From fc94c57d7f4da9043fa1a817562bb3104d66a0de Mon Sep 17 00:00:00 2001
From: dgtlmoon <dgtlmoon@gmail.com>
Date: Mon, 5 Dec 2022 16:36:00 +0100
Subject: [PATCH] Extract text as CSV - Extra validation (#1192)

---
 changedetectionio/__init__.py | 35 +++++++++++++++++++----------------
 changedetectionio/forms.py    |  3 +--
 2 files changed, 20 insertions(+), 18 deletions(-)

diff --git a/changedetectionio/__init__.py b/changedetectionio/__init__.py
index 167e1ffb5..4aa69713b 100644
--- a/changedetectionio/__init__.py
+++ b/changedetectionio/__init__.py
@@ -814,11 +814,12 @@ def changedetection_app(config=None, datastore_o=None):
     @login_required
     def diff_history_page(uuid):
 
+        from changedetectionio import forms
+
         # More for testing, possible to return the first/only
         if uuid == 'first':
             uuid = list(datastore.data['watching'].keys()).pop()
 
-
         extra_stylesheets = [url_for('static_content', group='styles', filename='diff.css')]
         try:
             watch = datastore.data['watching'][uuid]
@@ -827,21 +828,26 @@ def changedetection_app(config=None, datastore_o=None):
             return redirect(url_for('index'))
 
         # For submission of requesting an extract
+        extract_form = forms.extractDataForm(request.form)
         if request.method == 'POST':
-            extract_regex = request.form.get('extract_regex').strip()
-            output = watch.extract_regex_from_all_history(extract_regex)
-            if output:
-                watch_dir = os.path.join(datastore_o.datastore_path, uuid)
-                response = make_response(send_from_directory(directory=watch_dir, path=output, as_attachment=True))
-                response.headers['Content-type'] = 'text/csv'
-                response.headers['Cache-Control'] = 'no-cache, no-store, must-revalidate'
-                response.headers['Pragma'] = 'no-cache'
-                response.headers['Expires'] = 0
-                return response
+            if not extract_form.validate():
+                flash("An error occurred, please see below.", "error")
+
+            else:
+                extract_regex = request.form.get('extract_regex').strip()
+                output = watch.extract_regex_from_all_history(extract_regex)
+                if output:
+                    watch_dir = os.path.join(datastore_o.datastore_path, uuid)
+                    response = make_response(send_from_directory(directory=watch_dir, path=output, as_attachment=True))
+                    response.headers['Content-type'] = 'text/csv'
+                    response.headers['Cache-Control'] = 'no-cache, no-store, must-revalidate'
+                    response.headers['Pragma'] = 'no-cache'
+                    response.headers['Expires'] = 0
+                    return response
 
 
-            flash('Nothing matches that RegEx', 'error')
-            redirect(url_for('diff_history_page', uuid=uuid)+'#extract')
+                flash('Nothing matches that RegEx', 'error')
+                redirect(url_for('diff_history_page', uuid=uuid)+'#extract')
 
         history = watch.history
         dates = list(history.keys())
@@ -884,9 +890,6 @@ def changedetection_app(config=None, datastore_o=None):
         is_html_webdriver = True if watch.get('fetch_backend') == 'html_webdriver' or (
                     watch.get('fetch_backend', None) is None and system_uses_webdriver) else False
 
-        from changedetectionio import forms
-        extract_form = forms.extractDataForm(request.form)
-
         output = render_template("diff.html",
                                  current_diff_url=watch['url'],
                                  current_previous_version=str(previous_version),
diff --git a/changedetectionio/forms.py b/changedetectionio/forms.py
index e127c5e01..06aedab2b 100644
--- a/changedetectionio/forms.py
+++ b/changedetectionio/forms.py
@@ -451,6 +451,5 @@ class globalSettingsForm(Form):
 
 
 class extractDataForm(Form):
-    extract_regex = StringField('RegEx to extract')
+    extract_regex = StringField('RegEx to extract', validators=[validators.Length(min=1, message="Needs a RegEx")])
     extract_submit_button = SubmitField('Extract as CSV', render_kw={"class": "pure-button pure-button-primary"})
-