Update YTBtoTSV.py

Conversion/ToTSV/YTBtoTSV/YTBtoTSV.py

@@ -13,11 +13,16 @@ def ytbSearch(search, n):
     videosSearch = VideosSearch(search, limit=n)
     result = videosSearch.result()["result"]
     videos = []
-    for video in result:
-        id = video["id"]
-        title = video["title"]
-        author = video["channel"]["name"]
-        videos.append([id, author, title])
+    while len(videos) < n:
+        for video in result:
+            id = video["id"]
+            title = video["title"]
+            author = video["channel"]["name"]
+            videos.append([id, author, title])
+            if len(videos) == n:
+                break
+        search.next()
+        result = search.result()["result"]
     return videos
 
 
@@ -27,7 +32,7 @@ def getLang(list):
 
 
 def translatedTranscript(lang, lst, title, manual):
-    if lang != "fr" and lang != "en":
+    if lang != "en":
         res = lst.find_transcript([lang])
         trans = res.translate("en").fetch()
         print("Subtitles are translated from " +
@@ -59,8 +64,8 @@ def ytbTranscript(id, title):
 def tsvAdd(tsv, abstract, author, title, date, count):
     part = title + " : Part " + str(count)
     tsv += correctedSequence(author, False) + "\t" + correctedSequence(
-        part, False) + "\t" + date + "\t" + "1" + "\t" + "1" + "\t"
-    tsv += correctedSequence(title, False) + "\t"
+        title, False) + "\t" + date + "\t" + "1" + "\t" + "1" + "\t"
+    tsv += correctedSequence(part, False) + "\t"
     tsv += correctedSequence(abstract, True)
     return tsv
 
@@ -71,7 +76,7 @@ def transcriptAutomaticToDoc(transcript, author, title, date):
     count = 1
     tmp = ""
     for part in transcript:
-        tmp += part["text"] + " "
+        tmp += part["text"] + "\n"
         time += int(part["duration"])
         if time >= 20:
             tsv = tsvAdd(tsv, tmp, author, title, date, count)
@@ -117,7 +122,7 @@ def transcriptManualToDoc(transcript, author, title, date):
                 tmp += text
                 time += int(part["duration"])
         else:
-            tmp += text + " "
+            tmp += text + "\n"
             time += int(part["duration"])
     with open("tsv/" + title + ".tsv", "w", encoding='utf-8-sig') as file:
         file.write(tsv)
@@ -139,7 +144,7 @@ def correctedSequence(text, last):
 
 def transcriptToTsv(search, nbVideos):
     tsv = "authors\tsource\tpublication_year\tpublication_month\tpublication_day\ttitle\tabstract\n"
-    videos = ytbSearch(search, nbVideos)
+    videos = ytbSearch(search, nbVideos * 4)
     for video in videos:
         id, author, title = video[0], video[1], video[2]
         print("Processing conversion for video : " + title)