rra/page_metadata
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

count all media on page, including background images, handle "i18n_subsites" plugin

Browse Source
This commit is contained in:
rra 2020-01-21 13:44:43 +01:00
parent 83d57027b1
commit 5e96200f71
1 changed files with 53 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
page_metadata.py
View File

@ -43,7 +43,7 @@ def get_printable_size(byte_size):
elif size_index == 0:
return str(size)
else:
return "{:.3f}".format(size)
return "{:.2f}".format(size)
current_size = byte_size
size_index = 0
@ -56,6 +56,13 @@ def get_printable_size(byte_size):
measure = MEASURE[size_index]
return size + measure
def get_assets(soup):
assets = []
for a in soup.findAll('link', {'rel':['apple-touch-icon','icon','stylesheet']}):
a = a['href'].split('?')[0]
if a not in assets:
assets.append(a)
return assets
def get_media(html_file):
"""
@ -64,36 +71,69 @@ def get_media(html_file):
html_file = open(html_file).read()
soup = BeautifulSoup(html_file, 'html.parser')
media = []
for img in soup(['img', 'object']):
media.append(img['src'])
media = list(set(media)) # duplicate media don't increase page size
return media
featured_images = soup.findAll('div', {'class':'featured-img'})
for fi in featured_images:
fi = fi['style']
start = fi.find("url('")
end = fi.find("');")
url = fi[start+len("url('"):end]
media.append(url)
assets = get_assets(soup)
media = list(set(media+assets)) # duplicate media don't increase page size
return media, soup
def generate_metadata(path, context):
output_path = context['OUTPUT_PATH']
output_file = context['output_file']
siteurl = context['SITEURL']
plugins = context['PLUGINS']
subsites = False
if 'i18n_subsites' in plugins:
subsites = True
lang = context['DEFAULT_LANG']
general_output_path = output_path.replace(lang, '').strip('/')
siteurl = siteurl.replace(lang,'').strip('/')
media_size = 0
# enumerate all media displayed on the page
for m in get_media(path):
m = os.path.join(output_path, m)
# filter out SITEURL to prevent trouble
m = m.replace(context['SITEURL']+'/', '')
media, soup = get_media(path) #reuse the same soup to limit calculation
for m in media:
# filter out SITEURL to prevent trouble
file_name = m.replace(context['SITEURL']+'/', '')
# join output path to file, need to strip any leading slash for os.path
if subsites:
m = os.path.join(general_output_path, file_name.strip('/'))
else:
m = os.path.join(output_path, file_name.strip('/'))
if os.path.exists(m):
media_size = media_size + os.path.getsize(m)
current_file = os.path.join(output_path, output_file)
file_size = os.path.getsize(current_file)
with open(current_file, 'a') as f:
file_size = file_size + media_size
metadata = output_file + ' ' + context['NOW'] + ' ' + get_printable_size(file_size)
metadata = output_file + ' ' + context['NOW'] + ' ' + get_printable_size(file_size+len(metadata)) # cursed code is cursed
f.write(metadata)
file_size = file_size + media_size
metadata = get_printable_size(file_size)
metadata = get_printable_size(file_size+len(metadata)) # cursed code is cursed
# TODO: add a way to nicely insert the meta-data into an element with id
insert_metadata(path, metadata, soup)
def insert_metadata(output_file, metadata, soup):
tag = soup.find('div', {'id':'page-size'})
if tag:
with open(output_file,'w') as f:
tag.string = '{}'.format(metadata)
f.write(str(soup))
def register():
signals.content_written.connect(generate_metadata)