From be739267c7aad3c907da7dfd34bf80165ff56738 Mon Sep 17 00:00:00 2001
From: nglk <diakaggel@gmail.com>
Date: Fri, 1 Apr 2022 11:56:47 +0200
Subject: [PATCH] unsuccessful download file

---
 app.py.copy            | 112 +++++++++++++++++++++++++++++++++++++++++
 parsepdf.py            |   8 +++
 templates/base.html    |   3 +-
 templates/results.html |   2 +-
 4 files changed, 123 insertions(+), 2 deletions(-)
 create mode 100644 app.py.copy
 create mode 100644 parsepdf.py

diff --git a/app.py.copy b/app.py.copy
new file mode 100644
index 0000000..2eecfbb
--- /dev/null
+++ b/app.py.copy
@@ -0,0 +1,112 @@
+import os
+import random
+import shutil
+import string
+import subprocess
+from pathlib import Path
+from flask import Flask, flash, redirect, render_template, request, url_for
+import urllib.request
+from hocrtransformpdf import *
+from werkzeug.utils import secure_filename
+from flask_basicauth import BasicAuth
+import pdftotree
+from parsepdf import hocr_result
+
+UPLOAD_FOLDER = 'static/uploads'
+ALLOWED_EXTENSIONS = {'pdf'}
+
+app = Flask(__name__)
+
+app.config['BASIC_AUTH_USERNAME'] = 'wordmord'
+app.config['BASIC_AUTH_PASSWORD'] = 'tentacles'
+
+basic_auth = BasicAuth(app)
+
+app.config['UPLOAD_FOLDER'] = "static/pdf"
+
+
+@app.route('/', methods=['GET', 'POST'])
+@basic_auth.required
+def run_script():
+    if request.method == 'POST':
+       # download PDF from url
+       file = request.form['file']
+       if file:
+          response = urllib.request.urlopen(file)
+          file = open("static/pdf/downloaded.pdf", 'wb')
+          file.write(response.read())
+          file.close()
+          #hocr = pdftotree.parse("static/pdf/downloaded.pdf")
+          #hocrfile=open("static/hocr/downloaded.hocr", 'wb')   
+          #hocrfile.write(hocr.read())
+          #hocrfile.close
+          #output = execute('python parsepdf.py')
+          #hocr_result = pdftotree.parse("static/pdf/downloaded.pdf")
+          #hocr_result=hocr_result()
+
+#hocr = HocrTransform(hocr_result, 300)
+          #hocr.to_pdf(
+           # out_filename='static/pdf/output-2.pdf',
+            #image_filename='static/images/blank.png',
+            #show_bounding_boxes=False,
+            #interword_spaces=False,
+          #)
+
+          #uploadfilepath=os.path.join(app.config['UPLOAD_FOLDER'], file)
+          #file.save(uploadfilepath)
+          #return redirect(url_for('uploaded_file',file=file))
+       # url = request.form['url']
+        #if not url:
+         #   flash('Url is required!')
+       # else:
+        #    messages.append({'url': url})
+         #   return redirect(url_for('index'))
+       
+        # check if the post request has the file part
+        #if 'file' not in request.files:
+         #   flash('No file part')
+          #  return redirect(request.url)
+        #file = request.files['file']
+        # if user does not select file, browser also
+        # submit an empty part without filename
+        #if file.filename == '':
+         #   flash('No selected file')
+          #  return redirect(request.url)
+        #if file and allowed_file(file.filename):
+         #   filename = secure_filename(file.filename)
+          #  uploadfilepath=os.path.join(app.config['UPLOAD_FOLDER'], filename)
+           # file.save(uploadfilepath)
+            # return redirect(url_for('uploaded_file',
+            #                         filename=filename))
+
+        
+      # hocr_result = pdftotree.parse("static/pdf/downloaded.pdf")
+       #hocr = HocrTransform(hocr_filename=hocr_result, dpi=300)
+       #hocr.to_pdf(
+        # out_filename='static/pdf/output-2.pdf',
+        # image_filename='static/images/blank.png',
+        # show_bounding_boxes=False,
+        # interword_spaces=False
+       #)
+
+    #hocrfile='static/hocr/gynaikoktonia.hocr'
+    #hocr = HocrTransform(hocr_filename=hocrfile, dpi=300)
+    #hocr = HocrTransform(hocr_filename=hocr_result, dpi=300)
+    #hocr.to_pdf(
+    #    out_filename='static/pdf/output.pdf',
+    #    image_filename='static/images/blank.png',
+    #    show_bounding_boxes=False,
+    #     interword_spaces=False,
+    #)
+    #subprocess.check_output("python hocrtransformpdf.py -i static/images/blank.png static/hocr/downloaded.hocr static/pdf/output-2.pdf", shell=True)
+    return render_template('results.html',**locals())
+
+
+def allowed_file(filename):
+    return '.' in filename and \
+        filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
+
+
+if __name__ == "__main__":
+    app.run()
+
diff --git a/parsepdf.py b/parsepdf.py
new file mode 100644
index 0000000..cd26866
--- /dev/null
+++ b/parsepdf.py
@@ -0,0 +1,8 @@
+#!venv/bin python3
+import pdftotree
+
+def hocr_result():
+    hocr_result = pdftotree.parse("static/pdf/downloaded.pdf")
+    with open('static/hocr/downloaded.hocr', 'w') as f:
+        f.write(hocr_result)
+    return hocr_result
diff --git a/templates/base.html b/templates/base.html
index b0c899c..83be2b9 100644
--- a/templates/base.html
+++ b/templates/base.html
@@ -3,7 +3,8 @@
 <head>
     <meta charset="UTF-8"> 
     <meta name="viewport" content="width=device-width, initial-scale=1" />
-    <link
+<script type="text/javascript" src="https://code.jquery.com/jquery-3.6.0.min.js"> </script>    
+<link
       rel="stylesheet"
       href="../static/styles.css"
     />
diff --git a/templates/results.html b/templates/results.html
index 4887939..45695c1 100644
--- a/templates/results.html
+++ b/templates/results.html
@@ -6,7 +6,7 @@
 <table>
   <tr>
     <td colspan="3">
-<form method="post">
+<form method="POST">
       <!-- <input type="text" id="my_field" value="" wrap="soft"> -->
      <textarea name="file" id="my_field" wrap="soft" placeholder="Επικόλληση του συνδέσμου του PDF του Ποινικού Κώδικα που βρήκατε στο ηλεκτρονικό ταχυδρομείο σας ή στο ΔΙΑΒΑΣΕΜΕ.txt, εδώ.
 "></textarea>