def getAuthors(self): if 'author' in self.json.keys(): return ', '.join(listify(self.json['author'])) elif 'editor' in self.json.keys(): return ', '.join(listify(self.json['editor'])) else: return ''
def getAuthors(self):
if 'author' in self.json.keys():
return ', '.join(listify(self.json['author']))
elif 'editor' in self.json.keys():
return ', '.join(listify(self.json['editor']))
else:
return ''
def getBoxLinks(self):
links = [[], [], []]
# Crossref to the parent
# DOC: real crossrefs are nice, but all the global searches slows the engine down
# DOC: (running time 5m11.559s vs 0m7.396s is 42x)
# if 'crossref' in self.json.keys():
# xref = self.top().seek(self.json['crossref'])
# if xref:
# links.append('<strong><a href="{}.html">{}</a></strong>'.format(\
# xref.getKey(),
# xref.getKey().replace('-', ' ')))
# else:
# links.append('Xref not found')
# else:
# links.append('no Xref')
# DOC: instead, we just link to the parent!
links[0].append('<strong><a href="{}.html">{}</a></strong>'.format(\
self.up().getKey(),
self.up().getKey().replace('-', ' ')))
# DBLP
if 'dblpkey' in self.json:
for dblpk in listify(self.json['dblpkey']):
links[1].append('<a href="http://dblp.uni-trier.de/rec/html/{}">DBLP</a>'.format(dblpk))
elif 'dblpurl' in self.json:
links[1].append('<a href="{}">DBLP</a>'.format(self.json['dblpurl']))
else:
links[1].append('no DBLP info')
# Scholar
if 'title' in self.json.keys():
links[1].append('<a href="https://scholar.google.com/scholar?q=%22{}%22">Scholar</a>'.format(\
str(self.json['title']).replace(' ', '+')))
# Some publishers
if 'ee' in self.json.keys():
for e in listify(self.json['ee']):
if e.startswith('http://dl.acm.org') \
or e.startswith('http://doi.acm.org')\
or e.startswith('http://portal.acm.org'):
links[2].append('<a href="{}">ACM DL</a>'.format(e))
elif e.startswith('http://ieeexplore.ieee.org'):
links[2].append('<a href="{}">IEEE Xplore</a>'.format(e))
elif e.startswith('http://ieeecomputersociety.org'):
links[2].append('<a href="{}">IEEE CS</a>'.format(e))
elif e.startswith('http://drops.dagstuhl.de'):
links[2].append('<a href="{}">Dagstuhl</a>'.format(e))
elif e.find('computer.org/csdl/proceedings') > 0:
links[2].append('<a href="{}">CSDL</a>'.format(e))
elif e.startswith('http://journal.ub.tu-berlin.de/eceasst'):
links[2].append('<a href="{}">EC-EASST</a>'.format(e))
elif e.startswith('http://ceur-ws.org'):
links[2].append('<a href="{}">CEUR</a>'.format(e))
elif e.startswith('http://dx.doi.org'):
pass
else:
links[2].append('<a href="{}">?EE?</a>'.format(e))
if 'doi' in self.json.keys():
links[2].append('<a href="http://dx.doi.org/{}">DOI</a>'.format(self.json['doi']))
return '<hr/>'.join(['<br/>\n'.join(lb) for lb in links if lb])
def getAbbrAuthors(self):
# <abbr title="Rainer Koschke">RK</abbr>
if 'author' not in self.json.keys():
return ''
return ' ('+', '.join(['<abbr title="{0}">{1}</abbr>'.format(a,\
''.join([w[0] for w in a.replace('-', ' ').split(' ') if w]))\
for a in listify(self.json['author'])])+')'
def getAbbrAuthors(self):
# <abbr title="Rainer Koschke">RK</abbr>
if 'author' not in self.json.keys():
return ''
return ' ('+', '.join(['<abbr title="{0}">{1}</abbr>'.format(a,\
''.join([w[0] for w in a.replace('-', ' ').split(' ') if w]))\
for a in listify(self.json['author'])])+')'
def offer(self, year, conf):
ckey = conf.getKey()
for rule in listify(self.json['select']):
if ckey.startswith(rule):
self.addConf(year, conf)
# print(C.red(conf.getKey()), ' was accepted by ', C.red(self.name))
return
def offer(self, year, conf): ckey = conf.getKey() for rule in listify(self.json['select']): if ckey.startswith(rule): self.addConf(year, conf) # print(C.red(conf.getKey()), ' was accepted by ', C.red(self.name)) return
def kv2link(k, v):
if k == 'g':
ico = makeimg('g', 'Google')
r = '<a href="https://www.google.com/search?q={}">{}</a>'.format(
escape(v), v)
elif k.endswith('.wp'):
lang = k.split('.')[0]
# Using ISO 639-1 language names
ico = makeimg('wp', 'Wikipedia') + makeimg(lang, ISONames[lang])
lang = k.split('.')[0]
r = '<a href="https://{}.wikipedia.org/wiki/{}">{}</a>'.format(\
lang, \
escape(v).replace('%20', '_'), \
v)
elif k.endswith('.wb'):
lang = k.split('.')[0]
# Using ISO 639-1 language names
ico = makeimg('wb', 'Wikibooks') + makeimg(lang, ISONames[lang])
lang = k.split('.')[0]
r = '<a href="https://{}.wikibooks.org/wiki/{}">{}</a>'.format(\
lang, \
escape(v).replace('%20', '_'), \
