code
string
signature
string
docstring
string
loss_without_docstring
float64
loss_with_docstring
float64
factor
float64
kernel_id = re.search( # type: ignore 'kernel-(.*).json', ipykernel.connect.get_connection_file() ).group(1) servers = list_running_servers() for server in servers: response = requests.get(urljoin(server['url'], 'api/sessions'), params={'token': server.get('token', '')}) for session in json.loads(response.text): if session['kernel']['id'] == kernel_id: relative_path = session['notebook']['path'] return pjoin(server['notebook_dir'], relative_path) raise Exception('Noteboook not found.')
def get_notebook_name() -> str
Return the full path of the jupyter notebook. References ---------- https://github.com/jupyter/notebook/issues/1000#issuecomment-359875246
3.260202
2.926751
1.113932
shell = InteractiveShell.instance() path = fullname # load the notebook object with open(path, 'r', encoding='utf-8') as f: notebook = read(f, 4) # create the module and add it to sys.modules mod = types.ModuleType(fullname) mod.__file__ = path # mod.__loader__ = self mod.__dict__['get_ipython'] = get_ipython sys.modules[fullname] = mod # extra work to ensure that magics that would affect the user_ns # actually affect the notebook module's ns save_user_ns = shell.user_ns shell.user_ns = mod.__dict__ try: for cell in notebook.cells: if cell.cell_type == 'code': try: # only run valid python code ast.parse(cell.source) except SyntaxError: continue try: # pylint: disable=exec-used exec(cell.source, mod.__dict__) except NameError: print(cell.source) raise finally: shell.user_ns = save_user_ns return mod
def load_notebook(fullname: str)
Import a notebook as a module.
2.884031
2.747874
1.04955
event = f'message.{status}' if flask.has_request_context(): emit(event, dict(data=pack(content))) else: sio = flask.current_app.extensions['socketio'] sio.emit(event, dict(data=pack(content))) eventlet.sleep()
def _message(status, content)
Send message interface. Parameters ---------- status : str The type of message content : str
4.4984
5.470826
0.822252
return [dict(title=str(c), dataIndex=str(c), key=str(c)) for c in columns]
def _make_columns(columns: List[Union[int, str]]) -> List[Dict]
Transform list of columns into AntTable format.
4.539893
3.591716
1.26399
jsdata = [] for idx, row in data.iterrows(): row.index = row.index.astype(str) rdict = row.to_dict() rdict.update(dict(key=str(idx))) jsdata.append(rdict) return jsdata, Table._make_columns(data.columns)
def _make_data(data) -> Tuple[List[Dict], List[Dict]]
Transform table data into JSON.
4.181808
3.604654
1.160114
# pylint: disable=protected-access,unused-variable nargs = numargs(func) if nargs > 0: raise WrongNumberOfArguments( f'Decorated function "{func.__name__}" should have no arguments, it has {nargs}.' ) app = func() if app is None: raise TypeError( 'No `App` instance was returned. ' 'In the function decorated with @command, ' 'return the `App` instance so it can be built.' ) if not isinstance(app, App): raise TypeError( f'Returned value {app} is of type {type(app)}, ' 'it needs to be a bowtie.App instance.' ) @click.group(options_metavar='[--help]') def cmd(): @cmd.command(add_help_option=False) def build(): app._build() @cmd.command(add_help_option=False) @click.option('--host', '-h', default='0.0.0.0', type=str) @click.option('--port', '-p', default=9991, type=int) def run(host, port): app._build() app._serve(host, port) @cmd.command(add_help_option=True) @click.option('--host', '-h', default='0.0.0.0', type=str) @click.option('--port', '-p', default=9991, type=int) def serve(host, port): app._serve(host, port) @cmd.command(context_settings=dict(ignore_unknown_options=True), add_help_option=False) @click.argument('extra', nargs=-1, type=click.UNPROCESSED) def dev(extra): line = (_WEBPACK, '--config', 'webpack.dev.js') + extra call(line, cwd=app._build_dir) @cmd.command(context_settings=dict(ignore_unknown_options=True), add_help_option=False) @click.argument('extra', nargs=-1, type=click.UNPROCESSED) def prod(extra): line = (_WEBPACK, '--config', 'webpack.prod.js', '--progress') + extra call(line, cwd=app._build_dir) locale = inspect.stack()[1][0].f_locals module = locale.get("__name__") if module == "__main__": try: arg = sys.argv[1:] except IndexError: arg = ('--help',) # pylint: disable=too-many-function-args sys.exit(cmd(arg)) return cmd
def command(func)
Command line interface decorator. Decorate a function for building a Bowtie application and turn it into a command line interface.
2.536038
2.534627
1.000557
targets = filter(lambda x: x != '.', targets) # First, convert the suite to a proto test list - proto tests nicely # parse things like the fully dotted name of the test and the # finest-grained module it belongs to, which simplifies our job. proto_test_list = toProtoTestList(suite) # Extract a list of the modules that all of the discovered tests are in modules = set([x.module for x in proto_test_list]) # Get the list of user-specified targets that are NOT modules non_module_targets = [] for target in targets: if not list(filter(None, [target in x for x in modules])): non_module_targets.append(target) # Main loop -- iterating through all loaded test methods parallel_targets = [] for test in proto_test_list: found = False for target in non_module_targets: # target is a dotted name of either a test case or test method # here test.dotted name is always a dotted name of a method if (target in test.dotted_name): if target not in parallel_targets: # Explicitly specified targets get their own entry to # run parallel to everything else parallel_targets.append(target) found = True break if found: continue # This test does not appear to be part of a specified target, so # its entire module must have been discovered, so just add the # whole module to the list if we haven't already. if test.module not in parallel_targets: parallel_targets.append(test.module) return parallel_targets
def toParallelTargets(suite, targets)
Produce a list of targets which should be tested in parallel. For the most part this will be a list of test modules. The exception is when a dotted name representing something more granular than a module was input (like an individal test case or test method)
5.14784
5.036218
1.022164
return (os.path.isdir(file_path) and os.path.isfile(os.path.join(file_path, '__init__.py')))
def isPackage(file_path)
Determine whether or not a given path is a (sub)package or not.
1.988337
2.019253
0.984689
if not os.path.isfile(file_path): raise ValueError("'{}' is not a file.".format(file_path)) parent_dir = os.path.dirname(os.path.abspath(file_path)) dotted_module = os.path.basename(file_path).replace('.py', '') while isPackage(parent_dir): dotted_module = os.path.basename(parent_dir) + '.' + dotted_module parent_dir = os.path.dirname(parent_dir) debug("Dotted module: {} -> {}".format( parent_dir, dotted_module), 2) return (dotted_module, parent_dir)
def findDottedModuleAndParentDir(file_path)
I return a tuple (dotted_module, parent_dir) where dotted_module is the full dotted name of the module with respect to the package it is in, and parent_dir is the absolute path to the parent directory of the package. If the python file is not part of a package, I return (None, None). For for filepath /a/b/c/d.py where b is the package, ('b.c.d', '/a') would be returned.
1.99718
2.12547
0.939641
default_filepath = os.path.join(home, ".green") if os.path.isfile(default_filepath): filepaths.append(default_filepath) # Low priority env_filepath = os.getenv("GREEN_CONFIG") if env_filepath and os.path.isfile(env_filepath): filepaths.append(env_filepath) # Medium priority for cfg_file in ("setup.cfg", ".green"): cwd_filepath = os.path.join(os.getcwd(), cfg_file) if os.path.isfile(cwd_filepath): filepaths.append(cwd_filepath) # High priority if filepath and os.path.isfile(filepath): filepaths.append(filepath) if filepaths: global files_loaded files_loaded = filepaths # Python 3 has parser.read_file(iterator) while Python2 has # parser.readfp(obj_with_readline) read_func = getattr(parser, 'read_file', getattr(parser, 'readfp')) for filepath in filepaths: # Users are expected to put a [green] section # only if they use setup.cfg if filepath.endswith('setup.cfg'): with open(filepath) as f: read_func(f) else: read_func(ConfigFile(filepath)) return parser
def getConfig(filepath=None): # pragma: no cover parser = configparser.ConfigParser() filepaths = [] # Lowest priority goes first in the list home = os.getenv("HOME") if home
Get the Green config file settings. All available config files are read. If settings are in multiple configs, the last value encountered wins. Values specified on the command-line take precedence over all config file settings. Returns: A ConfigParser object.
2.906389
2.876828
1.010275
err = ''.join(traceback.format_exception(*err)) else: err = '' sys.__stdout__.write("({}) {} {}".format(os.getpid(), msg, err)+'\n') sys.__stdout__.flush()
def ddebug(msg, err=None): # pragma: no cover import os if err
err can be an instance of sys.exc_info() -- which is the latest traceback info
5.161487
4.097863
1.259556
for i in range(self._processes - len(self._pool)): w = self.Process(target=worker, args=(self._inqueue, self._outqueue, self._initializer, self._initargs, self._maxtasksperchild, self._wrap_exception, self._finalizer, self._finalargs) ) self._pool.append(w) w.name = w.name.replace('Process', 'PoolWorker') w.daemon = True w.start() util.debug('added worker')
def _repopulate_pool(self)
Bring the number of pool processes up to the specified number, for use after reaping workers which have exited.
