server.py 24.2 KB
Newer Older
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
#
# Copyright (C) 2015 INRA
# 
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
# 
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
# 
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#

import cherrypy
import cgi
import tempfile
import json
import sys
import datetime
from functools import wraps
import time
import os
import argparse

29
30
import logging

31
32
from argparse import ArgumentTypeError

33
34
from collections import OrderedDict

35
36
37
from .workflows_manager import WorkflowsManager
from .config_reader import JFlowConfigReader
from .workflow import Workflow
Ibouniyamine Nabihoudine's avatar
Ibouniyamine Nabihoudine committed
38
from .parameter import browsefile, localfile, urlfile, inputfile, create_test_function, MiltipleAction, MiltipleAppendAction, MultipleParameters
39
from workflows.types import *
40
from . import utils
41
from cctools.util import time_format
Ibouniyamine Nabihoudine's avatar
Ibouniyamine Nabihoudine committed
42
from .utils import get_octet_string_representation
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

# function in charge to upload large files
class UploadFieldStorage(cgi.FieldStorage):
    """Our version uses a named temporary file instead of the default
    non-named file; keeping it visibile (named), allows us to create a
    2nd link after the upload is done, thus avoiding the overhead of
    making a copy to the destination filename."""

    def get_tmp_directory(self):
        jflowconf = JFlowConfigReader()
        return jflowconf.get_tmp_directory()

    def get_file_name(self):
        self.tmpfile = None
        # if this is a file object, just return the name of the file
        if hasattr( self.file, 'name' ):
            return self.file.name
        # if not, this is a cStringIO.StringO, write it down
        # and return the file name
        else:
            tmp_folder = self.get_tmp_directory()
            if not os.path.exists( tmp_folder ):
                try : os.mkdir(tmp_folder)
                except : pass
67
68
            fh = open(os.path.join(tmp_folder, self.filename), "wb+")
            fh_name = fh.name
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
            fh.write(self.file.getvalue())
            fh.close()
            return fh_name

    def __del__(self):
        try:
            self.file.close()
        except AttributeError:
            pass
        try:
            tmp_folder = self.get_tmp_directory()
            os.remove(os.path.join(tmp_folder, self.filename))
        except:
            pass

    def make_file(self, binary=None):
        tmp_folder = self.get_tmp_directory()
        if not os.path.exists( tmp_folder ):
            try : os.mkdir(tmp_folder)
            except : pass
        return tempfile.NamedTemporaryFile(dir=tmp_folder)


def noBodyProcess():
    """Sets cherrypy.request.process_request_body = False, giving
    us direct control of the file upload destination. By default
    cherrypy loads it to memory, we are directing it to disk."""
    cherrypy.request.process_request_body = False

cherrypy.tools.noBodyProcess = cherrypy.Tool('before_request_body', noBodyProcess)

# define functions in charge to handle cross domain calls
def CORS():
    cherrypy.response.headers['Access-Control-Allow-Origin'] = '*'
    cherrypy.response.headers['Access-Control-Allow-Methods'] = 'OPTIONS, GET, POST'
    cherrypy.response.headers['Access-Control-Allow-Headers'] = 'Content-Type, Content-Range, Content-Disposition'
cherrypy.tools.CORS = cherrypy.Tool('before_finalize', CORS)

class JFlowJSONEncoder (json.JSONEncoder):
    def default(self, obj):
        if isinstance(obj, (datetime.date, datetime.datetime)):
            return obj.strftime( JFlowConfigReader().get_date_format() )
        else:
            return json.JSONEncoder.default(self, obj)

class JFlowServer (object):

    MULTIPLE_TYPE_SPLITER = "___"
    APPEND_PARAM_SPLITER = "::-::"
    JFLOW_WDATA = "data"

    def __init__(self):
        # Create a workflow manager to get access to our workflows
        self.wfmanager = WorkflowsManager()
        self.jflow_config_reader = JFlowConfigReader()
    