v)
elif k == 'wd':
ico = makeimg('wd', 'Wikidata')
r = '<a href="https://www.wikidata.org/wiki/{0}">{0}</a>'.format(v)
elif k == 'hwiki':
ico = makeimg('h', 'Haskell Wiki')
r = '<a href="https://wiki.haskell.org/{}">{}</a>'.format(escape(v), v)
elif k == 'so':
ico = makeimg('so', 'Stack Overflow')
r = '<a href="http://stackoverflow.com/questions/tagged?tagnames={0}">{0}</a>'.format(
v)
elif k == 'www':
ico = '' #makeimg('www', 'Homepage')
if not v.startswith('http'):
y = v
v = 'http://' + v
else:
y = v.replace('http://', '').replace('https://', '')
r = '<a href="{0}">{1}</a>'.format(v, y)
elif k == 'aka':
ico = ''
r = '<br/>'.join(['a.k.a.: “{}”'.format(z) for z in listify(v)])
else:
ico = ''
r = '?{}?{}?'.format(k, v)
return ico + ' ' + r + '<br/>'
def getPage(self):
if self.getTags():
cnt = '<h3>Tags:</h3><ul class="tri">'
cnt += '\n'.join([
'<li class="tag"><a href="tag/{}.html">#{}</a></li>'.format(
escape(t), t) for t in self.tags
])
cnt += '</ul><hr/>'
else:
cnt = ''
if 'stemmed' in self.json.keys():
stemmed = listify(self.json['stemmed'])
title = self.get('title')
ltitle = title.lower()
words = string2words(title)
words.reverse()
fancytitle = ''
for w in words:
i = ltitle.rindex(w)
try:
stem = stemmed[len(words) - words.index(w) - 1]
except:
print('Abnormal title in', self.getKey())
print('\tCould not get', w, 'from', stemmed)
break
if ifApproved(stem):
fancytitle = '<a href="word/{}.html">{}</a>{}'.format(\
stem, title[i:i+len(w)], title[i+len(w):]) + fancytitle
else:
fancytitle = title[i:] + fancytitle
ltitle = ltitle[:i]
title = title[:i]
fancytitle = title + fancytitle
else:
fancytitle = self.get('title')
return bibHTML.format(\
filename=self.getJsonName(),
title=self.get('title'),
stemmedTitle=fancytitle,
img=self.get('venue').lower(), # geticon?
authors=self.getAuthors(),
short='{}, {}'.format(self.get('venue'), self.get('year')),
code=self.getCode(),
bib=self.getBib(),
boxlinks=self.getBoxLinks(),
contents=cnt\
)
def kv2link(k, v):
if k == 'g':
ico = makeimg('g', 'Google')
r = '<a href="https://www.google.com/search?q={}">{}</a>'.format(escape(v), v)
elif k.endswith('.wp'):
lang = k.split('.')[0]
# Using ISO 639-1 language names
ico = makeimg('wp', 'Wikipedia') + makeimg(lang, ISONames[lang])
lang = k.split('.')[0]
r = '<a href="https://{}.wikipedia.org/wiki/{}">{}</a>'.format(\
lang, \
escape(v).replace('%20', '_'), \
v)
elif k.endswith('.wb'):
lang = k.split('.')[0]
# Using ISO 639-1 language names
ico = makeimg('wb', 'Wikibooks') + makeimg(lang, ISONames[lang])
lang = k.split('.')[0]
r = '<a href="https://{}.wikibooks.org/wiki/{}">{}</a>'.format(\
lang, \
escape(v).replace('%20', '_'), \
v)
elif k == 'wd':
ico = makeimg('wd', 'Wikidata')
r = '<a href="https://www.wikidata.org/wiki/{0}">{0}</a>'.format(v)
elif k == 'hwiki':
ico = makeimg('h', 'Haskell Wiki')
r = '<a href="https://wiki.haskell.org/{}">{}</a>'.format(escape(v), v)
elif k == 'so':
ico = makeimg('so', 'Stack Overflow')
r = '<a href="http://stackoverflow.com/questions/tagged?tagnames={0}">{0}</a>'.format(v)
elif k == 'www':
ico = ''#makeimg('www', 'Homepage')
if not v.startswith('http'):
y = v
v = 'http://'+v
else:
y = v.replace('http://', '').replace('https://', '')
r = '<a href="{0}">{1}</a>'.format(v, y)
elif k == 'aka':
ico = ''
r = '<br/>'.join(['a.k.a.: “{}”'.format(z) for z in listify(v)])
else:
ico = ''
r = '?{}?{}?'.format(k, v)
return ico + ' ' + r + '<br/>'
def getPage(self):
if self.getTags():
cnt = '<h3>Tags:</h3><ul class="tri">'
cnt += '\n'.join(['<li class="tag"><a href="tag/{}.html">#{}</a></li>'.format(escape(t), t) for t in self.tags])
cnt += '</ul><hr/>'
else:
cnt = ''
if 'stemmed' in self.json.keys():
stemmed = listify(self.json['stemmed'])
title = self.get('title')
ltitle = title.lower()
words = string2words(title)
words.reverse()
fancytitle = ''
for w in words:
i = ltitle.rindex(w)
try:
stem = stemmed[len(words) - words.index(w)-1]
except:
print('Abnormal title in', self.getKey())
break
if ifApproved(stem):
fancytitle = '<a href="word/{}.html">{}</a>{}'.format(\
stem, title[i:i+len(w)], title[i+len(w):]) + fancytitle
else:
fancytitle = title[i:] + fancytitle
ltitle = ltitle[:i]
title = title[:i]
fancytitle = title + fancytitle
else:
fancytitle = self.get('title')
return bibHTML.format(\
filename=self.getJsonName(),
title=self.get('title'),
stemmedTitle=fancytitle,
img=self.get('venue').lower(), # geticon?
