def get_unannotated_entries(databank_name):
databank = storage.find_one('databanks',{'name':databank_name})
if not databank:
raise Exception ("no such databank: " + databank_name)
# Needs a parent to determine what's missing
if 'parent_name' not in databank:
return []
entries = entries_by_pdbid(storage.find('entries',
{'databank_name': databank_name,
'$or': [{'filepath': {'$exists': True}}, {'comment': {'$exists': True }}]
}))
parent_entries = storage.find('entries',
{'databank_name': databank['parent_name'],
'filepath': {'$exists': True}
}, {'pdbid': 1})
unannotated = []
for parent_entry in parent_entries:
pdbid = parent_entry['pdbid']
if pdbid not in entries:
entry = {'pdbid': pdbid, 'databank_name': databank_name}
unannotated.append(entry)
return unannotated
def search_results_for(pdbid):
part = pdbid[1:3]
entries = entries_by_databank(storage.find('entries', {'pdbid': pdbid}))
databanks = databanks_by_name(storage.find('databanks', {}))
results = {}
for databank_name in databanks.keys():
databank = databanks[databank_name]
if databank_name in entries:
entry = entries[databank_name]
if 'filepath' in entry:
results[databank_name] = get_file_link(databank, pdbid)
elif 'comment' in entry:
results[databank_name] = entry['comment']
else:
results[databank_name] = 'Not available'
if 'parent_name' in databank:
parent_name = databank['parent_name']
if parent_name not in entries or 'comment' in entries[
parent_name]:
results[databank_name] += ', depends on %s' % parent_name
return results
def get_databank_hierarchy(name=None):
if name is None:
databanks = storage.find('databanks',
{'parent_name': {
'$exists': False
}}, {
'name': 1,
'_id': 0
})
else:
databanks = storage.find('databanks', {'parent_name': name}, {
'name': 1,
'_id': 0
})
tree = {}
for databank in databanks:
name = databank['name']
branch = get_databank_hierarchy(name)
tree[name] = branch
return tree
def search_results_for (pdbid):
part = pdbid [1:3]
entries = entries_by_databank (storage.find ('entries', {'pdbid': pdbid}))
databanks = databanks_by_name (storage.find ('databanks', {}))
results = {}
for databank_name in databanks.keys():
databank = databanks [databank_name]
if databank_name in entries:
entry = entries [databank_name]
if 'filepath' in entry:
results [databank_name] = get_file_link (databank, pdbid)
elif 'comment' in entry:
results [databank_name] = entry ['comment']
else:
results [databank_name] = 'Not available'
if 'parent_name' in databank:
parent_name = databank ['parent_name']
if parent_name not in entries or 'comment' in entries [parent_name]:
results [databank_name] += ', depends on %s' % parent_name
return results
def count_summary (databank_name):
databank = storage.find_one('databanks',{'name': databank_name})
if not databank:
raise Exception("no such databank: " + databank_name)
projection = {'pdbid':1, '_id':0}
count = {}
pdbids = Set ()
for entry in storage.find ('entries', {'databank_name': databank_name,'filepath': {'$exists': True}}, projection):
pdbids.add (entry ['pdbid'])
count ['present'] = len (pdbids)
if 'parent_name' in databank:
parent_name = databank ['parent_name']
parent_pdbids = Set()
missing_pdbids = Set()
parent_entries = storage.find ('entries', {'databank_name': parent_name,'filepath': {'$exists': True}}, projection)
comment_entries = storage.find('entries', {'databank_name': databank_name, 'comment': {'$exists': True}}, projection)
for entry in parent_entries:
parent_pdbids.add (entry ['pdbid'])
if entry ['pdbid'] not in pdbids:
missing_pdbids.add (entry ['pdbid'])
count ['missing'] = len (missing_pdbids)
count ['annotated'] = 0
for entry in comment_entries:
if entry ['pdbid'] in missing_pdbids:
count ['annotated'] += 1
# missing = annotated + unannotated
count ['unannotated'] = count ['missing'] - count ['annotated']
count ['obsolete'] = 0
for pdbid in pdbids:
if pdbid not in parent_pdbids:
count ['obsolete'] += 1
count ['valid'] = count ['present'] - count ['obsolete']
else: # no parent, so nothing is missing or obsolete
count ['missing'] = 0
count ['valid'] = count ['present']
count ['obsolete'] = 0
count ['annotated'] = 0
count ['unannotated'] = 0
return count
def comment_summary():
comments = {}
for entry in storage.find('entries', {
'comment': {
'$exists': True
},
'mtime': {
'$exists': True
}
}, {
'mtime': 1,
'comment': 1,
'_id': 0
}):
text = entry['comment']
if text not in comments:
comments[text] = {
'text': text,
'n_entries': 0,
'mtime': entry['mtime']
}
comments[text]['n_entries'] += 1
if comments[text]['mtime'] < entry['mtime']:
comments[text]['mtime'] = entry['mtime']
return comments.values()
def get_databank_hierarchy (name = None):
if name is None:
databanks = storage.find ('databanks', {'parent_name': {'$exists': False}}, {'name': 1, '_id': 0})
else:
databanks = storage.find ('databanks', {'parent_name': name}, {'name': 1, '_id': 0})
tree = {}
for databank in databanks:
name = databank ['name']
branch = get_databank_hierarchy (name)
tree [name] = branch
return tree
def get_missing_entries(databank_name):
databank = storage.find_one('databanks', {'name': databank_name})
if not databank:
raise Exception("no such databank: " + databank_name)
# Needs a parent to determine what's missing
if 'parent_name' not in databank:
return []
entries = entries_by_pdbid(
storage.find('entries', {'databank_name': databank_name}))
missing = []
for entry in get_present_entries(databank['parent_name']):
pdbid = entry['pdbid']
if pdbid in entries:
if 'filepath' not in entries[pdbid] or 'mtime' not in entries[
pdbid]:
missing.append(entries[pdbid])
else:
entry = {'pdbid': pdbid, 'databank_name': databank_name}
missing.append(entry)
return missing
def crawl_files (databank, pathnames):
present_entries_bypdbid = entries_by_pdbid (get_present_entries (databank ['name']))
record_pdbids = entries_by_pdbid (storage.find ('entries', {'databank_name': databank ['name']}, {'pdbid':1}))
pattern = parse_regex (databank['regex'])
for f in pathnames:
# Only use files that match the databank's pattern.
m = pattern.search(f)
if not m:
continue
# For disk files take their mtimes, for urls take current time.
mtime = time ()
if os.path.isfile (f):
mtime = os.path.getmtime (f)
entry = {
'databank_name': databank['name'],
'pdbid': m.group(1).lower(),
'filepath': f,
'mtime': mtime
}
if entry ['pdbid'] in present_entries_bypdbid:
continue
if entry ['pdbid'] in record_pdbids:
storage.update ('entries', {'databank_name': databank ['name'], 'pdbid': entry ['pdbid']}, entry)
else:
storage.insert ('entries', entry)
def crawl_lines (databank, filepath, lines):
present_entries_bypdbid = entries_by_pdbid(get_present_entries(databank['name']))
record_pdbids = entries_by_pdbid(storage.find('entries',{'databank_name':databank['name']}, {'pdbid':1}))
pattern = parse_regex(databank['regex'])
# If it's a disk file take its mtime, for urls take current time.
mtime = time()
if os.path.isfile (filepath):
mtime = os.path.getmtime (filepath)
for line in lines:
# Only use lines that match the databank's pattern
m = pattern.search (line)
if not m:
continue
entry = {
'databank_name': databank['name'],
'pdbid': m.group(1).lower(),
'filepath': filepath,
'mtime': mtime
}
if entry['pdbid'] in present_entries_bypdbid:
continue
if entry['pdbid'] in record_pdbids:
storage.update('entries', {'databank_name':databank['name'], 'pdbid':entry['pdbid']}, entry)
else:
storage.insert('entries', entry)
def get_entries_with_comment(databank_name, comment):
# ordering was found to make it take longer!
return storage.find('entries', {
'databank_name': databank_name,
'comment': comment
},
order=[("pdbid", pymongo.ASCENDING)])
def get_entries_with_pdbid(databank_name, pdbid):