Jerome Mariette's avatar
Jerome Mariette committed
125
    @staticmethod
Jerome Mariette's avatar
Jerome Mariette committed
126
    def quickstart(server_class, config=None, daemon=False):
Jerome Mariette's avatar
Jerome Mariette committed
127
128
129
130
131
132
133
134
135
        
        # daemonize the server if asked to
        if daemon:
            from cherrypy.process.plugins import Daemonizer
            Daemonizer(cherrypy.engine).subscribe()
        
        # define the socket host and port
        jflowconf = JFlowConfigReader()
        socket_opts = jflowconf.get_socket_options()
Jerome Mariette's avatar
Jerome Mariette committed
136
137
        
        # add the result directory
138
        if config is None or not '/' in config:
Jerome Mariette's avatar
Jerome Mariette committed
139
140
            config['/'] = {'tools.staticdir.root': jflowconf.get_work_directory()}
        else:
Jerome Mariette's avatar
Jerome Mariette committed
141
142
143
            link = os.path.join(config['/']['tools.staticdir.root'], "data")
            if not os.path.islink(link):
                os.symlink(jflowconf.get_work_directory(), link)
Jerome Mariette's avatar
Jerome Mariette committed
144
145
146
            
        config[os.path.join('/', JFlowServer.JFLOW_WDATA)] = {'tools.staticdir.on'  : True,
                                                              'tools.staticdir.dir' : jflowconf.get_work_directory()}
Jerome Mariette's avatar
Jerome Mariette committed
147
148
149
150
151
152
153
154
155
156
157
158
159
160
    
        # remove any limit on the request body size; cherrypy's default is 100MB
        # (maybe we should just increase it ?)
        cherrypy.server.max_request_body_size = 0
    
        # increase server socket timeout to 60s; we are more tolerant of bad
        # quality client-server connections (cherrypy's default is 10s)
        cherrypy.server.socket_timeout = 60
    
        cherrypy.config.update({'server.socket_host': socket_opts[0],
                                'server.socket_port': socket_opts[1]})
        # start the server
        cherrypy.quickstart(server_class(), config=config)
    
161
162
163
164
165
166
167
    def jsonify(func):
        '''JSON and JSONP decorator for CherryPy'''
        @wraps(func)
        def wrapper(*args, **kw):
            value = func(*args, **kw)
            cherrypy.response.headers["Content-Type"] = "application/json"
            # if JSONP request
168
169
            if "callback" in kw: 
                return ('%s(%s)' % (kw["callback"], json.dumps(value, cls=JFlowJSONEncoder))).encode('utf8')
170
            # else return the JSON
171
            else: return json.dumps(value, cls=JFlowJSONEncoder).encode('utf8')
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
        return wrapper

    def jsonify_workflow_status(self, workflow, init_to_zero=False):
        if workflow.start_time: start_time = time.asctime(time.localtime(workflow.start_time))
        else: start_time = "-"
        if workflow.start_time and workflow.end_time: elapsed_time = str(workflow.end_time-workflow.start_time)
        elif workflow.start_time: elapsed_time = str(time.time()-workflow.start_time)
        else: elapsed_time = "-"
        if workflow.end_time: end_time = time.asctime(time.localtime(workflow.end_time))
        else: end_time = "-"
        if init_to_zero:
            return {"id":utils.get_nb_string(workflow.id),
                    "name": workflow.name,
                    "status": Workflow.STATUS_STARTED,
                    "elapsed_time": str(elapsed_time),
                    "start_time": start_time,
                    "end_time": end_time,
                    "components": []}
        else:
            components = []
192
            components_status = workflow.get_components_status()
193
            for i, component in enumerate(workflow.get_components_nameid()):
194
                status_info = components_status[component]
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
                try: perc_waiting = (status_info["waiting"]*100.0)/status_info["tasks"]
                except: perc_waiting = 0
                try: perc_running = (status_info["running"]*100.0)/status_info["tasks"]
                except: perc_running = 0
                try: perc_failed = (status_info["failed"]*100.0)/status_info["tasks"]
                except: perc_failed = 0
                try: perc_aborted = (status_info["aborted"]*100.0)/status_info["tasks"]
                except: perc_aborted = 0
                try: perc_completed = (status_info["completed"]*100.0)/status_info["tasks"]
                except: perc_completed = 0
                components.append({"name": component,
                                  "elapsed_time": time_format(status_info["time"]),
                                  "total": status_info["tasks"],
                                  "waiting": status_info["waiting"],
                                  "failed": status_info["failed"],
                                  "running": status_info["running"],
                                  "aborted": status_info["aborted"],
                                  "completed": status_info["completed"]})

