EbookFoundation
/
regluit
mirror of https://github.com/EbookFoundation/regluit.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

update frontiersin harvest

pull/94/head
eric 2021-03-19 13:19:22 -04:00
parent 4ff2a66001
commit 6e9e9b4c20
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
core/loaders/harvest.py
View File

@ -142,7 +142,7 @@ def harvesters(ebook):
yield ebook.provider == 'ksp.kit.edu', harvest_ksp yield ebook.provider == 'ksp.kit.edu', harvest_ksp
yield ebook.provider == 'digitalis.uc.pt', harvest_digitalis yield ebook.provider == 'digitalis.uc.pt', harvest_digitalis
yield ebook.provider == 'nomos-elibrary.de', harvest_nomos yield ebook.provider == 'nomos-elibrary.de', harvest_nomos
yield ebook.provider == 'frontiersin.org', harvest_frontiersin yield 'frontiersin.org' in ebook.provider, harvest_frontiersin
yield ebook.provider in ['Palgrave Connect', 'Springer', 'springer.com'], harvest_springerlink yield ebook.provider in ['Palgrave Connect', 'Springer', 'springer.com'], harvest_springerlink
yield ebook.provider == 'pulp.up.ac.za', harvest_pulp yield ebook.provider == 'pulp.up.ac.za', harvest_pulp
yield ebook.provider == 'bloomsburycollections.com', harvest_bloomsbury yield ebook.provider == 'bloomsburycollections.com', harvest_bloomsbury
@ -531,6 +531,15 @@ def harvest_nomos(ebook):
return None, 0 return None, 0
def harvest_frontiersin(ebook): def harvest_frontiersin(ebook):
if 'GetFile.aspx' in ebook.url:
ebook.delete()
rl.last.pop(ebook.provider, 0)
return None, 0
if ebook.provider == 'journal.frontiersin.org':
ebook, status = redirect_ebook(ebook)
if status < 1:
return None, -1 if status < 0 else 0
num = 0 num = 0
harvested = None harvested = None
doc = get_soup(ebook.url) doc = get_soup(ebook.url)