authors=self.getAuthors(),
short='{}, {}'.format(self.get('venue'), self.get('year')),
code=self.getCode(),
bib=self.getBib(),
boxlinks=self.getBoxLinks(),
contents=cnt\
)
# if a manual rule does the same as the other heuristic, it’s dumb
for a in sorted(aka.keys()):
if len(aka[a]) == 1 and aka[a][0] in (nodiaLatin(a), simpleLatin(a)):
print('[ {} ]'.format(C.blue('DUMB')), simpleLatin(a), 'aliasing was unnecessary manual work')
elif len(aka[a]) == 2 and (aka[a] == [nodiaLatin(a), simpleLatin(a)] \
or aka[a] == [simpleLatin(a), nodiaLatin(a)]):
print('[ {} ]'.format(C.blue('DUMB')), simpleLatin(a), 'aliasing was a lot of unnecessary manual work')
elif nodiaLatin(a) in aka[a] or simpleLatin(a) in aka[a]:
print('[ {} ]'.format(C.blue('DUMB')), simpleLatin(a), 'aliasing contains some unnecessary manual work')
# auto-aliasing heuristic:
# for each author with diacritics, its non-diacritic twin is considered harmful
people = set()
for v in sleigh.venues:
for c in v.getConfs():
if 'editor' in c.json:
people.update(listify(c.json['editor']))
for p in c.papers:
if 'author' in p.json:
people.update(listify(p.json['author']))
for a in people:
for na in (nodiaLatin(a), simpleLatin(a)):
if na != a:
aka.setdefault(a, [])
aka[a].append(na)
# invert aliasing
for akey in aka:
if akey in ('ZZZZZZZZZZ', 'FILE'):
continue
for aval in aka[akey]:
renameto[aval] = akey
f = open('_renameto.json', 'w', encoding='utf-8')
# cs =
ads = [c.json['address'][-1] \
for c in {bykey[p].up() for p in persondef['authored'] if p in bykey} \
if 'address' in c.json]
if ads:
clist = {a: ads.count(a) for a in ads}
adds = '<code>Travelled to:</code><hr/>' \
+ '<br/>\n'.join(sorted(['{} × {}'.format(clist[a], a) for a in clist]))
boxlinks += adds
# collaborated with...
clist = {}
for p in persondef['authored']:
if p not in bykey:
continue
if 'author' in bykey[p].json.keys():
coas = listify(bykey[p].get('author'))
if ' ' in coas:
print('ERROR in [{}] - [{}] - [{}]'.format(
p, bykey[p].getKey(), coas))
D = len(coas)
if D == 1:
# solo papers count as coauthoring with yourself
a = '∅'
if a not in clist:
clist[a] = 0
if persondef['name'] in name2file.keys():
name2file['∅'] = name2file[persondef['name']]
else:
name2file['∅'] = persondef['name']
clist[a] += 1 / 2
continue
# if a manual rule does the same as the other heuristic, it’s dumb
for a in sorted(aka.keys()):
if len(aka[a]) == 1 and aka[a][0] in (nodiaLatin(a), simpleLatin(a)):
print('[ {} ]'.format(C.blue('DUMB')), a, 'aliasing was unnecessary manual work')
elif len(aka[a]) == 2 and (aka[a] == [nodiaLatin(a), simpleLatin(a)] \
or aka[a] == [simpleLatin(a), nodiaLatin(a)]):
print('[ {} ]'.format(C.blue('DUMB')), a, 'aliasing was a lot of unnecessary manual work')
elif nodiaLatin(a) in aka[a] or simpleLatin(a) in aka[a]:
print('[ {} ]'.format(C.blue('DUMB')), a, 'aliasing contains some unnecessary manual work')
# auto-aliasing heuristic:
# for each author with diacritics, its non-diacritic twin is considered harmful
people = set()
for v in sleigh.venues:
for c in v.getConfs():
if 'editor' in c.json:
people.update(listify(c.json['editor']))
for p in c.papers:
if 'author' in p.json:
people.update(listify(p.json['author']))
for a in people:
for na in (nodiaLatin(a), simpleLatin(a)):
if na != a:
aka.setdefault(a, [])
aka[a].append(na)
# invert aliasing
for akey in aka:
if akey in ('ZZZZZZZZZZ', 'FILE'):
continue
for aval in aka[akey]:
renameto[aval] = akey
f = open('_renameto.json', 'w', encoding='utf8')
if p['name'] in people.keys():
cx[report(C.red('duplicate') + ' ' + C.yellow(p), 1)] += 1
continue
people[p['name']] = p
print('{}: {} venues, {} papers written by {} people\n{}'.format(\
C.purple('BibSLEIGH'),
C.red(len(sleigh.venues)),
C.red(sleigh.numOfPapers()),
C.red(len(people)),
C.purple('='*42)))