            status = {"id":utils.get_nb_string(workflow.id),
                      "errors": workflow.get_errors(),
                      "name": workflow.name,
                      "metadata": workflow.metadata,
                      "status": workflow.get_status(),
                      "elapsed_time": "-" if elapsed_time == "-" else str(datetime.timedelta(seconds=int(str(elapsed_time).split(".")[0]))),
                      "start_time": start_time,
                      "end_time": end_time,
                      "components": components}

            return status

    @cherrypy.expose
    @jsonify
    def get_available_workflows(self, **kwargs):
        workflows = []
230
231
232
233
        filter_groups = None
        select = False
        if 'filter_groups' in kwargs : filter_groups = kwargs['filter_groups'].split(',')
        if 'select' in kwargs : select = kwargs['select'] in ['True', 'true', '1', 1]
234

235
        wf_instances, wf_methodes = self.wfmanager.get_available_workflows(filter_groups = filter_groups  , select = select)
236
        for instance in wf_instances:
237
            parameters, parameters_per_groups, ordered_groups = [], {}, ["default"]
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
            for param in instance.get_parameters():
                # if it's a multiple action change the action by the name
                if param.action == MiltipleAction:
                    action = "MiltipleAction"
                elif param.action == MiltipleAppendAction:
                    action = "MiltipleAppendAction"
                else:
                    action = param.action
                try:
                    cparam_help = param.global_help
                except:
                    cparam_help = param.help
                hash_param = {"help": cparam_help,
                              "required": param.required,
                              "default": param.default,
                              "choices": param.choices,
                              "action": action,
                              "type": param.get_type(),
                              "name": param.name,
                              "display_name": param.display_name,
                              "group": param.group}
                if hash_param["type"] == "date":
                    hash_param["format"] = self.jflow_config_reader.get_date_format()
                    if hash_param["format"] == '%d/%m/%Y':
                        hash_param["format"] = 'dd/mm/yyyy'
                    elif hash_param["format"] == '%d/%m/%y':
                        hash_param["format"] = 'dd/mm/yy'
                    elif hash_param["format"] == '%Y/%m/%d':
                        hash_param["format"] = 'yyyy/mm/dd'
                    elif hash_param["format"] == '%y/%m/%d':
                        hash_param["format"] = 'yy/mm/dd'
269
270
                if param.rules is not None:
                    hash_param["rules"] = param.rules
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
                # if it's a multiple type add sub parameters
                if type(param.type) == MultipleParameters:
                    hash_param["sub_parameters"] = []
                    for sub_param in param.sub_parameters:
                        hash_param["sub_parameters"].append({"help": sub_param.help,
                              "required": sub_param.required,
                              "default": sub_param.default,
                              "choices": sub_param.choices,
                              "action": sub_param.action,
                              "type": sub_param.get_type(),
                              "name": param.name + JFlowServer.MULTIPLE_TYPE_SPLITER + sub_param.flag,
                              "display_name": sub_param.display_name,
                              "group": sub_param.group})
                        if hash_param["sub_parameters"][-1]["type"] == "date":
                            hash_param["sub_parameters"][-1]["format"] = self.jflow_config_reader.get_date_format()
                            if hash_param["sub_parameters"][-1]["format"] == '%d/%m/%Y':
                                hash_param["sub_parameters"][-1]["format"] = 'dd/mm/yyyy'
                            elif hash_param["sub_parameters"][-1]["format"] == '%d/%m/%y':
                                hash_param["sub_parameters"][-1]["format"] = 'dd/mm/yy'
                            elif hash_param["sub_parameters"][-1]["format"] == '%Y/%m/%d':
                                hash_param["sub_parameters"][-1]["format"] = 'yyyy/mm/dd'
                            elif hash_param["sub_parameters"][-1]["format"] == '%y/%m/%d':
                                hash_param["sub_parameters"][-1]["format"] = 'yy/mm/dd'
294
295
                        if sub_param.rules is not None:
                            hash_param["sub_parameters"][-1]["rules"] = sub_param.rules
296
                parameters.append(hash_param)
297
                if param.group in parameters_per_groups:
298
299
                    parameters_per_groups[param.group].append(hash_param)
                else: parameters_per_groups[param.group] = [hash_param]
300
301
                if param.group not in ordered_groups:
                    ordered_groups.append(param.group)
302
303
            workflows.append({"name": instance.name,
                              "help": instance.description,
Ibouniyamine Nabihoudine's avatar
Ibouniyamine Nabihoudine committed
304
                              "class": instance.__class__.__name__,
305
306
                              "parameters": parameters,
                              "parameters_per_groups": parameters_per_groups,
307
                              "groups": ordered_groups})
308
309
310
311
312
313
314
315
316
317
        return workflows

