def moveissn(origin='.',prefixdir=True): '''Move files to their ISSN/Volume/Issue folder Files can be in prefix dir (prefixdir) or just in a folder''' targetfiles= '10.*/10.*.pdf' if prefixdir else '10.*.pdf' for f in glob.iglob(origin+os.sep+targetfiles): doi=DOI(filebasename(f)) os.renames(f,origin+os.sep+doi.prefix+os.sep+getpdfdir(doi)+doi.quote()+'.pdf')
def moveissn(origin='.', prefixdir=True):
'''Move files to their ISSN/Volume/Issue folder
Files can be in prefix dir (prefixdir) or just in a folder'''
targetfiles = '10.*/10.*.pdf' if prefixdir else '10.*.pdf'
for f in glob.iglob(origin + os.sep + targetfiles):
doi = DOI(filebasename(f))
os.renames(
f, origin + os.sep + doi.prefix + os.sep + getpdfdir(doi) +
doi.quote() + '.pdf')
def moveprefix(origin='.'):
'''Just move the PDFs in the origin folder to their prefix sub-folder'''
pdffiles = glob.glob(origin + os.sep + "10.*.pdf")
for pdf in pdffiles:
fname = os.path.split(pdf)[1]
doi = DOI(filebasename(pdf))
if (not os.path.exists(origin + os.sep + doi.prefix)):
os.makedirs(origin + os.sep + doi.prefix)
if (not os.path.exists(origin + os.sep + doi.prefix + os.sep + fname)):
os.renames(pdf, origin + os.sep + doi.prefix + os.sep + fname)
def comparepdfsizeMove(totalfiles, sfresult, target='Done', check=False):
# Start to compare
for fg in totalfiles: #glob.iglob(origin+os.sep+'10.*.pdf'):
fnamesplit = os.path.split(fg)
doi = DOI(filebasename(fg))
if (doi):
fpath = fg.strip().split('@', 1)
fsize = os.path.getsize(fg)
printout = ""
if (not sfresult.has_key(doi)):
if (not check or ph.FastCheck(fg)):
targetfname = target + os.sep + fnamesplit[1]
try:
os.renames(fg, targetfname)
except:
if (os.path.exists(targetfname)):
if (similarsize(
fsize, os.path.getsize(targetfname)) >= 0):
os.remove(fg)
else:
print "Move fail...", fg
else:
print 'File maybe wrong..', fg
else:
for line in sfresult[doi]:
s = similarsize(fsize, str(line).strip())
if (s >= 0):
os.remove(fg)
printout = ""
break
#elif ( s is 1 and int(fsize)<int(line.strip())):
# os.remove(fg)
# printout=""
# break
else:
printout = fg + " now size: " + str(
fsize) + "; in lib: " + str(line).strip()
if (printout): print printout
else:
print "File is not in doi style!", fg
def filter(self,doi): '''Use to filter the publisher''' doi=DOI(doi) return doi.publisherstyle(self.publisher)
def acsstyledoi(doi): doi=DOI(doi) if (doi.prefix == '10.1021'): return True else: return False
def genpdfsize(jsfnames):
'''Generate PDF information on SF library'''
sfurl = "http://oapdf.sourceforge.net/cgi-bin/pdfsize.cgi?owner=oapdf"
count = 0
tmpdoisize = {}
tmpdois = set()
toappend = {}
if isinstance(jsfnames, str):
jsfnames = [jsfnames]
for jf in jsfnames:
f = open(jf)
j = json.loads(f.read())
f.close()
for pdf, fs in j['items'].items():
count += 1
if (count % 50 == 0):
# Firstly, query the records
r = requests.post(sfurl,
params={'dois': json.dumps(list(tmpdois))},
timeout=120)
result = r.json()
for d, s in tmpdoisize.items():
if (not set(s).issubset(set(result.get(d, [])))):
toappend[d] = list(set(s + result.get(d, [])))
if (len(toappend) > 0):
rr = requests.post(sfurl,
params={
'doisize': json.dumps(toappend),
'update': "True"
},
timeout=120)
if (rr.status_code == 200):
#print rr.text
#if fail, submit at next time
tmpdoisize.clear()
tmpdois.clear()
toappend.clear()
else:
tmpdoisize.clear()
tmpdois.clear()
toappend.clear()
time.sleep(2)
doi = DOI(filebasename(pdf))
tmpdois.add(doi)
tmpdoisize.setdefault(doi, []).append(fs)
if (len(tmpdois) > 0):
# Firstly, query the records
r = requests.post(sfurl,
params={'dois': json.dumps(list(tmpdois))},
timeout=120)
result = r.json()
for d, s in tmpdoisize.items():
if (not set(s).issubset(set(result.get(d, [])))):
toappend[d] = list(set(s + result.get(d, [])))
if (len(toappend) > 0):
rr = requests.post(sfurl,
params={
'doisize': json.dumps(toappend),
'update': "True"
},
timeout=120)
if (rr.status_code == 200):
#print rr.text
#if fail, submit at next time
tmpdoisize.clear()
tmpdois.clear()
toappend.clear()
else:
tmpdoisize.clear()
tmpdois.clear()
toappend.clear()
print "Total done for:", count
def comparepdfsize(origin='.', target='Done', check=False, totalsize=None):
'''Compare PDF size in origin to online library.
