Merge pull request #22 from caugner/batch-translation

Add support for batch translation

README.md

@@ -92,6 +92,7 @@ docker-compose up -d --build
 | --port        | Set port to bind the server to | `5000`               |
 | --char-limit        | Set character limit | `No limit`               |
 | --req-limit        | Set maximum number of requests per minute per client | `No limit`               |
+| --batch-limit        | Set maximum number of texts to translate in a batch request | `No limit`               |
 | --ga-id        | Enable Google Analytics on the API client page by providing an ID | `No tracking`               |
 | --debug      | Enable debug environment | `False`           |
 | --ssl        | Whether to enable SSL | `False`               |

app/app.py

@@ -13,10 +13,10 @@ def get_remote_address():
 
     return ip
 
-def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_language_source="en", frontend_language_target="en"):
+def create_app(char_limit=-1, req_limit=-1, batch_limit=-1, ga_id=None, debug=False, frontend_language_source="en", frontend_language_target="en"):
     from app.init import boot
     boot()
-    
+
     from app.language import languages
     app = Flask(__name__)
 
@@ -36,9 +36,9 @@ def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_la
         })
     else:
         frontend_argos_language_source = next(iter([l for l in languages if l.code == frontend_language_source]), None)
-    
+
     frontend_argos_language_target = next(iter([l for l in languages if l.code == frontend_language_target]), None)
-    
+
     # Raise AttributeError to prevent app startup if user input is not valid.
     if frontend_argos_language_source is None:
         raise AttributeError(f"{frontend_language_source} as frontend source language is not supported.")
@@ -126,17 +126,20 @@ def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_la
           - in: formData
             name: q
             schema:
-              type: string
+              oneOf:
-              example: Hello world!
+                - type: string
+                  example: Hello world!
+                - type: array
+                  example: ['Hello world!']
             required: true
-            description: Text to translate
+            description: Text(s) to translate
           - in: formData
             name: source
             schema:
               type: string
               example: en
             required: true
-            description: Source language code      
+            description: Source language code
           - in: formData
             name: target
             schema:
@@ -152,8 +155,10 @@ def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_la
               type: object
               properties:
                 translatedText:
-                  type: string
+                  oneOf:
-                  description: Translated text
+                    - type: string
+                    - type: array
+                  description: Translated text(s)
           400:
             description: Invalid request
             schema:
@@ -200,8 +205,21 @@ def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_la
         if not target_lang:
             abort(400, description="Invalid request: missing target parameter")
 
+        batch = isinstance(q, list)
+
+        if batch and batch_limit != -1:
+          batch_size = len(q)
+          if batch_limit < batch_size:
+            abort(400, description="Invalid request: Request (%d) exceeds text limit (%d)" % (batch_size, batch_limit))
+
         if char_limit != -1:
-            q = q[:char_limit]
+            if batch:
+              chars = sum([len(text) for text in q])
+            else:
+              chars = len(q)
+
+            if char_limit < chars:
+              abort(400, description="Invalid request: Request (%d) exceeds character limit (%d)" % (chars, char_limit))
 
         if source_lang == 'auto':
             candidate_langs = list(filter(lambda l: l.lang in language_map, detect_langs(q)))
@@ -217,20 +235,24 @@ def create_app(char_limit=-1, req_limit=-1, ga_id=None, debug=False, frontend_la
                     source_lang = 'en'
             else:
                 source_lang = 'en'
-            
+
             if debug:
                 print("Auto detected: %s" % source_lang)
 
         src_lang = next(iter([l for l in languages if l.code == source_lang]), None)
         tgt_lang = next(iter([l for l in languages if l.code == target_lang]), None)
-        
+
         if src_lang is None:
             abort(400, description="%s is not supported" % source_lang)
         if tgt_lang is None:
             abort(400, description="%s is not supported" % target_lang)
 
         translator = src_lang.get_translation(tgt_lang)
+
         try:
+          if batch:
+            return jsonify({"translatedText": [translator.translate(text) for text in q] })
+          else:
             return jsonify({"translatedText": translator.translate(q) })
         except Exception as e:
             abort(500, description="Cannot translate text: %s" % str(e))

app/templates/index.html

@@ -247,9 +247,9 @@
 window.Prism = window.Prism || {};
 window.Prism.manual = true;
 </script>
-	
+
 <script src="https://cdnjs.cloudflare.com/ajax/libs/prism/1.22.0/prism.min.js"></script>
-	
+
 <script>
 // API host/endpoint
 var BaseUrl = window.location.protocol + "//" + window.location.host;
@@ -317,7 +317,7 @@ document.addEventListener('DOMContentLoaded', function(){
 				return;
 			}
 
-		    self.loading = false;			
+		    self.loading = false;
 		  } else {
 			self.error = "Cannot load /languages";
 			self.loading = false;
@@ -346,8 +346,8 @@ document.addEventListener('DOMContentLoaded', function(){
 			if (this.charactersLimit !== -1 && this.inputText.length >= this.charactersLimit){
 				this.inputText = this.inputText.substring(0, this.charactersLimit);
 			}
-            
+
-			
+
 	  },
 	  computed: {
 	  	requestCode: function(){
@@ -423,7 +423,7 @@ document.addEventListener('DOMContentLoaded', function(){
 				  	var res = JSON.parse(this.response);
 					// Success!
 					if (res.translatedText !== undefined){
-						self.translatedText = res.translatedText; 
+						self.translatedText = res.translatedText;
 					    self.loadingTranslation = false;
 					    self.output = JSON.stringify(res, null, 4);
 					}else{
@@ -450,7 +450,7 @@ document.addEventListener('DOMContentLoaded', function(){
 			this.$refs.translatedTextarea.select();
 			this.$refs.translatedTextarea.setSelectionRange(0, 9999999); /* For mobile devices */
 			document.execCommand("copy");
-			
+
 			if (this.copyTextLabel === "Copy Text"){
 				this.copyTextLabel = "Copied!";
 				var self = this;

main.py

@@ -10,6 +10,8 @@ parser.add_argument('--char-limit', default=-1, type=int, metavar="<number of ch
                     help='Set character limit (%(default)s)')
 parser.add_argument('--req-limit', default=-1, type=int, metavar="<number>",
                     help='Set maximum number of requests per minute per client (%(default)s)')
+parser.add_argument('--batch-limit', default=-1, type=int, metavar="<number of texts>",
+                    help='Set maximum number of texts to translate in a batch request (%(default)s)')
 parser.add_argument('--ga-id', type=str, default=None, metavar="<GA ID>",
                     help='Enable Google Analytics on the API client page by providing an ID (%(default)s)')
 parser.add_argument('--debug', default=False, action="store_true",
@@ -25,8 +27,9 @@ args = parser.parse_args()
 
 
 if __name__ == "__main__":
-    app = create_app(char_limit=args.char_limit, 
+    app = create_app(char_limit=args.char_limit,
                      req_limit=args.req_limit,
+                     batch_limit=args.batch_limit,
                      ga_id=args.ga_id,
                      debug=args.debug,
                      frontend_language_source=args.frontend_language_source,