    @cherrypy.expose
    @jsonify
    def run_workflow(self, **kwargs):
        try:
            kwargs_modified = {}

            # handle MultiParameterList
            multi_sub_params = {}
318
            for key in list(kwargs.keys()):
319
320
                parts = key.split(JFlowServer.MULTIPLE_TYPE_SPLITER)
                if len(parts) == 3:
321
                    if not parts[0] in kwargs_modified:
322
323
                        kwargs_modified[parts[0]] = []
                        multi_sub_params[parts[0]] = {}
324
                    if parts[2] in multi_sub_params[parts[0]]:
325
                        multi_sub_params[parts[0]][parts[2]].append((parts[1], kwargs[key]))
326
                    else:
327
                        multi_sub_params[parts[0]][parts[2]] = [(parts[1], kwargs[key])]
328

329
            for key in list(kwargs.keys()):
330
331
                parts = key.split(JFlowServer.MULTIPLE_TYPE_SPLITER)
                # split append values
332
                new_values = kwargs[key].split(JFlowServer.APPEND_PARAM_SPLITER)
333
334
335
336
337
338
339
                if len(new_values) == 1:
                    new_values = new_values[0]
                # if this is a classic Parameter
                if len(parts) == 1:
                    kwargs_modified[key] = new_values
                # if this is a MultiParameter
                elif len(parts) == 2:
340
                    if parts[0] in kwargs_modified:
341
342
343
344
345
346
347
348
349
350
                        kwargs_modified[parts[0]].append((parts[1], new_values))
                    else:
                        kwargs_modified[parts[0]] = [(parts[1], new_values)]

            # handle MultiParameterList
            for param in multi_sub_params:
                kwargs_modified[param] = []
                for sub_param in multi_sub_params[param]:
                    kwargs_modified[param].append(multi_sub_params[param][sub_param])

351
352
353
354
            for param in kwargs_modified:
                logging.debug(param)
                logging.debug(str(kwargs_modified[param]))

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
            workflow = self.wfmanager.run_workflow(kwargs_modified["workflow_class"], kwargs_modified)
            return { "status" : 0, "content" : self.jsonify_workflow_status(workflow, True) }
        except Exception as err:
            return { "status" : 1, "content" : str(err) }

