-
Notifications
You must be signed in to change notification settings - Fork 17
/
kernel.py
1284 lines (1202 loc) · 57.7 KB
/
kernel.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#!/usr/bin/env python3
#
# Copyright (c) Bo Peng and the University of Texas MD Anderson Cancer Center
# Distributed under the terms of the 3-clause BSD License.
import asyncio
import contextlib
import inspect
import logging
import os
import pprint
import subprocess
import sys
import threading
import time
from collections import defaultdict
from textwrap import dedent
import comm
import pandas as pd
import pkg_resources
from ipykernel._version import version_info as ipykernel_version_info
from ipykernel.ipkernel import IPythonKernel
from IPython.utils.tokenutil import line_at_cursor, token_at_cursor
from jupyter_client import manager
from sos._version import __sos_version__, __version__
from sos.eval import SoS_eval, interpolate
from sos.executor_utils import prepare_env
from sos.syntax import SOS_DIRECTIVE, SOS_SECTION_HEADER
from sos.utils import env, load_config_files, short_repr
from ._version import __version__ as __notebook_version__
from .comm_manager import SoSCommManager
from .completer import SoS_Completer
from .inspector import SoS_Inspector
from .magics import SoS_Magics
from .workflow_executor import (NotebookLoggingHandler, execute_scratch_cell, run_sos_workflow, start_controller)
from .subkernel import Subkernels
class FlushableStringIO:
def __init__(self, kernel, name, *args, **kwargs):
self.kernel = kernel
self.name = name
def write(self, content):
if content.startswith('HINT: '):
content = content.splitlines()
hint_line = content[0][6:].strip()
content = '\n'.join(content[1:])
self.kernel.send_response(self.kernel.iopub_socket, 'display_data', {
'metadata': {},
'data': {
'text/html': f'<div class="sos_hint">{hint_line}</div>'
}
})
if content:
if self.kernel._meta['capture_result'] is not None:
self.kernel._meta['capture_result'].append(('stream', {'name': self.name, 'text': content}))
if self.kernel._meta['render_result'] is False:
self.kernel.send_response(self.kernel.iopub_socket, 'stream', {'name': self.name, 'text': content})
def flush(self):
pass
__all__ = ['SoS_Kernel']
# translate a message to transient_display_data message
def make_transient_msg(msg_type, content):
if msg_type == 'display_data':
return {'data': content.get('data', {}), 'metadata': content.get('metadata', {})}
if msg_type == 'stream':
if content['name'] == 'stdout':
return {
'data': {
'text/plain': content['text'],
'application/vnd.jupyter.stdout': content['text']
},
'metadata': {}
}
return {
'data': {
'text/plain': content['text'],
'application/vnd.jupyter.stderr': content['text']
},
'metadata': {}
}
raise ValueError(f"failed to translate message {msg_type} to transient_display_data message")
class SoS_Kernel(IPythonKernel):
implementation = 'SOS'
implementation_version = __version__
language = 'sos'
language_version = __sos_version__
language_info = {
'mimetype': 'text/x-sos',
'name': 'sos',
'file_extension': '.sos',
'pygments_lexer': 'sos',
'codemirror_mode': 'sos',
'nbconvert_exporter': 'sos_notebook.converter.SoS_Exporter',
}
banner = "SoS kernel - script of scripts"
def get_supported_languages(self):
if self._supported_languages is not None:
return self._supported_languages
group = 'sos_languages'
self._supported_languages = {}
for entrypoint in pkg_resources.iter_entry_points(group=group):
# Grab the function that is the actual plugin.
name = entrypoint.name
env.log_to_file('KERNEL', f'Found registered language {name}')
try:
plugin = entrypoint.load()
self._supported_languages[name] = plugin
except Exception as e:
env.log_to_file('KERNEL', f'Failed to load registered language {name}: {e}')
self._failed_languages[name] = e
return self._supported_languages
supported_languages = property(lambda self: self.get_supported_languages())
def get_kernel_list(self):
if not hasattr(self, '_subkernels'):
self._subkernels = Subkernels(self)
# sort kernel list by name to avoid unnecessary change of .ipynb files
return self._subkernels
subkernels = property(lambda self: self.get_kernel_list())
def get_completer(self):
if self._completer is None:
self._completer = SoS_Completer(self)
return self._completer
completer = property(lambda self: self.get_completer())
def get_inspector(self):
if self._inspector is None:
self._inspector = SoS_Inspector(self)
return self._inspector
inspector = property(lambda self: self.get_inspector())
def __init__(self, **kwargs):
super().__init__(**kwargs)
self.options = ''
self.kernel = 'SoS'
# a dictionary of started kernels, with the format of
#
# 'R': ['ir', 'sos.R.sos_R', '#FFEEAABB']
#
# Note that:
#
# 'R' is the displayed name of the kernel.
# 'ir' is the kernel name.
# 'sos.R.sos_R' is the language module.
# '#FFEEAABB' is the background color
#
env.log_to_file('KERNEL', f'Starting SoS Kernel version {__notebook_version__} with SoS {__version__}')
self.kernels = {}
# self.shell = InteractiveShell.instance()
self.format_obj = self.shell.display_formatter.format
self._meta = {
'workflow': '',
'workflow_mode': False,
'render_result': False,
'capture_result': None,
'cell_id': '0',
'notebook_name': '',
'notebook_path': '',
'use_panel': True,
'use_iopub': False,
'default_kernel': 'SoS',
'cell_kernel': 'SoS',
'batch_mode': False
}
self._debug_mode = False
self._supported_languages = None
self._completer = None
self._inspector = None
self._real_execution_count = 1
self._execution_count = 1
self.frontend_comm = None
self.frontend_comm_cache = []
#
self.comm_manager = comm.get_comm_manager()
# remove the old comm_manager
self.shell.configurables.pop()
self.shell.configurables.append(self.comm_manager)
for msg_type in ['comm_open', 'comm_msg', 'comm_close']:
self.shell_handlers[msg_type] = getattr(self.comm_manager, msg_type)
self.comm_manager.register_target('sos_comm', self.sos_comm)
self.my_tasks = {}
self.magics = SoS_Magics(self)
self._failed_languages = {}
# enable matplotlib by default #77
self.shell.enable_gui = lambda gui: None
self.editor_kernel = 'sos'
# initialize env
prepare_env('')
self.original_keys = set(env.sos_dict._dict.keys()) | {
'SOS_VERSION', 'CONFIG', 'step_name', '__builtins__', 'input', 'output', 'depends'
}
env.logger.handlers = [x for x in env.logger.handlers if not isinstance(x, logging.StreamHandler)]
env.logger.addHandler(
NotebookLoggingHandler(
{
0: logging.ERROR,
1: logging.WARNING,
2: logging.INFO,
3: logging.DEBUG,
4: logging.DEBUG,
None: logging.INFO
}[env.verbosity],
kernel=self))
env.logger.print = lambda cell_id, msg, *args: \
self.send_response(self.iopub_socket, 'stream', {'name': 'stdout', 'text': msg}) if self._meta['batch_mode'] else self.send_frontend_msg('print', [cell_id, msg])
self.controller = None
cell_id = property(lambda self: self._meta['cell_id'])
_workflow_mode = property(lambda self: self._meta['workflow_mode'])
def sos_comm(self, comm, msg):
# record frontend_comm to send messages
self.frontend_comm = comm
@comm.on_msg
def handle_frontend_msg(msg):
content = msg['content']['data']
# log_to_file(msg)
for k, v in content.items():
if k == 'list-kernel':
if v:
self.subkernels.update(v)
self.subkernels.notify_frontend()
elif k == 'set-editor-kernel':
self.editor_kernel = v
elif k == 'cancel-workflow':
from .workflow_executor import cancel_workflow
cancel_workflow(v[0], self)
elif k == 'execute-workflow':
from .workflow_executor import execute_pending_workflow
execute_pending_workflow(v, self)
elif k == 'update-task-status':
if not isinstance(v, list):
env.log_to_file('KERNEL', f'Failed to parse message for update-task-status {v}')
continue
# split by host ...
host_status = defaultdict(list)
for name in v:
try:
tqu, tid, _ = name.rsplit('_', 2)
except Exception as e:
env.log_to_file('KERNEL', f'Failed to parse task ID {name}: {e}')