# traverse ALL the papers!
for v in sleigh.venues:
for c in v.getConfs():
for p in c.papers:
if 'author' in p.json.keys():
for a in listify(p.json['author']):
if a in people.keys():
if 'authored' not in people[a].keys():
people[a]['authored'] = []
if p.getKey() not in people[a]['authored']:
people[a]['authored'].append(p.getKey())
else:
report(C.yellow('Author not found: ') + a, 0)
if 'editor' in c.json.keys():
for e in listify(c.json['editor']):
if e in people.keys():
if 'edited' not in people[e].keys():
people[e]['edited'] = []
if c.getKey() not in people[e]['edited']:
people[e]['edited'].append(c.getKey())
else:
def checkon(fn, o):
if not os.path.exists(fn) or os.path.isdir(fn):
fn = fn + '.json'
f = open(fn, 'r')
lines = f.readlines()[1:-1]
f.close()
flines = json2lines(lines)
plines = sorted(json2lines(o.getJSON().split('\n')))
# "url" from DBLP are useless
if 'url' in o.json.keys():
o.json['url'] = [link.replace('https://', 'http://')\
for link in listify(o.json['url'])\
if not link.startswith('db/conf/')\
and not link.startswith('db/series/')\
and not link.startswith('db/books/')\
and not link.startswith('db/journals/')]
if not o.json['url']:
del o.json['url']
elif len(o.json['url']) == 1:
o.json['url'] = o.json['url'][0]
if 'ee' in o.json.keys() and 'doi' not in o.json.keys():
if isinstance(o.json['ee'], list):
if verbose:
print(C.red('Manylink:'), o.json['ee'])
newee = []
for onelink in listify(o.json['ee']):
if onelink.startswith('http://dx.doi.org/'):
o.json['doi'] = onelink[18:]
elif onelink.startswith('http://doi.acm.org/'):
o.json['doi'] = onelink[19:]
elif onelink.startswith('http://doi.ieeecomputersociety.org/'):
o.json['doi'] = onelink[35:]
elif onelink.startswith('http://dl.acm.org/citation.cfm?id='):
o.json['acmid'] = onelink[34:]
elif onelink.startswith('http://portal.acm.org/citation.cfm?id='):
o.json['acmid'] = onelink[38:]
elif onelink.startswith('http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=')\
or onelink.startswith('http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber='):
o.json['ieeearid'] = onelink.split('=')[-1]
elif onelink.startswith('http://ieeexplore.ieee.org/xpls/abs_all.jsp?isnumber=')\
and onelink.find('arnumber') > -1:
o.json['ieeearid'] = onelink.split('arnumber=')[-1].split('&')[0]
elif onelink.startswith('http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber='):
o.json['ieeepuid'] = onelink.split('=')[-1]
elif onelink.startswith('http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber='):
o.json['ieeeisid'] = onelink.split('=')[-1]
elif onelink.startswith('http://eceasst.cs.tu-berlin.de/index.php/eceasst/article/view/'):
newee.append('http://journal.ub.tu-berlin.de/eceasst/article/view/' + onelink.split('/')[-1])
elif onelink.endswith('.pdf') and \
(onelink.startswith('http://computer.org/proceedings/')\
or onelink.startswith('http://csdl.computer.org/')):
# Bad: http://computer.org/proceedings/icsm/1189/11890007.pdf
# Bad: http://csdl.computer.org/comp/proceedings/date/2003/1870/02/187020040.pdf
# Good: http://www.computer.org/csdl/proceedings/icsm/2001/1189/00/11890004.pdf
if onelink.startswith('http://csdl'):
cname, _, cid, mid, pid = onelink.split('/')[5:10]
else:
cname, cid, pid = onelink.split('/')[4:7]
# heuristic
if pid.startswith(cid):
mid = pid[len(cid):len(cid)+2]
else:
mid = '00'
newee.append('http://www.computer.org/csdl/proceedings/{}/{}/{}/{}/{}'.format(\
cname,
o.get('year'),
cid,
mid,
pid))
else:
if onelink.find('ieee') > -1:
print(C.purple('IEEE'), onelink)
if verbose:
print(C.yellow('Missed opportunity:'), onelink)
# nothing matches => preserve
newee.append(onelink)
if len(newee) == 0:
del o.json['ee']
elif len(newee) == 1:
o.json['ee'] = newee[0]
else:
o.json['ee'] = newee
# post-processing normalisation
if 'acmid' in o.json.keys() and not isinstance(o.json['acmid'], int) and o.json['acmid'].isdigit():
o.json['acmid'] = int(o.json['acmid'])
if 'eventuri' in o.json.keys():
o.json['eventurl'] = o.json['eventuri']
del o.json['eventuri']
if 'eventurl' in o.json.keys() and o.json['eventurl'].startswith('https://'):
o.json['eventurl'] = o.json['eventurl'].replace('https://', 'http://')
nlines = sorted(json2lines(o.getJSON().split('\n')))
if flines != plines:
return 1
elif plines != nlines:
f = open(fn, 'w')
f.write(o.getJSON())
f.close()
return 2
else:
return 0
def getBoxLinks(self):
links = [[], [], []]