# ordering was found to make it take longer!
return storage.find('entries', {
'databank_name': databank_name,
'pdbid': pdbid
},
order=[("pdbid", pymongo.ASCENDING)])
def databanks(name=None):
start_time = time()
if name is None:
databanks = storage.find('databanks', {})
else:
databanks = [ storage.find_one('databanks', {'name': name}) ]
end_time = time()
return render_template('databank/DatabankPage.html', db_tree=db_tree, nav_disabled='databanks', databanks=databanks)
def load_statistics():
_log.info("request for statistics")
#TODO: speed up this method
ndb = storage.count('databanks', {})
ne = 0
na = 0
nf = 0
nc = 0
unique_comments = Set()
recent_files = top_highest(10)
recent_annotations = top_highest(10)
for entry in storage.find('entries', {}):
ne += 1
if 'mtime' in entry:
if 'filepath' in entry:
nf += 1
recent_files.add(entry['mtime'], entry)
elif 'comment' in entry:
na += 1
unique_comments.add(entry['comment'])
recent_annotations.add(entry['mtime'], entry)
# Perform time-consuming operations only on the last 10 files and annotations
files = []
for f in recent_files.get():
files.append({
'path': f['filepath'],
'date': strftime(date_format, gmtime(f['mtime']))
})
annotations = []
for a in recent_annotations.get():
annotations.append({
'comment': a['comment'],
'pdbid': a['pdbid'],
'databank_name': a['databank_name'],
'date': strftime(date_format, gmtime(a['mtime']))
})
nc = len(unique_comments)
statistics = {}
statistics['total_databanks'] = ndb
statistics['total_entries'] = ne
statistics['total_files'] = nf
statistics['total_annotations'] = na
statistics['total_comments'] = nc
statistics['annotations'] = annotations
statistics['files'] = files
return jsonify(statistics)
def databanks(name=None):
start_time = time()
if name is None:
databanks = storage.find('databanks', {})
else:
databanks = [storage.find_one('databanks', {'name': name})]
end_time = time()
return render_template('databank/DatabankPage.html',
db_tree=db_tree,
nav_disabled='databanks',
databanks=databanks)
def get_annotated_entries(databank_name):
return storage.find('entries', {
'databank_name': databank_name,
'comment': {
'$exists': True
},
'filepath': {
'$exists': False
}
},
order=[("pdbid", pymongo.ASCENDING)])
def get_present_entries(databank_name, ordered=False):
# ordering was found to make it take longer!
return storage.find('entries', {
'databank_name': databank_name,
'filepath': {
'$exists': True
},
'mtime': {
'$exists': True
}
},
order=[("pdbid", pymongo.ASCENDING)])
def comment_summary ():
comments = {}
for entry in storage.find ('entries', {'comment': {'$exists': True}, 'mtime': {'$exists': True}}, {'mtime':1, 'comment':1, '_id':0}):
text = entry ['comment']
if text not in comments:
comments [text] = {'text':text, 'n_entries': 0, 'mtime': entry ['mtime']}
comments [text]['n_entries'] += 1
if comments [text]['mtime'] < entry ['mtime']:
comments [text]['mtime'] = entry ['mtime']
return comments.values ()
def load_statistics():
_log.info("request for statistics")
#TODO: speed up this method
ndb = storage.count('databanks', {})
ne = 0
na = 0
nf = 0
nc = 0
unique_comments = Set()
recent_files = top_highest(10)
recent_annotations = top_highest(10)
for entry in storage.find('entries', {}):
ne += 1
if 'mtime' in entry:
if 'filepath' in entry:
nf += 1
recent_files.add(entry ['mtime'], entry)
elif 'comment' in entry:
na += 1
unique_comments.add(entry ['comment'])
recent_annotations.add(entry ['mtime'], entry)
# Perform time-consuming operations only on the last 10 files and annotations
files = []
for f in recent_files.get():
files.append({'path': f['filepath'], 'date': strftime(date_format, gmtime(f['mtime']))})
annotations = []
for a in recent_annotations.get():
annotations.append({'comment': a ['comment'], 'pdbid': a ['pdbid'],
'databank_name': a ['databank_name'],
'date': strftime(date_format, gmtime(a['mtime']))})
nc = len(unique_comments)
statistics = {}
statistics ['total_databanks'] = ndb
statistics ['total_entries'] = ne
statistics ['total_files'] = nf
statistics ['total_annotations'] = na
statistics ['total_comments'] = nc
statistics ['annotations'] = annotations
statistics ['files'] = files
return jsonify(statistics)
def crawl_lines(databank, filepath, lines):
present_entries_bypdbid = entries_by_pdbid(
get_present_entries(databank['name']))
record_pdbids = entries_by_pdbid(
storage.find('entries', {'databank_name': databank['name']},
{'pdbid': 1}))
pattern = parse_regex(databank['regex'])