If new, move to target folder'''
if not os.path.exists(origin):
print "The origin folder not exist! Exit!"
sys.exit(1)
if (not os.path.exists(target)):
os.makedirs(target)
if (totalsize and float(totalsize) > 0):
totalsize = float(totalsize) * 1000000000
sfurl = "http://oapdf.sourceforge.net/cgi-bin/pdfsize.cgi?owner=oapdf"
workingdir = os.path.abspath('.')
# Get online informations
count = 0
sfresult = {}
tmpdois = set()
totalfiles = []
nowsizes = 0
for fg in glob.iglob(origin + os.sep + '10.*.pdf'):
if nowsizes > totalsize:
break
doi = DOI(filebasename(fg))
if (doi):
count += 1
totalfiles.append(fg)
nowsizes += os.path.getsize(fg)
if (count % 50 == 0):
# Firstly, query the records
try:
r = requests.post(
sfurl,
params={'dois': json.dumps(list(tmpdois))},
timeout=120)
except requests.RequestException:
r = requests.post(
sfurl,
params={'dois': json.dumps(list(tmpdois))},
timeout=120)
if (r.status_code == 200):
sfresult.update(r.json())
tmpdois.clear()
comparepdfsizeMove(totalfiles,
sfresult,
target=target,
check=check)
del totalfiles[:]
sfresult.clear()
#time.sleep(2)
tmpdois.add(doi)
if (len(tmpdois) > 0):
r = requests.post(sfurl,
params={'dois': json.dumps(list(tmpdois))},
timeout=120)
if (r.status_code == 200):
sfresult.update(r.json())
tmpdois.clear()
comparepdfsizeMove(totalfiles,
sfresult,
target=target,
check=check)
del totalfiles[:]
sfresult.clear()
def touchpage(origin='.', doilink='../doilink', pdf=True, force=False):
# Use to save local page record
if not os.path.exists(doilink):
os.makedirs(doilink + os.sep + 'pages')
doilink = doilink.rstrip('/').rstrip('\\')
sfurl = "http://oapdf.sourceforge.net/cgi-bin/touchdoi.cgi?owner=oapdf"
workdir = os.path.abspath(origin).rstrip('\\').rstrip('/')
count = 0
touchcount = 1 # avoid submit when start
forcesf = force # force to overwrite the exist doilink page
if (pdf):
result = (chain.from_iterable(
glob.iglob(os.path.join(x[0], '10.*.pdf'))
for x in os.walk(workdir)))
else:
result = (chain.from_iterable(
glob.iglob(os.path.join(x[0], '10.*.html'))
for x in os.walk(workdir)))
toappend = []
newtouch = 0
for f in result:
if (touchcount % 50 == 0):
r = requests.post(sfurl,
params={'dois': json.dumps(toappend)},
timeout=120)
if (r.status_code == 200):
bs = BeautifulSoup(r.text, "html.parser")
totaldid = bs.findChild('span', attrs={'id': 'total'})
if totaldid and totaldid.text:
newtouch += int(totaldid.text)
del toappend[:]
else:
print "Maybe Error when submit to SF-OAPDF.."