# Crossref to the parent
# DOC: real crossrefs are nice, but all the global searches slows the engine down
# DOC: (running time 5m11.559s vs 0m7.396s is 42x)
# if 'crossref' in self.json.keys():
# xref = self.top().seek(self.json['crossref'])
# if xref:
# links.append('<strong><a href="{}.html">{}</a></strong>'.format(\
# xref.getKey(),
# xref.getKey().replace('-', ' ')))
# else:
# links.append('Xref not found')
# else:
# links.append('no Xref')
# DOC: instead, we just link to the parent!
links[0].append('<strong><a href="{}.html">{}</a></strong>'.format(\
self.up().getKey(),
self.up().getKey().replace('-', ' ')))
# DBLP
if 'dblpkey' in self.json:
for dblpk in listify(self.json['dblpkey']):
links[1].append(
'<a href="http://dblp.org/rec/html/{}">DBLP</a>'.format(
dblpk))
elif 'dblpurl' in self.json:
links[1].append('<a href="{}">DBLP</a>'.format(
self.json['dblpurl']))
else:
links[1].append('no DBLP info')
# Scholar
if 'title' in self.json.keys():
links[1].append('<a href="https://scholar.google.com/scholar?q=%22{}%22">Scholar</a>'.format(\
str(self.json['title']).replace(' ', '+')))
# Some publishers
if 'ee' in self.json.keys():
for e in listify(self.json['ee']):
if e.startswith('http://dl.acm.org') \
or e.startswith('http://doi.acm.org')\
or e.startswith('http://portal.acm.org'):
links[2].append('<a href="{}">ACM DL</a>'.format(e))
elif e.startswith('http://ieeexplore.ieee.org'):
links[2].append('<a href="{}">IEEE Xplore</a>'.format(e))
elif e.startswith('http://ieeecomputersociety.org'):
links[2].append('<a href="{}">IEEE CS</a>'.format(e))
elif e.startswith('http://drops.dagstuhl.de'):
links[2].append('<a href="{}">Dagstuhl</a>'.format(e))
elif e.find('computer.org/csdl/proceedings') > 0:
links[2].append('<a href="{}">CSDL</a>'.format(e))
elif e.startswith('http://journal.ub.tu-berlin.de/eceasst'):
links[2].append('<a href="{}">EC-EASST</a>'.format(e))
elif e.startswith('http://ceur-ws.org'):
links[2].append('<a href="{}">CEUR</a>'.format(e))
elif e.startswith('http://dx.doi.org'):
pass
else:
links[2].append('<a href="{}">?EE?</a>'.format(e))
if 'doi' in self.json.keys():
links[2].append('<a href="http://dx.doi.org/{}">DOI</a>'.format(
self.json['doi']))
return '<hr/>'.join(['<br/>\n'.join(lb) for lb in links if lb])
if p['name'] in people.keys():
cx[report(C.red('duplicate')+' '+C.yellow(p), 1)] += 1
continue
people[p['name']] = p
print('{}: {} venues, {} papers written by {} people\n{}'.format(\
C.purple('BibSLEIGH'),
C.red(len(sleigh.venues)),
C.red(sleigh.numOfPapers()),
C.red(len(people)),
C.purple('='*42)))
# traverse ALL the papers!
for v in sleigh.venues:
for c in v.getConfs():
for p in c.papers:
if 'author' in p.json.keys():
for a in listify(p.json['author']):
if a in people.keys():
if 'authored' not in people[a].keys():
people[a]['authored'] = []
if p.getKey() not in people[a]['authored']:
people[a]['authored'].append(p.getKey())
else:
report(C.yellow('Author not found: ') + a, 0)
if 'editor' in c.json.keys():
for e in listify(c.json['editor']):
if e in people.keys():
if 'edited' not in people[e].keys():
people[e]['edited'] = []
if c.getKey() not in people[e]['edited']:
people[e]['edited'].append(c.getKey())
else:
def checkon(fn, o):
if os.path.isdir(fn):
fn = fn + '.json'
f = open(fn, 'r')
lines = f.readlines()[1:-1]
f.close()
flines = [strictstrip(s) for s in lines]
plines = sorted([strictstrip(s) for s in o.getJSON().split('\n')[1:-1]])
if flines != plines:
return 1
ts = []
# precise case-sensitive match
mcs = o.get('title')
# precise match for substrings
mes = baretext(mcs)
# precise match for words
mew = mes.split(' ')
# imprecise match for substrings
mis = superbaretext(mes)
# imprecise match for words
miw = mis.split(' ')
