From 8c79609f1646fce6c1a8748c24dbe3a2cde670a0 Mon Sep 17 00:00:00 2001
From: Carlos Wu <carlos.fei@canonical.com>
Date: Mon, 6 Dec 2021 10:13:53 +0000
Subject: [PATCH] Fix tutorials key errors (#117)

---
 .../discourse/parsers/base_parser.py          | 26 ++++++++++------
 .../discourse/parsers/tutorials.py            | 31 ++++++++++++++++---
 setup.py                                      |  5 ++-
 3 files changed, 47 insertions(+), 15 deletions(-)

diff --git a/canonicalwebteam/discourse/parsers/base_parser.py b/canonicalwebteam/discourse/parsers/base_parser.py
index 7edeaa0..f1b0256 100644
--- a/canonicalwebteam/discourse/parsers/base_parser.py
+++ b/canonicalwebteam/discourse/parsers/base_parser.py
@@ -306,11 +306,14 @@ def _parse_metadata(self, index_soup, section_name):
                                 value.contents[0] = value.find("a").text
 
                         else:
-                            error_message = "Warning: Link not found when "
-                            f"parsing row {index + 1}"
-                            f"\"{row_dict['topic_name']}\" {titles[index]}."
-                            "This row has been skipped."
-                            self.metadata_errors.append(error_message)
+                            error_message = (
+                                f"Warning: Link not found when parsing row"
+                                f' {index + 1} "{row_dict["topic_name"]}"'
+                                f" {titles[index]}. This row has been skipped."
+                            )
+                            if error_message not in self.metadata_errors:
+                                self.metadata_errors.append(error_message)
+
                             row_dict = None
                             MissingContentError(error_message)
                             break
@@ -324,10 +327,15 @@ def _parse_metadata(self, index_soup, section_name):
                     if (
                         (titles[index] == "path") or (titles[index] == "type")
                     ) and ((value.text == "") or (value.text is None)):
-                        error_message = "Warning: Link not found when"
-                        f" parsing row {index + 1} {row_dict['topic_name']}\""
-                        f"{titles[index]}. This row has been skipped."
-                        self.metadata_errors.append(error_message)
+                        error_message = (
+                            f"Warning: Title not found when parsing row"
+                            f' {index + 1} "{row_dict["topic_name"]}"'
+                            f" {titles[index]}."
+                        )
+
+                        if error_message not in self.metadata_errors:
+                            self.metadata_errors.append(error_message)
+
                         row_dict = None
                         MissingContentError(error_message)
                         break
diff --git a/canonicalwebteam/discourse/parsers/tutorials.py b/canonicalwebteam/discourse/parsers/tutorials.py
index c63b5bb..1b382e1 100644
--- a/canonicalwebteam/discourse/parsers/tutorials.py
+++ b/canonicalwebteam/discourse/parsers/tutorials.py
@@ -1,3 +1,5 @@
+import flask
+
 # Packages
 from bs4 import BeautifulSoup
 from datetime import datetime, timedelta
@@ -5,11 +7,13 @@
 # Local
 from canonicalwebteam.discourse.parsers.base_parser import BaseParser
 
+allowed_tutorial_keys = ["summary", "categories", "difficulty", "author"]
+
 
 class TutorialParser(BaseParser):
     def __init__(self, api, index_topic_id, url_prefix):
         self.tutorials = None
-
+        self.errors = []
         return super().__init__(api, index_topic_id, url_prefix)
 
     def parse(self):
@@ -111,6 +115,7 @@ def _get_tutorials_topics(self):
 
         response = self.api.get_topics(topics)
         tutorial_data = []
+        self.errors = []
 
         for topic in response:
             topic_soup = BeautifulSoup(
@@ -132,12 +137,28 @@ def _get_tutorials_topics(self):
             )
 
             metadata = {"id": topic[0], "title": topic[1], "link": link}
+            error_message = None
             for row in rows:
                 key = row.select_one("td:first-child").text.lower()
-                value = row.select_one("td:last-child").text
-                metadata[key] = value
-
-            tutorial_data.append(metadata)
+                # Markdown errors made by discourse users
+                if key not in allowed_tutorial_keys:
+                    error_message = (
+                        f'The tutorial "{topic[1]}" contains an incorrect'
+                        f' key error "{key}", only'
+                        f' {", ".join(allowed_tutorial_keys)} are allowed.'
+                        f" This tutorial has been skipped"
+                    )
+                    flask.current_app.extensions["sentry"].captureMessage(
+                        error_message
+                    )
+                    if error_message not in self.errors:
+                        self.errors.append(error_message)
+                    break
+                else:
+                    value = row.select_one("td:last-child").text
+                    metadata[key] = value
+            if not error_message:
+                tutorial_data.append(metadata)
 
         # Tutorial will be in the same order as in the URLs table
         return sorted(tutorial_data, key=lambda x: topics.index(x["id"]))
diff --git a/setup.py b/setup.py
index 83ac85e..b1f0c1d 100755
--- a/setup.py
+++ b/setup.py
@@ -7,7 +7,10 @@
     version="4.0.6",
     author="Canonical webteam",
     author_email="webteam@canonical.com",
-    url="https://github.com/canonical-webteam/canonicalwebteam.docs",
+    url=(
+        "https://github.com/canonical-web-and-design/"
+        "canonicalwebteam.discourse"
+    ),
     description=(
         "Flask extension to integrate discourse content generated "
         "to docs to your website."