1
0
Fork 0
mirror of https://github.com/kemayo/leech synced 2025-12-06 08:22:56 +01:00

fix(Partial-Fix-to-Issue-#2): Leech can now download images however there is no way of disabling this option and this was only tested with stories from fiction.live

BREAKING CHANGE:
This commit is contained in:
Emmanuel Jemeni 2023-02-22 20:58:53 +01:00 committed by David Lynch
parent 1edde92a9d
commit 71345b2658
2 changed files with 148 additions and 16 deletions

View file

@ -1,6 +1,8 @@
from .epub import make_epub, EpubFile
from .cover import make_cover
from .cover import make_cover_from_url
from .cover import make_cover, make_cover_from_url
from .image import get_image_from_url
from sites import Image
from bs4 import BeautifulSoup
import html
import unicodedata
@ -72,7 +74,8 @@ class CoverOptions:
height = attr.ib(default=None, converter=attr.converters.optional(int))
wrapat = attr.ib(default=None, converter=attr.converters.optional(int))
bgcolor = attr.ib(default=None, converter=attr.converters.optional(tuple))
textcolor = attr.ib(default=None, converter=attr.converters.optional(tuple))
textcolor = attr.ib(
default=None, converter=attr.converters.optional(tuple))
cover_url = attr.ib(default=None, converter=attr.converters.optional(str))
@ -82,8 +85,18 @@ def chapter_html(story, titleprefix=None, normalize=False):
title = chapter.title or f'#{i}'
if hasattr(chapter, '__iter__'):
# This is a Section
chapters.extend(chapter_html(chapter, titleprefix=title, normalize=normalize))
chapters.extend(chapter_html(
chapter, titleprefix=title, normalize=normalize))
else:
soup = BeautifulSoup(chapter.contents, 'html5lib')
for count, img in enumerate(soup.find_all('img')):
img_contents = get_image_from_url(img['src']).read()
chapter.images.append(Image(
path=f"images/ch{i}_leechimage_{count}.png",
contents=img_contents,
content_type='image/png'
))
img['src'] = f"../images/ch{i}_leechimage_{count}.png"
# Add all pictures on this chapter as well.
for image in chapter.images:
# For/else syntax, check if the image path already exists, if it doesn't add the image.
@ -92,20 +105,23 @@ def chapter_html(story, titleprefix=None, normalize=False):
if other_file.path == image.path:
break
else:
chapters.append(EpubFile(path=image.path, contents=image.contents, filetype=image.content_type))
chapters.append(EpubFile(
path=image.path, contents=image.contents, filetype=image.content_type))
title = titleprefix and f'{titleprefix}: {title}' or title
contents = chapter.contents
contents = str(soup)
if normalize:
title = unicodedata.normalize('NFKC', title)
contents = unicodedata.normalize('NFKC', contents)
chapters.append(EpubFile(
title=title,
path=f'{story.id}/chapter{i + 1}.html',
contents=html_template.format(title=html.escape(title), text=contents)
contents=html_template.format(
title=html.escape(title), text=contents)
))
if story.footnotes:
chapters.append(EpubFile(title="Footnotes", path=f'{story.id}/footnotes.html', contents=html_template.format(title="Footnotes", text='\n\n'.join(story.footnotes))))
chapters.append(EpubFile(title="Footnotes", path=f'{story.id}/footnotes.html', contents=html_template.format(
title="Footnotes", text='\n\n'.join(story.footnotes))))
return chapters
@ -127,14 +143,19 @@ def generate_epub(story, cover_options={}, output_filename=None, output_dir=None
extra_metadata['Tags'] = ', '.join(story.tags)
if extra_metadata:
metadata['extra'] = '\n '.join(f'<dt>{k}</dt><dd>{v}</dd>' for k, v in extra_metadata.items())
metadata['extra'] = '\n '.join(
f'<dt>{k}</dt><dd>{v}</dd>' for k, v in extra_metadata.items())
valid_cover_options = ('fontname', 'fontsize', 'width', 'height', 'wrapat', 'bgcolor', 'textcolor', 'cover_url')
cover_options = CoverOptions(**{k: v for k, v in cover_options.items() if k in valid_cover_options})
cover_options = attr.asdict(cover_options, filter=lambda k, v: v is not None, retain_collection_types=True)
valid_cover_options = ('fontname', 'fontsize', 'width',
'height', 'wrapat', 'bgcolor', 'textcolor', 'cover_url')
cover_options = CoverOptions(
**{k: v for k, v in cover_options.items() if k in valid_cover_options})
cover_options = attr.asdict(
cover_options, filter=lambda k, v: v is not None, retain_collection_types=True)
if cover_options and "cover_url" in cover_options:
image = make_cover_from_url(cover_options["cover_url"], story.title, story.author)
image = make_cover_from_url(
cover_options["cover_url"], story.title, story.author)
elif story.cover_url:
image = make_cover_from_url(story.cover_url, story.title, story.author)
else:
@ -145,10 +166,17 @@ def generate_epub(story, cover_options={}, output_filename=None, output_dir=None
[
# The cover is static, and the only change comes from the image which we generate
EpubFile(title='Cover', path='cover.html', contents=cover_template),
EpubFile(title='Front Matter', path='frontmatter.html', contents=frontmatter_template.format(now=datetime.datetime.now(), **metadata)),
EpubFile(title='Front Matter', path='frontmatter.html', contents=frontmatter_template.format(
now=datetime.datetime.now(), **metadata)),
*chapter_html(story, normalize=normalize),
EpubFile(path='Styles/base.css', contents=requests.Session().get('https://raw.githubusercontent.com/mattharrison/epub-css-starter-kit/master/css/base.css').text, filetype='text/css'),
EpubFile(path='images/cover.png', contents=image.read(), filetype='image/png'),
EpubFile(
path='Styles/base.css',
contents=requests.Session().get(
'https://raw.githubusercontent.com/mattharrison/epub-css-starter-kit/master/css/base.css').text,
filetype='text/css'
),
EpubFile(path='images/cover.png',
contents=image.read(), filetype='image/png'),
],
metadata,
output_dir=output_dir

