INSTRUCTION
stringlengths
1
8.43k
RESPONSE
stringlengths
75
104k
Generates python code calling a hook.
def visit_Hook(self, node: parsing.Hook) -> ast.expr: """Generates python code calling a hook. self.evalHook('hookname', self.ruleNodes[-1]) """ return ast.Call( ast.Attribute( ast.Name('self', ast.Load()), 'evalHook', ast.Load()), [ ast.Str(node.name), ast.Subscript( ast.Attribute( ast.Name('self', ast.Load()), 'ruleNodes', ast.Load()), ast.Index(ast.UnaryOp(ast.USub(), ast.Num(1))), ast.Load())], [], None, None)
Generates python code calling a rule.
def visit_Rule(self, node: parsing.Rule) -> ast.expr: """Generates python code calling a rule. self.evalRule('rulename') """ return ast.Call( ast.Attribute(ast.Name('self', ast.Load()), 'evalRule', ast.Load()), [ast.Str(node.name)], [], None, None)
Generates python code to capture text consumed by a clause.
def visit_Capture(self, node: parsing.Capture) -> [ast.stmt] or ast.expr: """Generates python code to capture text consumed by a clause. #If all clauses can be inlined self.beginTag('tagname') and clause and self.endTag('tagname') if not self.beginTag('tagname'): return False <code for the clause> if not self.endTag('tagname'): return False """ begintag = ast.Attribute( ast.Name('self', ast.Load()), 'beginTag', ast.Load()) endtag = ast.Attribute( ast.Name('self', ast.Load()), 'endTag', ast.Load()) begin = ast.Call(begintag, [ast.Str(node.tagname)], [], None, None) end = ast.Call(endtag, [ast.Str(node.tagname)], [], None, None) result = [begin, self.visit(node.pt), end] for clause in result: if not isinstance(clause, ast.expr): break else: return ast.BoolOp(ast.And(), result) res = [] for stmt in map(self._clause, result): res.extend(stmt) return res
Generates python code for a scope.
def visit_Scope(self, node: parsing.Capture) -> [ast.stmt] or ast.expr: """Generates python code for a scope. if not self.begin(): return False res = self.pt() if not self.end(): return False return res """ return ast.Name('scope_not_implemented', ast.Load()) raise NotImplementedError()
Generates python code for alternatives.
def visit_Alt(self, node: parsing.Alt) -> [ast.stmt]: """Generates python code for alternatives. try: try: <code for clause> #raise AltFalse when alternative is False raise AltTrue() except AltFalse: pass return False except AltTrue: pass """ clauses = [self.visit(clause) for clause in node.ptlist] for clause in clauses: if not isinstance(clause, ast.expr): break else: return ast.BoolOp(ast.Or(), clauses) res = ast.Try([], [ast.ExceptHandler( ast.Name('AltTrue', ast.Load()), None, [ast.Pass()])], [], []) alt_true = [ast.Raise(ast.Call( ast.Name('AltTrue', ast.Load()), [], [], None, None), None)] alt_false = [ast.ExceptHandler( ast.Name('AltFalse', ast.Load()), None, [ast.Pass()])] self.in_try += 1 for clause in node.ptlist: res.body.append( ast.Try(self._clause(self.visit(clause)) + alt_true, alt_false, [], [])) self.in_try -= 1 res.body.append(self.__exit_scope()) return [res]
Generates python code for clauses.
def visit_Seq(self, node: parsing.Seq) -> [ast.stmt] or ast.expr: """Generates python code for clauses. #Continuous clauses which can can be inlined are combined with and clause and clause if not clause: return False if not clause: return False """ exprs, stmts = [], [] for clause in node.ptlist: clause_ast = self.visit(clause) if isinstance(clause_ast, ast.expr): exprs.append(clause_ast) else: if exprs: stmts.extend(self.combine_exprs_for_clauses(exprs)) exprs = [] stmts.extend(self._clause(clause_ast)) if not stmts: return ast.BoolOp(ast.And(), exprs) if exprs: stmts.extend(self.combine_exprs_for_clauses(exprs)) return stmts
Generates python code for an optional clause.
def visit_RepOptional(self, node: parsing.RepOptional) -> ([ast.stmt] or ast.expr): """Generates python code for an optional clause. <code for the clause> """ cl_ast = self.visit(node.pt) if isinstance(cl_ast, ast.expr): return ast.BoolOp(ast.Or(), [cl_ast, ast.Name('True', ast.Load())]) self.in_optional += 1 cl_ast = self.visit(node.pt) self.in_optional -= 1 return cl_ast
Generates python code for a clause repeated 0 or more times.
def visit_Rep0N(self, node: parsing.Rep0N) -> [ast.stmt]: """Generates python code for a clause repeated 0 or more times. #If all clauses can be inlined while clause: pass while True: <code for the clause> """ cl_ast = self.visit(node.pt) if isinstance(cl_ast, ast.expr): return [ast.While(cl_ast, [ast.Pass()], [])] self.in_loop += 1 clause = self._clause(self.visit(node.pt)) self.in_loop -= 1 return [ast.While(ast.Name('True', ast.Load()), clause, [])]
Generates python code for a clause repeated 1 or more times.
def visit_Rep1N(self, node: parsing.Rep0N) -> [ast.stmt]: """Generates python code for a clause repeated 1 or more times. <code for the clause> while True: <code for the clause> """ clause = self.visit(node.pt) if isinstance(clause, ast.expr): return (self._clause(clause) + self.visit_Rep0N(node)) self.in_loop += 1 clause = self._clause(self.visit(node.pt)) self.in_loop -= 1 return self._clause(self.visit(node.pt)) + [ ast.While(ast.Name('True', ast.Load()), clause, [])]
month format: YYYYMM
def synthese(self, month=None): """ month format: YYYYMM """ if month is None and self.legislature == '2012-2017': raise AssertionError('Global Synthesis on legislature does not work, see https://github.com/regardscitoyens/nosdeputes.fr/issues/69') if month is None: month = 'data' url = '%s/synthese/%s/%s' % (self.base_url, month, self.format) data = requests.get(url).json() return [depute[self.ptype] for depute in data[self.ptype_plural]]
cat two strings but handle \ n for tabulation
def catend(dst: str, src: str, indent) -> str: """cat two strings but handle \n for tabulation""" res = dst txtsrc = src if not isinstance(src, str): txtsrc = str(src) for c in list(txtsrc): if len(res) > 0 and res[-1] == '\n': res += (indentable.char_indent * indentable.num_indent) * \ (indent - 1) + c else: res += c return res
recurs into list for indentation
def list_set_indent(lst: list, indent: int=1): """recurs into list for indentation""" for i in lst: if isinstance(i, indentable): i.set_indent(indent) if isinstance(i, list): list_set_indent(i, indent)
recurs into list for string computing
def list_to_str(lst: list, content: str, indent: int=1): """recurs into list for string computing """ for i in lst: if isinstance(i, indentable): content = i.to_str(content, indent) elif isinstance(i, list): content = list_to_str(i, content, indent) elif isinstance(i, str): content = catend(content, i, indent) return content
Print nodes.