    @cherrypy.expose
    @jsonify
    def delete_workflow(self, **kwargs):
        self.wfmanager.delete_workflow(kwargs["workflow_id"])

    @cherrypy.expose
    @jsonify
    def rerun_workflow(self, **kwargs):
        workflow = self.wfmanager.rerun_workflow(kwargs["workflow_id"])
        return self.jsonify_workflow_status(workflow)

    @cherrypy.expose
    @jsonify
    def reset_workflow_component(self, **kwargs):
        workflow = self.wfmanager.reset_workflow_component(kwargs["workflow_id"], kwargs["component_name"])
        return self.jsonify_workflow_status(workflow)

    @cherrypy.expose
    def upload_light(self, **kwargs):
        uniq_directory = ""
380
        for key in list(kwargs.keys()):
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
            if key == "uniq_directory":
                uniq_directory = kwargs['uniq_directory']
            else:
                file_param = key

        # the file transfer can take a long time; by default cherrypy
        # limits responses to 300s; we increase it to 1h
        cherrypy.response.timeout = 3600

        # upload file by chunks
        file_dir = os.path.join( self.jflow_config_reader.get_tmp_directory(), uniq_directory )
        os.mkdir( file_dir )

        if isinstance(kwargs[file_param], list):
            for cfile in kwargs[file_param]:
396
                FH_sever_file = open(os.path.join(file_dir, cfile.filename), "w")
397
398
399
400
401
402
403
                while True:
                    data = cfile.file.read(8192)
                    if not data:
                        break
                    FH_sever_file.write(data)
                FH_sever_file.close()
        else:
404
            FH_sever_file = open(os.path.join(file_dir, kwargs[file_param].filename), "w")
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
            while True:
                data = kwargs[file_param].file.read(8192)
                if not data:
                    break
                FH_sever_file.write(data)
            FH_sever_file.close()

    @cherrypy.expose
    @cherrypy.tools.noBodyProcess()
    @cherrypy.tools.CORS()
    def upload(self):
        # the file transfer can take a long time; by default cherrypy
        # limits responses to 300s; we increase it to 1h
        cherrypy.response.timeout = 3600

        # convert the header keys to lower case
        lcHDRS = {}
422
        for key, val in cherrypy.request.headers.items():
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
            lcHDRS[key.lower()] = val

        # at this point we could limit the upload on content-length...
        # incomingBytes = int(lcHDRS['content-length'])

        # create our version of cgi.FieldStorage to parse the MIME encoded
        # form data where the file is contained
        formFields = UploadFieldStorage(fp=cherrypy.request.rfile,
                                        headers=lcHDRS,
                                        environ={'REQUEST_METHOD':'POST'},
                                        keep_blank_values=True)

        # we now create a link to the file, using the submitted
        # filename; if we renamed, there would be a failure because
        # the NamedTemporaryFile, used by our version of cgi.FieldStorage,
        # explicitly deletes the original filename
439
        for current in list(formFields.keys()):
440
441
442
443
444
445
446
447
            if current != 'uniq_directory':
                currentFile = formFields[current]
                fileDir = os.path.join(self.jflow_config_reader.get_tmp_directory(), formFields.getvalue("uniq_directory"))
                os.mkdir(fileDir)
                if isinstance(currentFile, list):
                    for cfile in currentFile:
                        os.link(
                                cfile.get_file_name(),
448
                                os.path.join(fileDir, cfile.filename)
449
450
451
452
                        )
                else:
                    os.link(
                            currentFile.get_file_name(),
453
                            os.path.join(fileDir, currentFile.filename)
454
455
456
457
458
459
460
461
                    )

