def SummarizeMetaData(X):
if 'image' in X.keys():
from PIL import Image
K = 125
try:
x = Image.open('..' + X['image'])
except:
print 'Importing image', X['image'], 'failed. here is the error:'
print_exc()
sizetag = ''
else:
(w,h) = x.size
if w > K or h > K:
r = float(max(w,h))
w = int(w * K/r)
h = int(h * K/r)
sizetag = 'width="' + str(w) + '" height="' + str(h) + '"'
image = '<img src="' + X['image'] + '" ' + sizetag + '/><br/>'
else:
image = ''
if 'description' in X.keys():
description = '<strong>Description: </strong>' + X['description'].replace('\n','<br/>')
else:
description = ''
if 'author' in X.keys():
author = '<strong>Author: </strong>' + X['author']
else:
author = ''
if 'title' in X.keys():
title = '<strong>Title: </strong>' + X['title']
else:
title = ''
if 'keywords' in X.keys():
keywords = '<strong>Keywords: </strong>' + X['keywords']
else:
keywords = ''
if 'signature' in X.keys():
if X['signature'] != 'directory':
signature = '<strong>Signature: </strong> This appears to be a ' + X['signature'] + ' file.'
elif 'DIR_signature' in X.keys():
signature = '<strong>Signature: </strong> This is a directory consisting of ' + X['DIR_signature'] + ' files.'
else:
signature = ''
else:
signature = ''
X['signature'] = ''
nr = [x for x in X.keys() if x.endswith('nrows')]
nc = [x for x in X.keys() if x.endswith('ncols')]
preamble = 'It has' if X['signature'] == 'tabular' else 'Its constituent datasets commonly have' if X['signature'] == 'directory' and 'DIR_signature' in X.keys() and X['DIR_signature'] == 'tabular' else 'This data has'
if len(nr) > 0 and len(nc) > 0:
ending = str(X[nr[0]]) + ' rows and ' + str(X[nc[0]]) + ' columns.'
elif len(nr) > 0:
ending = str(X[nr[0]]) + ' rows.'
elif len(nc) > 0:
ending = str(X[nc[0]]) + ' columns.'
else:
ending = ''
if ending != '':
tabulartext = preamble + ' ' + ending
else:
tabulartext = ''
nn = [x for x in X.keys() if x.endswith('colnames')]
if len(nn) > 0:
names = X[nn[0]]
nt = [x for x in X.keys() if x.endswith('coltypes') and len(X[x]) == len(X[nn[0]])]
if len(nt):
types = [' (' + t + ')' for t in X[nt[0]]]
else:
types = ['']*len(names)
nd = [x for x in X.keys() if x.endswith('coldescrs') and isinstance(X[x],dict) and set(X[x].keys()).intersection(names)]
if len(nd) > 0:
descrs = X[nd[0]]
descrs = [': ' + descrs[n] if n in descrs.keys() else '' for n in names]
else:
descrs = ['']*len(names)
coltext = 'The columns are:<br/>' + '<br/>'.join(['<strong>'+n+'</strong>' + t + d for (n,t,d) in zip(names,types,descrs)])
else:
coltext = ''
nt = []
nd = []
#frequentwords
if 'frequentwords' in X.keys():
frequentwords = 'Frequent words in this data include: ' + repr(X['frequentwords']) + '.'
else:
frequentwords = ''
text = '<br/><br/>'.join([x for x in [image,title,author,description,signature,tabulartext,coltext,frequentwords,keywords] if x != ''])
OtherKeys = set(X.keys()).difference(['image','coloring','description','author','title','keywords','signature','frequentwords','colformats','nfiles','npaths'] + nr + nc + nn + nt + nd)
if OtherKeys:
text += '<br/><br/><strong>Other MetaData</strong>:' + '<br/><br/>'.join(['<strong>' + k + ': </strong>' + (X[k] if is_string_like(X[k]) else repr(X[k])) for k in OtherKeys])
return text
def ConsolidateSources(metapath,objname=None,extensions=None):
consolidated = {}
if extensions is None:
extensions = ['Attached','Associated','Automatic','Inherited']
combined = CombineSources(metapath,extensions=extensions)
if 'Resources' in combined.keys():
consolidated['Resources'] = uniqify(ListUnion(combined['Resources'].values()))
if 'image' in combined.keys():
consolidated['image'] = ListUnion([x.split(',') if is_string_like(x) else x for x in combined['image'].values()])
if 'author' in combined.keys():
consolidated['author'] = '; '.join(combined['author'].values())
if 'title' in combined.keys():
consolidated['title'] = '; '.join(combined['title'].values())
if 'description' in combined.keys():
descrs = combined['description'].items()
if len(descrs) == 1:
consolidated['description'] = descrs[0][1]
else:
consolidated['description'] = '\n\n'.join([e + ': ' + d for (e,d) in descrs])
elif 'Verbose' in combined.keys():
descrs = combined['Verbose'].items()
if len(descrs) == 1:
consolidated['description'] = descrs[0][1]
else:
consolidated['description'] = '\n\n'.join([e + ': ' + d for (e,d) in descrs])
if 'keywords' in combined.keys():
for k in combined['keywords'].keys():
if not is_string_like(combined['keywords'][k]):
combined['keywords'][k] = ','.join(combined['keywords'][k])
consolidated['keywords'] = ','.join([x.strip() for x in uniqify((','.join(combined['keywords'].values())).split(','))])
if 'signature' in combined.keys():
s = uniqify(combined['signature'].values())
if len(s) == 1:
consolidated['signature'] = s[0]
else:
consolidated['signature'] = ''
L = ['nrows','ncols','coloring','wordassoc','colformats','coltypes','colnames','wordassoc','frequentwords','nfiles','npaths']
LL = L + [x for x in combined.keys() if x.startswith('DIR_')]
for x in LL:
if x in combined.keys() and 'Automatic' in combined[x].keys():
consolidated[x] = combined[x]['Automatic']
elif x in combined.keys() and 'Attached' in combined[x].keys():
consolidated[x] = combined[x]['Automatic']
elif x in combined.keys() and 'Associated' in combined[x].keys():
consolidated[x] = combined[x]['Associated']
elif x in combined.keys() and 'Inherited' in combined[x].keys():
consolidated[x] = combined[x]['Inherited']
if 'coldescrs' in combined.keys():
coldescrs = {}
for c in combined['coldescrs'].values():
if isinstance(c,dict):
for k in c.keys():
if k in coldescrs.keys():
coldescrs[k] += (c[k],)
else:
coldescrs[k] = (c[k],)
for k in coldescrs.keys():
coldescrs[k] = '\n'.join(coldescrs[k])
consolidated['coldescrs'] = coldescrs
OtherKeys = set(combined.keys()).difference(consolidated.keys())
for k in OtherKeys:
consolidated[k] = ' '.join([x if is_string_like(x) else repr(x) for x in combined[k].values()])
return consolidated