def echo_nodes(self, *rest): """ Print nodes. example:: R = [ In : node #echo("coucou", 12, node) ] """ txt = "" for thing in rest: if isinstance(thing, Node): txt += self.value(thing) else: txt += str(thing) print(txt) return True
function that connect each other one sequence of MatchExpr.
def populate_from_sequence(seq: list, r: ref(Edge), sr: state.StateRegister): """ function that connect each other one sequence of MatchExpr. """ base_state = r # we need to detect the last state of the sequence idxlast = len(seq) - 1 idx = 0 for m in seq: # alternatives are represented by builtin list if isinstance(m, list): # so recursively connect all states of each alternative sequences. for item in m: populate_from_sequence(item, r, sr) elif isinstance(m, MatchExpr): # from the current state, have we a existing edge for this event? eX = r().get_next_edge(m) if eX is None: sX = None if idx != idxlast: sX = state.State(sr) sX.matchDefault(base_state().s) else: # last state of sequence return to the base sX = base_state().s eX = Edge(sX) r().next_edge[id(sX)] = eX m.attach(r().s, sX, sr) r = ref(eX) idx += 1
function that create a state for all instance of MatchExpr in the given list and connect each others.
def populate_state_register(all_seq: [list], sr: state.StateRegister) -> Edge: """ function that create a state for all instance of MatchExpr in the given list and connect each others. """ # Basic State s0 = state.State(sr) # loop on himself s0.matchDefault(s0) # this is default sr.set_default_state(s0) # use Edge to store connection e0 = Edge(s0) for seq in all_seq: r = ref(e0) # merge all sequences into one tree automata populate_from_sequence(seq, r, sr) # return edge for debug purpose return e0
main function for creating a bottom - up tree automata for a block of matching statements.
def build_state_tree(self, tree: list, sr: state.StateRegister): """ main function for creating a bottom-up tree automata for a block of matching statements. """ all_seq = [] # for all statements populate a list # from deeper to nearer of MatchExpr instances. for stmt in self.stmts: part_seq = list() stmt.build_state_tree(part_seq) all_seq.append(part_seq) # Walk on all MatchExpr instance # and create State instance into the StateRegister self.root_edge = populate_state_register(all_seq, sr)
Test if a node set with setint or setstr equal a certain value
def pred_eq(self, n, val): """ Test if a node set with setint or setstr equal a certain value example:: R = [ __scope__:n ['a' #setint(n, 12) | 'b' #setint(n, 14)] C [#eq(n, 12) D] ] """ v1 = n.value v2 = val if hasattr(val, 'value'): v2 = val.value if isinstance(v1, int) and not isinstance(v2, int): return v1 == int(v2) return v1 == v2
Create a Grammar from a string
def from_string(bnf: str, entry=None, *optional_inherit) -> Grammar: """ Create a Grammar from a string """ inherit = [Grammar] + list(optional_inherit) scope = {'grammar': bnf, 'entry': entry} return build_grammar(tuple(inherit), scope)
Create a Grammar from a file
def from_file(fn: str, entry=None, *optional_inherit) -> Grammar: """ Create a Grammar from a file """ import os.path if os.path.exists(fn): f = open(fn, 'r') bnf = f.read() f.close() inherit = [Grammar] + list(optional_inherit) scope = {'grammar': bnf, 'entry': entry, 'source': fn} return build_grammar(tuple(inherit), scope) raise Exception("File not Found!")
Parse source using the grammar
def parse(self, source: str=None, entry: str=None) -> parsing.Node: """Parse source using the grammar""" self.from_string = True if source is not None: self.parsed_stream(source) if entry is None: entry = self.entry if entry is None: raise ValueError("No entry rule name defined for {}".format( self.__class__.__name__)) return self._do_parse(entry)
Parse filename using the grammar
def parse_file(self, filename: str, entry: str=None) -> parsing.Node: """Parse filename using the grammar""" self.from_string = False import os.path with open(filename, 'r') as f: self.parsed_stream(f.read(), os.path.abspath(filename)) if entry is None: entry = self.entry if entry is None: raise ValueError("No entry rule name defined for {}".format( self.__class__.__name__)) return self._do_parse(entry)
Basically copy one node to another. usefull to transmit a node from a terminal rule as result of the current rule.
def set_node(self, dst, src): """ Basically copy one node to another. usefull to transmit a node from a terminal rule as result of the current rule. example:: R = [ In : node #set(_, node) ] here the node return by the rule In is also the node return by the rule R """ if not isinstance(src, Node): dst.value = src else: dst.set(src) idsrc = id(src) iddst = id(dst) if iddst not in self.id_cache: print("DST: %s" % repr(dst)) print("RULE_NODES %s" % repr(self.rule_nodes)) print("IDCACHE %s" % repr(self.id_cache)) if idsrc in self.id_cache: k = self.id_cache[idsrc] k2 = self.id_cache[iddst] if k in self.rule_nodes: self.tag_cache[k2] = self.tag_cache[k] return True
Set a node to a value captured from another node
def set_node_as_int(self, dst, src): """ Set a node to a value captured from another node example:: R = [ In : node #setcapture(_, node) ] """ dst.value = self.value(src) return True
get the value of subnode
def get_subnode(self, dst, ast, expr): """ get the value of subnode example:: R = [ __scope__:big getsomethingbig:>big #get(_, big, '.val') // copy big.val into _ ] """ dst.value = eval('ast' + expr) return True
Default serializer for json.
def default_serializer(o): """Default serializer for json.""" defs = ( ((datetime.date, datetime.time), lambda x: x.isoformat(), ), ((datetime.datetime, ), lambda x: dt2utc_timestamp(x), ), ) for types, fun in defs: if isinstance(o, types): return fun(o)
Get list of depositions ( as iterator ).
def _get_depositions(user=None, type=None): """Get list of depositions (as iterator). This is redefined Deposition.get_depositions classmethod without order-by for better performance. """ from invenio.modules.workflows.models import BibWorkflowObject, Workflow from invenio.modules.deposit.models import InvalidDepositionType from flask import current_app from invenio.ext.sqlalchemy import db from invenio.modules.deposit.models import Deposition params = [ Workflow.module_name == 'webdeposit', ] if user: params.append(BibWorkflowObject.id_user == user.get_id()) else: params.append(BibWorkflowObject.id_user != 0) if type: params.append(Workflow.name == type.get_identifier()) objects = BibWorkflowObject.query.join("workflow").options( db.contains_eager('workflow')).filter(*params) def _create_obj(o): try: obj = Deposition(o) except InvalidDepositionType as err: current_app.logger.exception(err) return None if type is None or obj.type == type: return obj return None def mapper_filter(objs): for o in objs: o = _create_obj(o) if o is not None: yield o return mapper_filter(objects)
Get deposits.
def get(query, from_date, limit=0, **kwargs): """Get deposits.""" dep_generator = _get_depositions() total_depids = 1 # Count of depositions is hard to determine # If limit provided, serve only first n=limit items if limit > 0: dep_generator = islice(dep_generator, limit) total_depids = limit return total_depids, dep_generator
Dump the deposition object as dictionary.
