196 lines
7.8 KiB
Python
196 lines
7.8 KiB
Python
import mimetypes
|
|
from lxml.etree import Element, CDATA, tostring
|
|
from typing import Optional, List, TypeVar, Dict, Any
|
|
from datetime import datetime
|
|
from collections import namedtuple
|
|
|
|
import pytz
|
|
|
|
__all__ = ('GenRSS', 'Enclosure',)
|
|
|
|
ElementT = TypeVar('ElementT')
|
|
|
|
Enclosure = namedtuple('Enclosure', ('url', 'size', 'type'))
|
|
Enclosure.__new__.__defaults__ = (None, None, None)
|
|
Enclosure.__doc__ = 'Creates information for enclosure tag.'
|
|
|
|
RSS_DEFAULT_GENERATOR = f'Generated by genrss for python'
|
|
|
|
|
|
def create_element(name: str, text: Any = None, children: List[ElementT] = None,
|
|
**kwargs) -> ElementT:
|
|
"""Creates xml node with text or children elements.
|
|
|
|
:param name: Tag name of node with namespace
|
|
:param text: Text of node
|
|
:param children: Appends elements as child nodes
|
|
"""
|
|
el = Element(name, **kwargs)
|
|
if text:
|
|
if isinstance(text, datetime):
|
|
text = text.replace(tzinfo=pytz.timezone('GMT')). \
|
|
strftime("%a, %d %b %Y %H:%M:%S %Z")
|
|
el.text = text
|
|
elif isinstance(children, (list, tuple)):
|
|
for child in children:
|
|
el.append(child)
|
|
return el
|
|
|
|
|
|
class GenRSS:
|
|
"""Generates RSS feed of channel.
|
|
|
|
:param title: Title of your site or feed
|
|
:param site_url: Absolute url to the site that the feed is for
|
|
:param feed_url: Absolute url to the rss feed
|
|
:param description: A short description of feed
|
|
:param image_url: Image absolute url for channel
|
|
:param author: Author of channel
|
|
:param pub_date: Datetime in utc when last item was published
|
|
:param copyright: Copyright information for this feed
|
|
:param language: The language of the content of this feed.
|
|
:param editor: Who manages content in this feed
|
|
:param webmaster: Who manages feed availability and technical support
|
|
:param generator: Feed generator
|
|
"""
|
|
|
|
def __init__(self, title: str, site_url: str, feed_url: str, **kwargs):
|
|
self.title: str = title
|
|
self.site_url: str = site_url
|
|
self.feed_url: str = feed_url
|
|
self.description: str = kwargs.pop('description', self.title)
|
|
self.image_url: Optional[str] = kwargs.pop('image_url', None)
|
|
self.author: Optional[str] = kwargs.pop('author', None)
|
|
self.pub_date: Optional[datetime] = kwargs.pop('pub_date', None)
|
|
self.copyright: Optional[str] = kwargs.pop('copyright', None)
|
|
self.language: Optional[str] = kwargs.pop('language', None)
|
|
self.editor: Optional[str] = kwargs.pop('editor', None)
|
|
self.webmaster: Optional[str] = kwargs.pop('webmaster', None)
|
|
self.docs_url: Optional[str] = kwargs.pop('docs_url', None)
|
|
self.categories: List[str] = kwargs.pop('categories', [])
|
|
|
|
self.items: List[Element] = []
|
|
self.generator: str = kwargs.pop('generator', RSS_DEFAULT_GENERATOR)
|
|
self.root_version: str = '2.0'
|
|
self.root_nsmap: Dict[str, str] = {
|
|
'atom': 'http://www.w3.org/2005/Atom'
|
|
}
|
|
|
|
def item(self, title: str, **kwargs):
|
|
"""Adds item to the feed.
|
|
|
|
An item can be used for recipes, blog entries, project update, log
|
|
entry, etc. Your RSS feed can have any number of items.
|
|
|
|
:param title: Title of this particular item
|
|
:param description: Content for the item. Can contain html but
|
|
link and image urls must be absolute path including hostname
|
|
:param url: Url to the item. This could be a blog entry
|
|
:param guid: A unique string feed readers use to know if an item
|
|
is new or has already been seen. If you use a guid never change
|
|
it. If you don't provide a guid then your item urls must be unique
|
|
:param author: If included it is the name of the item's creator.