# now match!
for t in tags:
# print('Check',t,'vs',mes)
if 'name' not in t.keys():
print(C.red('ERROR:'), 'no name for tag from file', t['FILE'])
continue
if all([not k.startswith('match') for k in t.keys()]):
print(C.red('ERROR:'), 'no match rules for tag', t['name'])
continue
for k in t.keys():
if k == 'matchentry':
if o.getKey() in t[k]:
ts += [t['name']]
elif k.startswith('match'):
ts += [t['name'] for s in listify(t[k]) if matchModes[k](s, mcs, mes, mew, mis, miw)]
# ts += [t['name'] for s in listify(t[k]) if fmm(t, k, s, mcs, mes, mew, mis, miw)]
# second pass: check reliefs
for t in tags:
if 'relieves' in t.keys():
for r in listify(t['relieves']):
if t['name'] in ts and r in ts:
ts.remove(r)
if t['name'] not in relieved.keys():
relieved[t['name']] = 0
relieved[t['name']] += 1
if ts:
if not o.tags:
o.tags = []
for t in ts:
if t not in o.tags:
o.tags.append(t)
# uncomment the following one line to overwrite all tags
o.tags = uniq(ts)
# let’s keep tags clean and sorted
o.tags = sorted(o.tags)
nlines = sorted([strictstrip(s) for s in o.getJSON().split('\n')[1:-1]])
if plines != nlines:
f = open(fn, 'w')
f.write(o.getJSON())
f.close()
return 2
else:
return 0
def getBib(self):
if len(self.json) < 1:
return '@misc{EMPTY,}'
s = '@%s{%s,\n' % (self.get('type'), self.getKey())
n2f = self.n2f if self.n2f else self.back.n2f
for k in sorted(self.json.keys()):
if k == k.upper() or k.endswith('short') or k == 'tag':
# secret key
continue
if k in ('author', 'editor'):
# TODO: add (correct!) links
aelinks = [\
'<a href="{}">{}</a>'.format(n2f[ae], ae)
if ae in n2f.keys()
else ae
for ae in listify(self.json[k])
]
s += '\t{:<13} = "{}",\n'.format(k, ' and '.join(aelinks))
elif k in ('title', 'booktitle', 'series', 'publisher', 'journal'):
if k + 'short' not in self.json.keys():
s += '\t{0:<13} = "{{{1}}}",\n'.format(k, self.json[k])
else:
s += '\t{0:<13} = "{{<span id="{0}">{1}</span>}}",\n'.format(
k, self.json[k])
elif k in ('crossref', 'key', 'type', 'venue', 'twitter', \
'eventtitle', 'eventurl', 'nondblpkey', 'dblpkey', 'dblpurl', \
'programchair', 'generalchair', 'roles', 'tagged', 'stemmed', \
'status', 'ieeepuid', 'ieeearid', 'ieeeisid', 'cite'):
# TODO: ban 'ee' as well
pass
elif k == 'doi':
s += '<span class="uri">\t{0:<13} = "<a href="http://dx.doi.org/{1}">{1}</a>",\n</span>'.format(
k, self.json[k])
elif k == 'acmid':
s += '<span class="uri">\t{0:<13} = "<a href="http://dl.acm.org/citation.cfm?id={1}">{1}</a>",\n</span>'.format(
k, self.json[k])
elif k == 'ieeearid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber={1}">{1}</a>",\n</span>'.format(
k, self.json[k])
elif k == 'ieeepuid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber={1}">{1}</a>",\n</span>'.format(
k, self.json[k])
elif k == 'ieeeisid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber={1}">{1}</a>",\n</span>'.format(
k, self.json[k])
elif k == 'dblpkey':
# Legacy!
# s += '\t{0:<13} = "<a href="http://dblp.uni-trier.de/db/{1}">{1}</a>",\n</span>'.format(k, self.json[k])
s += '\t{0:<13} = "<a href="http://dblp.uni-trier.de/rec/html/{1}">{1}</a>",\n'.format(
k, self.json[k])
elif k == 'isbn':
s += '<span id="isbn">\t{:<13} = "{}",\n</span>'.format(
k, self.json[k])
elif k in ('ee', 'url'):
for e in listify(self.json[k]):
# VVZ: eventually would like to get rid of EE completely
# VVZ: limiting it for now to possibly interesting cases
if k == 'ee' and (e.startswith('http://dx.doi.org') or \
e.startswith('http://dl.acm.org') or\
e.startswith('http://doi.ieeecomputersociety.org')\
):
continue
s += '<span class="uri">\t{0:<13} = "<a href=\"{1}\">{1}</a>",\n</span>'.format(
k, e)
elif k in ('year', 'volume', 'issue', 'number') and isinstance(
self.json[k], int):
s += '\t{0:<13} = {1},\n'.format(k, self.json[k])
elif k == 'pages':
s += '\t{0:<13} = "{1}",\n'.format(k, self.getPagesBib())
elif k == 'address':
if isinstance(self.json[k], str):
a = self.json[k]
elif self.json[k][1]:
a = ', '.join(self.json[k])
else:
a = self.json[k][0] + ', ' + self.json[k][2]
s += '\t{0:<13} = "{1}",\n'.format(k, a)
else:
s += '\t{0:<13} = "{1}",\n'.format(k, self.json[k])
s += '}'
return s.replace('<i>', '\\emph{').replace('</i>', '}')
def checkon(fn, o):
if os.path.isdir(fn):
fn = fn + '.json'
f = open(fn, 'r', encoding='utf-8')
lines = f.readlines()[1:-1]
f.close()
flines = [strictstrip(s) for s in lines]
plines = sorted([strictstrip(s) for s in o.getJSON().split('\n')[1:-1]])
if flines != plines:
return 1
ts = []
# precise case-sensitive match
mcs = o.get('title')
# precise match for substrings
mes = baretext(mcs)
# precise match for words
mew = mes.split(' ')
# imprecise match for substrings
mis = superbaretext(mes)
# imprecise match for words
miw = mis.split(' ')
