"""Models related to the news section in Tibia.com."""
import datetime
import re
import urllib.parse
from typing import List, Optional
from tibiapy import abc
from tibiapy.enums import NewsCategory, NewsType
from tibiapy.errors import InvalidContent
from tibiapy.utils import get_tibia_url, parse_form_data, parse_tibia_date, parse_tibiacom_content, try_enum
__all__ = (
"News",
"NewsArchive",
"NewsEntry",
)
ICON_PATTERN = re.compile(r"newsicon_([^_]+)_(?:small|big)")
[docs]class NewsArchive(abc.Serializable):
"""Represents the news archive.
.. versionadded:: 5.0.0
Attributes
----------
start_date: :class:`datetime.date`
The start date to show news for.
end_date: :class:`datetime.date`
The end date to show news for.
types: :class:`list` of :class:`NewsType`
The type of news to show.
categories: :class:`list` of :class:`NewsCategory`.
The categories to show.
entries: :class:`list` of :class:`NewsEntry`
The news matching the provided parameters.
"""
__slots__ = (
"start_date",
"end_date",
"types",
"categories",
"entries",
)
def __init__(self, **kwargs):
self.start_date: datetime.date = kwargs.get("start_date")
self.end_date: datetime.date = kwargs.get("end_date")
self.types: List[NewsType] = kwargs.get("types", [])
self.categories: List[NewsCategory] = kwargs.get("categories", [])
self.entries: List[NewsEntry] = kwargs.get("entries", [])
def __repr__(self):
return f"<{self.__class__.__name__} start_date={self.start_date!r} end_date={self.end_date!r}>"
@property
def url(self):
return self.get_url()
@property
def form_data(self):
return self.get_form_data(self.start_date, self.end_date, self.categories, self.types)
[docs] @classmethod
def get_url(cls):
"""Get the URL to Tibia.com's news archive page.
Notes
-----
It is not possible to perform a search using query parameters.
News searches can only be performed using POST requests sending the parameters as form-data.
Returns
-------
:class:`str`
The URL to the news archive page on Tibia.com.
"""
return get_tibia_url("news", "newsarchive")
[docs] @classmethod
def from_content(cls, content):
"""Get a list of news from the HTML content of the news search page.
Parameters
----------
content: :class:`str`
The HTML content of the page.
Returns
-------
:class:`NewsArchive`
The news archive with the news found.
Raises
------
InvalidContent
If content is not the HTML of a news search's page.
"""
try:
parsed_content = parse_tibiacom_content(content)
tables = parsed_content.find_all("table", attrs={"width": "100%"})
news_table, filters_table = tables
form = parsed_content.find("form")
news_archive = cls._parse_filtering(form)
title_row = news_table.find("td", attrs={"class": "white", "colspan": "3"})
if title_row.text != "Search Results":
raise InvalidContent("content is not from the news archive section in Tibia.com")
rows = news_table.find_all("tr", attrs={"class": ["Odd", "Even"]})
for row in rows:
cols_raw = row.find_all('td')
if len(cols_raw) != 3:
continue
entry = cls._parse_entry(cols_raw)
news_archive.entries.append(entry)
return news_archive
except (AttributeError, IndexError, ValueError) as e:
raise InvalidContent("content is not from the news archive section in Tibia.com", e)
@classmethod
def _parse_filtering(cls, form):
form_data = parse_form_data(form)
start_date = datetime.date(
int(form_data.pop("filter_begin_year")),
int(form_data.pop("filter_begin_month")),
int(form_data.pop("filter_begin_day")),
)
end_date = datetime.date(
int(form_data.pop("filter_end_year")),
int(form_data.pop("filter_end_month")),
int(form_data.pop("filter_end_day")),
)
news_types = []
for news_type in NewsType:
value = form_data.pop(news_type.filter_name, None)
if value:
news_types.append(news_type)
categories = []
for category in NewsCategory:
value = form_data.pop(category.filter_name, None)
if value:
categories.append(category)
return cls(start_date=start_date, end_date=end_date, categories=categories, types=news_types)
@classmethod
def _parse_entry(cls, cols_raw):
img = cols_raw[0].find('img')
img_url = img["src"]
category_name = ICON_PATTERN.search(img_url)
category = try_enum(NewsCategory, category_name.group(1))
for br in cols_raw[1].find_all("br"):
br.replace_with("\n")
date_str, news_type_str = cols_raw[1].text.splitlines()
date = parse_tibia_date(date_str)
news_type_str = news_type_str.replace('\xa0', ' ')
news_type = try_enum(NewsType, news_type_str)
title = cols_raw[2].text
news_link = cols_raw[2].find('a')
url = urllib.parse.urlparse(news_link["href"])
query = urllib.parse.parse_qs(url.query)
news_id = int(query["id"][0])
return NewsEntry(news_id, title, news_type, category, date, category_icon=img_url)
[docs]class News(abc.BaseNews, abc.Serializable):
"""Represents a news entry.