|
|
If not provided the item author will be the same as the feed
|
|
author. This is typical except on multi-author blogs
|
|
:param categories: If provided, each array item will be added as a
|
|
category element
|
|
:param enclosure: An enclosure object
|
|
:param pub_date: The date and time of when the item was created.
|
|
Feed readers use this to determine the sort order. Some readers
|
|
will also use it to determine if the content should be presented
|
|
as unread
|
|
"""
|
|
description: str = kwargs.pop('description', '')
|
|
url: Optional[str] = kwargs.pop('url', None)
|
|
guid: Optional[str] = kwargs.pop('guid', None)
|
|
author: Optional[str] = kwargs.pop('author', None)
|
|
categories: List[str] = kwargs.pop('categories', [])
|
|
enclosure: Optional[Enclosure] = kwargs.pop('enclosure', None)
|
|
pub_date: Optional[datetime] = kwargs.pop('pub_date', None)
|
|
|
|
item = create_element('item', children=[
|
|
create_element('title', CDATA(title)),
|
|
create_element('description', CDATA(description)),
|
|
])
|
|
|
|
if url:
|
|
item.append(create_element('link', url))
|
|
|
|
item.append(create_element(
|
|
'guid',
|
|
attrib={'isPermaLink': str(bool(not guid and url)).lower()},
|
|
text=(guid or url or CDATA(title))
|
|
))
|
|
|
|
if author or self.author:
|
|
if 'dc' not in self.root_nsmap:
|
|
self.root_nsmap['dc'] = 'http://purl.org/dc/elements/1.1/'
|
|
|
|
item.append(create_element(
|
|
'{http://purl.org/dc/elements/1.1/}creator',
|
|
CDATA(author or self.author)
|
|
))
|
|
|
|
for category in categories:
|
|
item.append(create_element('category', CDATA(category)))
|
|
|
|
if enclosure:
|
|
item.append(create_element(
|
|
'enclosure',
|
|
url=enclosure.url,
|
|
length=str(enclosure.size or 0),
|
|
type=enclosure.type or mimetypes.guess_type(enclosure.url)[0]
|
|
))
|
|
|
|
if pub_date:
|
|
item.append(create_element('pubDate', pub_date))
|
|
|
|
self.items.append(item)
|
|
|
|
def xml(self, pretty: bool = False) -> str:
|
|
"""Returns the XML as a string.
|
|
|
|
:param pretty: Pretty print xml
|
|
"""
|
|
root = Element('rss', nsmap=self.root_nsmap, version=self.root_version)
|
|
channel = create_element('channel', children=[
|
|
create_element('title', CDATA(self.title)),
|
|
create_element('description', CDATA(self.description)),
|
|
create_element('link', self.site_url),
|
|
create_element('{http://www.w3.org/2005/Atom}link',
|
|
href=self.feed_url, rel='self',
|
|
type='application/rss+xml'),
|
|
create_element('generator', self.generator),
|
|
create_element('lastBuildDate', datetime.utcnow())
|
|
])
|
|
|
|
if self.image_url:
|
|
channel.append(create_element('image', children=[
|
|
create_element('url', self.image_url),
|
|
create_element('title', CDATA(self.title)),
|
|
create_element('link', self.site_url)
|
|
]))
|
|
for category in self.categories:
|
|
channel.append(create_element('category', CDATA(category)))
|
|
if self.pub_date:
|
|
channel.append(create_element('pubDate', self.pub_date))
|
|
if self.copyright:
|
|
channel.append(create_element('copyright', CDATA(self.copyright)))
|
|
if self.language:
|
|
channel.append(create_element('language', CDATA(self.language)))
|
|
if self.editor:
|
|
channel.append(create_element('managingEditor', CDATA(self.editor)))
|
|
if self.webmaster:
|
|
channel.append(create_element('webMaster', CDATA(self.webmaster)))
|
|
if self.docs_url:
|
|
channel.append(create_element('docs', self.docs_url))
|
|
|
|
for item in self.items:
|
|
channel.append(item)
|
|
|
|
root.append(channel)
|
|
|
|
return '<?xml version="1.0" encoding="UTF-8"?>\n' \
|
|
+ ('\n' if pretty else '') \
|
|
+ tostring(root, pretty_print=pretty).decode('utf-8')
|