# now match!
for t in tags:
# print('Check',t,'vs',mes)
if 'name' not in t.keys():
print(C.red('ERROR:'), 'no name for tag from file', t['FILE'])
continue
if all([not k.startswith('match') for k in t.keys()]):
print(C.red('ERROR:'), 'no match rules for tag', t['name'])
continue
for k in t.keys():
if k == 'matchentry':
if o.getKey() in t[k]:
ts += [t['name']]
elif k.startswith('match'):
ts += [
t['name'] for s in listify(t[k])
if matchModes[k](s, mcs, mes, mew, mis, miw)
]
# ts += [t['name'] for s in listify(t[k]) if fmm(t, k, s, mcs, mes, mew, mis, miw)]
# second pass: check reliefs
for t in tags:
if 'relieves' in t.keys():
for r in listify(t['relieves']):
if t['name'] in ts and r in ts:
ts.remove(r)
if t['name'] not in relieved.keys():
relieved[t['name']] = 0
relieved[t['name']] += 1
if ts:
if not o.tags:
o.tags = []
for t in ts:
if t not in o.tags:
o.tags.append(t)
# uncomment the following one line to overwrite all tags
o.tags = uniq(ts)
# let’s keep tags clean and sorted
o.tags = sorted(o.tags)
nlines = sorted([strictstrip(s) for s in o.getJSON().split('\n')[1:-1]])
if plines != nlines:
f = open(fn, 'w', encoding='utf-8')
f.write(o.getJSON())
f.close()
return 2
else:
return 0
print('[', C.red('NOGO'), ']', 'No name in', fn)
continue
people[p['name']] = p
print('{}: {} venues, {} papers\n{}'.format(\
C.purple('BibSLEIGH'),
C.red(len(sleigh.venues)),
C.red(sleigh.numOfPapers()),
C.purple('='*42)))
# All people who ever contributed
names = []
for v in sleigh.venues:
for c in v.getConfs():
for p in c.papers:
for k in ('author', 'editor'):
if k in p.json.keys():
names += [a for a in listify(p.json[k]) if a not in names]
# caching
peoplekeys = people.keys()
if os.path.exists('_established.json'):
established = json.load(open('_established.json', 'r'))
else:
established = {}
# print(people)
CXread = len(people)
for name in names:
if name not in peoplekeys:
p = {'name': name,\
'FILE': ienputdir + '/people/' + fileify(name) + '.json',\
'dblp': dblpify(name)}
people[p['name']] = p
# flatten conferences for easy lookup
def getBib(self):
if len(self.json) < 1:
return '@misc{EMPTY,}'
s = '@%s{%s,\n' % (self.get('type'), self.getKey())
n2f = self.n2f if self.n2f else self.back.n2f
for k in sorted(self.json.keys()):
if k == k.upper() or k.endswith('short') or k == 'tag':
# secret key
continue
if k in ('author', 'editor'):
# TODO: add (correct!) links
aelinks = [\
'<a href="{}">{}</a>'.format(n2f[ae], ae)
if ae in n2f.keys()
else ae
for ae in listify(self.json[k])
]
s += '\t{:<13} = "{}",\n'.format(k, ' and '.join(aelinks))
elif k in ('title', 'booktitle', 'series', 'publisher', 'journal'):
if k+'short' not in self.json.keys():
s += '\t{0:<13} = "{{{1}}}",\n'.format(k, self.json[k])
else:
s += '\t{0:<13} = "{{<span id="{0}">{1}</span>}}",\n'.format(k, self.json[k])
elif k in ('crossref', 'key', 'type', 'venue', 'twitter', \
'eventtitle', 'eventurl', 'nondblpkey', 'dblpkey', 'dblpurl', \
'programchair', 'generalchair', 'roles', 'tagged', 'stemmed', \
'status', 'ieeepuid', 'ieeearid', 'ieeeisid', 'cite'):
# TODO: ban 'ee' as well
pass
elif k == 'doi':
s += '<span class="uri">\t{0:<13} = "<a href="http://dx.doi.org/{1}">{1}</a>",\n</span>'.format(k, self.json[k])
elif k == 'acmid':
s += '<span class="uri">\t{0:<13} = "<a href="http://dl.acm.org/citation.cfm?id={1}">{1}</a>",\n</span>'.format(k, self.json[k])
elif k == 'ieeearid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber={1}">{1}</a>",\n</span>'.format(k, self.json[k])
elif k == 'ieeepuid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber={1}">{1}</a>",\n</span>'.format(k, self.json[k])
elif k == 'ieeeisid':
s += '<span class="uri">\t{0:<13} = "<a href="http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber={1}">{1}</a>",\n</span>'.format(k, self.json[k])
elif k == 'dblpkey':
# Legacy!
# s += '\t{0:<13} = "<a href="http://dblp.uni-trier.de/db/{1}">{1}</a>",\n</span>'.format(k, self.json[k])
s += '\t{0:<13} = "<a href="http://dblp.uni-trier.de/rec/html/{1}">{1}</a>",\n'.format(k, self.json[k])
elif k == 'isbn':
s += '<span id="isbn">\t{:<13} = "{}",\n</span>'.format(k, self.json[k])
elif k in ('ee', 'url'):
for e in listify(self.json[k]):
# VVZ: eventually would like to get rid of EE completely
# VVZ: limiting it for now to possibly interesting cases
if k == 'ee' and (e.startswith('http://dx.doi.org') or \
e.startswith('http://dl.acm.org') or\
e.startswith('http://doi.ieeecomputersociety.org')\
):
continue
s += '<span class="uri">\t{0:<13} = "<a href=\"{1}\">{1}</a>",\n</span>'.format(k, e)
elif k in ('year', 'volume', 'issue', 'number') and isinstance(self.json[k], int):
s += '\t{0:<13} = {1},\n'.format(k, self.json[k])
elif k == 'pages':
s += '\t{0:<13} = "{1}",\n'.format(k, self.getPagesBib())
elif k == 'address':
if isinstance(self.json[k], str):
a = self.json[k]
elif self.json[k][1]:
a = ', '.join(self.json[k])
else:
a = self.json[k][0] + ', ' + self.json[k][2]
s += '\t{0:<13} = "{1}",\n'.format(k, a)
else:
s += '\t{0:<13} = "{1}",\n'.format(k, self.json[k])
s += '}'
return s.replace('<i>', '\\emph{').replace('</i>', '}')
def checkon(fn, o):
if not os.path.exists(fn) or os.path.isdir(fn):
fn = fn + '.json'
f = open(fn, 'r', encoding='utf-8')
lines = f.readlines()[1:-1]
f.close()
flines = json2lines(lines)
plines = sorted(json2lines(o.getJSON().split('\n')))
# "url" from DBLP are useless
if 'url' in o.json.keys():
o.json['url'] = [link.replace('https://', 'http://')\
for link in listify(o.json['url'])\
if not link.startswith('db/conf/')\
and not link.startswith('db/series/')\
and not link.startswith('db/books/')\
and not link.startswith('db/journals/')]
if not o.json['url']:
del o.json['url']
elif len(o.json['url']) == 1:
o.json['url'] = o.json['url'][0]
if 'ee' in o.json.keys() and 'doi' not in o.json.keys():
if isinstance(o.json['ee'], list):
if verbose:
print(C.red('Manylink:'), o.json['ee'])
newee = []
for onelink in listify(o.json['ee']):
if onelink.startswith('http://dx.doi.org/'):
o.json['doi'] = onelink[18:]
elif onelink.startswith('http://doi.acm.org/'):
o.json['doi'] = onelink[19:]
elif onelink.startswith('http://doi.ieeecomputersociety.org/'):
o.json['doi'] = onelink[35:]
elif onelink.startswith('http://dl.acm.org/citation.cfm?id='):
o.json['acmid'] = onelink[34:]
elif onelink.startswith('http://portal.acm.org/citation.cfm?id='):
o.json['acmid'] = onelink[38:]
elif onelink.startswith('http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=')\
or onelink.startswith('http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber='):
o.json['ieeearid'] = onelink.split('=')[-1]
elif onelink.startswith('http://ieeexplore.ieee.org/xpls/abs_all.jsp?isnumber=')\
and onelink.find('arnumber') > -1:
o.json['ieeearid'] = onelink.split('arnumber=')[-1].split(
'&')[0]
elif onelink.startswith(
'http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber='
):
o.json['ieeepuid'] = onelink.split('=')[-1]
elif onelink.startswith(
'http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber='):
o.json['ieeeisid'] = onelink.split('=')[-1]
elif onelink.startswith(
'http://eceasst.cs.tu-berlin.de/index.php/eceasst/article/view/'
):
newee.append(
'http://journal.ub.tu-berlin.de/eceasst/article/view/' +
onelink.split('/')[-1])
elif onelink.endswith('.pdf') and \
(onelink.startswith('http://computer.org/proceedings/')\
or onelink.startswith('http://csdl.computer.org/')):
# Bad: http://computer.org/proceedings/icsm/1189/11890007.pdf
# Bad: http://csdl.computer.org/comp/proceedings/date/2003/1870/02/187020040.pdf
# Good: http://www.computer.org/csdl/proceedings/icsm/2001/1189/00/11890004.pdf
if onelink.startswith('http://csdl'):
cname, _, cid, mid, pid = onelink.split('/')[5:10]
else:
cname, cid, pid = onelink.split('/')[4:7]
# heuristic
if pid.startswith(cid):
mid = pid[len(cid):len(cid) + 2]
else:
mid = '00'
newee.append('http://www.computer.org/csdl/proceedings/{}/{}/{}/{}/{}'.format(\
cname,
o.get('year'),
cid,
mid,
pid))
else:
if onelink.find('ieee') > -1:
print(C.purple('IEEE'), onelink)
if verbose:
print(C.yellow('Missed opportunity:'), onelink)
# nothing matches => preserve
newee.append(onelink)
if len(newee) == 0:
del o.json['ee']
elif len(newee) == 1:
o.json['ee'] = newee[0]
else:
o.json['ee'] = newee
# post-processing normalisation
if 'acmid' in o.json.keys() and not isinstance(
o.json['acmid'], int) and o.json['acmid'].isdigit():
o.json['acmid'] = int(o.json['acmid'])
if 'eventuri' in o.json.keys():
o.json['eventurl'] = o.json['eventuri']
del o.json['eventuri']
if 'eventurl' in o.json.keys() and o.json['eventurl'].startswith(
'https://'):
o.json['eventurl'] = o.json['eventurl'].replace('https://', 'http://')
nlines = sorted(json2lines(o.getJSON().split('\n')))
if flines != plines:
return 1
elif plines != nlines:
f = open(fn, 'w', encoding='utf-8')
f.write(o.getJSON())
f.close()
return 2
else:
return 0
# travelled to...
# NB: code clone of AST::Venue
# cs =
ads = [c.json['address'][-1] \
for c in {bykey[p].up() for p in persondef['authored']} \
if 'address' in c.json]
if ads:
clist = {a:ads.count(a) for a in ads}
adds = '<code>Travelled to:</code><hr/>' \
+ '<br/>\n'.join(sorted(['{} × {}'.format(clist[a], a) for a in clist]))
boxlinks += adds
# collaborated with...
clist = {}
for p in persondef['authored']:
if 'author' in bykey[p].json.keys():
coas = listify(bykey[p].get('author'))
if ' ' in coas:
print('ERROR in [{}] - [{}] - [{}]'.format(p, bykey[p].getKey(), coas))
D = len(coas)
if D == 1:
# solo papers count as coauthoring with yourself
a = '∅'
if a not in clist:
clist[a] = 0
if persondef['name'] in name2file.keys():
name2file['∅'] = name2file[persondef['name']]
else:
name2file['∅'] = persondef['name']
clist[a] += 1/2
continue
for a in coas:
