def search_authors(self, query=""):
ret_limit = 100
c = Commits(index.Connector())
ret = c.es.search(index=c.index,
q=query,
df="author_name",
size=10000,
default_operator="AND",
_source_includes=["author_name", "author_email"])
ret = ret['hits']['hits']
if not len(ret):
return {}
idents = Contributors()
authors = dict([(d['_source']['author_email'],
d['_source']['author_name']) for d in ret])
result = {}
_idents = idents.get_idents_by_emails(list(authors.keys())[:ret_limit])
for iid, ident in _idents.items():
email = ident['default-email']
name = ident['name'] or authors[email]
result[utils.encrypt(xorkey, iid)] = {
'name': name,
'gravatar':
hashlib.md5(email.encode(errors='ignore')).hexdigest()
}
result = OrderedDict(
sorted(list(result.items()), key=lambda t: t[1]['name']))
return result
def contributor(self, cid=None):
if not cid:
abort(404,
detail="No contributor specified")
c = Commits(index.Connector())
idents = Contributors()
try:
cid = utils.decrypt(xorkey, cid)
except Exception:
abort(404,
detail="The cid is incorrectly formated")
_, ident = idents.get_ident_by_id(cid)
if not ident:
# No ident has been declared for that contributor
ident = list(idents.get_idents_by_emails(cid).values())[0]
mails = ident['emails']
name = ident['name']
if not name:
raw_names = c.get_commits_author_name_by_emails([cid])
if cid not in raw_names:
# TODO: get_commits_author_name_by_emails must
# support look by committer email too
name = 'Unnamed'
else:
name = raw_names[cid]
infos = {}
infos['name'] = name
infos['mails_amount'] = len(mails)
infos['gravatar'] = hashlib.md5(
ident['default-email'].encode(errors='ignore')).hexdigest()
return infos
def search_authors(self, query=""):
ret_limit = 100
c = Commits(index.Connector())
ret = c.es.search(
index=c.index, doc_type=c.dbname,
q=query, df="author_name", size=10000,
default_operator="AND",
_source_include=["author_name", "author_email"])
ret = ret['hits']['hits']
if not len(ret):
return {}
idents = Contributors()
authors = dict([(d['_source']['author_email'],
d['_source']['author_name']) for d in ret])
result = {}
_idents = idents.get_idents_by_emails(list(authors.keys())[:ret_limit])
for iid, ident in _idents.items():
email = ident['default-email']
name = ident['name'] or authors[email]
result[utils.encrypt(xorkey, iid)] = {
'name': name,
'gravatar': hashlib.md5(
email.encode(errors='ignore')).hexdigest()}
result = OrderedDict(
sorted(list(result.items()), key=lambda t: t[1]['name']))
return result
def commits(self, pid=None, tid=None, cid=None, gid=None,
start=0, limit=10,
dfrom=None, dto=None, inc_merge_commit=None,
inc_repos=None, metadata=None, exc_groups=None,
inc_groups=None):
c = Commits(index.Connector())
projects_index = Projects()
idents = Contributors()
query_kwargs = utils.resolv_filters(
projects_index, idents, pid, tid, cid, gid,
dfrom, dto, inc_repos, inc_merge_commit,
metadata, exc_groups, inc_groups)
query_kwargs.update(
{'start': start, 'limit': limit})
resp = c.get_commits(**query_kwargs)
for cmt in resp[2]:
# Get extra metadata keys
extra = set(cmt.keys()) - set(PROPERTIES.keys())
cmt['metadata'] = list(extra)
cmt['repos'] = [r for r in cmt['repos']
if not r.startswith('meta_ref: ')]
# Compute link to access commit diff based on the
# URL template provided in projects.yaml
cmt['gitwebs'] = [
projects_index.get_gitweb_link(r) %
{'sha': cmt['sha']} for r in cmt['repos']]
cmt['projects'] = utils.get_projects_from_references(
projects_index, cmt['repos'])
# Also remove the URI part
cmt['repos'] = [":".join(p.split(':')[-2:]) for
p in cmt['repos']]
# Request the ident index to fetch author/committer name/email
for elm in ('author', 'committer'):
ident = list(idents.get_idents_by_emails(
cmt['%s_email' % elm]).values())[0]
cmt['%s_email' % elm] = ident['default-email']
if ident['name']:
cmt['%s_name' % elm] = ident['name']
# Convert the TTL to something human readable
cmt['ttl'] = str((datetime.fromtimestamp(cmt['ttl']) -
datetime.fromtimestamp(0)))
cmt['author_gravatar'] = \
hashlib.md5(cmt['author_email'].encode(
errors='ignore')).hexdigest()
cmt['committer_gravatar'] = \
hashlib.md5(cmt['committer_email'].encode(
errors='ignore')).hexdigest()
if len(cmt['commit_msg']) > 80:
cmt['commit_msg'] = cmt['commit_msg'][0:76] + '...'