def dump(deposition, from_date, with_json=True, latest_only=False, **kwargs): """Dump the deposition object as dictionary.""" # Serialize the __getstate__ and fall back to default serializer dep_json = json.dumps(deposition.__getstate__(), default=default_serializer) dep_dict = json.loads(dep_json) dep_dict['_p'] = {} dep_dict['_p']['id'] = deposition.id dep_dict['_p']['created'] = dt2utc_timestamp(deposition.created) dep_dict['_p']['modified'] = dt2utc_timestamp(deposition.modified) dep_dict['_p']['user_id'] = deposition.user_id dep_dict['_p']['state'] = deposition.state dep_dict['_p']['has_sip'] = deposition.has_sip() dep_dict['_p']['submitted'] = deposition.submitted return dep_dict
Get BibDocs for Invenio 1.
def _get_recids_invenio12(from_date): """Get BibDocs for Invenio 1.""" from invenio.dbquery import run_sql return (id[0] for id in run_sql( 'select id_bibrec from ' 'bibrec_bibdoc as r join bibdoc as d on r.id_bibdoc=d.id ' 'where d.modification_date >=%s', (from_date, ), run_on_slave=True))
Get BibDocs for Invenio 2.
def _get_recids_invenio2(from_date): """Get BibDocs for Invenio 2.""" from invenio.legacy.dbquery import run_sql return (id[0] for id in run_sql( 'select id_bibrec from ' 'bibrec_bibdoc as r join bibdoc as d on r.id_bibdoc=d.id ' 'where d.modification_date >=%s', (from_date, ), run_on_slave=True))
Import BibDocFile.
def _import_bibdoc(): """Import BibDocFile.""" try: from invenio.bibdocfile import BibRecDocs, BibDoc except ImportError: from invenio.legacy.bibdocfile.api import BibRecDocs, BibDoc return BibRecDocs, BibDoc
Dump all BibDoc metadata.
def dump_bibdoc(recid, from_date, **kwargs): """Dump all BibDoc metadata. :param docid: BibDoc ID :param from_date: Dump only BibDoc revisions newer than this date. :returns: List of version of the BibDoc formatted as a dict """ BibRecDocs, BibDoc = _import_bibdoc() bibdocfile_dump = [] date = datetime.datetime.strptime(from_date, '%Y-%m-%d %H:%M:%S') for bibdoc in BibRecDocs(recid).list_bibdocs(): for version in bibdoc.list_versions(): bibdoc_version = bibdoc.list_version_files(version) for f in bibdoc_version: if f.is_icon() or f.md < date: # Don't care about icons # Don't care about files not modified since from_date continue bibdocfile_dump.append(dict( bibdocid=f.get_bibdocid(), checksum=f.get_checksum(), comment=f.get_comment(), copyright=( f.get_copyright() if hasattr(f, 'get_copyright') else None), creation_date=datetime_toutc(f.cd).isoformat(), description=f.get_description(), encoding=f.encoding, etag=f.etag, flags=f.flags, format=f.get_format(), full_name=f.get_full_name(), full_path=f.get_full_path(), hidden=f.hidden, license=( f.get_license()if hasattr(f, 'get_license') else None), modification_date=datetime_toutc(f.md).isoformat(), name=f.get_name(), mime=f.mime, path=f.get_path(), recid=f.get_recid(), recids_doctype=f.recids_doctypes, size=f.get_size(), status=f.get_status(), subformat=f.get_subformat(), superformat=f.get_superformat(), type=f.get_type(), url=f.get_url(), version=f.get_version(), )) return bibdocfile_dump
Get bibdocs to check.
def get_check(): """Get bibdocs to check.""" try: from invenio.dbquery import run_sql except ImportError: from invenio.legacy.dbquery import run_sql return ( run_sql('select count(id) from bibdoc', run_on_slave=True)[0][0], [id[0] for id in run_sql('select id from bibdoc', run_on_slave=True)], )
Check bibdocs.
def check(id_): """Check bibdocs.""" BibRecDocs, BibDoc = _import_bibdoc() try: BibDoc(id_).list_all_files() except Exception: click.secho("BibDoc {0} failed check.".format(id_), fg='red')
Dump the oauth2server tokens.
def dump(obj, from_date, with_json=True, latest_only=False, **kwargs): """Dump the oauth2server tokens.""" return dict(id=obj.id, client_id=obj.client_id, user_id=obj.user_id, token_type=obj.token_type, access_token=obj.access_token, refresh_token=obj.refresh_token, expires=dt2iso_or_empty(obj.expires), _scopes=obj._scopes, is_personal=obj.is_personal, is_internal=obj.is_internal)
Get UserEXT objects.
def get(*args, **kwargs): """Get UserEXT objects.""" try: from invenio.modules.accounts.models import UserEXT except ImportError: from invenio_accounts.models import UserEXT q = UserEXT.query return q.count(), q.all()
Dump the UserEXt objects as a list of dictionaries.
def dump(u, from_date, with_json=True, latest_only=False, **kwargs): """Dump the UserEXt objects as a list of dictionaries. :param u: UserEXT to be dumped. :type u: `invenio_accounts.models.UserEXT [Invenio2.x]` :returns: User serialized to dictionary. :rtype: dict """ return dict(id=u.id, method=u.method, id_user=u.id_user)
Get communities.
def get(*args, **kwargs): """Get communities.""" from invenio.modules.communities.models import FeaturedCommunity q = FeaturedCommunity.query return q.count(), q.all()
Dump the community object as dictionary.
def dump(fc, from_date, with_json=True, latest_only=False, **kwargs): """Dump the community object as dictionary. :param fc: Community featuring to be dumped. :type fc: `invenio_communities.models.FeaturedCommunity [Invenio2.x]` :returns: Community serialized to dictionary. :rtype: dict """ return dict(id=fc.id, id_community=fc.id_community, start_date=fc.start_date.isoformat())
Get record ids for Invenio 1.
def _get_modified_recids_invenio12(from_date): """Get record ids for Invenio 1.""" from invenio.search_engine import search_pattern from invenio.dbquery import run_sql return set((id[0] for id in run_sql( 'select id from bibrec where modification_date >= %s', (from_date, ), run_on_slave=True))), search_pattern
Get record ids for Invenio 2.
def _get_modified_recids_invenio2(from_date): """Get record ids for Invenio 2.""" from invenio.legacy.search_engine import search_pattern from invenio.modules.records.models import Record date = datetime.datetime.strptime(from_date, '%Y-%m-%d %H:%M:%S') return set( (x[0] for x in Record.query.filter(Record.modification_date >= date).values( Record.id))), search_pattern
Get all restrictions for a given collection users and fireroles.
def _get_collection_restrictions(collection): """Get all restrictions for a given collection, users and fireroles.""" try: from invenio.dbquery import run_sql from invenio.access_control_firerole import compile_role_definition except ImportError: from invenio.modules.access.firerole import compile_role_definition from invenio.legacy.dbquery import run_sql res = run_sql( 'SELECT r.firerole_def_src, email ' 'FROM accROLE as r ' 'JOIN accROLE_accACTION_accARGUMENT ON r.id=id_accROLE ' 'JOIN accARGUMENT AS a ON a.id=id_accARGUMENT ' 'JOIN user_accROLE AS u ON r.id=u.id_accROLE ' 'JOIN user ON user.id=u.id_user ' 'WHERE a.keyword="collection" AND ' 'a.value=%s AND ' 'id_accACTION=(select id from accACTION where name="viewrestrcoll")', (collection, ), run_on_slave=True ) fireroles = set() users = set() for f, u in res: fireroles.add(compile_role_definition(f)) users.add(u) return {'fireroles': list(fireroles), 'users': users}
Get record revisions.
def get_record_revisions(recid, from_date): """Get record revisions.""" try: from invenio.dbquery import run_sql except ImportError: from invenio.legacy.dbquery import run_sql return run_sql( 'SELECT job_date, marcxml ' 'FROM hstRECORD WHERE id_bibrec = %s AND job_date >= %s ' 'ORDER BY job_date ASC', (recid, from_date), run_on_slave=True)
Get all collections the record belong to.
def get_record_collections(recid): """Get all collections the record belong to.""" try: from invenio.search_engine import ( get_all_collections_of_a_record, get_restricted_collections_for_recid) except ImportError: from invenio.legacy.search_engine import ( get_all_collections_of_a_record, get_restricted_collections_for_recid) collections = { 'all': get_all_collections_of_a_record(recid, recreate_cache_if_needed=False), } collections['restricted'] = dict( (coll, _get_collection_restrictions(coll)) for coll in get_restricted_collections_for_recid( recid, recreate_cache_if_needed=False)) return collections
Dump JSON of record.
def dump_record_json(marcxml): """Dump JSON of record.""" try: from invenio.modules.records.api import Record d = Record.create(marcxml, 'marc') return d.dumps(clean=True) except ImportError: from invenio.bibfield import create_record d = create_record(marcxml, master_format='marc') return d.dumps()
Get recids matching query and with changes.
def get(query, from_date, **kwargs): """Get recids matching query and with changes.""" recids, search_pattern = get_modified_recids(from_date) recids = recids.union(get_modified_bibdoc_recids(from_date)) if query: recids = recids.intersection( set(search_pattern(p=query.encode('utf-8')))) return len(recids), recids
Dump MARCXML and JSON representation of a record.
def dump(recid, from_date, with_json=False, latest_only=False, with_collections=False, **kwargs): """Dump MARCXML and JSON representation of a record. :param recid: Record identifier :param from_date: Dump only revisions from this date onwards. :param with_json: If ``True`` use old ``Record.create`` to generate the JSON representation of the record. :param latest_only: Dump only the last revision of the record metadata. :param with_collections: If ``True`` dump the list of collections that the record belongs to. :returns: List of versions of the record. """ # Grab latest only if latest_only: revision_iter = [get_record_revisions(recid, from_date)[-1]] else: revision_iter = get_record_revisions(recid, from_date) # Dump revisions record_dump = dict( record=[], files=[], recid=recid, collections=get_record_collections(recid) if with_collections else None, ) for revision_date, revision_marcxml in revision_iter: marcxml = zlib.decompress(revision_marcxml) record_dump['record'].append( dict( modification_datetime=datetime_toutc(revision_date) .isoformat(), marcxml=marcxml, json=dump_record_json(marcxml) if with_json else None, )) record_dump['files'] = dump_bibdoc(recid, from_date) return record_dump
Dump the remote accounts as a list of dictionaries.
def dump(ra, from_date, with_json=True, latest_only=False, **kwargs): """Dump the remote accounts as a list of dictionaries. :param ra: Remote account to be dumped. :type ra: `invenio_oauthclient.models.RemoteAccount [Invenio2.x]` :returns: Remote accounts serialized to dictionary. :rtype: dict """ return dict(id=ra.id, user_id=ra.user_id, client_id=ra.client_id, extra_data=ra.extra_data)
Helper function for loading JSON data verbatim into model.
def load_common(model_cls, data): """Helper function for loading JSON data verbatim into model.""" obj = model_cls(**data) db.session.add(obj) db.session.commit()
Collect entry points.
def collect_things_entry_points(): """Collect entry points.""" things = dict() for entry_point in iter_entry_points(group='invenio_migrator.things'): things[entry_point.name] = entry_point.load() return things
Initialize app context for Invenio 2. x.
def init_app_context(): """Initialize app context for Invenio 2.x.""" try: from invenio.base.factory import create_app app = create_app() app.test_request_context('/').push() app.preprocess_request() except ImportError: pass
Cache for heavy function calls.
def memoize(func): """Cache for heavy function calls.""" cache = {} @wraps(func) def wrap(*args, **kwargs): key = '{0}{1}'.format(args, kwargs) if key not in cache: cache[key] = func(*args, **kwargs) return cache[key] return wrap
Import run_sql.
def _get_run_sql(): """Import ``run_sql``.""" try: from invenio.dbquery import run_sql except ImportError: from invenio.legacy.dbquery import run_sql return run_sql
Get roles connected to an action.
def get_connected_roles(action_id): """Get roles connected to an action.""" try: from invenio.access_control_admin import compile_role_definition except ImportError: from invenio.modules.access.firerole import compile_role_definition run_sql = _get_run_sql() roles = {} res = run_sql( 'select r.id, r.name, r.description, r.firerole_def_src, ' 'a.keyword, a.value, email from accROLE as r ' 'join accROLE_accACTION_accARGUMENT on r.id=id_accROLE ' 'join accARGUMENT as a on a.id=id_accARGUMENT ' 'join user_accROLE as u on r.id=u.id_accROLE ' 'join user on user.id=u.id_user ' 'where id_accACTION=%s', (action_id, ) ) for r in res: role = roles.setdefault( r[0], { 'id': r[0], 'name': r[1], 'description': r[2], 'firerole_def': r[3], 'compiled_firerole_def': compile_role_definition(r[3]), 'users': set(), 'parameters': {} } ) param = role['parameters'].setdefault(r[4], set()) param.add(r[5]) role['users'].add(r[6]) return six.itervalues(roles)
Get action definitions to dump.
def get(query, *args, **kwargs): """Get action definitions to dump.""" run_sql = _get_run_sql() actions = [ dict(id=row[0], name=row[1], allowedkeywords=row[2], optional=row[3]) for action in query.split(',') for row in run_sql( 'select id, name, description, allowedkeywords, optional ' 'from accACTION where name like %s', (action, ), run_on_slave=True) ] return len(actions), actions
Get users.
def get(*args, **kwargs): """Get users.""" from invenio.modules.oauthclient.models import RemoteToken q = RemoteToken.query return q.count(), q.all()
Dump the remote tokens as a list of dictionaries.