3.842238
3.376584
1.137907
if level <= debug_level: logging.debug(' ' * (level - 1) * 2 + str(message))
def debug(message, level=1)
So we can tune how much debug output we get when we turn it on.
4.202891
4.306973
0.975834
actual_spaces = (indent * self.indent_spaces) - len(outcome_char) return (outcome_char + ' ' * actual_spaces + line)
def formatLine(self, line, indent=0, outcome_char='')
Takes a single line, optionally adds an indent and/or outcome character to the beginning of the line.
5.82238
5.028763
1.157815
if not issubclass(GreenStream, type(stream)): stream = GreenStream(stream, disable_windows=args.disable_windows, disable_unidecode=args.disable_unidecode) result = GreenTestResult(args, stream) # Note: Catching SIGINT isn't supported by Python on windows (python # "WONTFIX" issue 18040) installHandler() registerResult(result) with warnings.catch_warnings(): if args.warnings: # pragma: no cover # if args.warnings is set, use it to filter all the warnings warnings.simplefilter(args.warnings) # if the filter is 'default' or 'always', special-case the # warnings from the deprecated unittest methods to show them # no more than once per module, because they can be fairly # noisy. The -Wd and -Wa flags can be used to bypass this # only when args.warnings is None. if args.warnings in ['default', 'always']: warnings.filterwarnings('module', category=DeprecationWarning, message='Please use assert\w+ instead.') result.startTestRun() pool = LoggingDaemonlessPool(processes=args.processes or None, initializer=InitializerOrFinalizer(args.initializer), finalizer=InitializerOrFinalizer(args.finalizer)) manager = multiprocessing.Manager() targets = [(target, manager.Queue()) for target in toParallelTargets(suite, args.targets)] if targets: for index, (target, queue) in enumerate(targets): if args.run_coverage: coverage_number = index + 1 else: coverage_number = None debug("Sending {} to runner {}".format(target, poolRunner)) pool.apply_async( poolRunner, (target, queue, coverage_number, args.omit_patterns, args.cov_config_file)) pool.close() for target, queue in targets: abort = False while True: msg = queue.get() # Sentinel value, we're done if not msg: break else: # Result guaranteed after this message, we're # currently waiting on this test, so print out # the white 'processing...' version of the output result.startTest(msg) proto_test_result = queue.get() result.addProtoTestResult(proto_test_result) if result.shouldStop: abort = True break if abort: break pool.close() pool.join() result.stopTestRun() removeResult(result) return result
def run(suite, stream, args, testing=False)
Run the given test case or test suite with the specified arguments. Any args.stream passed in will be wrapped in a GreenStream
5.782552
5.751722
1.00536
nineq, nz, neq, nBatch = get_sizes(G, A) eps = 1e-7 Q_tilde = Q + eps * torch.eye(nz).type_as(Q).repeat(nBatch, 1, 1) D_tilde = D + eps * torch.eye(nineq).type_as(Q).repeat(nBatch, 1, 1) dx, ds, dz, dy = factor_solve_kkt_reg( Q_tilde, D_tilde, G, A, rx, rs, rz, ry, eps) res = kkt_resid_reg(Q, D, G, A, eps, dx, ds, dz, dy, rx, rs, rz, ry) resx, ress, resz, resy = res res = resx for k in range(niter): ddx, dds, ddz, ddy = factor_solve_kkt_reg(Q_tilde, D_tilde, G, A, -resx, -ress, -resz, -resy if resy is not None else None, eps) dx, ds, dz, dy = [v + dv if v is not None else None for v, dv in zip((dx, ds, dz, dy), (ddx, dds, ddz, ddy))] res = kkt_resid_reg(Q, D, G, A, eps, dx, ds, dz, dy, rx, rs, rz, ry) resx, ress, resz, resy = res # res = torch.cat(resx) res = resx return dx, ds, dz, dy
def solve_kkt_ir(Q, D, G, A, rx, rs, rz, ry, niter=1)
Inefficient iterative refinement.
2.512223
2.491721
1.008228
nineq, nz, neq, nBatch = get_sizes(G, A) invQ_rx = rx.btrisolve(*Q_LU) if neq > 0: h = torch.cat((invQ_rx.unsqueeze(1).bmm(A.transpose(1, 2)).squeeze(1) - ry, invQ_rx.unsqueeze(1).bmm(G.transpose(1, 2)).squeeze(1) + rs / d - rz), 1) else: h = invQ_rx.unsqueeze(1).bmm(G.transpose(1, 2)).squeeze(1) + rs / d - rz w = -(h.btrisolve(*S_LU)) g1 = -rx - w[:, neq:].unsqueeze(1).bmm(G).squeeze(1) if neq > 0: g1 -= w[:, :neq].unsqueeze(1).bmm(A).squeeze(1) g2 = -rs - w[:, neq:] dx = g1.btrisolve(*Q_LU) ds = g2 / d dz = w[:, neq:] dy = w[:, :neq] if neq > 0 else None return dx, ds, dz, dy
def solve_kkt(Q_LU, d, G, A, S_LU, rx, rs, rz, ry)
Solve KKT equations for the affine step
3.2918
3.243233
1.014975
nineq, nz, neq, nBatch = get_sizes(G, A) try: Q_LU = btrifact_hack(Q) except: raise RuntimeError() # S = [ A Q^{-1} A^T A Q^{-1} G^T ] # [ G Q^{-1} A^T G Q^{-1} G^T + D^{-1} ] # # We compute a partial LU decomposition of the S matrix # that can be completed once D^{-1} is known. # See the 'Block LU factorization' part of our website # for more details. G_invQ_GT = torch.bmm(G, G.transpose(1, 2).btrisolve(*Q_LU)) R = G_invQ_GT.clone() S_LU_pivots = torch.IntTensor(range(1, 1 + neq + nineq)).unsqueeze(0) \ .repeat(nBatch, 1).type_as(Q).int() if neq > 0: invQ_AT = A.transpose(1, 2).btrisolve(*Q_LU) A_invQ_AT = torch.bmm(A, invQ_AT) G_invQ_AT = torch.bmm(G, invQ_AT) LU_A_invQ_AT = btrifact_hack(A_invQ_AT) P_A_invQ_AT, L_A_invQ_AT, U_A_invQ_AT = torch.btriunpack(*LU_A_invQ_AT) P_A_invQ_AT = P_A_invQ_AT.type_as(A_invQ_AT) S_LU_11 = LU_A_invQ_AT[0] U_A_invQ_AT_inv = (P_A_invQ_AT.bmm(L_A_invQ_AT) ).btrisolve(*LU_A_invQ_AT) S_LU_21 = G_invQ_AT.bmm(U_A_invQ_AT_inv) T = G_invQ_AT.transpose(1, 2).btrisolve(*LU_A_invQ_AT) S_LU_12 = U_A_invQ_AT.bmm(T) S_LU_22 = torch.zeros(nBatch, nineq, nineq).type_as(Q) S_LU_data = torch.cat((torch.cat((S_LU_11, S_LU_12), 2), torch.cat((S_LU_21, S_LU_22), 2)), 1) S_LU_pivots[:, :neq] = LU_A_invQ_AT[1] R -= G_invQ_AT.bmm(T) else: S_LU_data = torch.zeros(nBatch, nineq, nineq).type_as(Q) S_LU = [S_LU_data, S_LU_pivots] return Q_LU, S_LU, R
def pre_factor_kkt(Q, G, A)
Perform all one-time factorizations and cache relevant matrix products
2.962658
2.969993
0.99753
nBatch, nineq = d.size() neq = S_LU[1].size(1) - nineq # TODO: There's probably a better way to add a batched diagonal. global factor_kkt_eye if factor_kkt_eye is None or factor_kkt_eye.size() != d.size(): # print('Updating batchedEye size.') factor_kkt_eye = torch.eye(nineq).repeat( nBatch, 1, 1).type_as(R).byte() T = R.clone() T[factor_kkt_eye] += (1. / d).squeeze().view(-1) T_LU = btrifact_hack(T) global shown_btrifact_warning if shown_btrifact_warning or not T.is_cuda: # TODO: Don't use pivoting in most cases because # torch.btriunpack is inefficient here: oldPivotsPacked = S_LU[1][:, -nineq:] - neq oldPivots, _, _ = torch.btriunpack( T_LU[0], oldPivotsPacked, unpack_data=False) newPivotsPacked = T_LU[1] newPivots, _, _ = torch.btriunpack( T_LU[0], newPivotsPacked, unpack_data=False) # Re-pivot the S_LU_21 block. if neq > 0: S_LU_21 = S_LU[0][:, -nineq:, :neq] S_LU[0][:, -nineq:, :neq] = newPivots.transpose(1, 2).bmm(oldPivots.bmm(S_LU_21)) # Add the new S_LU_22 block pivots. S_LU[1][:, -nineq:] = newPivotsPacked + neq # Add the new S_LU_22 block. S_LU[0][:, -nineq:, -nineq:] = T_LU[0]
def factor_kkt(S_LU, R, d)
Factor the U22 block that we can only do after we know D.
