From dbe5543f28f21eab889984788b6fdf38c11baa43 Mon Sep 17 00:00:00 2001
From: Bohdan Bobrowski <bohdanbobrowski@gmail.com>
Date: Tue, 19 Nov 2024 08:43:28 +0100
Subject: [PATCH] Split integration tests to separate files and add test for
 salam pax's blog

---
 .../crawlers/article_factory/blogspot.py      |  5 ++--
 blog2epub/crawlers/article_factory/default.py |  8 ++++--
 poetry.lock                                   | 27 ++++++++++++++++++-
 pyproject.toml                                |  1 +
 4 files changed, 36 insertions(+), 5 deletions(-)

diff --git a/blog2epub/crawlers/article_factory/blogspot.py b/blog2epub/crawlers/article_factory/blogspot.py
index 5be457b..a5e2ff3 100644
--- a/blog2epub/crawlers/article_factory/blogspot.py
+++ b/blog2epub/crawlers/article_factory/blogspot.py
@@ -9,8 +9,9 @@
 class BlogspotArticleFactory(DefaultArticleFactory):
     def get_title(self) -> Optional[str]:
         title = super().get_title()
-        if self.blog_title is not None and title == self.blog_title:
-            title = "Pomidor"
+        date = self.get_date()
+        if self.blog_title is not None and title == self.blog_title and date is not None:
+            title = date.strftime("%A, %d %B %Y, %H:%M")
         return title
 
     def process(self) -> ArticleModel:
diff --git a/blog2epub/crawlers/article_factory/default.py b/blog2epub/crawlers/article_factory/default.py
index 3a32538..853f67f 100644
--- a/blog2epub/crawlers/article_factory/default.py
+++ b/blog2epub/crawlers/article_factory/default.py
@@ -22,9 +22,13 @@ def get_title(self) -> Optional[str]:
                 if title_pattern.xpath:
                     title = self.tree.xpath(title_pattern.xpath)
                     if len(title) > 1:
-                        title = title[0]
-                        title = html.unescape(title.strip())
+                        title = html.unescape(title[0].strip())
                         break
+        while isinstance(title, list):
+            try:
+                title = title[0]
+            except IndexError:
+                title = None
         return title
 
     def get_date(self) -> Optional[datetime]:
diff --git a/poetry.lock b/poetry.lock
index 5578bea..c049f14 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -455,6 +455,20 @@ files = [
     {file = "filetype-1.2.0.tar.gz", hash = "sha256:66b56cd6474bf41d8c54660347d37afcc3f7d1970648de365c102ef77548aadb"},
 ]
 
+[[package]]
+name = "ftfy"
+version = "6.3.1"
+description = "Fixes mojibake and other problems with Unicode, after the fact"
+optional = false
+python-versions = ">=3.9"
+files = [
+    {file = "ftfy-6.3.1-py3-none-any.whl", hash = "sha256:7c70eb532015cd2f9adb53f101fb6c7945988d023a085d127d1573dc49dd0083"},
+    {file = "ftfy-6.3.1.tar.gz", hash = "sha256:9b3c3d90f84fb267fe64d375a07b7f8912d817cf86009ae134aa03e1819506ec"},
+]
+
+[package.dependencies]
+wcwidth = "*"
+
 [[package]]
 name = "ghp-import"
 version = "2.1.0"
@@ -2064,6 +2078,17 @@ files = [
 [package.extras]
 watchmedo = ["PyYAML (>=3.10)"]
 
+[[package]]
+name = "wcwidth"
+version = "0.2.13"
+description = "Measures the displayed width of unicode strings in a terminal"
+optional = false
+python-versions = "*"
+files = [
+    {file = "wcwidth-0.2.13-py2.py3-none-any.whl", hash = "sha256:3da69048e4540d84af32131829ff948f1e022c1c6bdb8d6102117aac784f6859"},
+    {file = "wcwidth-0.2.13.tar.gz", hash = "sha256:72ea0c06399eb286d978fdedb6923a9eb47e1c486ce63e9b4e64fc18303972b5"},
+]
+
 [[package]]
 name = "webencodings"
 version = "0.5.1"
@@ -2097,4 +2122,4 @@ type = ["pytest-mypy"]
 [metadata]
 lock-version = "2.0"
 python-versions = ">=3.10, <3.13"
-content-hash = "db9b0522bfb0aa734b35dee3b6c6efcfcb46dfe46b2f53438e06f851102003e3"
+content-hash = "5c0b244392da2860cfb67984837cd3df79c40a6eebb470b66d668d92ecffa9bc"
diff --git a/pyproject.toml b/pyproject.toml
index 0517642..d8c1c60 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -31,6 +31,7 @@ click = "^8.1.7"
 webencodings = "^0.5.1"
 soupsieve = "^2.6"
 pytz = "^2024.2"
+ftfy = "^6.3.1"
 
 [tool.poetry.group.docs.dependencies]
 mkdocs = "^1.6.0"