    @cherrypy.expose
    @jsonify
    def get_workflows_status(self, **kwargs):
        status = []
        workflows = self.wfmanager.get_workflows(use_cache=True)
        for workflow in workflows:
462
            if "metadata_filter" in kwargs:
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
                is_ok = False
                for wf_meta in workflow.metadata:
                    for metadata in kwargs["metadata_filter"].split(","):
                        if wf_meta == metadata:
                            is_ok = True
                            break
                    if is_ok: break
                if is_ok: status.append(self.jsonify_workflow_status(workflow))
            else:
                status.append(self.jsonify_workflow_status(workflow))
        return status

    @cherrypy.expose
    @jsonify
    def get_workflow_status(self, **kwargs):
        workflow = self.wfmanager.get_workflow(kwargs["workflow_id"])
        if kwargs["display"] == "list":
            return self.jsonify_workflow_status(workflow)
        elif kwargs["display"] == "graph":
            g = workflow.get_execution_graph()
            status = self.jsonify_workflow_status(workflow)
            nodes = []
            for node in g.nodes():
                if Workflow.INPUTFILE_GRAPH_LABEL in g.node_attributes(node):
                    nodes.append({"name": node, "display_name": g.node_attributes(node)[1], "type": "inputfile"})
                elif Workflow.INPUTFILES_GRAPH_LABEL in g.node_attributes(node):
                    nodes.append({"name": node, "display_name": g.node_attributes(node)[1], "type": "inputfiles"})
                elif Workflow.INPUTDIRECTORY_GRAPH_LABEL in g.node_attributes(node):
                    nodes.append({"name": node, "display_name": g.node_attributes(node)[1], "type": "inputdirectory"})    
                elif Workflow.COMPONENT_GRAPH_LABEL in g.node_attributes(node):
                    nodes.append({"name": node, "display_name": g.node_attributes(node)[1], "type": "component"})
            status["nodes"] = nodes
            status["edges"] = g.edges()
            return status

    def _webify_outputs(self, web_path, path):
        work_dir  = self.jflow_config_reader.get_work_directory()
        if work_dir.endswith("/"): work_dir = work_dir[:-1]
        socket_opt = self.jflow_config_reader.get_socket_options()
502
503
504
505
506
507
        if not os.path.isdir(path):
            return {
                    'url':'http://' + socket_opt[0] + ':' + str(socket_opt[1]) + '/' + path.replace(work_dir, web_path),
                    'size': get_octet_string_representation(os.path.getsize(os.path.abspath(path))),
                    'extension': os.path.splitext(path)[1]
                   }
508
509
        else:  # Remove directories from the list
            return None
Floreal Cabanettes's avatar
Floreal Cabanettes committed
510
        # TODO: show output directories and add the possibility to browse them
511
512
513
514

    @cherrypy.expose
    @jsonify
    def get_workflow_outputs(self, **kwargs):
515
        on_disk_outputs, on_web_outputs = self.wfmanager.get_workflow_outputs(kwargs["workflow_id"]), OrderedDict()
516
        for cpt_name in list(on_disk_outputs.keys()):
517
518
519
            on_web_outputs[cpt_name] = {}
            for outf in on_disk_outputs[cpt_name]:
                on_web_outputs[cpt_name][outf] = self._webify_outputs(JFlowServer.JFLOW_WDATA, on_disk_outputs[cpt_name][outf])
520
521
                if on_web_outputs[cpt_name][outf] is None:  # Remove directories from the list
                    del on_web_outputs[cpt_name][outf]
522
523
524
525
526
527
528
        return on_web_outputs

    @cherrypy.expose
    @jsonify
    def validate_field(self, **kwargs):
        try:
            value_key = None
529
            for key in list(kwargs.keys()):
530
531
532
533
534
535
                if key != "type" and key != "callback" and key != "_" and key != "action": 
                    value_key = key
                    break
            # if it's an append parameter, let's check each value
            if kwargs["action"] == "append":
                for cval in kwargs[value_key].split("\n"):
536
                    create_test_function(kwargs["type"])(cval)
537
            else:
538
                create_test_function(kwargs["type"])(kwargs[value_key])
539
            return True
540
        except Exception as e:
541
            return str(e)