4.54381
4.420864
1.02781
nineq, nz, neq, _ = get_sizes(G, A) invQ_rx = torch.potrs(rx.view(-1, 1), U_Q).view(-1) if neq > 0: h = torch.cat([torch.mv(A, invQ_rx) - ry, torch.mv(G, invQ_rx) + rs / d - rz], 0) else: h = torch.mv(G, invQ_rx) + rs / d - rz w = -torch.potrs(h.view(-1, 1), U_S).view(-1) g1 = -rx - torch.mv(G.t(), w[neq:]) if neq > 0: g1 -= torch.mv(A.t(), w[:neq]) g2 = -rs - w[neq:] dx = torch.potrs(g1.view(-1, 1), U_Q).view(-1) ds = g2 / d dz = w[neq:] dy = w[:neq] if neq > 0 else None # if np.all(np.array([x.norm() for x in [rx, rs, rz, ry]]) != 0): if dbg: import IPython import sys IPython.embed() sys.exit(-1) # if rs.norm() > 0: import IPython, sys; IPython.embed(); sys.exit(-1) return dx, ds, dz, dy
def solve_kkt(U_Q, d, G, A, U_S, rx, rs, rz, ry, dbg=False)
Solve KKT equations for the affine step
3.218962
3.189029
1.009386
nineq, nz, neq, _ = get_sizes(G, A) # S = [ A Q^{-1} A^T A Q^{-1} G^T ] # [ G Q^{-1} A^T G Q^{-1} G^T + D^{-1} ] U_Q = torch.potrf(Q) # partial cholesky of S matrix U_S = torch.zeros(neq + nineq, neq + nineq).type_as(Q) G_invQ_GT = torch.mm(G, torch.potrs(G.t(), U_Q)) R = G_invQ_GT if neq > 0: invQ_AT = torch.potrs(A.t(), U_Q) A_invQ_AT = torch.mm(A, invQ_AT) G_invQ_AT = torch.mm(G, invQ_AT) # TODO: torch.potrf sometimes says the matrix is not PSD but # numpy does? I filed an issue at # https://github.com/pytorch/pytorch/issues/199 try: U11 = torch.potrf(A_invQ_AT) except: U11 = torch.Tensor(np.linalg.cholesky( A_invQ_AT.cpu().numpy())).type_as(A_invQ_AT) # TODO: torch.trtrs is currently not implemented on the GPU # and we are using gesv as a workaround. U12 = torch.gesv(G_invQ_AT.t(), U11.t())[0] U_S[:neq, :neq] = U11 U_S[:neq, neq:] = U12 R -= torch.mm(U12.t(), U12) return U_Q, U_S, R
def pre_factor_kkt(Q, G, A)
Perform all one-time factorizations and cache relevant matrix products
3.709364
3.71829
0.997599
nineq = R.size(0) U_S[-nineq:, -nineq:] = torch.potrf(R + torch.diag(1 / d.cpu()).type_as(d))
def factor_kkt(U_S, R, d)
Factor the U22 block that we can only do after we know D.
7.498366
7.296558
1.027658
def decorate(function): if not seconds: return function if use_signals: def handler(signum, frame): _raise_exception(timeout_exception, exception_message) @wraps(function) def new_function(*args, **kwargs): new_seconds = kwargs.pop('timeout', seconds) if new_seconds: old = signal.signal(signal.SIGALRM, handler) signal.setitimer(signal.ITIMER_REAL, new_seconds) try: return function(*args, **kwargs) finally: if new_seconds: signal.setitimer(signal.ITIMER_REAL, 0) signal.signal(signal.SIGALRM, old) return new_function else: @wraps(function) def new_function(*args, **kwargs): timeout_wrapper = _Timeout(function, timeout_exception, exception_message, seconds) return timeout_wrapper(*args, **kwargs) return new_function return decorate
def timeout(seconds=None, use_signals=True, timeout_exception=TimeoutError, exception_message=None)
Add a timeout parameter to a function and return it. :param seconds: optional time limit in seconds or fractions of a second. If None is passed, no timeout is applied. This adds some flexibility to the usage: you can disable timing out depending on the settings. :type seconds: float :param use_signals: flag indicating whether signals should be used for timing function out or the multiprocessing When using multiprocessing, timeout granularity is limited to 10ths of a second. :type use_signals: bool :raises: TimeoutError if time limit is reached It is illegal to pass anything other than a function as the first parameter. The function is wrapped and returned to the caller.
1.910564
2.043271
0.935051
try: queue.put((True, function(*args, **kwargs))) except: queue.put((False, sys.exc_info()[1]))
def _target(queue, function, *args, **kwargs)
Run a function with arguments and return output via a queue. This is a helper function for the Process created in _Timeout. It runs the function with positional arguments and keyword arguments and then returns the function's output by way of a queue. If an exception gets raised, it is returned to _Timeout to be raised by the value property.
2.371452
2.874043
0.825128
if self.__process.is_alive(): self.__process.terminate() _raise_exception(self.__timeout_exception, self.__exception_message)
def cancel(self)
Terminate any possible execution of the embedded function.
7.931096
6.677467
1.18774
if self.__timeout < time.time(): self.cancel() return self.__queue.full() and not self.__queue.empty()
def ready(self)
Read-only property indicating status of "value" property.
7.772804
8.221615
0.945411
if self.ready is True: flag, load = self.__queue.get() if flag: return load raise load
def value(self)
Read-only property containing data returned from function.
16.731735
13.611957
1.229194
bbs = [] for thing in paths_n_stuff: if is_path_segment(thing) or isinstance(thing, Path): bbs.append(thing.bbox()) elif isinstance(thing, complex): bbs.append((thing.real, thing.real, thing.imag, thing.imag)) else: try: complexthing = complex(thing) bbs.append((complexthing.real, complexthing.real, complexthing.imag, complexthing.imag)) except ValueError: raise TypeError( "paths_n_stuff can only contains Path, CubicBezier, " "QuadraticBezier, Line, and complex objects.") xmins, xmaxs, ymins, ymaxs = list(zip(*bbs)) xmin = min(xmins) xmax = max(xmaxs) ymin = min(ymins) ymax = max(ymaxs) return xmin, xmax, ymin, ymax
def big_bounding_box(paths_n_stuff)
Finds a BB containing a collection of paths, Bezier path segments, and points (given as complex numbers).
2.30513
2.171829
1.061377
disvg(paths, colors=colors, filename=filename, stroke_widths=stroke_widths, nodes=nodes, node_colors=node_colors, node_radii=node_radii, openinbrowser=openinbrowser, timestamp=timestamp, margin_size=margin_size, mindim=mindim, dimensions=dimensions, viewbox=viewbox, text=text, text_path=text_path, font_size=font_size, attributes=attributes, svg_attributes=svg_attributes, svgwrite_debug=svgwrite_debug, paths2Drawing=paths2Drawing)
def wsvg(paths=None, colors=None, filename=os_path.join(getcwd(), 'disvg_output.svg'), stroke_widths=None, nodes=None, node_colors=None, node_radii=None, openinbrowser=False, timestamp=False, margin_size=0.1, mindim=600, dimensions=None, viewbox=None, text=None, text_path=None, font_size=None, attributes=None, svg_attributes=None, svgwrite_debug=False, paths2Drawing=False)
Convenience function; identical to disvg() except that openinbrowser=False by default. See disvg() docstring for more info.
1.262309
1.283589
0.983421
roots = np.roots(p) if realroots: roots = [r.real for r in roots if isclose(r.imag, 0)] roots = [r for r in roots if condition(r)] duplicates = [] for idx, (r1, r2) in enumerate(combinations(roots, 2)): if isclose(r1, r2): duplicates.append(idx) return [r for idx, r in enumerate(roots) if idx not in duplicates]
def polyroots(p, realroots=False, condition=lambda r: True)
Returns the roots of a polynomial with coefficients given in p. p[0] * x**n + p[1] * x**(n-1) + ... + p[n-1]*x + p[n] INPUT: p - Rank-1 array-like object of polynomial coefficients. realroots - a boolean. If true, only real roots will be returned and the condition function can be written assuming all roots are real. condition - a boolean-valued function. Only roots satisfying this will be returned. If realroots==True, these conditions should assume the roots are real. OUTPUT: A list containing the roots of the polynomial. NOTE: This uses np.isclose and np.roots
2.293101
2.643953
0.8673
assert isinstance(f, np.poly1d) and isinstance(g, np.poly1d) assert g != np.poly1d([0]) if g(t0) != 0: return f(t0)/g(t0) elif f(t0) == 0: return rational_limit(f.deriv(), g.deriv(), t0) else: raise ValueError("Limit does not exist.")
def rational_limit(f, g, t0)
Computes the limit of the rational function (f/g)(t) as t approaches t0.
2.657671
2.737225
0.970936
b = Line(xmin + 1j*ymin, xmax + 1j*ymin) t = Line(xmin + 1j*ymax, xmax + 1j*ymax) r = Line(xmax + 1j*ymin, xmax + 1j*ymax) l = Line(xmin + 1j*ymin, xmin + 1j*ymax) return Path(b, r, t.reversed(), l.reversed())
def bbox2path(xmin, xmax, ymin, ymax)
Converts a bounding box 4-tuple to a Path object.
2.36759
2.306506
1.026483
return Path(*[Line(points[i], points[i+1]) for i in range(len(points) - 1)])
def polyline(*points)
Converts a list of points to a Path composed of lines connecting those points (i.e. a linear spline or polyline). See also `polygon()`.
3.551376
3.295019
1.077801
return Path(*[Line(points[i], points[(i + 1) % len(points)]) for i in range(len(points))])
def polygon(*points)
Converts a list of points to a Path composed of lines connecting those points, then closes the path by connecting the last point to the first. See also `polyline()`.
3.143064
2.782976
1.129389
order = len(bpoints) - 1 if order == 3: return CubicBezier(*bpoints) elif order == 2: return QuadraticBezier(*bpoints) elif order == 1: return Line(*bpoints) else: assert len(bpoints) in {2, 3, 4}
def bpoints2bezier(bpoints)
Converts a list of length 2, 3, or 4 to a CubicBezier, QuadraticBezier, or Line object, respectively. See also: poly2bez.
2.297036
2.19537
1.046309
bpoints = polynomial2bezier(poly) if return_bpoints: return bpoints else: return bpoints2bezier(bpoints)
def poly2bez(poly, return_bpoints=False)
Converts a cubic or lower order Polynomial object (or a sequence of coefficients) to a CubicBezier, QuadraticBezier, or Line object as appropriate. If return_bpoints=True then this will instead only return the control points of the corresponding Bezier curve. Note: The inverse operation is available as a method of CubicBezier, QuadraticBezier and Line objects.
3.21234
3.889879
0.82582
if is_bezier_segment(bez): bez = bez.bpoints() return bezier2polynomial(bez, numpy_ordering=numpy_ordering, return_poly1d=return_poly1d)
def bez2poly(bez, numpy_ordering=True, return_poly1d=False)
Converts a Bezier object or tuple of Bezier control points to a tuple of coefficients of the expanded polynomial. return_poly1d : returns a numpy.poly1d object. This makes computations of derivatives/anti-derivatives and many other operations quite quick. numpy_ordering : By default (to accommodate numpy) the coefficients will be output in reverse standard order. Note: This function is redundant thanks to the .poly() method included with all bezier segment classes.
3.723709
3.692133
1.008552
def transform(z): return exp(1j*radians(degs))*(z - origin) + origin if origin is None: if isinstance(curve, Arc): origin = curve.center else: origin = curve.point(0.5) if isinstance(curve, Path): return Path(*[rotate(seg, degs, origin=origin) for seg in curve]) elif is_bezier_segment(curve): return bpoints2bezier([transform(bpt) for bpt in curve.bpoints()]) elif isinstance(curve, Arc): new_start = transform(curve.start) new_end = transform(curve.end) new_rotation = curve.rotation + degs return Arc(new_start, radius=curve.radius, rotation=new_rotation, large_arc=curve.large_arc, sweep=curve.sweep, end=new_end) else: raise TypeError("Input `curve` should be a Path, Line, " "QuadraticBezier, CubicBezier, or Arc object.")
def rotate(curve, degs, origin=None)
Returns curve rotated by `degs` degrees (CCW) around the point `origin` (a complex number). By default origin is either `curve.point(0.5)`, or in the case that curve is an Arc object, `origin` defaults to `curve.center`.
3.044178
2.772934
1.097819
if isinstance(curve, Path): return Path(*[translate(seg, z0) for seg in curve]) elif is_bezier_segment(curve): return bpoints2bezier([bpt + z0 for bpt in curve.bpoints()]) elif isinstance(curve, Arc): new_start = curve.start + z0 new_end = curve.end + z0 return Arc(new_start, radius=curve.radius, rotation=curve.rotation, large_arc=curve.large_arc, sweep=curve.sweep, end=new_end) else: raise TypeError("Input `curve` should be a Path, Line, " "QuadraticBezier, CubicBezier, or Arc object.")
def translate(curve, z0)
Shifts the curve by the complex quantity z such that translate(curve, z0).point(t) = curve.point(t) + z0
2.945862
2.868009
1.027145
if sy is None: isy = 1j*sx else: isy = 1j*sy def _scale(z): if sy is None: return sx*z return sx*z.real + isy*z.imag def scale_bezier(bez): p = [_scale(c) for c in bez2poly(bez)] p[-1] += origin - _scale(origin) return poly2bez(p) if isinstance(curve, Path): return Path(*[scale(seg, sx, sy, origin) for seg in curve]) elif is_bezier_segment(curve): return scale_bezier(curve) elif isinstance(curve, Arc): if sy is None or sy == sx: return Arc(start=sx*(curve.start - origin) + origin, radius=sx*curve.radius, rotation=curve.rotation, large_arc=curve.large_arc, sweep=curve.sweep, end=sx*(curve.end - origin) + origin) else: raise Exception("\nFor `Arc` objects, only scale transforms " "with sx==sy are implemented.\n") else: raise TypeError("Input `curve` should be a Path, Line, " "QuadraticBezier, CubicBezier, or Arc object.")
def scale(curve, sx, sy=None, origin=0j)
Scales `curve`, about `origin`, by diagonal matrix `[[sx,0],[0,sy]]`. Notes: ------ * If `sy` is not specified, it is assumed to be equal to `sx` and a scalar transformation of `curve` about `origin` will be returned. I.e. scale(curve, sx, origin).point(t) == ((curve.point(t) - origin) * sx) + origin
3.270763
3.373415
0.96957
def to_point(p): return np.array([[p.real], [p.imag], [1.0]]) def to_vector(z): return np.array([[z.real], [z.imag], [0.0]]) def to_complex(v): return v.item(0) + 1j * v.item(1) if isinstance(curve, Path): return Path(*[transform(segment, tf) for segment in curve]) elif is_bezier_segment(curve): return bpoints2bezier([to_complex(tf.dot(to_point(p))) for p in curve.bpoints()]) elif isinstance(curve, Arc): new_start = to_complex(tf.dot(to_point(curve.start))) new_end = to_complex(tf.dot(to_point(curve.end))) new_radius = to_complex(tf.dot(to_vector(curve.radius))) return Arc(new_start, radius=new_radius, rotation=curve.rotation, large_arc=curve.large_arc, sweep=curve.sweep, end=new_end) else: raise TypeError("Input `curve` should be a Path, Line, " "QuadraticBezier, CubicBezier, or Arc object.")
def transform(curve, tf)
Transforms the curve by the homogeneous transformation matrix tf
2.516097
2.512986
1.001238
assert 0 <= t <= 1 dseg = seg.derivative(t) # Note: dseg might be numpy value, use np.seterr(invalid='raise') try: unit_tangent = dseg/abs(dseg) except (ZeroDivisionError, FloatingPointError): # This may be a removable singularity, if so we just need to compute # the limit. # Note: limit{{dseg / abs(dseg)} = sqrt(limit{dseg**2 / abs(dseg)**2}) dseg_poly = seg.poly().deriv() dseg_abs_squared_poly = (real(dseg_poly) ** 2 + imag(dseg_poly) ** 2) try: unit_tangent = csqrt(rational_limit(dseg_poly**2, dseg_abs_squared_poly, t)) except ValueError: bef = seg.poly().deriv()(t - 1e-4) aft = seg.poly().deriv()(t + 1e-4) mes = ("Unit tangent appears to not be well-defined at " "t = {}, \n".format(t) + "seg.poly().deriv()(t - 1e-4) = {}\n".format(bef) + "seg.poly().deriv()(t + 1e-4) = {}".format(aft)) raise ValueError(mes) return unit_tangent
def bezier_unit_tangent(seg, t)
Returns the unit tangent of the segment at t. Notes ----- If you receive a RuntimeWarning, try the following: >>> import numpy >>> old_numpy_error_settings = numpy.seterr(invalid='raise') This can be undone with: >>> numpy.seterr(**old_numpy_error_settings)
4.514592
4.734145
0.953624
dz = self.derivative(t) ddz = self.derivative(t, n=2) dx, dy = dz.real, dz.imag ddx, ddy = ddz.real, ddz.imag old_np_seterr = np.seterr(invalid='raise') try: kappa = abs(dx*ddy - dy*ddx)/sqrt(dx*dx + dy*dy)**3 except (ZeroDivisionError, FloatingPointError): # tangent vector is zero at t, use polytools to find limit p = self.poly() dp = p.deriv() ddp = dp.deriv() dx, dy = real(dp), imag(dp) ddx, ddy = real(ddp), imag(ddp) f2 = (dx*ddy - dy*ddx)**2 g2 = (dx*dx + dy*dy)**3 lim2 = rational_limit(f2, g2, t) if lim2 < 0: # impossible, must be numerical error return 0 kappa = sqrt(lim2) finally: np.seterr(**old_np_seterr) return kappa
def segment_curvature(self, t, use_inf=False)
returns the curvature of the segment at t. Notes ----- If you receive a RuntimeWarning, run command >>> old = np.seterr(invalid='raise') This can be undone with >>> np.seterr(**old)
3.538768
3.345453
1.057784
def _radius(tau): return abs(seg.point(tau) - origin) shifted_seg_poly = seg.poly() - origin r_squared = real(shifted_seg_poly) ** 2 + \ imag(shifted_seg_poly) ** 2 extremizers = [0, 1] + polyroots01(r_squared.deriv()) extrema = [(_radius(t), t) for t in extremizers] if return_all_global_extrema: raise NotImplementedError else: seg_global_min = min(extrema, key=itemgetter(0)) seg_global_max = max(extrema, key=itemgetter(0)) return seg_global_min, seg_global_max
def bezier_radialrange(seg, origin, return_all_global_extrema=False)
returns the tuples (d_min, t_min) and (d_max, t_max) which minimize and maximize, respectively, the distance d = |self.point(t)-origin|. return_all_global_extrema: Multiple such t_min or t_max values can exist. By default, this will only return one. Set return_all_global_extrema=True to return all such global extrema.