Attributes
----------
id: :class:`int`
The internal ID of the news entry.
title: :class:`str`
The title of the news entry.
category: :class:`NewsCategory`
The category this belongs to.
category_icon: :class:`str`
The URL of the icon corresponding to the category.
date: :class:`datetime.date`
The date when the news were published.
content: :class:`str`, optional
The raw html content of the entry.
thread_id: :class:`int`, optional
The thread id of the designated discussion thread for this entry.
"""
def __init__(self, news_id, title, content, date, category, **kwargs):
self.id: int = news_id
self.title: str = title
self.content: str = content
self.date: datetime.date = date
self.category: NewsCategory = category
self.thread_id: Optional[int] = kwargs.get("thread_id", None)
self.category_icon: Optional[str] = kwargs.get("category_icon")
# id, title, category and date inherited from BaseNews.
__slots__ = (
"id",
"title",
"category",
"category_icon",
"date",
"content",
"thread_id",
)
@property
def thread_url(self):
""":class:`str`: The URL to the thread discussing this news entry, if any."""
return abc.BaseThread.get_url(self.thread_id) if self.thread_id else None
[docs] @classmethod
def from_content(cls, content, news_id=0):
"""Get a news entry by its HTML content from Tibia.com.
Notes
-----
Since there's no way to obtain the entry's Id from the page contents, it will always be 0.
A news_id can be passed to set the news_id of the resulting object.
Parameters
----------
content: :class:`str`
The HTML content of the page.
news_id: :class:`int`, optional
The news_id belonging to the content being parsed.
Returns
-------
:class:`News`
The news article found in the page.
Raises
------
InvalidContent
If content is not the HTML of a news' page.
"""
if "(no news with id " in content:
return None
try:
parsed_content = parse_tibiacom_content(content)
# Read Information from the headline
headline = parsed_content.find("div", attrs={"class": "NewsHeadline"})
img = headline.find('img')
img_url = img["src"]
category_name = ICON_PATTERN.search(img_url)
category = try_enum(NewsCategory, category_name.group(1))
title_div = headline.find("div", attrs={"class": "NewsHeadlineText"})
title = title_div.text.replace('\xa0', ' ')
date_div = headline.find("div", attrs={"class": "NewsHeadlineDate"})
date_str = date_div.text.replace('\xa0', ' ').replace('-', '').strip()
date = parse_tibia_date(date_str)
# Read the page's content.
content_table = parsed_content.find("table")
content_row = content_table.find("td")
content = content_row.encode_contents().decode()
thread_id = None
thread_div = content_table.find("div")
if thread_div:
news_link = thread_div.find('a')
url = urllib.parse.urlparse(news_link["href"])
query = urllib.parse.parse_qs(url.query)
thread_id = int(query["threadid"][0])
return cls(news_id, title, content, date, category, thread_id=thread_id, category_icon=img_url)
except AttributeError:
raise InvalidContent("content is not from the news archive section in Tibia.com")
[docs]class NewsEntry(abc.BaseNews, abc.Serializable):
"""A news entry from the news archive.
Attributes
----------
id: :class:`int`
The internal ID of the news entry.
title: :class:`str`
The title of the news entry.
News tickers have a fragment of their content as a title.
category: :class:`NewsCategory`
The category this belongs to.
category_icon: :class:`str`
The URL of the icon corresponding to the category.
date: :class:`datetime.date`
The date when the news were published.
type: :class:`NewsType`
The type of news of this list entry.
"""
__slots__ = (
"id",
"title",
"category",
"category_icon",
"date",
"type",
)
def __init__(self, news_id, title, news_type, category, date, **kwargs):
self.id: int = news_id
self.title: str = title
self.type: NewsType = news_type
self.category: NewsCategory = category
self.date: datetime.datetime = date
self.category_icon: Optional[str] = kwargs.get("category_icon", None)
def __repr__(self):
return f"<{self.__class__.__name__} id={self.id} title={self.title!r} type={self.type!r} " \
f"category={self.category!r} date={self.date!r}>"
