioutil.WriteFile(metadataPath, metadataJson, 0644)
+	if err != nil {
+		panic(err)
+	}
+}
+
+func writeCSV(data [][]string) {
+	f, err := os.Create(metadataPathCSV)
+	defer f.Close()
+
+	if err != nil {
+		panic(err)
+	}
+
+	w := csv.NewWriter(f)
+	w.Comma = '\t'
+	defer w.Flush()
+
+	if err := w.WriteAll(data); err != nil {
+		panic(err)
+	}
+}
diff --git a/main.go b/main.go
index a2eae60..13729b9 100644
--- a/main.go
+++ b/main.go
@@ -1,18 +1,15 @@
 package main
 
 import (
-	"bufio"
-	"encoding/csv"
-	"encoding/json"
 	"flag"
 	"fmt"
-	"io/ioutil"
 	"os"
 	"path"
 	"path/filepath"
 	"strings"
 )
 
+// TODO: to config file
 const (
 	subExt          = ".vtt"
 	outdir          = "data/utterances"
@@ -76,7 +73,7 @@ func linesToUtterances(lines []string, fd *FileData) []*Utterance {
 			RightTime: strings.TrimSpace(splitted[1]),
 			FD:        fd,
 		}
-		u.OutPath = fmt.Sprintf("%s/%s_%s_%s.wav", outdir, fd.AudioBase,
+		u.OutPath = fmt.Sprintf("%s/%s_%s_%s.opus", outdir, fd.AudioBase,
 			u.LeftTime, u.RightTime)
 
 		if u.LeftTime == u.RightTime {
@@ -88,80 +85,6 @@ func linesToUtterances(lines []string, fd *FileData) []*Utterance {
 	return resp
 }
 
-func readLines(filepath string) []string {
-	file, err := os.Open(filepath)
-	if err != nil {
-		panic(err)
-	}
-	defer file.Close()
-
-	resp := []string{}
-	scanner := bufio.NewScanner(file)
-	for scanner.Scan() {
-		resp = append(resp, scanner.Text())
-	}
-
-	if err := scanner.Err(); err != nil {
-		panic(err)
-	}
-	return resp
-}
-
-// writeLines writes the lines to the given file.
-func writeLines(lines []string, path string) error {
-	file, err := os.Create(path)
-	// file, err := os.OpenFile(path,
-	// 	os.O_APPEND|os.O_WRONLY|os.O_CREATE, 0666)
-	if err != nil {
-		return err
-	}
-	defer file.Close()
-
-	w := bufio.NewWriter(file)
-	for _, line := range lines {
-		fmt.Fprintln(w, line)
-	}
-	return w.Flush()
-}
-
-func readJson(filepath string) map[string]string {
-	data := make(map[string]string)
-	plan, err := ioutil.ReadFile(filepath)
-	if err != nil {
-		return data
-	}
-	err = json.Unmarshal(plan, &data)
-	if err != nil {
-		panic(err)
-	}
-	return data
-}
-
-func writeJson(data map[string]string) {
-	metadataJson, _ := json.MarshalIndent(data, "", "  ")
-	err := ioutil.WriteFile(metadataPath, metadataJson, 0644)
-	if err != nil {
-		panic(err)
-	}
-}
-
-func writeCSV(data [][]string) {
-	f, err := os.Create(metadataPathCSV)
-	defer f.Close()
-
-	if err != nil {
-		panic(err)
-	}
-
-	w := csv.NewWriter(f)
-	w.Comma = '\t'
-	defer w.Flush()
-
-	if err := w.WriteAll(data); err != nil {
-		panic(err)
-	}
-}
-
 func buildFFmpegCall(ut *Utterance) string {
 	return fmt.Sprintf(
 		`yes no | ffmpeg -i "%s" -ss %s -to %s \
diff --git a/readme.md b/readme.md
index 8e5c4e7..a79b840 100644
--- a/readme.md
+++ b/readme.md
@@ -6,3 +6,4 @@
 - add config file; move constance to config file;
 - ffmpeg call should be unique;
 - ffmpeg commands are slow, split to clips inside of this module;
+- wav take a lot of space; no need for conversion, pytorch supports opus (convert to opus);