3.882716
3.740372
1.038056
assert path.isclosed() intersections = Path(Line(pt, opt)).intersect(path) if len(intersections) % 2: return True else: return False
def path_encloses_pt(pt, opt, path)
returns true if pt is a point enclosed by path (which must be a Path object satisfying path.isclosed==True). opt is a point you know is NOT enclosed by path.
6.050798
5.068597
1.193782
mid = (start + end)/2 mid_point = curve.point(mid) length = abs(end_point - start_point) first_half = abs(mid_point - start_point) second_half = abs(end_point - mid_point) length2 = first_half + second_half if (length2 - length > error) or (depth < min_depth): # Calculate the length of each segment: depth += 1 return (segment_length(curve, start, mid, start_point, mid_point, error, min_depth, depth) + segment_length(curve, mid, end, mid_point, end_point, error, min_depth, depth)) # This is accurate enough. return length2
def segment_length(curve, start, end, start_point, end_point, error=LENGTH_ERROR, min_depth=LENGTH_MIN_DEPTH, depth=0)
Recursively approximates the length by straight lines
2.591853
2.543261
1.019106
curve_length = curve.length(error=error, min_depth=min_depth) assert curve_length > 0 if not 0 <= s <= curve_length: raise ValueError("s is not in interval [0, curve.length()].") if s == 0: return 0 if s == curve_length: return 1 if isinstance(curve, Path): seg_lengths = [seg.length(error=error, min_depth=min_depth) for seg in curve] lsum = 0 # Find which segment the point we search for is located on for k, len_k in enumerate(seg_lengths): if lsum <= s <= lsum + len_k: t = inv_arclength(curve[k], s - lsum, s_tol=s_tol, maxits=maxits, error=error, min_depth=min_depth) return curve.t2T(k, t) lsum += len_k return 1 elif isinstance(curve, Line): return s / curve.length(error=error, min_depth=min_depth) elif (isinstance(curve, QuadraticBezier) or isinstance(curve, CubicBezier) or isinstance(curve, Arc)): t_upper = 1 t_lower = 0 iteration = 0 while iteration < maxits: iteration += 1 t = (t_lower + t_upper)/2 s_t = curve.length(t1=t, error=error, min_depth=min_depth) if abs(s_t - s) < s_tol: return t elif s_t < s: # t too small t_lower = t else: # s < s_t, t too big t_upper = t if t_upper == t_lower: warn("t is as close as a float can be to the correct value, " "but |s(t) - s| = {} > s_tol".format(abs(s_t-s))) return t raise Exception("Maximum iterations reached with s(t) - s = {}." "".format(s_t - s)) else: raise TypeError("First argument must be a Line, QuadraticBezier, " "CubicBezier, Arc, or Path object.")
def inv_arclength(curve, s, s_tol=ILENGTH_S_TOL, maxits=ILENGTH_MAXITS, error=ILENGTH_ERROR, min_depth=ILENGTH_MIN_DEPTH)
INPUT: curve should be a CubicBezier, Line, of Path of CubicBezier and/or Line objects. OUTPUT: Returns a float, t, such that the arc length of curve from 0 to t is approximately s. s_tol - exit when |s(t) - s| < s_tol where s(t) = seg.length(0, t, error, min_depth) and seg is either curve or, if curve is a Path object, then seg is a segment in curve. error - used to compute lengths of cubics and arcs min_depth - used to compute lengths of cubics and arcs Note: This function is not designed to be efficient, but if it's slower than you need, make sure you have scipy installed.
2.625599
2.57426
1.019943
assert t0 < t1 if t0 == 0: cropped_seg = seg.split(t1)[0] elif t1 == 1: cropped_seg = seg.split(t0)[1] else: pt1 = seg.point(t1) # trim off the 0 <= t < t0 part trimmed_seg = crop_bezier(seg, t0, 1) # find the adjusted t1 (i.e. the t1 such that # trimmed_seg.point(t1) ~= pt))and trim off the t1 < t <= 1 part t1_adj = trimmed_seg.radialrange(pt1)[0][1] cropped_seg = crop_bezier(trimmed_seg, 0, t1_adj) return cropped_seg
def crop_bezier(seg, t0, t1)
returns a cropped copy of this segment which starts at self.point(t0) and ends at self.point(t1).
4.093462
3.867905
1.058315
if wrt_parameterization: return self.start == previous.end and np.isclose( self.derivative(0), previous.derivative(1)) else: return self.start == previous.end and np.isclose( self.unit_tangent(0), previous.unit_tangent(1))
def joins_smoothly_with(self, previous, wrt_parameterization=False)
Checks if this segment joins smoothly with previous segment. By default, this only checks that this segment starts moving (at t=0) in the same direction (and from the same positive) as previous stopped moving (at t=1). To check if the tangent magnitudes also match, set wrt_parameterization=True.
2.936291
2.240697
1.310437
distance = self.end - self.start return self.start + distance*t
def point(self, t)
returns the coordinates of the Bezier curve evaluated at t.
7.792747
6.886596
1.131582
return abs(self.end - self.start)*(t1-t0)
def length(self, t0=0, t1=1, error=None, min_depth=None)
returns the length of the line segment between t0 and t1.
8.326482
6.006574
1.386228
return inv_arclength(self, s, s_tol=s_tol, maxits=maxits, error=error, min_depth=min_depth)
def ilength(self, s, s_tol=ILENGTH_S_TOL, maxits=ILENGTH_MAXITS, error=ILENGTH_ERROR, min_depth=ILENGTH_MIN_DEPTH)
Returns a float, t, such that self.length(0, t) is approximately s. See the inv_arclength() docstring for more details.
2.96678
1.922493
1.543194
assert self.end != self.start if n == 1: return self.end - self.start elif n > 1: return 0 else: raise ValueError("n should be a positive integer.")
def derivative(self, t=None, n=1)
returns the nth derivative of the segment at t.
3.726334
3.483768
1.069628
assert self.end != self.start dseg = self.end - self.start return dseg/abs(dseg)
def unit_tangent(self, t=None)
returns the unit tangent of the segment at t.
7.868661
6.28594
1.251787
if isinstance(other_seg, Line): assert other_seg.end != other_seg.start and self.end != self.start assert self != other_seg # Solve the system [p1-p0, q1-q0]*[t1, t2]^T = q0 - p0 # where self == Line(p0, p1) and other_seg == Line(q0, q1) a = (self.start.real, self.end.real) b = (self.start.imag, self.end.imag) c = (other_seg.start.real, other_seg.end.real) d = (other_seg.start.imag, other_seg.end.imag) denom = ((a[1] - a[0])*(d[0] - d[1]) - (b[1] - b[0])*(c[0] - c[1])) if np.isclose(denom, 0): return [] t1 = (c[0]*(b[0] - d[1]) - c[1]*(b[0] - d[0]) - a[0]*(d[0] - d[1]))/denom t2 = -(a[1]*(b[0] - d[0]) - a[0]*(b[1] - d[0]) - c[0]*(b[0] - b[1]))/denom if 0 <= t1 <= 1 and 0 <= t2 <= 1: return [(t1, t2)] return [] elif isinstance(other_seg, QuadraticBezier): t2t1s = bezier_by_line_intersections(other_seg, self) return [(t1, t2) for t2, t1 in t2t1s] elif isinstance(other_seg, CubicBezier): t2t1s = bezier_by_line_intersections(other_seg, self) return [(t1, t2) for t2, t1 in t2t1s] elif isinstance(other_seg, Arc): t2t1s = other_seg.intersect(self) return [(t1, t2) for t2, t1 in t2t1s] elif isinstance(other_seg, Path): raise TypeError( "other_seg must be a path segment, not a Path object, use " "Path.intersect().") else: raise TypeError("other_seg must be a path segment.")
def intersect(self, other_seg, tol=None)
Finds the intersections of two segments. returns a list of tuples (t1, t2) such that self.point(t1) == other_seg.point(t2). Note: This will fail if the two segments coincide for more than a finite collection of points. tol is not used.