104
ebook/image.py Normal file
View file

@ -0,0 +1,104 @@
# Basically the same as cover.py with some minor differences
from PIL import Image, ImageDraw, ImageFont
from io import BytesIO
import textwrap
import requests
import logging
logger = logging.getLogger(__name__)
def make_image(
message: str,
width=600,
height=300,
fontname="Helvetica",
font_size=40,
bg_color=(0, 0, 0),
textcolor=(255, 255, 255),
wrap_at=30
):
"""
This function should only be called if get_image_from_url() fails
"""
img = Image.new("RGBA", (width, height), bg_color)
draw = ImageDraw.Draw(img)
message = textwrap.fill(message, wrap_at)
font = _safe_font(fontname, size=font_size)
message_size = draw.textsize(message, font=font)
draw_text_outlined(
draw, ((width - message_size[0]) / 2, 100), message, textcolor, font=font)
# draw.text(((width - title_size[0]) / 2, 100), title, textcolor, font=font)
output = BytesIO()
img.save(output, "PNG")
output.name = 'cover.png'
# writing left the cursor at the end of the file, so reset it
output.seek(0)
return output
def get_image_from_url(url: str):
"""
Basically the same as make_cover_from_url()
"""
try:
logger.info("Downloading image from " + url)
img = requests.Session().get(url)
cover = BytesIO(img.content)
img_format = Image.open(cover).format
# The `Image.open` read a few bytes from the stream to work out the
# format, so reset it:
cover.seek(0)
if img_format != "PNG":
cover = _convert_to_png(cover)
except Exception as e:
logger.info("Encountered an error downloading cover: " + str(e))
cover = make_image("There was a problem downloading this image.")
return cover
def _convert_to_png(image_bytestream):
png_image = BytesIO()
Image.open(image_bytestream).save(png_image, format="PNG")
png_image.name = 'cover.png'
png_image.seek(0)
return png_image
def _safe_font(preferred, *args, **kwargs):
for font in (preferred, "Helvetica", "FreeSans", "Arial"):
try:
return ImageFont.truetype(*args, font=font, **kwargs)
except IOError:
pass
# This is pretty terrible, but it'll work regardless of what fonts the
# system has. Worst issue: can't set the size.
return ImageFont.load_default()
def draw_text_outlined(draw, xy, text, fill=None, font=None, anchor=None):
x, y = xy
# Outline
draw.text((x - 1, y), text=text, fill=(0, 0, 0), font=font, anchor=anchor)
draw.text((x + 1, y), text=text, fill=(0, 0, 0), font=font, anchor=anchor)
draw.text((x, y - 1), text=text, fill=(0, 0, 0), font=font, anchor=anchor)
draw.text((x, y + 1), text=text, fill=(0, 0, 0), font=font, anchor=anchor)
# Fill
draw.text(xy, text=text, fill=fill, font=font, anchor=anchor)
if __name__ == '__main__':
f = make_image(
'Test of a Title which is quite long and will require multiple lines')
with open('output.png', 'wb') as out:
out.write(f.read())