Update PDF_to_TSV.py

Streamlit/pages/PDF_to_TSV.py

@@ -65,7 +65,7 @@ def estimateLanguagesPercentage(languages):
     for l in languages:
         total += int(languages[l])
     for l in languages:
-        tmp = int(languages[l]) / total * 100
+        tmp = round(int(languages[l]) / total * 100, 1)
         if tmp > max:
             max = tmp
             principal = l
@@ -167,15 +167,20 @@ def correctedSequence(text, last):
 
 
 def getInfo():
-    return st.session_state.author, st.session_state.title, st.session_state.watermark
+    title = st.session_state.title
+    if title == "":
+        title = st.session_state.fileName.replace(".pdf", "")
+    return st.session_state.author, title, st.session_state.watermark
 
 
 def txtToTSV(fileName, fileAddress, pdfDir):
     st.session_state.page = 1
     author, title, watermark = getInfo()
     tsv = "authors\tsource\tpublication_year\tpublication_month\tpublication_day\ttitle\tabstract\n"
-    tsv, languages = segmentAbstract(fileName, fileAddress, tsv, author, title,
-                                     str(date.today().year), "1", "1", watermark)
+    with st.spinner(st.session_state.general_text_dict['loading']):
+        tsv, languages = segmentAbstract(fileName, fileAddress, tsv, author, title,
+                                         str(date.today().year), "1", "1", watermark)
+
     if '/' in fileName:
         fileName = fileName.split('/')[1]
     with open(pdfDir + "/" + fileName.replace(".pdf", ".tsv"), "w", encoding="utf-8-sig") as file:
@@ -224,6 +229,7 @@ def upPage():
 
 def uploadZip():
     st.write(st.session_state.general_text_dict['text'])
+    st.write(st.session_state.general_text_dict['text2'])
     return st.file_uploader(
         st.session_state.general_text_dict['file'], type=["zip"], key='file')
 
@@ -231,6 +237,7 @@ def uploadZip():
 def askPDF(fileName):
     with st.form("Submit"):
         st.write(fileName)
+        st.write(st.session_state.general_text_dict['text3'])
         col1, col2 = st.columns(2)
         st.session_state.author = ""
         st.session_state.title = ""
@@ -279,6 +286,7 @@ if st.session_state.page == 2:
 
 if st.session_state.page == 1:
     fileName = os.listdir(st.session_state.zipDir.name)[st.session_state.nbDoc]
+    st.session_state.fileName = fileName
     if '/' in fileName:
         fileName = fileName.split('/')[1]
     askPDF(fileName)