2.10761
2.031216
1.03761
xmin = min(self.start.real, self.end.real) xmax = max(self.start.real, self.end.real) ymin = min(self.start.imag, self.end.imag) ymax = max(self.start.imag, self.end.imag) return xmin, xmax, ymin, ymax
def bbox(self)
returns the bounding box for the segment in the form (xmin, xmax, ymin, ymax).
1.820305
1.676629
1.085693
# Single-precision floats have only 7 significant figures of # resolution, so test that we're within 6 sig figs. if np.isclose(point, self.start, rtol=0, atol=1e-6): return 0.0 elif np.isclose(point, self.end, rtol=0, atol=1e-6): return 1.0 # Finding the point "by hand" here is much faster than calling # radialrange(), see the discussion on PR #40: # https://github.com/mathandy/svgpathtools/pull/40#issuecomment-358134261 p = self.poly() # p(t) = (p_1 * t) + p_0 = point # t = (point - p_0) / p_1 t = (point - p[0]) / p[1] if np.isclose(t.imag, 0) and (t.real >= 0.0) and (t.real <= 1.0): return t.real return None
def point_to_t(self, point)
If the point lies on the Line, returns its `t` parameter. If the point does not lie on the Line, returns None.
4.284103
4.051685
1.057363
return Line(self.point(t0), self.point(t1))
def cropped(self, t0, t1)
returns a cropped copy of this segment which starts at self.point(t0) and ends at self.point(t1).
6.004348
4.132162
1.453077
pt = self.point(t) return Line(self.start, pt), Line(pt, self.end)
def split(self, t)
returns two segments, whose union is this segment and which join at self.point(t).
5.497168
3.749544
1.46609
return bezier_radialrange(self, origin, return_all_global_extrema=return_all_global_extrema)
def radialrange(self, origin, return_all_global_extrema=False)
returns the tuples (d_min, t_min) and (d_max, t_max) which minimize and maximize, respectively, the distance d = |self.point(t)-origin|.
3.882297
3.991108
0.972737
return scale(self, sx=sx, sy=sy, origin=origin)
def scaled(self, sx, sy=None, origin=0j)
Scale transform. See `scale` function for further explanation.
5.154077
3.192759
1.614302
if warning_on: warn(_is_smooth_from_warning) if isinstance(previous, QuadraticBezier): return (self.start == previous.end and (self.control - self.start) == ( previous.end - previous.control)) else: return self.control == self.start
def is_smooth_from(self, previous, warning_on=True)
[Warning: The name of this method is somewhat misleading (yet kept for compatibility with scripts created using svg.path 2.0). This method is meant only for d string creation and should not be used to check for kinks. To check a segment for differentiability, use the joins_smoothly_with() method instead.]
4.356516
3.673711
1.185862
return (1 - t)**2*self.start + 2*(1 - t)*t*self.control + t**2*self.end
def point(self, t)
returns the coordinates of the Bezier curve evaluated at t.
3.158515
2.601938
1.213909
p = self.bpoints() coeffs = (p[0] - 2*p[1] + p[2], 2*(p[1] - p[0]), p[0]) if return_coeffs: return coeffs else: return np.poly1d(coeffs)
def poly(self, return_coeffs=False)
returns the quadratic as a Polynomial object.
2.944818
2.808818
1.048419
p = self.bpoints() if n == 1: return 2*((p[1] - p[0])*(1 - t) + (p[2] - p[1])*t) elif n == 2: return 2*(p[2] - 2*p[1] + p[0]) elif n > 2: return 0 else: raise ValueError("n should be a positive integer.")
def derivative(self, t, n=1)
returns the nth derivative of the segment at t. Note: Bezier curves can have points where their derivative vanishes. If you are interested in the tangent direction, use the unit_tangent() method instead.
2.765072
2.535073
1.090727
new_quad = QuadraticBezier(self.end, self.control, self.start) if self._length_info['length']: new_quad._length_info = self._length_info new_quad._length_info['bpoints'] = ( self.end, self.control, self.start) return new_quad
def reversed(self)
returns a copy of the QuadraticBezier object with its orientation reversed.
4.30211
3.571423
1.204593
bpoints1, bpoints2 = split_bezier(self.bpoints(), t) return QuadraticBezier(*bpoints1), QuadraticBezier(*bpoints2)
def split(self, t)
returns two segments, whose union is this segment and which join at self.point(t).
5.445813
4.776465
1.140135
if warning_on: warn(_is_smooth_from_warning) if isinstance(previous, CubicBezier): return (self.start == previous.end and (self.control1 - self.start) == ( previous.end - previous.control2)) else: return self.control1 == self.start
def is_smooth_from(self, previous, warning_on=True)
[Warning: The name of this method is somewhat misleading (yet kept for compatibility with scripts created using svg.path 2.0). This method is meant only for d string creation and should not be used to check for kinks. To check a segment for differentiability, use the joins_smoothly_with() method instead.]
4.177956
3.64568
1.146002
# algebraically equivalent to # P0*(1-t)**3 + 3*P1*t*(1-t)**2 + 3*P2*(1-t)*t**2 + P3*t**3 # for (P0, P1, P2, P3) = self.bpoints() return self.start + t*( 3*(self.control1 - self.start) + t*( 3*(self.start + self.control2) - 6*self.control1 + t*( -self.start + 3*(self.control1 - self.control2) + self.end )))
def point(self, t)
Evaluate the cubic Bezier curve at t using Horner's rule.
3.4592
2.944725
1.174711
if t0 == 0 and t1 == 1: if self._length_info['bpoints'] == self.bpoints() \ and self._length_info['error'] >= error \ and self._length_info['min_depth'] >= min_depth: return self._length_info['length'] # using scipy.integrate.quad is quick if _quad_available: s = quad(lambda tau: abs(self.derivative(tau)), t0, t1, epsabs=error, limit=1000)[0] else: s = segment_length(self, t0, t1, self.point(t0), self.point(t1), error, min_depth, 0) if t0 == 0 and t1 == 1: self._length_info['length'] = s self._length_info['bpoints'] = self.bpoints() self._length_info['error'] = error self._length_info['min_depth'] = min_depth return self._length_info['length'] else: return s
def length(self, t0=0, t1=1, error=LENGTH_ERROR, min_depth=LENGTH_MIN_DEPTH)
Calculate the length of the path up to a certain position
2.482484
2.519341
0.98537
return self.start, self.control1, self.control2, self.end
def bpoints(self)
returns the Bezier control points of the segment.
7.949986
3.49295
2.276009
p = self.bpoints() if n == 1: return 3*(p[1] - p[0])*(1 - t)**2 + 6*(p[2] - p[1])*(1 - t)*t + 3*( p[3] - p[2])*t**2 elif n == 2: return 6*( (1 - t)*(p[2] - 2*p[1] + p[0]) + t*(p[3] - 2*p[2] + p[1])) elif n == 3: return 6*(p[3] - 3*(p[2] - p[1]) - p[0]) elif n > 3: return 0 else: raise ValueError("n should be a positive integer.")
def derivative(self, t, n=1)
returns the nth derivative of the segment at t. Note: Bezier curves can have points where their derivative vanishes. If you are interested in the tangent direction, use the unit_tangent() method instead.
2.220928
2.098512
1.058335
new_cub = CubicBezier(self.end, self.control2, self.control1, self.start) if self._length_info['length']: new_cub._length_info = self._length_info new_cub._length_info['bpoints'] = ( self.end, self.control2, self.control1, self.start) return new_cub
def reversed(self)
returns a copy of the CubicBezier object with its orientation reversed.
4.056022
3.284256
1.234989
if isinstance(other_seg, Line): return bezier_by_line_intersections(self, other_seg) elif (isinstance(other_seg, QuadraticBezier) or isinstance(other_seg, CubicBezier)): assert self != other_seg longer_length = max(self.length(), other_seg.length()) return bezier_intersections(self, other_seg, longer_length=longer_length, tol=tol, tol_deC=tol) elif isinstance(other_seg, Arc): t2t1s = other_seg.intersect(self) return [(t1, t2) for t2, t1 in t2t1s] elif isinstance(other_seg, Path): raise TypeError( "other_seg must be a path segment, not a Path object, use " "Path.intersect().") else: raise TypeError("other_seg must be a path segment.")
def intersect(self, other_seg, tol=1e-12)
Finds the intersections of two segments. returns a list of tuples (t1, t2) such that self.point(t1) == other_seg.point(t2). Note: This will fail if the two segments coincide for more than a finite collection of points.
3.356851
3.097383
1.08377
bpoints1, bpoints2 = split_bezier(self.bpoints(), t) return CubicBezier(*bpoints1), CubicBezier(*bpoints2)
def split(self, t)
returns two segments, whose union is this segment and which join at self.point(t).
5.527829
4.864524
1.136355
x = real(zeta) y = imag(zeta) z = x*self.radius.real + y*self.radius.imag return self.rot_matrix*z + self.center
def iu1transform(self, zeta)
This is an affine transformation, the inverse of self.u1transform().