def dump(rt, from_date, with_json=True, latest_only=False, **kwargs): """Dump the remote tokens as a list of dictionaries. :param ra: Remote toekn to be dumped. :type ra: `invenio_oauthclient.models.RemoteToken [Invenio2.x]` :returns: Remote tokens serialized to dictionary. :rtype: dict """ return dict(id_remote_account=rt.id_remote_account, token_type=rt.token_type, access_token=rt.access_token, secret=rt.secret)
Load the oauth2server token from data dump.
def load_token(data): """Load the oauth2server token from data dump.""" from invenio_oauth2server.models import Token data['expires'] = iso2dt_or_none(data['expires']) load_common(Token, data)
Migrate a record from a migration dump.
def import_record(data, source_type=None, latest_only=False): """Migrate a record from a migration dump. :param data: Dictionary for representing a single record and files. :param source_type: Determines if the MARCXML or the JSON dump is used. Default: ``marcxml``. :param latest_only: Determine is only the latest revision should be loaded. """ source_type = source_type or 'marcxml' assert source_type in ['marcxml', 'json'] recorddump = current_migrator.records_dump_cls( data, source_type=source_type, pid_fetchers=current_migrator.records_pid_fetchers, ) try: current_migrator.records_dumploader_cls.create(recorddump) db.session.commit() except Exception: db.session.rollback() raise
Import config var import path or use default value.
def config_imp_or_default(app, config_var_imp, default): """Import config var import path or use default value.""" imp = app.config.get(config_var_imp) return import_string(imp) if imp else default
Flask application initialization.
def init_app(self, app): """Flask application initialization.""" self.init_config(app.config) state = _InvenioMigratorState(app) app.extensions['invenio-migrator'] = state app.cli.add_command(dumps) return state
Get users.
def get(*args, **kwargs): """Get users.""" from invenio.modules.oauth2server.models import Client q = Client.query return q.count(), q.all()
Dump the oauth2server Client.
def dump(obj, from_date, with_json=True, latest_only=False, **kwargs): """Dump the oauth2server Client.""" return dict(name=obj.name, description=obj.description, website=obj.website, user_id=obj.user_id, client_id=obj.client_id, client_secret=obj.client_secret, is_confidential=obj.is_confidential, is_internal=obj.is_internal, _redirect_uris=obj._redirect_uris, _default_scopes=obj._default_scopes)
Get user accounts Invenio 1.
def _get_users_invenio12(*args, **kwargs): """Get user accounts Invenio 1.""" from invenio.dbquery import run_sql, deserialize_via_marshal User = namedtuple('User', [ 'id', 'email', 'password', 'password_salt', 'note', 'full_name', 'settings', 'nickname', 'last_login' ]) users = run_sql( 'SELECT id, email, password, note, settings, nickname, last_login' ' FROM user', run_on_slave=True) return len(users), [ User( id=user[0], email=user[1], password=user[2].decode('latin1'), password_salt=user[1], note=user[3], full_name=user[5], settings=deserialize_via_marshal(user[4]) if user[4] else {}, # we don't have proper nicknames on Invenio v1 nickname='id_{0}'.format(user[0]), last_login=user[6]) for user in users ]
Get user accounts from Invenio 2.
def _get_users_invenio2(*args, **kwargs): """Get user accounts from Invenio 2.""" from invenio.modules.accounts.models import User q = User.query return q.count(), q.all()
Get users.
def get(*args, **kwargs): """Get users.""" try: return _get_users_invenio12(*args, **kwargs) except ImportError: return _get_users_invenio2(*args, **kwargs)
Dump the users as a list of dictionaries.
def dump(u, *args, **kwargs): """Dump the users as a list of dictionaries. :param u: User to be dumped. :type u: `invenio.modules.accounts.models.User [Invenio2.x]` or namedtuple. :returns: User serialized to dictionary. :rtype: dict """ return dict( id=u.id, email=u.email, password=u.password, password_salt=u.password_salt, note=u.note, full_name=u.full_name if hasattr(u, 'full_name') else '{0} {1}'.format( u.given_names, u.family_name), settings=u.settings, nickname=u.nickname, last_login=dt2iso_or_empty(u.last_login))
Load the raw JSON dump of the Deposition.
def load_deposit(data): """Load the raw JSON dump of the Deposition. Uses Record API in order to bypass all Deposit-specific initialization, which are to be done after the final stage of deposit migration. :param data: Dictionary containing deposition data. :type data: dict """ from invenio_db import db deposit, dep_pid = create_record_and_pid(data) deposit = create_files_and_sip(deposit, dep_pid) db.session.commit()
Create the deposit record metadata and persistent identifier.
def create_record_and_pid(data): """Create the deposit record metadata and persistent identifier. :param data: Raw JSON dump of the deposit. :type data: dict :returns: A deposit object and its pid :rtype: (`invenio_records.api.Record`, `invenio_pidstore.models.PersistentIdentifier`) """ from invenio_records.api import Record from invenio_pidstore.models import PersistentIdentifier, PIDStatus, \ RecordIdentifier deposit = Record.create(data=data) created = arrow.get(data['_p']['created']).datetime deposit.model.created = created.replace(tzinfo=None) depid = deposit['_p']['id'] pid = PersistentIdentifier.create( pid_type='depid', pid_value=str(depid), object_type='rec', object_uuid=str(deposit.id), status=PIDStatus.REGISTERED ) if RecordIdentifier.query.get(int(depid)) is None: RecordIdentifier.insert(int(depid)) deposit.commit() return deposit, pid
Create deposit Bucket Files and SIPs.
