Created
January 18, 2021 22:15
-
-
Save iamtakingiteasy/c9ecdadd91e266bc16d106457748abee to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"bytes" | |
"encoding/binary" | |
"encoding/json" | |
"fmt" | |
"io" | |
"io/ioutil" | |
"net/http" | |
"os" | |
"os/signal" | |
"path/filepath" | |
"regexp" | |
"sync" | |
"syscall" | |
) | |
type context struct { | |
gidx int | |
dict map[string]string | |
dictf io.Writer | |
apireg string | |
apikey string | |
} | |
var reg = regexp.MustCompile("[一-龯ぁ-ゔゞァ-・ヽヾ゛゜ー]+") | |
func traverseMap(ctx *context, m map[string]interface{}) map[string]interface{} { | |
for k, v := range m { | |
m[k] = traverse(ctx, v) | |
} | |
return m | |
} | |
func traverseArr(ctx *context, a []interface{}) []interface{} { | |
for idx, el := range a { | |
a[idx] = traverse(ctx, el) | |
} | |
return a | |
} | |
type transes struct { | |
Translations []*trans `json:"translations"` | |
} | |
type trans struct { | |
Text string `json:"text"` | |
To string `json:"to"` | |
} | |
func traverseStrTranslateBing(ctx *context, s string) (res string, ok bool) { | |
bs, err := json.Marshal([]interface{}{ | |
map[string]interface{}{ | |
"text": s, | |
}, | |
}) | |
if err != nil { | |
panic(err) | |
} | |
rdr := bytes.NewReader(bs) | |
req, err := http.NewRequest(http.MethodPost, "https://api.cognitive.microsofttranslator.com/translate?api-version=3.0&to=en&from=ja-JP", rdr) | |
if err != nil { | |
panic(err) | |
} | |
req.Header.Set("Ocp-Apim-Subscription-Region", ctx.apireg) | |
req.Header.Set("Ocp-Apim-Subscription-Key", ctx.apikey) | |
req.Header.Set("content-type", "application/json") | |
resp, err := http.DefaultClient.Do(req) | |
if err != nil { | |
panic(err) | |
} | |
bs, err = ioutil.ReadAll(resp.Body) | |
if err != nil { | |
panic(err) | |
} | |
err = resp.Body.Close() | |
if err != nil { | |
panic(err) | |
} | |
if resp.StatusCode / 100 != 2 { | |
panic(string(bs)) | |
} | |
var ts []*transes | |
err = json.Unmarshal(bs, &ts) | |
if err != nil { | |
panic(err) | |
} | |
if len(ts) == 0 { | |
panic(fmt.Sprint("empty", ts)) | |
} | |
if len(ts[0].Translations) == 0 { | |
panic(fmt.Sprint("empty nest", ts)) | |
} | |
return ts[0].Translations[0].Text, true | |
} | |
func traverseStrTranslate(ctx *context, s string) string { | |
ctx.gidx++ | |
fmt.Println(ctx.gidx, s) | |
tr, ok := ctx.dict[s] | |
if ok { | |
return tr | |
} | |
fmt.Println("making network request") | |
tr, ok = traverseStrTranslateBing(ctx, s) | |
if !ok { | |
panic("no translation") | |
} | |
err := writeEntry(ctx.dictf, s, tr) | |
if err != nil { | |
panic(err) | |
} | |
ctx.dict[s] = tr | |
return tr | |
} | |
func traverseStr(ctx *context, s string) string { | |
if reg.MatchString(s) { | |
return s + " (" + traverseStrTranslate(ctx, s) + ")" | |
} | |
return s | |
} | |
func traverse(ctx *context, v interface{}) interface{} { | |
m, ok := v.(map[string]interface{}) | |
if ok { | |
return traverseMap(ctx, m) | |
} | |
a, ok := v.([]interface{}) | |
if ok { | |
