Improved feedback to the user

gradio_app.py

@@ -13,17 +13,22 @@ translator = SalamandraTA7bTranslator(hf_token)
 
 def upload_file(filepath, source_lang, target_lang):
     aligner = Aligner(config_folder, source_lang, target_lang, temp_folder)
-    translated_file_name = translate_document(filepath, source_lang, target_lang, translator, aligner)
-    return [gr.UploadButton(visible=False),
-            gr.DownloadButton(label=f"Download {translated_file_name}", value=translated_file_name, visible=True,
-                              interactive=True)]
+    for status, translated_file_name in translate_document(filepath, source_lang, target_lang, translator, aligner):
+        if translated_file_name:  # finished
+            yield [gr.UploadButton(visible=False),
+                   gr.DownloadButton(label=f"Download {translated_file_name}", value=translated_file_name,
+                                     visible=True, interactive=True),
+                   gr.Textbox(visible=False)]
+        else:
+            yield [gr.UploadButton(visible=False),
+                   gr.DownloadButton(visible=False),
+                   gr.Textbox(value=status, visible=True)]
 
 
 def before_processing():
     return [
         gr.UploadButton(visible=False),
-        gr.DownloadButton("Processing...", visible=True, interactive=False)
-        # Keep download hidden until processing finishes
+        gr.Textbox(value="Processing...", visible=True),
     ]
 
 
@@ -42,8 +47,11 @@ with gr.Blocks() as demo:
         with gr.Row():
             u = gr.UploadButton("Upload a file", file_count="single")
             d = gr.DownloadButton("Download the file", visible=False)
+            status_text = gr.Textbox(label="Status", visible=False)
 
-        u.upload(fn=before_processing, inputs=None, outputs=[u, d]).then(upload_file, [u, dropdown1, dropdown2], [u, d])
+        u.upload(fn=before_processing, inputs=None, outputs=[u, status_text]).then(upload_file,
+                                                                                   [u, dropdown1, dropdown2],
+                                                                                   [u, d, status_text])
         d.click(download_file, None, [u, d])
 if __name__ == "__main__":
     demo.launch(server_name="0.0.0.0", server_port=7860)

src/translate_any_doc.py

@@ -1,5 +1,6 @@
 import shutil
 import string
+import sys
 import time
 import os
 from itertools import groupby
@@ -12,6 +13,8 @@ import glob
 import spacy
 from spacy.tokens import Doc
 
+from gradio_client.exceptions import AppError
+
 import tqdm
 
 # Load multilingual model to use as sentence tokenizer
@@ -366,7 +369,8 @@ def translate_document(input_file: str, source_lang: str, target_lang: str,
                        translator,
                        aligner: Aligner,
                        temp_folder: str = "tmp",
-                       tikal_folder: str = "okapi-apps_gtk2-linux-x86_64_1.47.0", with_format: bool = True) -> str:
+                       tikal_folder: str = "okapi-apps_gtk2-linux-x86_64_1.47.0", with_format: bool = True) -> (str,
+                                                                                                                str):
     input_filename = input_file.split("/")[-1]
     os.makedirs(temp_folder, exist_ok=True)
 
@@ -390,20 +394,28 @@ def translate_document(input_file: str, source_lang: str, target_lang: str,
         original_spacing += spaces
 
     translated_sentences = []
-    for sentence, spacing in tqdm.tqdm(zip(original_tokenized_sentences_with_style, original_spacing),
-                                       desc="Translating paragraphs...",
-                                       total=len(original_tokenized_sentences_with_style)):
+    yield "Translating 0%...", None
+    total = len(original_tokenized_sentences_with_style)
+    pbar = tqdm.tqdm(desc="Translating paragraphs...", total=total)
+
+    for i, (sentence, spacing) in enumerate(zip(original_tokenized_sentences_with_style, original_spacing)):
         text = Doc(spacy_nlp.vocab, words=[token["text"] for token in sentence], spaces=spacing).text
 
         while True:
             try:
                 translated_sentences.append(translator.translate(text, source_lang, target_lang))
                 break
-            except:
-                continue
+            except AppError as e:
+                print(e)
+                sys.exit()
+
+        pbar.update(1)
+        percent_complete = int(((i + 1) / total) * 100)
+        yield f"Translating {percent_complete}%...", None
 
     # time to align the translation with the original
     print("Generating alignments...")
+    yield "Aligning...", None
     start_time = time.time()
     translated_sentences_with_style, translated_sentences_spacing = generate_alignments(
         original_tokenized_sentences_with_style,
@@ -468,4 +480,4 @@ def translate_document(input_file: str, source_lang: str, target_lang: str,
     translated_file_path = re.search(r'(?<=Output:\s)(.*)', output)[0]
 
     print(f"Saved file in {translated_file_path}")
-    return translated_file_path
+    yield "", translated_file_path