def create_files_and_sip(deposit, dep_pid): """Create deposit Bucket, Files and SIPs.""" from invenio_pidstore.errors import PIDDoesNotExistError from invenio_pidstore.models import PersistentIdentifier, PIDStatus, \ RecordIdentifier from invenio_sipstore.errors import SIPUserDoesNotExist from invenio_sipstore.models import SIP, RecordSIP, SIPFile from invenio_files_rest.models import Bucket, FileInstance, ObjectVersion from invenio_records_files.models import RecordsBuckets from invenio_db import db buc = Bucket.create() recbuc = RecordsBuckets(record_id=deposit.id, bucket_id=buc.id) db.session.add(recbuc) deposit.setdefault('_deposit', dict()) deposit.setdefault('_buckets', dict(deposit=str(buc.id))) deposit.setdefault('_files', list()) files = deposit.get('files', []) sips = deposit.get('sips', []) # Look for prereserved DOI (and recid) if 'drafts' in deposit: drafts = list(deposit['drafts'].items()) if len(drafts) != 1: logger.exception('Deposit {dep_pid} has multiple drafts'.format( dep_pid=dep_pid)) if len(drafts) == 1: draft_type, draft = drafts[0] draft_v = draft['values'] if 'prereserve_doi' in draft_v: pre_recid = str(draft_v['prereserve_doi']['recid']) pre_doi = str(draft_v['prereserve_doi']['doi']) # If pre-reserve info available, try to reserve 'recid' try: pid = PersistentIdentifier.get(pid_type='recid', pid_value=str(pre_recid)) except PIDDoesNotExistError: # Reserve recid pid = PersistentIdentifier.create( pid_type='recid', pid_value=str(pre_recid), object_type='rec', status=PIDStatus.RESERVED ) # If pre-reserve info available, try to reserve 'doi' try: pid = PersistentIdentifier.get(pid_type='doi', pid_value=str(pre_doi)) except PIDDoesNotExistError: # Reserve DOI pid = PersistentIdentifier.create( pid_type='doi', pid_value=str(pre_doi), object_type='rec', status=PIDStatus.RESERVED ) if RecordIdentifier.query.get(int(pre_recid)) is None: RecordIdentifier.insert(int(pre_recid)) # Store the path -> FileInstance mappings for SIPFile creation later dep_file_instances = list() for file_ in files: size = file_['size'] key = file_['name'] # Warning: Assumes all checksums are MD5! checksum = 'md5:{0}'.format(file_['checksum']) fi = FileInstance.create() fi.set_uri(file_['path'], size, checksum) ov = ObjectVersion.create(buc, key, _file_id=fi.id) ext = splitext(ov.key)[1].lower() if ext.startswith('.'): ext = ext[1:] file_meta = dict( bucket=str(ov.bucket.id), key=ov.key, checksum=ov.file.checksum, size=ov.file.size, version_id=str(ov.version_id), type=ext, ) deposit['_files'].append(file_meta) dep_file_instances.append((file_['path'], fi)) # Get a recid from SIP information recid = None if sips: recids = [int(sip['metadata']['recid']) for sip in sips] if len(set(recids)) > 1: logger.error('Multiple recids ({recids}) found in deposit {depid}' ' does not exists.'.format(recids=recids, depid=dep_pid.pid_value)) raise DepositMultipleRecids(dep_pid.pid_value, list(set(recids))) elif recids: # If only one recid recid = recids[0] for idx, sip in enumerate(sips): agent = None user_id = None if sip['agents']: agent = dict( ip_address=empty_str_if_none( sip['agents'][0].get('ip_address', "")), email=empty_str_if_none( sip['agents'][0].get('email_address', "")), ) user_id = sip['agents'][0]['user_id'] if user_id == 0: user_id = None content = sip['package'] sip_format = 'marcxml' try: sip = SIP.create(sip_format, content, user_id=user_id, agent=agent) except SIPUserDoesNotExist: logger.exception('User ID {user_id} referred in deposit {depid} ' 'does not exists.'.format( user_id=user_id, depid=dep_pid.pid_value)) sip = SIP.create(sip_format, content, agent=agent) # Attach recid to SIP if recid: try: pid = PersistentIdentifier.get(pid_type='recid', pid_value=str(recid)) record_sip = RecordSIP(sip_id=sip.id, pid_id=pid.id) db.session.add(record_sip) except PIDDoesNotExistError: logger.exception('Record {recid} referred in ' 'Deposit {depid} does not exists.'.format( recid=recid, depid=dep_pid.pid_value)) if deposit['_p']['submitted'] is True: logger.exception('Pair {recid}/{depid} was submitted,' ' (should it be unpublished?).'.format( recid=recid, depid=dep_pid.pid_value)) else: msg = 'Pair {recid}/{depid} was not submitted.'.format( recid=recid, depid=dep_pid.pid_value) logger.exception(msg) # Reserve recid pid = PersistentIdentifier.create( pid_type='recid', pid_value=str(recid), object_type='rec', status=PIDStatus.RESERVED ) if RecordIdentifier.query.get(int(recid)) is None: RecordIdentifier.insert(int(recid)) if idx == 0: for fp, fi in dep_file_instances: sipf = SIPFile(sip_id=sip.id, filepath=fp, file_id=fi.id) db.session.add(sipf) deposit.commit() return deposit
Dump the community object as dictionary.
def dump(c, from_date, with_json=True, latest_only=False, **kwargs): """Dump the community object as dictionary. :param c: Community to be dumped. :type c: `invenio.modules.communities.models.Community` :returns: Community serialized to dictionary. :rtype: dict """ return dict(id=c.id, id_user=c.id_user, title=c.title, description=c.description, page=c.page, curation_policy=c.curation_policy, last_record_accepted=dt2iso_or_empty(c.last_record_accepted), logo_ext=c.logo_ext, logo_url=c.logo_url, ranking=c.ranking, fixed_points=c.fixed_points, created=c.created.isoformat(), last_modified=c.last_modified.isoformat(), id_collection_provisional=c.id_collection_provisional, id_oairepository=c.id_oairepository, id_collection=c.id_collection)
Load a single record into the database.
def _loadrecord(record_dump, source_type, eager=False): """Load a single record into the database. :param record_dump: Record dump. :type record_dump: dict :param source_type: 'json' or 'marcxml' :param eager: If ``True`` execute the task synchronously. """ if eager: import_record.s(record_dump, source_type=source_type).apply(throw=True) elif current_migrator.records_post_task: chain( import_record.s(record_dump, source_type=source_type), current_migrator.records_post_task.s() )() else: import_record.delay(record_dump, source_type=source_type)
Load records migration dump.
def loadrecords(sources, source_type, recid): """Load records migration dump.""" # Load all record dumps up-front and find the specific JSON if recid is not None: for source in sources: records = json.load(source) for item in records: if str(item['recid']) == str(recid): _loadrecord(item, source_type, eager=True) click.echo("Record '{recid}' loaded.".format(recid=recid)) return click.echo("Record '{recid}' not found.".format(recid=recid)) else: for idx, source in enumerate(sources, 1): click.echo('Loading dump {0} of {1} ({2})'.format( idx, len(sources), source.name)) data = json.load(source) with click.progressbar(data) as records: for item in records: _loadrecord(item, source_type)
Inspect records in a migration dump.
def inspectrecords(sources, recid, entity=None): """Inspect records in a migration dump.""" for idx, source in enumerate(sources, 1): click.echo('Loading dump {0} of {1} ({2})'.format(idx, len(sources), source.name)) data = json.load(source) # Just print record identifiers if none are selected. if not recid: click.secho('Record identifiers', fg='green') total = 0 for r in (d['recid'] for d in data): click.echo(r) total += 1 click.echo('{0} records found in dump.'.format(total)) return data = list(filter(lambda d: d['recid'] == recid, data)) if not data: click.secho("Record not found.", fg='yellow') return for record in data: if entity is None: click.echo(json.dumps(record, indent=2)) if entity == 'files': click.secho('Files', fg='green') click.echo( json.dumps(record['files'], indent=2)) if entity == 'json': click.secho('Records (JSON)', fg='green') for revision in record['record']: click.secho('Revision {0}'.format( revision['modification_datetime']), fg='yellow') click.echo(json.dumps(revision['json'], indent=2)) if entity == 'marcxml': click.secho('Records (MARCXML)', fg='green') for revision in record['record']: click.secho( 'Revision {0}'.format(revision['marcxml']), fg='yellow') click.echo(revision)
Common helper function for load simple objects.