# If it's a disk file take its mtime, for urls take current time.
mtime = time()
if os.path.isfile(filepath):
mtime = os.path.getmtime(filepath)
for line in lines:
# Only use lines that match the databank's pattern
m = pattern.search(line)
if not m:
continue
entry = {
'databank_name': databank['name'],
'pdbid': m.group(1).lower(),
'filepath': filepath,
'mtime': mtime
}
if entry['pdbid'] in present_entries_bypdbid:
continue
if entry['pdbid'] in record_pdbids:
storage.update('entries', {
'databank_name': databank['name'],
'pdbid': entry['pdbid']
}, entry)
else:
storage.insert('entries', entry)
def crawl_files(databank, pathnames):
present_entries_bypdbid = entries_by_pdbid(
get_present_entries(databank['name']))
record_pdbids = entries_by_pdbid(
storage.find('entries', {'databank_name': databank['name']},
{'pdbid': 1}))
pattern = parse_regex(databank['regex'])
for f in pathnames:
# Only use files that match the databank's pattern.
m = pattern.search(f)
if not m:
continue
# For disk files take their mtimes, for urls take current time.
mtime = time()
if os.path.isfile(f):
mtime = os.path.getmtime(f)
entry = {
'databank_name': databank['name'],
'pdbid': m.group(1).lower(),
'filepath': f,
'mtime': mtime
}
if entry['pdbid'] in present_entries_bypdbid:
continue
if entry['pdbid'] in record_pdbids:
storage.update('entries', {
'databank_name': databank['name'],
'pdbid': entry['pdbid']
}, entry)
else:
storage.insert('entries', entry)
#!/usr/bin/python
import sys
import os
import commands
sys.path.append(os.path.join(os.path.dirname(__file__), '..'))
from storage import storage
from sets import Set
from httplib import HTTPConnection
from utils import has_annotated_parent, has_present_parent, databanks_by_name, valid_path, get_entry
databanks = databanks_by_name(storage.find('databanks', {}))
regexes = {}
for name in databanks:
regexes[name] = databanks[name]['regex'].try_compile()
psql_defs = 'export PGPASSWORD=oon6oo4J'
psql_call = 'psql whynot2 -h cmbi11 -U whynotuser'
# Compare entries
entries_mongo = {}
files_mongo = {}
comments_mongo = {}
entries_psql = {}
comments_psql = {}
files_psql = {}
def get_entries_with_pdbid (databank_name, pdbid):
# ordering was found to make it take longer!
return storage.find ('entries', {'databank_name': databank_name, 'pdbid': pdbid}, order=[("pdbid", pymongo.ASCENDING)])
from urllib2 import urlopen
from storage import storage
import pymongo
from sets import Set
from httplib import HTTPConnection
def databanks_by_name (databanks):
d = {}
for databank in databanks:
d [databank ['name']] = databank
return d
databanks = databanks_by_name (storage.find ('databanks', {}))
databank_regexes = {}
for name in databanks:
databank_regexes [name] = databanks [name]['regex'].try_compile ()
# Verifies that the path contains the regex that the databank prescribes.
# For a file path, checks that the file exists
# For an url, checks that the url points to a vallid location. (hssp response 200)
def valid_path (databank_name, path):
if not databank_regexes [databank_name].search (path):
return False
if path.startswith ('http://') or path.startswith ('ftp://'):
#!/usr/bin/python
import sys
import os
import commands
sys.path.append (os.path.join(os.path.dirname (__file__), '..'))