5.154815
4.575392
1.126639
assert 0 <= t0 <= 1 and 0 <= t1 <= 1 if _quad_available: return quad(lambda tau: abs(self.derivative(tau)), t0, t1, epsabs=error, limit=1000)[0] else: return segment_length(self, t0, t1, self.point(t0), self.point(t1), error, min_depth, 0)
def length(self, t0=0, t1=1, error=LENGTH_ERROR, min_depth=LENGTH_MIN_DEPTH)
The length of an elliptical large_arc segment requires numerical integration, and in that case it's simpler to just do a geometric approximation, as for cubic bezier curves.
4.027458
3.90178
1.03221
angle = radians(self.theta + t*self.delta) phi = radians(self.rotation) rx = self.radius.real ry = self.radius.imag k = (self.delta*2*pi/360)**n # ((d/dt)angle)**n if n % 4 == 0 and n > 0: return rx*cos(phi)*cos(angle) - ry*sin(phi)*sin(angle) + 1j*( rx*sin(phi)*cos(angle) + ry*cos(phi)*sin(angle)) elif n % 4 == 1: return k*(-rx*cos(phi)*sin(angle) - ry*sin(phi)*cos(angle) + 1j*( -rx*sin(phi)*sin(angle) + ry*cos(phi)*cos(angle))) elif n % 4 == 2: return k*(-rx*cos(phi)*cos(angle) + ry*sin(phi)*sin(angle) + 1j*( -rx*sin(phi)*cos(angle) - ry*cos(phi)*sin(angle))) elif n % 4 == 3: return k*(rx*cos(phi)*sin(angle) + ry*sin(phi)*cos(angle) + 1j*( rx*sin(phi)*sin(angle) - ry*cos(phi)*cos(angle))) else: raise ValueError("n should be a positive integer.")
def derivative(self, t, n=1)
returns the nth derivative of the segment at t.
2.013442
1.973687
1.020143
dseg = self.derivative(t) return dseg/abs(dseg)
def unit_tangent(self, t)
returns the unit tangent vector of the segment at t (centered at the origin and expressed as a complex number).
9.418225
5.324557
1.768828
return Arc(self.end, self.radius, self.rotation, self.large_arc, not self.sweep, self.start)
def reversed(self)
returns a copy of the Arc object with its orientation reversed.
6.354658
4.448014
1.428651
def _deg(rads, domain_lower_limit): # Convert rads to degrees in [0, 360) domain degs = degrees(rads % (2*pi)) # Convert to [domain_lower_limit, domain_lower_limit + 360) domain k = domain_lower_limit // 360 degs += k * 360 if degs < domain_lower_limit: degs += 360 return degs if self.delta > 0: degs = _deg(psi, domain_lower_limit=self.theta) else: degs = _deg(psi, domain_lower_limit=self.theta) return (degs - self.theta)/self.delta
def phase2t(self, psi)
Given phase -pi < psi <= pi, returns the t value such that exp(1j*psi) = self.u1transform(self.point(t)).
3.351231
3.2176
1.041531
# a(t) = radians(self.theta + self.delta*t) # = (2*pi/360)*(self.theta + self.delta*t) # x'=0: ~~~~~~~~~ # -rx*cos(phi)*sin(a(t)) = ry*sin(phi)*cos(a(t)) # -(rx/ry)*cot(phi)*tan(a(t)) = 1 # a(t) = arctan(-(ry/rx)tan(phi)) + pi*k === atan_x # y'=0: ~~~~~~~~~~ # rx*sin(phi)*sin(a(t)) = ry*cos(phi)*cos(a(t)) # (rx/ry)*tan(phi)*tan(a(t)) = 1 # a(t) = arctan((ry/rx)*cot(phi)) # atanres = arctan((ry/rx)*cot(phi)) === atan_y # ~~~~~~~~ # (2*pi/360)*(self.theta + self.delta*t) = atanres + pi*k # Therfore, for both x' and y', we have... # t = ((atan_{x/y} + pi*k)*(360/(2*pi)) - self.theta)/self.delta # for all k s.t. 0 < t < 1 from math import atan, tan if cos(self.phi) == 0: atan_x = pi/2 atan_y = 0 elif sin(self.phi) == 0: atan_x = 0 atan_y = pi/2 else: rx, ry = self.radius.real, self.radius.imag atan_x = atan(-(ry/rx)*tan(self.phi)) atan_y = atan((ry/rx)/tan(self.phi)) def angle_inv(ang, k): # inverse of angle from Arc.derivative() return ((ang + pi*k)*(360/(2*pi)) - self.theta)/self.delta xtrema = [self.start.real, self.end.real] ytrema = [self.start.imag, self.end.imag] for k in range(-4, 5): tx = angle_inv(atan_x, k) ty = angle_inv(atan_y, k) if 0 <= tx <= 1: xtrema.append(self.point(tx).real) if 0 <= ty <= 1: ytrema.append(self.point(ty).imag) xmin = max(xtrema) return min(xtrema), max(xtrema), min(ytrema), max(ytrema)
def bbox(self)
returns a bounding box for the segment in the form (xmin, xmax, ymin, ymax).
3.231119
3.15939
1.022703
if abs(self.delta*(t1 - t0)) <= 180: new_large_arc = 0 else: new_large_arc = 1 return Arc(self.point(t0), radius=self.radius, rotation=self.rotation, large_arc=new_large_arc, sweep=self.sweep, end=self.point(t1), autoscale_radius=self.autoscale_radius)
def cropped(self, t0, t1)
returns a cropped copy of this segment which starts at self.point(t0) and ends at self.point(t1).
3.563088
3.264268
1.091543
newpath = [seg.reversed() for seg in self] newpath.reverse() return Path(*newpath)
def reversed(self)
returns a copy of the Path object with its orientation reversed.
6.117809
4.362813
1.402263
return all(self[i].end == self[i+1].start for i in range(len(self) - 1))
def iscontinuous(self)
Checks if a path is continuous with respect to its parameterization.
4.486453
4.209183
1.065872
subpaths = [] subpath_start = 0 for i in range(len(self) - 1): if self[i].end != self[(i+1) % len(self)].start: subpaths.append(Path(*self[subpath_start: i+1])) subpath_start = i+1 subpaths.append(Path(*self[subpath_start: len(self)])) return subpaths
def continuous_subpaths(self)
Breaks self into its continuous components, returning a list of continuous subpaths. I.e. (all(subpath.iscontinuous() for subpath in self.continuous_subpaths()) and self == concatpaths(self.continuous_subpaths())) )
2.232546
2.073667
1.076618
assert len(self) != 0 assert self.iscontinuous() return self.start == self.end
def isclosed(self)
This function determines if a connected path is closed.
8.540316
8.233519
1.037262
mes = ("This attribute is deprecated, consider using isclosed() " "method instead.\n\nThis attribute is kept for compatibility " "with scripts created using svg.path (v2.0). You can prevent " "this warning in the future by setting " "CLOSED_WARNING_ON=False.") if warning_on: warn(mes) return self._closed and self._is_closable()
def closed(self, warning_on=CLOSED_WARNING_ON)
The closed attribute is deprecated, please use the isclosed() method instead. See _closed_warning for more information.
9.413644
7.809728
1.205374
if use_closed_attrib: self_closed = self.closed(warning_on=False) if self_closed: segments = self[:-1] else: segments = self[:] else: self_closed = False segments = self[:] current_pos = None parts = [] previous_segment = None end = self[-1].end for segment in segments: seg_start = segment.start # If the start of this segment does not coincide with the end of # the last segment or if this segment is actually the close point # of a closed path, then we should start a new subpath here. if current_pos != seg_start or \ (self_closed and seg_start == end and use_closed_attrib): parts.append('M {},{}'.format(seg_start.real, seg_start.imag)) if isinstance(segment, Line): args = segment.end.real, segment.end.imag parts.append('L {},{}'.format(*args)) elif isinstance(segment, CubicBezier): if useSandT and segment.is_smooth_from(previous_segment, warning_on=False): args = (segment.control2.real, segment.control2.imag, segment.end.real, segment.end.imag) parts.append('S {},{} {},{}'.format(*args)) else: args = (segment.control1.real, segment.control1.imag, segment.control2.real, segment.control2.imag, segment.end.real, segment.end.imag) parts.append('C {},{} {},{} {},{}'.format(*args)) elif isinstance(segment, QuadraticBezier): if useSandT and segment.is_smooth_from(previous_segment, warning_on=False): args = segment.end.real, segment.end.imag parts.append('T {},{}'.format(*args)) else: args = (segment.control.real, segment.control.imag, segment.end.real, segment.end.imag) parts.append('Q {},{} {},{}'.format(*args)) elif isinstance(segment, Arc): args = (segment.radius.real, segment.radius.imag, segment.rotation,int(segment.large_arc), int(segment.sweep),segment.end.real, segment.end.imag) parts.append('A {},{} {} {:d},{:d} {},{}'.format(*args)) current_pos = segment.end previous_segment = segment if self_closed: parts.append('Z') return ' '.join(parts)
def d(self, useSandT=False, use_closed_attrib=False)
Returns a path d-string for the path object. For an explanation of useSandT and use_closed_attrib, see the compatibility notes in the README.