def loadcommon(sources, load_task, asynchronous=True, predicate=None, task_args=None, task_kwargs=None): """Common helper function for load simple objects. .. note:: Keyword arguments ``task_args`` and ``task_kwargs`` are passed to the ``load_task`` function as ``*task_args`` and ``**task_kwargs``. .. note:: The `predicate` argument is used as a predicate function to load only a *single* item from across all dumps (this CLI function will return after loading the item). This is primarily used for debugging of the *dirty* data within the dump. The `predicate` should be a function with a signature ``f(dict) -> bool``, i.e. taking a single parameter (an item from the dump) and return ``True`` if the item should be loaded. See the ``loaddeposit`` for a concrete example. :param sources: JSON source files with dumps :type sources: list of str (filepaths) :param load_task: Shared task which loads the dump. :type load_task: function :param asynchronous: Flag for serial or asynchronous execution of the task. :type asynchronous: bool :param predicate: Predicate for selecting only a single item from the dump. :type predicate: function :param task_args: positional arguments passed to the task. :type task_args: tuple :param task_kwargs: named arguments passed to the task. :type task_kwargs: dict """ # resolve the defaults for task_args and task_kwargs task_args = tuple() if task_args is None else task_args task_kwargs = dict() if task_kwargs is None else task_kwargs click.echo('Loading dumps started.') for idx, source in enumerate(sources, 1): click.echo('Opening dump file {0} of {1} ({2})'.format( idx, len(sources), source.name)) data = json.load(source) with click.progressbar(data) as data_bar: for d in data_bar: # Load a single item from the dump if predicate is not None: if predicate(d): load_task.s(d, *task_args, **task_kwargs).apply( throw=True) click.echo("Loaded a single record.") return # Load dumps normally else: if asynchronous: load_task.s(d, *task_args, **task_kwargs).apply_async() else: load_task.s(d, *task_args, **task_kwargs).apply( throw=True)
Load communities.
def loadcommunities(sources, logos_dir): """Load communities.""" from invenio_migrator.tasks.communities import load_community loadcommon(sources, load_community, task_args=(logos_dir, ))
Load users.
def loadusers(sources): """Load users.""" from .tasks.users import load_user # Cannot be executed asynchronously due to duplicate emails and usernames # which can create a racing condition. loadcommon(sources, load_user, asynchronous=False)
Load deposit.
def loaddeposit(sources, depid): """Load deposit. Usage: invenio dumps loaddeposit ~/data/deposit_dump_*.json invenio dumps loaddeposit -d 12345 ~/data/deposit_dump_*.json """ from .tasks.deposit import load_deposit if depid is not None: def pred(dep): return int(dep["_p"]["id"]) == depid loadcommon(sources, load_deposit, predicate=pred, asynchronous=False) else: loadcommon(sources, load_deposit)
Return profiler statistics.
def get_profiler_statistics(sort="cum_time", count=20, strip_dirs=True): """Return profiler statistics. :param str sort: dictionary key to sort by :param int|None count: the number of results to return, None returns all results. :param bool strip_dirs: if True strip the directory, otherwise return the full path """ json_stats = [] pstats = yappi.convert2pstats(yappi.get_func_stats()) if strip_dirs: pstats.strip_dirs() for func, func_stat in pstats.stats.iteritems(): path, line, func_name = func cc, num_calls, total_time, cum_time, callers = func_stat json_stats.append({ "path": path, "line": line, "func_name": func_name, "num_calls": num_calls, "total_time": total_time, "total_time_per_call": total_time/num_calls if total_time else 0, "cum_time": cum_time, "cum_time_per_call": cum_time/num_calls if cum_time else 0 }) return sorted(json_stats, key=itemgetter(sort), reverse=True)[:count]
Run as sample test server.
def main(port=8888): """Run as sample test server.""" import tornado.ioloop routes = [] + TornadoProfiler().get_routes() app = tornado.web.Application(routes) app.listen(port) tornado.ioloop.IOLoop.current().start()
Return current profiler statistics.
def get(self): """Return current profiler statistics.""" sort = self.get_argument('sort', 'cum_time') count = self.get_argument('count', 20) strip_dirs = self.get_argument('strip_dirs', True) error = '' sorts = ('num_calls', 'cum_time', 'total_time', 'cum_time_per_call', 'total_time_per_call') if sort not in sorts: error += "Invalid `sort` '%s', must be in %s." % (sort, sorts) try: count = int(count) except (ValueError, TypeError): error += "Can't cast `count` '%s' to int." % count if count <= 0: count = None strip_dirs = str(strip_dirs).lower() not in ('false', 'no', 'none', 'null', '0', '') if error: self.write({'error': error}) self.set_status(400) self.finish() return try: statistics = get_profiler_statistics(sort, count, strip_dirs) self.write({'statistics': statistics}) self.set_status(200) except TypeError: logger.exception('Error while retrieving profiler statistics') self.write({'error': 'No stats available. Start and stop the profiler before trying to retrieve stats.'}) self.set_status(404) self.finish()
Start a new profiler.
def post(self): """Start a new profiler.""" if is_profiler_running(): self.set_status(201) self.finish() return start_profiling() self.set_status(201) self.finish()
Dump current profiler statistics into a file.
def post(self): """Dump current profiler statistics into a file.""" filename = self.get_argument('filename', 'dump.prof') CProfileWrapper.profiler.dump_stats(filename) self.finish()
Return current profiler statistics.
def get(self): """Return current profiler statistics.""" CProfileWrapper.profiler.print_stats() s = StringIO.StringIO() sortby = 'cumulative' ps = pstats.Stats(CProfileWrapper.profiler, stream=s).sort_stats(sortby) ps.print_stats() self.set_status(200) self.write(s.getvalue()) self.finish()
Clear profiler statistics.
def delete(self): """Clear profiler statistics.""" CProfileWrapper.profiler.create_stats() self.enable() self.set_status(204) self.finish()
Start a new profiler.
def post(self): """Start a new profiler.""" CProfileWrapper.profiler = cProfile.Profile() CProfileWrapper.profiler.enable() self.running = True self.set_status(201) self.finish()
Stop the profiler.
def delete(self): """Stop the profiler.""" CProfileWrapper.profiler.disable() self.running = False self.set_status(204) self.finish()
Check if the profiler is running.
def get(self): """Check if the profiler is running.""" self.write({"running": self.running}) self.set_status(200) self.finish()
Disable timestamp update per method.
def disable_timestamp(method): """Disable timestamp update per method.""" @wraps(method) def wrapper(*args, **kwargs): result = None with correct_date(): result = method(*args, **kwargs) return result return wrapper
Load user from data dump.