# Add cid and ccid
cmt['cid'] = utils.encrypt(xorkey, cmt['author_email'])
cmt['ccid'] = utils.encrypt(xorkey, cmt['committer_email'])
# Remove email details
del cmt['author_email']
del cmt['committer_email']
return resp
def authors(self, pid=None, tid=None, cid=None, gid=None,
dfrom=None, dto=None, inc_merge_commit=None,
inc_repos=None, metadata=None, exc_groups=None,
inc_groups=None):
projects_index = Projects()
idents = Contributors()
query_kwargs = utils.resolv_filters(
projects_index, idents, pid, tid, cid, gid,
dfrom, dto, inc_repos, inc_merge_commit,
metadata, exc_groups, inc_groups)
c = Commits(index.Connector())
if not c.get_commits_amount(**query_kwargs):
return []
ret = c.get_authors_histo(**query_kwargs)[1]
for bucket in ret:
_idents = idents.get_idents_by_emails(bucket['authors_email'])
bucket['value'] = len(_idents)
bucket['date'] = bucket['key_as_string']
del bucket['authors_email']
del bucket['doc_count']
del bucket['key_as_string']
del bucket['key']
return ret
def contributor(self, cid=None):
if not cid:
abort(404, detail="No contributor specified")
c = Commits(index.Connector())
idents = Contributors()
try:
cid = utils.decrypt(xorkey, cid)
except Exception:
abort(404, detail="The cid is incorrectly formated")
_, ident = idents.get_ident_by_id(cid)
if not ident:
# No ident has been declared for that contributor
ident = list(idents.get_idents_by_emails(cid).values())[0]
mails = ident['emails']
name = ident['name']
if not name:
raw_names = c.get_commits_author_name_by_emails([cid])
if cid not in raw_names:
# TODO: get_commits_author_name_by_emails must
# support look by committer email too
name = 'Unnamed'
else:
name = raw_names[cid]
infos = {}
infos['name'] = name
infos['mails_amount'] = len(mails)
infos['gravatar'] = hashlib.md5(
ident['default-email'].encode(errors='ignore')).hexdigest()
return infos
def authors(self,
pid=None,
tid=None,
cid=None,
gid=None,
dfrom=None,
dto=None,
inc_merge_commit=None,
inc_repos=None,
metadata=None,
exc_groups=None,
inc_groups=None):
projects_index = Projects()
idents = Contributors()
query_kwargs = utils.resolv_filters(projects_index, idents, pid, tid,
cid, gid, dfrom, dto, inc_repos,
inc_merge_commit, metadata,
exc_groups, inc_groups)
c = Commits(index.Connector())
if not c.get_commits_amount(**query_kwargs):
return []
ret = c.get_authors_histo(**query_kwargs)[1]
for bucket in ret:
_idents = idents.get_idents_by_emails(bucket['authors_email'])
bucket['value'] = len(_idents)
bucket['date'] = bucket['key_as_string']
del bucket['authors_email']
del bucket['doc_count']
del bucket['key_as_string']
del bucket['key']
return ret
def index(self, prefix=None, nameonly='false', withstats='false'):
ci = Commits(index.Connector())
contributors_index = Contributors()
groups = contributors_index.get_groups()
if nameonly == 'true':
ret = dict([(k, None) for k in groups.keys()])
if prefix:
ret = dict([(k, None) for k in ret.keys() if
k.lower().startswith(prefix)])
return ret
ret_groups = {}
for group, data in groups.items():
if prefix and not group.lower().startswith(prefix.lower()):
continue
rg = {'members': {},
'description': data['description'],
'domains': data.get('domains', [])}
emails = data['emails'].keys()
members = contributors_index.get_idents_by_emails(emails)
for id, member in members.items():
member['gravatar'] = hashlib.md5(
member['default-email']).hexdigest()
# TODO(fbo): bounces should be a list of bounce
# Let's deactivate that for now
# member['bounces'] = bounces
del member['emails']
if not member['name']:
# Try to find it among commits
suggested = ci.get_commits_author_name_by_emails(
[member['default-email']])
name = suggested.get(member['default-email'],
'Unknown name')
member['name'] = name
del member['default-email']
rg['members'][utils.encrypt(xorkey, id)] = member
if withstats == 'true':
# TODO(fbo): This endpoint needs to handle some filters like
# dates bounces to return more accurate stats
# Fetch the number of projects and repos contributed to
p_filter = {}
query_kwargs = {
'mails': data['emails'],
'merge_commit': False,
'repos': p_filter,
}
projects = Projects()
tops_ctl = tops.TopProjectsController()
top_projects = tops_ctl.gbycommits(
ci, projects, query_kwargs, False)
top_repos = tops_ctl.gbycommits(
ci, projects, query_kwargs, True)
rg['projects_amount'] = len(top_projects)
rg['repos_amount'] = len(top_repos)
ret_groups[group] = rg
return ret_groups
def index(self, prefix=None, nameonly='false', withstats='false',
pid=None, dfrom=None, dto=None, inc_merge_commit=None):
ci = Commits(index.Connector())
contributors_index = Contributors()
groups = contributors_index.get_groups()