2.044586
2.001348
1.021605
if T == 1: return len(self)-1, 1 if T == 0: return 0, 0 self._calc_lengths() # Find which segment self.point(T) falls on: T0 = 0 # the T-value the current segment starts on for seg_idx, seg_length in enumerate(self._lengths): T1 = T0 + seg_length # the T-value the current segment ends on if T1 >= T: # This is the segment! t = (T - T0)/seg_length return seg_idx, t T0 = T1 assert 0 <= T <= 1 raise BugException
def T2t(self, T)
returns the segment index, `seg_idx`, and segment parameter, `t`, corresponding to the path parameter `T`. In other words, this is the inverse of the `Path.t2T()` method.
4.517387
4.058393
1.113097
self._calc_lengths() # Accept an index or a segment for seg if isinstance(seg, int): seg_idx = seg else: try: seg_idx = self.index(seg) except ValueError: assert is_path_segment(seg) or isinstance(seg, int) raise segment_start = sum(self._lengths[:seg_idx]) segment_end = segment_start + self._lengths[seg_idx] T = (segment_end - segment_start)*t + segment_start return T
def t2T(self, seg, t)
returns the path parameter T which corresponds to the segment parameter t. In other words, for any Path object, path, and any segment in path, seg, T(t) = path.t2T(seg, t) is the unique reparameterization such that path.point(T(t)) == seg.point(t) for all 0 <= t <= 1. Input Note: seg can be a segment in the Path object or its corresponding index.
4.047261
3.541396
1.142843
seg_idx, t = self.T2t(T) seg = self._segments[seg_idx] return seg.derivative(t, n=n)/seg.length()**n
def derivative(self, T, n=1)
returns the tangent vector of the Path at T (centered at the origin and expressed as a complex number). Note: Bezier curves can have points where their derivative vanishes. If you are interested in the tangent direction, use unit_tangent() method instead.
6.684678
5.763305
1.159869
seg_idx, t = self.T2t(T) return self._segments[seg_idx].unit_tangent(t)
def unit_tangent(self, T)
returns the unit tangent vector of the Path at T (centered at the origin and expressed as a complex number). If the tangent vector's magnitude is zero, this method will find the limit of self.derivative(tau)/abs(self.derivative(tau)) as tau approaches T.
6.97909
6.746915
1.034412
seg_idx, t = self.T2t(T) seg = self[seg_idx] if np.isclose(t, 0) and (seg_idx != 0 or self.end==self.start): previous_seg_in_path = self._segments[ (seg_idx - 1) % len(self._segments)] if not seg.joins_smoothly_with(previous_seg_in_path): return float('inf') elif np.isclose(t, 1) and (seg_idx != len(self) - 1 or self.end == self.start): next_seg_in_path = self._segments[ (seg_idx + 1) % len(self._segments)] if not next_seg_in_path.joins_smoothly_with(seg): return float('inf') dz = self.derivative(T) ddz = self.derivative(T, n=2) dx, dy = dz.real, dz.imag ddx, ddy = ddz.real, ddz.imag return abs(dx*ddy - dy*ddx)/(dx*dx + dy*dy)**1.5
def curvature(self, T)
returns the curvature of this Path object at T and outputs float('inf') if not differentiable at T.
2.636595
2.500969
1.05423
def area_without_arcs(path): area_enclosed = 0 for seg in path: x = real(seg.poly()) dy = imag(seg.poly()).deriv() integrand = x*dy integral = integrand.integ() area_enclosed += integral(1) - integral(0) return area_enclosed def seg2lines(seg): num_lines = int(ceil(seg.length() / chord_length)) pts = [seg.point(t) for t in np.linspace(0, 1, num_lines+1)] return [Line(pts[i], pts[i+1]) for i in range(num_lines)] assert self.isclosed() bezier_path_approximation = [] for seg in self: if isinstance(seg, Arc): bezier_path_approximation += seg2lines(seg) else: bezier_path_approximation.append(seg) return area_without_arcs(Path(*bezier_path_approximation))
def area(self, chord_length=1e-4)
Find area enclosed by path. Approximates any Arc segments in the Path with lines approximately `chord_length` long, and returns the area enclosed by the approximated Path. Default chord length is 0.01. If Arc segments are included in path, to ensure accurate results, make sure this `chord_length` is set to a reasonable value (e.g. by checking curvature). Notes ----- * Negative area results from clockwise (as opposed to counter-clockwise) parameterization of the input Path. To Contributors --------------- This is one of many parts of `svgpathtools` that could be improved by a noble soul implementing a piecewise-linear approximation scheme for paths (one with controls to guarantee a desired accuracy).
3.723205
3.365097
1.106418
path1 = self if isinstance(other_curve, Path): path2 = other_curve else: path2 = Path(other_curve) assert path1 != path2 intersection_list = [] for seg1 in path1: for seg2 in path2: if justonemode and intersection_list: return intersection_list[0] for t1, t2 in seg1.intersect(seg2, tol=tol): T1 = path1.t2T(seg1, t1) T2 = path2.t2T(seg2, t2) intersection_list.append(((T1, seg1, t1), (T2, seg2, t2))) if justonemode and intersection_list: return intersection_list[0] # Note: If the intersection takes place at a joint (point one seg ends # and next begins in path) then intersection_list may contain a # redundant intersection. This code block checks for and removes said # redundancies. if intersection_list: pts = [seg1.point(_t1) for _T1, _seg1, _t1 in list(zip(*intersection_list))[0]] indices2remove = [] for ind1 in range(len(pts)): for ind2 in range(ind1 + 1, len(pts)): if abs(pts[ind1] - pts[ind2]) < tol: # then there's a redundancy. Remove it. indices2remove.append(ind2) intersection_list = [inter for ind, inter in enumerate(intersection_list) if ind not in indices2remove] return intersection_list
def intersect(self, other_curve, justonemode=False, tol=1e-12)
returns list of pairs of pairs ((T1, seg1, t1), (T2, seg2, t2)) giving the intersection points. If justonemode==True, then returns just the first intersection found. tol is used to check for redundant intersections (see comment above the code block where tol is used). Note: If the two path objects coincide for more than a finite set of points, this code will fail.
3.199713
2.846908
1.123926
bbs = [seg.bbox() for seg in self._segments] xmins, xmaxs, ymins, ymaxs = list(zip(*bbs)) xmin = min(xmins) xmax = max(xmaxs) ymin = min(ymins) ymax = max(ymaxs) return xmin, xmax, ymin, ymax
def bbox(self)
returns a bounding box for the input Path object in the form (xmin, xmax, ymin, ymax).
2.167362
1.941472
1.11635
assert 0 <= T0 <= 1 and 0 <= T1<= 1 assert T0 != T1 assert not (T0 == 1 and T1 == 0) if T0 == 1 and 0 < T1 < 1 and self.isclosed(): return self.cropped(0, T1) if T1 == 1: seg1 = self[-1] t_seg1 = 1 i1 = len(self) - 1 else: seg1_idx, t_seg1 = self.T2t(T1) seg1 = self[seg1_idx] if np.isclose(t_seg1, 0): i1 = (self.index(seg1) - 1) % len(self) seg1 = self[i1] t_seg1 = 1 else: i1 = self.index(seg1) if T0 == 0: seg0 = self[0] t_seg0 = 0 i0 = 0 else: seg0_idx, t_seg0 = self.T2t(T0) seg0 = self[seg0_idx] if np.isclose(t_seg0, 1): i0 = (self.index(seg0) + 1) % len(self) seg0 = self[i0] t_seg0 = 0 else: i0 = self.index(seg0) if T0 < T1 and i0 == i1: new_path = Path(seg0.cropped(t_seg0, t_seg1)) else: new_path = Path(seg0.cropped(t_seg0, 1)) # T1<T0 must cross discontinuity case if T1 < T0: if not self.isclosed(): raise ValueError("This path is not closed, thus T0 must " "be less than T1.") else: for i in range(i0 + 1, len(self)): new_path.append(self[i]) for i in range(0, i1): new_path.append(self[i]) # T0<T1 straight-forward case else: for i in range(i0 + 1, i1): new_path.append(self[i]) if t_seg1 != 0: new_path.append(seg1.cropped(0, t_seg1)) return new_path
def cropped(self, T0, T1)
returns a cropped copy of the path.
2.090693
2.018829
1.035597
if return_all_global_extrema: raise NotImplementedError else: global_min = (np.inf, None, None) global_max = (0, None, None) for seg_idx, seg in enumerate(self): seg_global_min, seg_global_max = seg.radialrange(origin) if seg_global_min[0] < global_min[0]: global_min = seg_global_min + (seg_idx,) if seg_global_max[0] > global_max[0]: global_max = seg_global_max + (seg_idx,) return global_min, global_max
def radialrange(self, origin, return_all_global_extrema=False)
returns the tuples (d_min, t_min, idx_min), (d_max, t_max, idx_max) which minimize and maximize, respectively, the distance d = |self[idx].point(t)-origin|.
1.826067
1.670842
1.092902
t1 = 1-t return [n_choose_k(n, k) * t1**(n-k) * t**k for k in range(n+1)]
def bernstein(n, t)
returns a list of the Bernstein basis polynomials b_{i, n} evaluated at t, for i =0...n
3.713827
3.978735
0.933419