def load_user(data): """Load user from data dump. NOTE: This task takes into account the possible duplication of emails and usernames, hence it should be called synchronously. In such case of collision it will raise UserEmailExistsError or UserUsernameExistsError, if email or username are already existing in the database. Caller of this task should take care to to resolve those collisions beforehand or after catching an exception. :param data: Dictionary containing user data. :type data: dict """ from invenio_accounts.models import User from invenio_userprofiles.api import UserProfile email = data['email'].strip() if User.query.filter_by(email=email).count() > 0: raise UserEmailExistsError( "User email '{email}' already exists.".format(email=email)) last_login = None if data['last_login']: last_login = arrow.get(data['last_login']).datetime confirmed_at = None if data['note'] == '1': confirmed_at = datetime.utcnow() salt = data['password_salt'] checksum = data['password'] if not checksum: new_password = None # Test if password hash is in Modular Crypt Format elif checksum.startswith('$'): new_password = checksum else: new_password = str.join('$', ['', u'invenio-aes', salt, checksum]) with db.session.begin_nested(): obj = User( id=data['id'], password=new_password, email=email, confirmed_at=confirmed_at, last_login_at=last_login, active=(data['note'] != '0'), ) db.session.add(obj) nickname = data['nickname'].strip() overwritten_username = ('username' in data and 'displayname' in data) # NOTE: 'username' and 'displayname' will exist in data dump only # if it was inserted there after dumping. It normally should not come from # Invenio 1.x or 2.x data dumper script. In such case, those values will # have precedence over the 'nickname' field. if nickname or overwritten_username: p = UserProfile(user=obj) p.full_name = data.get('full_name', '').strip() if overwritten_username: p._username = data['username'].lower() p._displayname = data['displayname'] elif nickname: if UserProfile.query.filter( UserProfile._username == nickname.lower()).count() > 0: raise UserUsernameExistsError( "Username '{username}' already exists.".format( username=nickname)) try: p.username = nickname except ValueError: current_app.logger.warn( u'Invalid username {0} for user_id {1}'.format( nickname, data['id'])) p._username = nickname.lower() p._displayname = nickname db.session.add(p) db.session.commit()
Calculate image translations in parallel.
def calc_translations_parallel(images): """Calculate image translations in parallel. Parameters ---------- images : ImageCollection Images as instance of ImageCollection. Returns ------- 2d array, (ty, tx) ty and tx is translation to previous image in respectively x or y direction. """ w = Parallel(n_jobs=_CPUS) res = w(delayed(images.translation)(img) for img in images) # save results to Image object, as Parallel is spawning another process for i,translation in enumerate(res): images[i].translation = translation return np.array(res)
Stitch regular spaced images.
def stitch(images): """Stitch regular spaced images. Parameters ---------- images : ImageCollection or list of tuple(path, row, column) Each image-tuple should contain path, row and column. Row 0, column 0 is top left image. Example: >>> images = [('1.png', 0, 0), ('2.png', 0, 1)] Returns ------- tuple (stitched, offset) Stitched image and registered offset (y, x). """ if type(images) != ImageCollection: images = ImageCollection(images) calc_translations_parallel(images) _translation_warn(images) yoffset, xoffset = images.median_translation() if xoffset != yoffset: warn('yoffset != xoffset: %s != %s' % (yoffset, xoffset)) # assume all images have the same shape y, x = imread(images[0].path).shape height = y*len(images.rows) + yoffset*(len(images.rows)-1) width = x*len(images.cols) + xoffset*(len(images.cols)-1) # last dimension is number of images on top of each other merged = np.zeros((height, width, 2), dtype=np.int) for image in images: r, c = image.row, image.col mask = _merge_slice(r, c, y, x, yoffset, xoffset) # last dim is used for averaging the seam img = _add_ones_dim(imread(image.path)) merged[mask] += img # average seam, possible improvement: use gradient merged[..., 0] /= merged[..., 1] return merged[..., 0].astype(np.uint8), (yoffset, xoffset)
Adds a dimensions with ones to array.
def _add_ones_dim(arr): "Adds a dimensions with ones to array." arr = arr[..., np.newaxis] return np.concatenate((arr, np.ones_like(arr)), axis=-1)
Create record based on dump.
def create(cls, dump): """Create record based on dump.""" # If 'record' is not present, just create the PID if not dump.data.get('record'): try: PersistentIdentifier.get(pid_type='recid', pid_value=dump.recid) except PIDDoesNotExistError: PersistentIdentifier.create( 'recid', dump.recid, status=PIDStatus.RESERVED ) db.session.commit() return None dump.prepare_revisions() dump.prepare_pids() dump.prepare_files() # Create or update? existing_files = [] if dump.record: existing_files = dump.record.get('_files', []) record = cls.update_record(revisions=dump.revisions, created=dump.created, record=dump.record) pids = dump.missing_pids else: record = cls.create_record(dump) pids = dump.pids if pids: cls.create_pids(record.id, pids) if dump.files: cls.create_files(record, dump.files, existing_files) # Update files. if dump.is_deleted(record): cls.delete_record(record) return record
Create a new record from dump.
def create_record(cls, dump): """Create a new record from dump.""" # Reserve record identifier, create record and recid pid in one # operation. timestamp, data = dump.latest record = Record.create(data) record.model.created = dump.created.replace(tzinfo=None) record.model.updated = timestamp.replace(tzinfo=None) RecordIdentifier.insert(dump.recid) PersistentIdentifier.create( pid_type='recid', pid_value=str(dump.recid), object_type='rec', object_uuid=str(record.id), status=PIDStatus.REGISTERED ) db.session.commit() return cls.update_record(revisions=dump.rest, record=record, created=dump.created)
Update an existing record.
def update_record(cls, revisions, created, record): """Update an existing record.""" for timestamp, revision in revisions: record.model.json = revision record.model.created = created.replace(tzinfo=None) record.model.updated = timestamp.replace(tzinfo=None) db.session.commit() return Record(record.model.json, model=record.model)
Create persistent identifiers.
def create_pids(cls, record_uuid, pids): """Create persistent identifiers.""" for p in pids: PersistentIdentifier.create( pid_type=p.pid_type, pid_value=p.pid_value, pid_provider=p.provider.pid_provider if p.provider else None, object_type='rec', object_uuid=record_uuid, status=PIDStatus.REGISTERED, ) db.session.commit()
Delete a record and it s persistent identifiers.
def delete_record(cls, record): """Delete a record and it's persistent identifiers.""" record.delete() PersistentIdentifier.query.filter_by( object_type='rec', object_uuid=record.id, ).update({PersistentIdentifier.status: PIDStatus.DELETED}) cls.delete_buckets(record) db.session.commit()
Create files.
def create_files(cls, record, files, existing_files): """Create files. This method is currently limited to a single bucket per record. """ default_bucket = None # Look for bucket id in existing files. for f in existing_files: if 'bucket' in f: default_bucket = f['bucket'] break # Create a bucket in default location if none is found. if default_bucket is None: b = Bucket.create() BucketTag.create(b, 'record', str(record.id)) default_bucket = str(b.id) db.session.commit() else: b = Bucket.get(default_bucket) record['_files'] = [] for key, meta in files.items(): obj = cls.create_file(b, key, meta) ext = splitext(obj.key)[1].lower() if ext.startswith('.'): ext = ext[1:] record['_files'].append(dict( bucket=str(obj.bucket.id), key=obj.key, version_id=str(obj.version_id), size=obj.file.size, checksum=obj.file.checksum, type=ext, )) db.session.add( RecordsBuckets(record_id=record.id, bucket_id=b.id) ) record.commit() db.session.commit() return [b]