# incorrect ID...
continue
host_status[tqu].append(tid)
# log_to_file(host_status)
#
from sos.hosts import Host
for tqu, tids in host_status.items():
try:
h = Host(tqu, start_engine=True)
except Exception as e:
env.log_to_file('KERNEL', f'Failed to connect to host {tqu}: {e}')
for tid in tids:
self.send_frontend_msg('task_status', {
'task_id': tid,
'queue': tqu,
'status': 'missing',
'duration': ''
})
continue
for tid, tst, tdt in h._task_engine.monitor_tasks(tids):
self.send_frontend_msg('task_status', {
'task_id': tid,
'queue': tqu,
'status': tst,
'duration': tdt
})
self.send_frontend_msg('update-duration', {})
elif k == 'paste-table':
try:
from tabulate import tabulate
df = pd.read_clipboard()
tbl = tabulate(df, headers='keys', tablefmt='pipe')
self.send_frontend_msg('paste-table', tbl)
except Exception as e:
self.send_frontend_msg('alert', f'Failed to paste clipboard as table: {e}')
elif k == 'notebook-version':
# send the version of notebook, right now we will not do anything to it, but
# we will send the version of sos-notebook there
self.send_frontend_msg('notebook-version', __notebook_version__)
else:
# this somehow does not work
self.warn(f'Unknown message {k}: {v}')
def notify_error(self, e):
msg = {
'status': 'error',
'ename': e.__class__.__name__,
'evalue': str(e),
'traceback': [f'\033[91m{e}\033[0m'],
'execution_count': self._execution_count,
}
if self._meta['suppress_error']:
self.send_response(self.iopub_socket, 'stream', {
'name': 'stderr',
'text': f"{msg['ename']}: {msg['evalue']}"
})
else:
self.send_response(self.iopub_socket, 'error', msg)
return msg
def send_frontend_msg(self, msg_type, msg=None):
# if comm is never created by frontend, the kernel is in test mode without frontend
if msg_type in ('display_data', 'stream'):
if self._meta['use_panel'] is False or self._meta['cell_id'] == -1:
self.send_response(self.iopub_socket, msg_type, {} if msg is None else msg)
elif self._meta['use_iopub']:
self.send_response(self.iopub_socket, 'transient_display_data', make_transient_msg(msg_type, msg))
elif self.frontend_comm:
if self.frontend_comm_cache:
for mt, mg in self.frontend_comm_cache:
self.frontend_comm.send(make_transient_msg(mt, mg), {'msg_type': 'transient_display_data'})
self.frontend_comm_cache = []
self.frontend_comm.send(make_transient_msg(msg_type, msg), {'msg_type': 'transient_display_data'})
elif self._meta['batch_mode']:
env.log_to_file('MESSAGE', f'frontend message of type {msg_type} is sent in batch mode.')
else:
self.frontend_comm_cache.append([msg_type, msg])
env.log_to_file('MESSAGE', f'fronten not ready or broken. Message of type {msg_type} is cached')
elif self.frontend_comm:
if self.frontend_comm_cache:
for mt, mg in self.frontend_comm_cache:
self.frontend_comm.send({} if mg is None else mg, {'msg_type': mt})
self.frontend_comm_cache = []
self.frontend_comm.send({} if msg is None else msg, {'msg_type': msg_type})
elif self._meta['batch_mode']:
env.log_to_file('MESSAGE', f'frontend message of type {msg_type} is sent in batch mode.')
else:
# frontend_comm is not ready
self.frontend_comm_cache.append([msg_type, msg])
env.log_to_file('MESSAGE', f'fronten not ready or broken. Message of type {msg_type} is cached')
@contextlib.contextmanager
def redirect_sos_io(self):
save_stdout = sys.stdout
save_stderr = sys.stderr
sys.stdout = FlushableStringIO(self, 'stdout')
sys.stderr = FlushableStringIO(self, 'stderr')
yield
sys.stdout = save_stdout
sys.stderr = save_stderr
async def get_vars_from(self, items, from_kernel=None, explicit=False, as_var=None):
if as_var is not None:
if not isinstance(as_var, str):
self.warn('Option --as should be a string.')
return
if len(items) > 1:
self.warn('Only one expression is allowed when option --as is used')
return
if from_kernel is None or from_kernel.lower() == 'sos':
# Feature removed #253
# autmatically get all variables with names start with 'sos'
# default_items = [
# x for x in env.sos_dict.keys()
# if x.startswith('sos') and x not in self.original_keys
# ]
if not items:
return
for item in items:
if item not in env.sos_dict:
self.warn(f'Variable {item} does not exist')
return
kinfo = self.subkernels.find(self.kernel)
if kinfo.language in self.supported_languages:
lan = self.supported_languages[kinfo.language]
try:
get_vars_func = lan(self, kinfo.kernel).get_vars
args = inspect.getfullargspec(get_vars_func).args
if 'as_var' in args:
await get_vars_func(items, as_var=as_var)
else:
if as_var is not None:
self.warn(f'Subkernel {kinfo.language} does not support option --as')
await get_vars_func(items)
except Exception as e:
self.warn(f'Failed to get variable: {e}\n')
return
elif self.kernel == 'SoS':
self.warn('Magic %get without option --kernel can only be executed by subkernels')
return
else:
if explicit:
self.warn(
f'Magic %get failed because the language module for {self.kernel} is not properly installed. Please install it according to language specific instructions on the Running SoS section of the SoS homepage and restart Jupyter server.'
)
return
elif self.kernel.lower() == 'sos':
# if another kernel is specified and the current kernel is sos
# we get from subkernel
try:
await self.switch_kernel(from_kernel)
await self.put_vars_to(items, as_var=as_var)
except Exception as e:
self.warn(f'Failed to get {", ".join(items)} from {from_kernel}: {e}')
finally:
await self.switch_kernel('SoS')
else:
# if another kernel is specified, we should try to let that kernel pass
# the variables to this one directly
try:
my_kernel = self.kernel
await self.switch_kernel(from_kernel)
# put stuff to sos or my_kernel directly
await self.put_vars_to(items, to_kernel=my_kernel, explicit=explicit, as_var=as_var)
except Exception as e:
self.warn(f'Failed to get {", ".join(items)} from {from_kernel}: {e}')
finally:
# then switch back
await self.switch_kernel(my_kernel)
async def put_vars_to(self, items, to_kernel=None, explicit=False, as_var=None):
if not items:
return
if as_var is not None:
if not isinstance(as_var, str):
self.warn('Option --as should be a string.')
return
if len(items) > 1:
self.warn('Only one expression is allowed when option --as is used')
return
if self.kernel.lower() == 'sos':
if to_kernel is None:
self.warn('Magic %put without option --kernel can only be executed by subkernels')
return
# if another kernel is specified and the current kernel is sos
try:
# switch to kernel and bring in items
await self.switch_kernel(to_kernel, in_vars=items, as_var=as_var)
except Exception as e:
self.warn(f'Failed to put {", ".join(items)} to {to_kernel}: {e}')
finally:
# switch back
await self.switch_kernel('SoS')
else:
# put to sos kernel or another kernel
kinfo = self.subkernels.find(self.kernel)
if kinfo.language not in self.supported_languages:
if explicit:
self.warn(f'Subkernel {self.kernel} does not support magic %put.')
return
#
lan = self.supported_languages[kinfo.language]
# pass language name to to_kernel
try:
put_vars_func = lan(self, kinfo.kernel).put_vars
args = inspect.getfullargspec(put_vars_func).args
to_kernel_name = self.subkernels.find(to_kernel).language if to_kernel else 'SoS'
if 'as_var' in args:
objects = put_vars_func(items, to_kernel=to_kernel_name, as_var=as_var)
else:
objects = put_vars_func(items, to_kernel=to_kernel_name)
except Exception as e:
# if somethign goes wrong in the subkernel does not matter
env.log_to_file('MAGIC', f'Failed to call put_var({items}) from {kinfo.kernel}: {e}')
objects = {}
if isinstance(objects, dict):
# returns a SOS dictionary
try:
# if the variable is passing through SoS, let us try to restore variables in SoS
if to_kernel is not None:
missing_vars = [x for x in objects.keys() if x not in env.sos_dict]
existing_vars = {x: env.sos_dict[x] for x in objects.keys() if x in env.sos_dict}
env.sos_dict.update(objects)
except Exception as e:
self.warn(f'Failed to put {", ".join(items)} to {to_kernel}: {e}')
return
if to_kernel is None:
return
# if another kernel is specified and the current kernel is not sos
# we need to first put to sos then to another kernel
my_kernel = self.kernel
try:
# switch to the destination kernel and bring in vars
await self.switch_kernel(to_kernel, in_vars=[as_var] if as_var else items)
except Exception as e:
self.warn(f'Failed to put {", ".join(items)} to {to_kernel}: {e}')
finally:
# switch back to the original kernel
await self.switch_kernel(my_kernel)
# restore sos_dict to avoid bypassing effect #252
for missing_var in missing_vars:
env.sos_dict.pop(missing_var)
env.sos_dict.update(existing_vars)
elif isinstance(objects, str):
# an statement that will be executed in the destination kernel
if to_kernel is None or to_kernel == 'SoS':
# evaluate in SoS, this should not happen or rarely happen
# because the subkernel should return a dictionary for SoS kernel
try:
exec(objects, env.sos_dict._dict)
except Exception as e:
self.warn(f'Failed to put variables {items} to SoS kernel: {e}')
return
try:
my_kernel = self.kernel
# switch to the destination kernel
await self.switch_kernel(to_kernel)
# execute the statement to pass variables directly to destination kernel
await self.run_cell(objects, True, False)
except Exception as e:
self.warn(f'Failed to put {", ".join(items)} to {to_kernel}: {e}')
finally:
# switch back to the original kernel
await self.switch_kernel(my_kernel)
else:
self.warn(f'Unrecognized return value of type {object.__class__.__name__} for action %put')
async def expand_text_in(self, text, sigil=None, kernel='SoS'):
'''
Expand a piece of (markdown) text in specified kernel, used by
magic %expand
'''
if not text:
return ''
if sigil is None:
sigil = '{ }'
if sigil.count(' ') != 1:
raise ValueError(f'Invalid interpolation delimiter "{sigil}": should be in the format of "L R"')
if sigil.split(' ')[0] not in text:
return text
if not kernel or kernel.lower() == 'sos':
if sigil != '{ }':
from sos.parser import replace_sigil
text = replace_sigil(text, sigil)
return interpolate(text, local_dict=env.sos_dict._dict)
# check if the language supports expand protocol
kinfo = self.subkernels.find(kernel)
if kinfo.language not in self.supported_languages:
self.warn(f'Subkernel {kernel} does not support magic %expand --in')
return text
lan = self.supported_languages[kinfo.language](self, kinfo.kernel)
if not hasattr(lan, 'expand'):
self.warn(f'Subkernel {kernel} does not support magic %expand --in')
return text
orig_kernel = self.kernel
try:
await self.switch_kernel(kernel)
return lan.expand(text, sigil)
except Exception as e:
self.warn(f'Failed to expand {text} with sigin {sigil} in kernel {kernel}: {e}')
return text
finally:
await self.switch_kernel(orig_kernel)
def do_is_complete(self, code):
'''check if new line is in order'''
code = code.strip()
if not code:
return {'status': 'complete', 'indent': ''}
env.log_to_file('MESSAGE', f'Checking is_complete of "{code}"')
lines = code.split('\n')
# first let us remove "valid" magics
while any(line.startswith('%') or line.startswith('!') for line in lines):
for idx, line in enumerate(lines):
if line.startswith('%') or line.startswith('!'):
# if the last line ending with \, incomplete
if line.endswith('\\'):
if idx == len(lines) - 1:
return {'status': 'incomplete', 'indent': ''}
lines[idx] = lines[idx][:-1] + lines[idx + 1]
lines[idx + 1] = ''
else:
# valid, complete, ignore
lines[idx] = ''
if self.kernel == 'SoS':
for idx, line in enumerate(lines):
# remove header
if SOS_SECTION_HEADER.match(line):
lines[idx] = ''
# remove input stuff?
if SOS_DIRECTIVE.match(line):
if any(line.startswith(x) for x in ('input:', 'output:', 'depends:', 'parameter:')):
# directive, remvoe them
lines[idx] = lines[idx].split(':', 1)[-1]
elif idx == len(lines) - 1:
# sh: with no script, incomplete
return {'status': 'incomplete', 'indent': ' '}
else:
# remove the rest of them because they are embedded script
for i in range(idx, len(lines)):
lines[i] = ''
# check the rest if it is ok
try:
from IPython.core.inputtransformer2 import \
TransformerManager as ipf
except ImportError:
from IPython.core.inputsplitter import InputSplitter as ipf
code = '\n'.join(lines) + '\n\n'
res = ipf().check_complete(code)
env.log_to_file('MESSAGE', f'SoS kernel returns {res} for code {code}')
return {'status': res[0], 'indent': res[1]}
# non-SoS kernels
try:
cell_kernel = self.subkernels.find(self.editor_kernel)
if cell_kernel.name not in self.kernels:
orig_kernel = self.kernel
try:
# switch to start the new kernel
asyncio.run(self.switch_kernel(cell_kernel.name))
finally:
asyncio.run(self.switch_kernel(orig_kernel))
KC = self.kernels[cell_kernel.name][1]
# clear the shell channel
while KC.shell_channel.msg_ready():
KC.shell_channel.get_msg()
code = '\n'.join(lines)
KC.is_complete(code)
msg = KC.shell_channel.get_msg()
if msg['header']['msg_type'] == 'is_complete_reply':
env.log_to_file('MESSAGE', f'{self.kernel} kernel returns {msg["content"]} for code {code}')
return msg['content']
raise RuntimeError(
f"is_complete_reply not obtained: {msg['header']['msg_type']} {msg['content']} returned instead")
except Exception as e:
env.logger.debug(f'Completion fail with exception: {e}')
return {'status': 'incomplete', 'indent': ''}
def do_inspect(self, code, cursor_pos, detail_level=0):
if self.editor_kernel.lower() == 'sos':
line, offset = line_at_cursor(code, cursor_pos)
name = token_at_cursor(code, cursor_pos)
data = self.inspector.inspect(name, line, cursor_pos - offset)
return {'status': 'ok', 'metadata': {}, 'found': True if data else False, 'data': data}
cell_kernel = self.subkernels.find(self.editor_kernel)
try:
_, KC = self.kernels[cell_kernel.name]
except Exception as e:
env.log_to_file('KERNEL', f'Failed to get subkernels {cell_kernel.name}: {e}')
KC = self.KC
try:
KC.inspect(code, cursor_pos)
while KC.shell_channel.msg_ready():
msg = KC.shell_channel.get_msg()
if msg['header']['msg_type'] == 'inspect_reply':
return msg['content']
# other messages, do not know what is going on but
# we should not wait forever and cause a deadloop here
env.log_to_file(
'MESSAGE',
f"complete_reply not obtained: {msg['header']['msg_type']} {msg['content']} returned instead")
break
except Exception as e:
env.log_to_file('KERNEL', f'Completion fail with exception: {e}')
async def do_complete(self, code, cursor_pos):
if self.editor_kernel.lower() == 'sos':
text, matches = self.completer.complete_text(code, cursor_pos)
return {
'matches': matches,
'cursor_end': cursor_pos,
'cursor_start': cursor_pos - len(text),
'metadata': {},
'status': 'ok'
}
try:
cell_kernel = self.subkernels.find(self.editor_kernel)
if cell_kernel.name not in self.kernels:
orig_kernel = self.kernel
try:
# switch to start the new kernel
await self.switch_kernel(cell_kernel.name)
finally:
await self.switch_kernel(orig_kernel)
KC = self.kernels[cell_kernel.name][1]
# clear the shell channel
while KC.shell_channel.msg_ready():
KC.shell_channel.get_msg()
KC.complete(code, cursor_pos)
msg = KC.shell_channel.get_msg()
if msg['header']['msg_type'] == 'complete_reply':
return msg['content']
raise RuntimeError(
f"complete_reply not obtained: {msg['header']['msg_type']} {msg['content']} returned instead")
except Exception as e:
env.logger.debug(f'Completion fail with exception: {e}')
return {
'matches': [],
'cursor_end': cursor_pos,
'cursor_start': cursor_pos,
'metadata': {},
'status': 'error'
}
def warn(self, message):
message = str(message).rstrip() + '\n'
if message.strip():
self.send_response(self.iopub_socket, 'stream', {'name': 'stderr', 'text': message})
async def run_cell(self, code, silent, store_history, on_error=None):
#
if not self.KM.is_alive():
self.send_response(self.iopub_socket, 'stream',
dict(name='stdout', text='Restarting kernel "{}"\n'.format(self.kernel)))
self.KM.restart_kernel(now=False)
self.KC = self.KM.client()
# flush stale replies, which could have been ignored, due to missed heartbeats
while self.KC.shell_channel.msg_ready():
self.KC.shell_channel.get_msg()
# executing code in another kernel.
# https://github.com/ipython/ipykernel/blob/604ee892623cca29eb495933eb5aa26bd166c7ff/ipykernel/inprocess/client.py#L94
content = dict(code=code, silent=silent, store_history=store_history, user_expressions={}, allow_stdin=False)
msg = self.KC.session.msg('execute_request', content)
# use the msg_id of the sos kernel for the subkernel to make sure that the messages sent
# from the subkernel has the correct msg_id in parent_header so that they can be
# displayed directly in the notebook (without using self._parent_header
if ipykernel_version_info[0] >= 6:
msg['msg_id'] = self.get_parent()['header']['msg_id']
else:
msg['msg_id'] = self._parent_header['header']['msg_id']
msg['header']['msg_id'] = msg['msg_id']
self.KC.shell_channel.send(msg)
# first thing is wait for any side effects (output, stdin, etc.)
iopub_started = False
iopub_ended = False
shell_ended = False
res = None
while not (iopub_started and iopub_ended and shell_ended):
try:
# display intermediate print statements, etc.
while self.KC.stdin_channel.msg_ready():
sub_msg = self.KC.stdin_channel.get_msg()
env.log_to_file('MESSAGE',
f"MSG TYPE {sub_msg['header']['msg_type']} CONTENT\n {pprint.pformat(sub_msg)}")
if sub_msg['header']['msg_type'] != 'input_request':
self.session.send(self.stdin_socket, sub_msg)
else:
content = sub_msg["content"]
if content['password']:
res = self.getpass(prompt=content['prompt'])
else:
res = self.raw_input(prompt=content['prompt'])
self.KC.input(res)
while self.KC.iopub_channel.msg_ready():
sub_msg = self.KC.iopub_channel.get_msg()
msg_type = sub_msg['header']['msg_type']
env.log_to_file(
'MESSAGE',
f"IOPUB MSG TYPE {sub_msg['header']['msg_type']} CONTENT \n {pprint.pformat(sub_msg)}")
if msg_type == 'status':
if sub_msg["content"]["execution_state"] == 'busy':
iopub_started = True
elif iopub_started and sub_msg["content"]["execution_state"] == 'idle':
iopub_ended = True
continue
if msg_type in ('execute_input', 'execute_result'):
# override execution count with the master count,
# not sure if it is needed
sub_msg['content']['execution_count'] = self._execution_count
#
if msg_type in ['display_data', 'stream', 'execute_result', 'update_display_data', 'error']:
if self._meta['capture_result'] is not None:
self._meta['capture_result'].append((msg_type, sub_msg['content']))
if msg_type == 'execute_result' or (not silent and self._meta['render_result'] is False):
if msg_type == 'error' and self._meta['suppress_error']:
self.send_response(
self.iopub_socket, 'stream', {
'name': 'stderr',
'text': f"{sub_msg['content']['ename']}: {sub_msg['content']['evalue']}"
})
else:
self.session.send(self.iopub_socket, sub_msg)
else:
# if the subkernel tried to create a customized comm
if msg_type == 'comm_open':
self.comm_manager.register_subcomm(sub_msg['content']['comm_id'], self.KC, self)
self.session.send(self.iopub_socket, sub_msg)
if self.KC.shell_channel.msg_ready():
# now get the real result
reply = self.KC.get_shell_msg()
reply['content']['execution_count'] = self._execution_count
env.log_to_file('MESSAGE', f'GET SHELL MSG {pprint.pformat(reply)}')
res = reply['content']
shell_ended = True
time.sleep(0.001)
except KeyboardInterrupt:
self.KM.interrupt_kernel()
return res
def get_info_of_subkernels(self):
from jupyter_client.kernelspec import KernelSpecManager
km = KernelSpecManager()
available_subkernels = '''<table>
<tr>
<th>Subkernel</th>
<th>Kernel Name</th>
<th>Language</th>
<th>Language Module</th>
<th style="text-align:left">Interpreter</th>
</tr>'''
for sk in self.subkernels.kernel_list():
spec = km.get_kernel_spec(sk.kernel)
if sk.name in ('SoS', 'Markdown'):
lan_module = ''
elif sk.language in self.supported_languages:
lan_module = f"<code>{self.supported_languages[sk.language].__module__.split('.')[0]}</code>"
else:
lan_module = '<font style="color:red">Unavailable</font>'
available_subkernels += f'''\
<tr>
<td>{sk.name}</td>
<td><code>{sk.kernel}</code></td>
<td>{spec.language}</td>
<td>{lan_module}</td>
<td style="text-align:left"><code>{spec.argv[0]}</code></td>
</tr>'''
available_subkernels += '</table>'
return available_subkernels
async def switch_kernel(self, kernel, in_vars=None, kernel_name=None, language=None, color=None, as_var=None):
# switching to a non-sos kernel
if not kernel:
kinfo = self.subkernels.find(self.kernel)
self.send_response(self.iopub_socket, 'display_data',
dict(metadata={}, data={'text/html': self.get_info_of_subkernels()}))
return
kinfo = self.subkernels.find(kernel, kernel_name, language, color)
if kinfo.name == self.kernel:
return
if kinfo.name == 'SoS':
# non-SoS to SoS
if in_vars:
await self.put_vars_to(in_vars, as_var=as_var)
self.kernel = 'SoS'
elif self.kernel != 'SoS':
# Non-SoS to Non-SoS
await self.switch_kernel('SoS', in_vars)
await self.switch_kernel(kinfo.name, in_vars)
else:
# SoS to non-SoS
env.log_to_file('KERNEL', f'Switch from {self.kernel} to {kinfo.name}')
# case when self.kernel == 'sos', kernel != 'sos'
# to a subkernel
new_kernel = False
if kinfo.name not in self.kernels:
# start a new kernel
try:
env.log_to_file('KERNEL', f'Starting subkernel {kinfo.name}')
self.kernels[kinfo.name] = manager.start_new_kernel(
startup_timeout=30, kernel_name=kinfo.kernel, cwd=os.getcwd())
new_kernel = True
except Exception as e:
env.log_to_file('KERNEL', f'Failed to start kernel {kinfo.kernel}. Trying again...')
# try toget error message
import tempfile
with tempfile.TemporaryFile() as ferr:
try:
# this should fail, but sometimes the second attempt will succeed #282
self.kernels[kinfo.name] = manager.start_new_kernel(
startup_timeout=60,
kernel_name=kinfo.kernel,
cwd=os.getcwd(),
stdout=subprocess.DEVNULL,
stderr=ferr)
new_kernel = True
env.log_to_file('KERNEL', f'Kernel {kinfo.kernel} started with the second attempt.')
except Exception as e:
ferr.seek(0)
raise RuntimeError(
f'Failed to start kernel "{kernel}". {e}\nError Message:\n{ferr.read().decode()}'
) from e
self.KM, self.KC = self.kernels[kinfo.name]
self.kernel = kinfo.name
if new_kernel and not kinfo.codemirror_mode:
self.KC.kernel_info()
kinfo.codemirror_mode = self.KC.get_shell_msg(timeout=10)['content']['language_info'].get(
'codemirror_mode', '')
self.subkernels.notify_frontend()
if new_kernel and kinfo.language in self.supported_languages:
lan_module = self.supported_languages[kinfo.language](self, kinfo.kernel)
init_stmts = lan_module.init_statements
if hasattr(lan_module, '__version__'):
module_version = f' (version {lan_module.__version__})'
else:
module_version = ' (version unavailable)'
env.log_to_file('KERNEL', f'Loading language module for kernel {kinfo.name}{module_version}')
if init_stmts:
await self.run_cell(init_stmts, True, False)
# passing
if in_vars:
await self.get_vars_from(in_vars, as_var=as_var)
def shutdown_kernel(self, kernel, restart=False):
kernel = self.subkernels.find(kernel).name
if kernel == 'SoS':
# cannot restart myself ...
self.warn('Cannot restart SoS kernel from within SoS.')
elif kernel:
if kernel not in self.kernels:
self.send_response(self.iopub_socket, 'stream', dict(name='stdout', text=f'{kernel} is not running'))
elif restart:
orig_kernel = self.kernel
try:
# shutdown
self.shutdown_kernel(kernel)
# switch back to kernel (start a new one)
asyncio.run(self.switch_kernel(kernel))
finally:
# finally switch to starting kernel
asyncio.run(self.switch_kernel(orig_kernel))
else:
# shutdown
if self.kernel == kernel:
asyncio.run(self.switch_kernel('SoS'))
try:
self.kernels[kernel][0].shutdown_kernel(restart=False)
except Exception as e:
self.warn(f'Failed to shutdown kernel {kernel}: {e}\n')
finally:
self.kernels.pop(kernel)
else:
self.send_response(
self.iopub_socket, 'stream',
dict(
name='stdout',
text='Specify one of the kernels to shutdown: SoS{}\n'.format(''.join(
f', {x}' for x in self.kernels))))
# stop_controller(self.controller)
# def get_response(self, statement, msg_types, name=None):
# return asyncio.run(self._async_get_response(statement, msg_types, name))
def get_response(self, statement, msg_types, name=None):
# get response of statement of specific msg types.
while self.KC.shell_channel.msg_ready():
self.KC.shell_channel.get_msg()
while self.KC.iopub_channel.msg_ready():
sub_msg = self.KC.iopub_channel.get_msg()
if sub_msg['header']['msg_type'] != 'status':
env.log_to_file('MESSAGE',
f"Overflow message in iopub {sub_msg['header']['msg_type']} {sub_msg['content']}")
responses = []
self.KC.execute(statement, silent=False, store_history=False)
# first thing is wait for any side effects (output, stdin, etc.)
iopub_started = False
iopub_ended = False
shell_ended = False
while not (iopub_started and iopub_ended and shell_ended):
# display intermediate print statements, etc.
while self.KC.iopub_channel.msg_ready():
sub_msg = self.KC.iopub_channel.get_msg()
msg_type = sub_msg['header']['msg_type']
env.log_to_file('MESSAGE', f'Received {msg_type} {sub_msg["content"]}')
if msg_type == 'status':
if sub_msg["content"]["execution_state"] == 'busy':
iopub_started = True
elif iopub_started and sub_msg["content"]["execution_state"] == 'idle':
iopub_ended = True
continue
if msg_type in msg_types and (name is None or sub_msg['content'].get('name', None) in name or
any(x in name for x in sub_msg['content'].keys())):
env.log_to_file('MESSAGE', f'Capture response: {msg_type}: {sub_msg["content"]}')
responses.append([msg_type, sub_msg['content']])
else:
env.log_to_file('MESSAGE', f'Non-response: {msg_type}: {sub_msg["content"]}')
#
# we ignore the messages we are not interested.
#
# self.send_response(
# self.iopub_socket, msg_type, sub_msg['content'])
if self.KC.shell_channel.msg_ready():
# now get the real result