from storage import storage
from sets import Set
from httplib import HTTPConnection
from utils import has_annotated_parent, has_present_parent, databanks_by_name, valid_path, get_entry
databanks = databanks_by_name (storage.find ('databanks', {}))
regexes = {}
for name in databanks:
regexes [name] = databanks [name]['regex'].try_compile ()
psql_defs = 'export PGPASSWORD=oon6oo4J'
psql_call = 'psql whynot2 -h cmbi11 -U whynotuser'
# Compare entries
entries_mongo = {}
files_mongo = {}
comments_mongo = {}
entries_psql = {}
comments_psql = {}
files_psql = {}
def count_summary(databank_name):
databank = storage.find_one('databanks', {'name': databank_name})
if not databank:
raise Exception("no such databank: " + databank_name)
projection = {'pdbid': 1, '_id': 0}
count = {}
pdbids = Set()
for entry in storage.find('entries', {
'databank_name': databank_name,
'filepath': {
'$exists': True
}
}, projection):
pdbids.add(entry['pdbid'])
count['present'] = len(pdbids)
if 'parent_name' in databank:
parent_name = databank['parent_name']
parent_pdbids = Set()
missing_pdbids = Set()
parent_entries = storage.find('entries', {
'databank_name': parent_name,
'filepath': {
'$exists': True
}
}, projection)
comment_entries = storage.find('entries', {
'databank_name': databank_name,
'comment': {
'$exists': True
}
}, projection)
for entry in parent_entries:
parent_pdbids.add(entry['pdbid'])
if entry['pdbid'] not in pdbids:
missing_pdbids.add(entry['pdbid'])
count['missing'] = len(missing_pdbids)
count['annotated'] = 0
for entry in comment_entries:
if entry['pdbid'] in missing_pdbids:
count['annotated'] += 1
# missing = annotated + unannotated
count['unannotated'] = count['missing'] - count['annotated']
count['obsolete'] = 0
for pdbid in pdbids:
if pdbid not in parent_pdbids:
count['obsolete'] += 1
count['valid'] = count['present'] - count['obsolete']
else: # no parent, so nothing is missing or obsolete
count['missing'] = 0
count['valid'] = count['present']
count['obsolete'] = 0
count['annotated'] = 0
count['unannotated'] = 0
return count
from storage import storage
import pymongo
from sets import Set
from httplib import HTTPConnection
def databanks_by_name(databanks):
d = {}
for databank in databanks:
d[databank['name']] = databank
return d
databanks = databanks_by_name(storage.find('databanks', {}))
databank_regexes = {}
for name in databanks:
databank_regexes[name] = databanks[name]['regex'].try_compile()
# Verifies that the path contains the regex that the databank prescribes.
# For a file path, checks that the file exists
# For an url, checks that the url points to a vallid location. (hssp response 200)
def valid_path(databank_name, path):
if not databank_regexes[databank_name].search(path):
return False
if path.startswith('http://') or path.startswith('ftp://'):
def get_all_entries_with_comment (comment):
# ordering was found to make it take longer!
return storage.find ('entries', {'comment': comment}, order=[("pdbid", pymongo.ASCENDING)])
def get_entries_with_comment (databank_name, comment):
# ordering was found to make it take longer!
return storage.find ('entries', {'databank_name': databank_name, 'comment': comment}, order=[("pdbid", pymongo.ASCENDING)])
def get_annotated_entries (databank_name):
return storage.find('entries', {'databank_name': databank_name, 'comment': {'$exists': True}, 'filepath': {'$exists': False}},
order=[("pdbid", pymongo.ASCENDING)])
def get_all_entries_with_comment(comment):
# ordering was found to make it take longer!
return storage.find('entries', {'comment': comment},
order=[("pdbid", pymongo.ASCENDING)])
def get_present_entries (databank_name, ordered=False):
# ordering was found to make it take longer!
return storage.find('entries', {'databank_name': databank_name,'filepath': {'$exists': True}, 'mtime': {'$exists': True}},
order=[("pdbid", pymongo.ASCENDING)])