sys.exit(1)
count += 1
fname = filebasename(f)
if (' ' in fname):
print "File name has blank!", f
os.renames(
f,
os.path.split(f)[0] + os.sep + fname.strip() +
os.path.splitext(f)[1])
fname = fname.strip()
doi = DOI(fname)
if (doi):
dirname = doilink + "/pages/" + doi.decompose(url=False,
outdir=True)
if (forcesf or not os.path.exists(dirname + fname + '.html')):
touchcount += 1
toappend.append(doi)
try:
if (not os.path.exists(dirname)): os.makedirs(dirname)
f = open(dirname + fname + '.html', "w")
f.close()
except WindowsError as e:
print e
except:
print "Something error for file:", f
else:
print "File name may be error (Not DOI name):", fname
r = requests.post(sfurl,
params={'dois': json.dumps(toappend)},
timeout=120)
if (r.status_code == 200):
bs = BeautifulSoup(r.text, "html.parser")
totaldid = bs.findChild('span', attrs={'id': 'total'})
if totaldid and totaldid.text:
newtouch += int(totaldid.text)
del toappend[:]
else:
print "Maybe Error when submit to SF-OAPDF.."
sys.exit(1)
print "Process total file:", count, "; local touch new:", touchcount - 1, "; remote touch:", newtouch
def touchpage(origin='.', doilink='../doilink',pdf=True,force=False):
# Use to save local page record
if not os.path.exists(doilink):
os.makedirs(doilink+os.sep+'pages')
doilink=doilink.rstrip('/').rstrip('\\')
sfurl="http://oapdf.sourceforge.net/cgi-bin/touchdoi.cgi?owner=oapdf"
workdir=os.path.abspath(origin).rstrip('\\').rstrip('/')
count=0
touchcount=1 # avoid submit when start
forcesf=force # force to overwrite the exist doilink page
if (pdf):
result = (chain.from_iterable(glob.iglob(os.path.join(x[0], '10.*.pdf')) for x in os.walk(workdir)))
else:
result = (chain.from_iterable(glob.iglob(os.path.join(x[0], '10.*.html')) for x in os.walk(workdir)))
toappend=[]
newtouch=0
for f in result:
if (touchcount%50==0):
r=requests.post(sfurl,params={'dois':json.dumps(toappend)},timeout=120)
if (r.status_code == 200):
bs=BeautifulSoup(r.text,"html.parser")
totaldid=bs.findChild('span',attrs={'id':'total'})
if totaldid and totaldid.text :
newtouch+=int(totaldid.text)
del toappend[:]
else:
print "Maybe Error when submit to SF-OAPDF.."
sys.exit(1)
count+=1
fname=filebasename(f)
if (' ' in fname):
print "File name has blank!",f
os.renames(f,os.path.split(f)[0]+os.sep+fname.strip()+os.path.splitext(f)[1])
fname=fname.strip()
doi=DOI(fname)
if (doi):
dirname=doilink+"/pages/"+doi.decompose(url=False, outdir=True)
if (forcesf or not os.path.exists(dirname+fname+'.html')):
touchcount+=1
toappend.append(doi)
try:
if (not os.path.exists(dirname)): os.makedirs(dirname)
f=open(dirname+fname+'.html',"w")
f.close()
except WindowsError as e:
print e
except:
print "Something error for file:",f
else:
print "File name may be error (Not DOI name):",fname
r=requests.post(sfurl,params={'dois':json.dumps(toappend)},timeout=120)
if (r.status_code == 200):
bs=BeautifulSoup(r.text,"html.parser")
totaldid=bs.findChild('span',attrs={'id':'total'})
if totaldid and totaldid.text :
newtouch+=int(totaldid.text)
del toappend[:]
else:
print "Maybe Error when submit to SF-OAPDF.."
sys.exit(1)
print "Process total file:",count,"; local touch new:",touchcount-1, "; remote touch:",newtouch
if (not os.path.exists('Done')):
os.makedirs('Done')
ph = PDFHandler()
nocheckpdf = False
targetdir = '.'
workingdir = os.path.abspath('.')
if (len(sys.argv) >= 2):
nocheckpdf = True
targetdir = sys.argv[1]
for fg in glob.iglob(targetdir + os.sep + '10.*.pdf'):
fnamesplit = os.path.split(fg)
doi = DOI(os.path.splitext(fnamesplit[1])[0])
if (doi):
fpath = fg.strip().split('@', 1)
fsize = os.path.getsize(fg)
fname = 'Decoy/' + doi.prefix + os.sep + fnamesplit[1]
printout = ""
if (not os.path.exists(fname)):
if (nocheckpdf or ph.FastCheck(fg)):
targetfname = 'Done' + os.sep + fnamesplit[1]
try:
os.renames(fg, targetfname)
except:
if (os.path.exists(targetfname)):
if (similarsize(fsize, os.path.getsize(targetfname)) >=
0):