if withstats == 'true':
projects_index = Projects()
if nameonly == 'true':
ret = dict([(k, None) for k in groups.keys()])
if prefix:
ret = dict([(k, None) for k in ret.keys() if
k.lower().startswith(prefix)])
return ret
ret_groups = {}
for group, data in groups.items():
if prefix and not group.lower().startswith(prefix.lower()):
continue
rg = {'members': {},
'description': data.get('description', ''),
'domains': data.get('domains', [])}
emails = list(data['emails'].keys())
members = contributors_index.get_idents_by_emails(emails)
for id, member in members.items():
member['gravatar'] = hashlib.md5(
member['default-email'].encode(
errors='ignore')).hexdigest()
# TODO(fbo): bounces should be a list of bounce
# Let's deactivate that for now
# member['bounces'] = bounces
del member['emails']
if not member['name']:
# Try to find it among commits
suggested = ci.get_commits_author_name_by_emails(
[member['default-email']])
name = suggested.get(member['default-email'],
'Unnamed')
member['name'] = name
del member['default-email']
rg['members'][utils.encrypt(xorkey, id)] = member
if withstats == 'true':
# Fetch the number of projects and repos contributed to
query_kwargs = utils.resolv_filters(
projects_index, contributors_index, pid, None, None, group,
dfrom, dto, None, inc_merge_commit, None, None, None)
repos = [r for r in ci.get_repos(**query_kwargs)[1]
if not r.startswith('meta_ref: ')]
projects = utils.get_projects_from_references(
projects_index, repos)
rg['repos_amount'] = len(repos)
rg['projects_amount'] = len(projects)
ret_groups[group] = rg
return ret_groups
def contributor(self, cid=None):
if not cid:
abort(404,
detail="No contributor specified")
try:
cid = utils.decrypt(xorkey, cid)
except Exception:
abort(404,
detail="The cid is incorrectly formated")
c = Commits(index.Connector())
idents = Contributors()
projects = Projects()
_, ident = idents.get_ident_by_id(cid)
if not ident:
# No ident has been declared for that contributor
ident = idents.get_idents_by_emails(cid).values()[0]
mails = ident['emails']
name = ident['name']
if not name:
raw_names = c.get_commits_author_name_by_emails([cid])
if cid not in raw_names:
# TODO: get_commits_author_name_by_emails must
# support look by committer email too
name = 'Unnamed'
else:
name = raw_names[cid]
p_filter = {}
query_kwargs = {
'mails': mails,
'merge_commit': False,
'repos': p_filter,
}
tops_ctl = tops.TopProjectsController()
top_projects = tops_ctl.gbycommits(c, projects, query_kwargs, False)
top_repos = tops_ctl.gbycommits(c, projects, query_kwargs, True)
infos = {}
infos['name'] = name
infos['mails_amount'] = len(mails)
infos['projects_amount'] = len(top_projects)
infos['repos_amount'] = len(top_repos)
infos['gravatar'] = hashlib.md5(ident['default-email']).hexdigest()
return infos
def index(self,
prefix=None,
nameonly='false',
withstats='false',
pid=None,
dfrom=None,
dto=None,
inc_merge_commit=None):
ci = Commits(index.Connector())
contributors_index = Contributors()
groups = contributors_index.get_groups()
if withstats == 'true':
projects_index = Projects()
if nameonly == 'true':
ret = dict([(k, None) for k in groups.keys()])
if prefix:
ret = dict([(k, None) for k in ret.keys()
if k.lower().startswith(prefix)])
return ret
ret_groups = {}
for group, data in groups.items():
if prefix and not group.lower().startswith(prefix.lower()):
continue
rg = {
'members': {},
'description': data.get('description', ''),
'domains': data.get('domains', [])
}
emails = list(data['emails'].keys())
members = contributors_index.get_idents_by_emails(emails)
for id, member in members.items():
member['gravatar'] = hashlib.md5(
member['default-email'].encode(
errors='ignore')).hexdigest()
# TODO(fbo): bounces should be a list of bounce
# Let's deactivate that for now
# member['bounces'] = bounces
del member['emails']
if not member['name']:
# Try to find it among commits
suggested = ci.get_commits_author_name_by_emails(
[member['default-email']])
name = suggested.get(member['default-email'], 'Unnamed')
member['name'] = name
del member['default-email']
rg['members'][utils.encrypt(xorkey, id)] = member
if withstats == 'true':
# Fetch the number of projects and repos contributed to
query_kwargs = utils.resolv_filters(
projects_index, contributors_index, pid, None, None, group,
dfrom, dto, None, inc_merge_commit, None, None, None)
repos = [
r for r in ci.get_repos(**query_kwargs)[1]
if not r.startswith('meta_ref: ')
]
projects = utils.get_projects_from_references(
projects_index, repos)
rg['repos_amount'] = len(repos)
rg['projects_amount'] = len(projects)
ret_groups[group] = rg
return ret_groups