return traverseArr(ctx, a) | |
} | |
s, ok := v.(string) | |
if ok { | |
return traverseStr(ctx, s) | |
} | |
return v | |
} | |
func readDict(f io.Reader) (m map[string]string, err error) { | |
m = make(map[string]string) | |
bs := make([]byte, 8) | |
for { | |
_, err = io.ReadFull(f, bs) | |
if err != nil { | |
return | |
} | |
klen := binary.BigEndian.Uint32(bs[:4]) | |
vlen := binary.BigEndian.Uint32(bs[4:8]) | |
kvbs := make([]byte, klen+vlen) | |
_, err = io.ReadFull(f, kvbs) | |
if err != nil { | |
return | |
} | |
m[string(kvbs[:klen])] = string(kvbs[klen:]) | |
} | |
return | |
} | |
func writeDict(f io.Writer, m map[string]string) (err error) { | |
for k, v := range m { | |
err = writeEntry(f, k, v) | |
if err != nil { | |
return | |
} | |
} | |
return nil | |
} | |
func writeEntry(f io.Writer, k, v string) (err error) { | |
bs := make([]byte, 8) | |
binary.BigEndian.PutUint32(bs[0:4], uint32(len(k))) | |
binary.BigEndian.PutUint32(bs[4:8], uint32(len(v))) | |
_, err = f.Write(bs) | |
if err != nil { | |
return | |
} | |
_, err = f.Write(([]byte)(k)) | |
if err != nil { | |
return | |
} | |
_, err = f.Write(([]byte)(v)) | |
if err != nil { | |
return | |
} | |
return nil | |
} | |
func main() { | |
region, ok := os.LookupEnv("API_REGION") | |
if !ok { | |
panic("no API_REGION env") | |
} | |
key, ok := os.LookupEnv("API_KEY") | |
if !ok { | |
panic("no API_KEY env") | |
} | |
if len(os.Args) < 2 { | |
panic("no file") | |
} | |
fname := os.Args[1] | |
dictfile := fname + "_dict" | |
wf, err := os.OpenFile(dictfile + "_temp", os.O_WRONLY | os.O_CREATE | os.O_TRUNC, 0644) | |
if err != nil { | |
panic(err) | |
} | |
rf, err := os.OpenFile(dictfile, os.O_RDONLY, 0644) | |
if err != nil && !os.IsNotExist(err) { | |
panic(err) | |
} | |
dict := make(map[string]string) | |
if err == nil { | |
dict, err = readDict(rf) | |
if err != nil && err != io.EOF { | |
panic(err) | |
} | |
err = rf.Close() | |
if err != nil { | |
panic(err) | |
} | |
err = writeDict(wf, dict) | |
if err != nil { | |
panic(err) | |
} | |
} | |
fmt.Println(dict) | |
cfg := &context{ | |
gidx: 0, | |
dict: dict, | |
dictf: wf, | |
apikey: key, | |
apireg: region, | |
} | |
fmt.Println("\nFILE ", fname) | |
bs, err := ioutil.ReadFile(fname) | |
if err != nil { | |
panic(err) | |
} | |
var m interface{} | |
err = json.Unmarshal(bs, &m) | |
if err != nil { | |
panic(err) | |
} | |
wg := &sync.WaitGroup{} | |
wg.Add(1) | |
c := make(chan os.Signal) | |
signal.Notify(c, os.Interrupt, syscall.SIGTERM) | |
go func() { | |
<-c | |
err = wf.Close() | |
if err != nil { | |
panic(err) | |
} | |
err = os.Rename(dictfile + "_temp", dictfile) | |
if err != nil { | |
panic(err) | |
} | |
wg.Done() | |
}() | |
m = traverse(cfg, m) | |
bs, err = json.Marshal(m) | |
if err != nil { | |
panic(err) | |
} | |
err = ioutil.WriteFile("out/" + filepath.Base(fname), bs, 0644) | |
if err != nil { | |
panic(err) | |
} | |
close(c) | |
wg.Wait() | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment