Blame view

web/run.py 32.4 KB
bde97e4d   Goutte   Add more changes ...
1
2
# coding=utf-8

9390ec89   Goutte   Initial experimen...
3
import StringIO
bc18b96c   Goutte   Implement first (...
4
import datetime
8644387c   Goutte   Use real data.
5
import gzip
bc18b96c   Goutte   Implement first (...
6
7
8
import json
import logging
import random
2fedd73b   Goutte   Initial implement...
9
import tarfile
bde97e4d   Goutte   Add more changes ...
10
import time
8644387c   Goutte   Use real data.
11
import urllib
9390ec89   Goutte   Initial experimen...
12
from csv import writer as csv_writer
bc18b96c   Goutte   Implement first (...
13
14
15
16
from math import sqrt
from os import environ, remove as removefile
from os.path import isfile, join, abspath, dirname

9390ec89   Goutte   Initial experimen...
17
from flask import Flask
9390ec89   Goutte   Initial experimen...
18
from flask import request
bc18b96c   Goutte   Implement first (...
19
from flask import url_for, send_from_directory, abort as abort_flask
bde97e4d   Goutte   Add more changes ...
20
from jinja2 import Environment, FileSystemLoader, Markup
57493104   Goutte   Add the time to t...
21
from netCDF4 import Dataset, date2num
bc18b96c   Goutte   Implement first (...
22
from yaml import load as yaml_load
9390ec89   Goutte   Initial experimen...
23
24
25
26
27
28
29

# PATH RELATIVITY #############################################################

THIS_DIRECTORY = dirname(abspath(__file__))


def get_path(relative_path):
a4a9ef03   Goutte   Cache generated C...
30
    """Get an absolute path from the relative path to this script directory."""
9390ec89   Goutte   Initial experimen...
31
32
33
34
35
36
37
38
39
40
41
42
43
    return abspath(join(THIS_DIRECTORY, relative_path))


# COLLECT GLOBAL INFORMATION FROM SOURCES #####################################

# VERSION
with open(get_path('../VERSION'), 'r') as version_file:
    version = version_file.read().strip()

# CONFIG
with open(get_path('../config.yml'), 'r') as config_file:
    config = yaml_load(config_file.read())

c0df94bc   Goutte   Adding more logs.
44
45
FILE_DATE_FMT = "%Y-%m-%dT%H:%M:%S"

9390ec89   Goutte   Initial experimen...
46

f75faf5f   Goutte   WIP
47
48
49
# LOGGING #####################################################################

log = logging.getLogger("HelioPropa")
9bfa6c42   Goutte   More bug hunting.
50
log.setLevel(logging.DEBUG)
077980eb   Goutte   Improve availabil...
51
# log.setLevel(logging.ERROR)                        # <-- set log level here !
b2837a08   Goutte   Add three retries...
52
53
54
55
56
logHandler = logging.FileHandler(get_path('run.log'))
logHandler.setFormatter(logging.Formatter(
    "%(asctime)s - %(levelname)s - %(message)s"
))
log.addHandler(logHandler)
f75faf5f   Goutte   WIP
57
58


e18701b6   Goutte   Cache clear (remo...
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
# HARDCODED CONFIGURATION #####################################################

# Absolute path to the data cache directory
CACHE_DIR = get_path('../cache')

# These two configs are not in the YAML config because adding a new parameter
# will not work as-is, you'll have to edit some netcdf-related code.

# The slugs of the available parameters in the generated CSV files.
# The order matters. If you change this you also need to change the
# innermost loop of `get_data_for_target`.
# The javascript knows the targets' properties under these names.
PROPERTIES = ('time', 'vrad', 'vtan', 'vtot', 'btan', 'temp', 'pdyn', 'dens',
              'angl', 'xhee', 'yhee')

# The parameters that the users can handle.
# The slug MUST be one of the properties above.
PARAMETERS = {
    'pdyn': {
        'slug': 'pdyn',
        'name': 'Dyn. Pressure',
        'title': 'The dynamic pressure.',
        'units': 'nPa',
        'active': True,
        'position': 10,
    },
    'vtot': {
        'slug': 'vtot',
        'name': 'Velocity',
        'title': 'The velocity of the particles.',
        'units': 'km/s',
        'active': False,
        'position': 20,
    },
    'btan': {
        'slug': 'btan',
        'name': 'B Tangential',
        'title': 'B Tangential.',
        'units': 'nT',
        'active': False,
        'position': 30,
    },
    'temp': {
        'slug': 'temp',
        'name': 'Temperature',
        'title': 'The absolute temperature.',
        'units': 'K',
        'active': False,
        'position': 40,
    },
    'dens': {
        'slug': 'dens',
        'name': 'Density',
        'title': 'The density N.',
        'units': u'cm⁻³',
        'active': False,
        'position': 50,
    },
    'angl': {
        'slug': 'angl',
        'name': 'Angle T-S-E',
        'title': 'Angle Target-Sun-Earth.',
        'units': 'deg',
        'active': False,
        'position': 60,
    },
}


9390ec89   Goutte   Initial experimen...
128
129
130
131
# SETUP FLASK ENGINE ##########################################################

app = Flask(__name__, root_path=THIS_DIRECTORY)
app.debug = environ.get('DEBUG') == 'true'
b2837a08   Goutte   Add three retries...
132
if app.debug:
2fedd73b   Goutte   Initial implement...
133
    log.info("Starting Flask app IN DEBUG MODE...")
b2837a08   Goutte   Add three retries...
134
135
else:
    log.info("Starting Flask app...")
9390ec89   Goutte   Initial experimen...
136
137
138
139
140
141
142
143
144
145
146


# SETUP JINJA2 TEMPLATE ENGINE ################################################

def static_global(filename):
    return url_for('static', filename=filename)


def shuffle_filter(seq):
    """
    This shuffles the sequence it is applied to.
2fedd73b   Goutte   Initial implement...
147
    Jinja2 _should_ provide this.
9390ec89   Goutte   Initial experimen...
148
149
150
151
152
153
154
155
156
157
158
    """
    try:
        result = list(seq)
        random.shuffle(result)
        return result
    except:
        return seq


def markdown_filter(value, nl2br=False, p=True):
    """
2fedd73b   Goutte   Initial implement...
159
    Converts markdown into html.
9390ec89   Goutte   Initial experimen...
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
    nl2br: set to True to replace line breaks with <br> tags
    p: set to False to remove the enclosing <p></p> tags
    """
    from markdown import markdown
    from markdown.extensions.nl2br import Nl2BrExtension
    from markdown.extensions.abbr import AbbrExtension
    extensions = [AbbrExtension()]
    if nl2br is True:
        extensions.append(Nl2BrExtension())
    markdowned = markdown(value, output_format='html5', extensions=extensions)
    if p is False:
        markdowned = markdowned.replace(r"<p>", "").replace(r"</p>", "")
    return markdowned


bde97e4d   Goutte   Add more changes ...
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
_js_escapes = {
        '\\': '\\u005C',
        '\'': '\\u0027',
        '"': '\\u0022',
        '>': '\\u003E',
        '<': '\\u003C',
        '&': '\\u0026',
        '=': '\\u003D',
        '-': '\\u002D',
        ';': '\\u003B',
        u'\u2028': '\\u2028',
        u'\u2029': '\\u2029'
}
# Escape every ASCII character with a value less than 32.
_js_escapes.update(('%c' % z, '\\u%04X' % z) for z in xrange(32))


def escapejs_filter(value):
    escaped = []
    for letter in value:
        if letter in _js_escapes:
            escaped.append(_js_escapes[letter])
        else:
            escaped.append(letter)

    return Markup("".join(escaped))

9390ec89   Goutte   Initial experimen...
202
203
204
205
206
207
208
209
210
211
212
213
tpl_engine = Environment(loader=FileSystemLoader([get_path('view')]),
                         trim_blocks=True,
                         lstrip_blocks=True)

tpl_engine.globals.update(
    url_for=url_for,
    static=static_global,
)

tpl_engine.filters['markdown'] = markdown_filter
tpl_engine.filters['md'] = markdown_filter
tpl_engine.filters['shuffle'] = shuffle_filter
bde97e4d   Goutte   Add more changes ...
214
tpl_engine.filters['escapejs'] = escapejs_filter
9390ec89   Goutte   Initial experimen...
215
216
217
218
219
220
221
222
223
224
225

tpl_global_vars = {
    'request': request,
    'version': version,
    'config': config,
    'now': datetime.datetime.now(),
}


# HELPERS #####################################################################

57f42bd7   Goutte   Log the abortions.
226
227
228
229
230
def abort(code, message):
    log.error(message)
    abort_flask(code, message)


9390ec89   Goutte   Initial experimen...
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
def render_view(view, context=None):
    """
    A simple helper to render [view] template with [context] vars.
    It automatically adds the global template vars defined above, too.
    It returns a string, usually the HTML contents to display.
    """
    context = {} if context is None else context
    return tpl_engine.get_template(view).render(
        dict(tpl_global_vars.items() + context.items())
    )


# def render_page(page, title="My Page", context=None):
#     """
#     A simple helper to render the md_page.html template with [context] vars &
#     the additional contents of `page/[page].md` in the `md_page` variable.
#     It automagically adds the global template vars defined above, too.
#     It returns a string, usually the HTML contents to display.
#     """
#     if context is None:
#         context = {}
#     context['title'] = title
#     context['md_page'] = ''
#     with file(get_path('page/%s.md' % page)) as f:
#         context['md_page'] = f.read()
#     return tpl_engine.get_template('md_page.html').render(
#         dict(tpl_global_vars.items() + context.items())
#     )

077980eb   Goutte   Improve availabil...
260

bc18b96c   Goutte   Implement first (...
261
262
263
264
265
266
267
def is_list_in_list(needle, haystack):
    for n in needle:
        if n not in haystack:
            return False
    return True


2d2af24b   Goutte   Add a basic orbit...
268
def datetime_from_list(time_list):
0b9821dd   Goutte   Clean up.
269
    """
2fedd73b   Goutte   Initial implement...
270
    Datetimes in retrieved CDFs are stored as lists of numbers,
80352490   Goutte   Multi model suppo...
271
272
    with DayOfYear starting at 0. We want it starting at 1 because it's what
    vendor parsers use, both in python and javascript.
0b9821dd   Goutte   Clean up.
273
    """
2d2af24b   Goutte   Add a basic orbit...
274
275
276
277
278
279
    # Day Of Year starts at 0, but for our datetime parser it starts at 1
    doy = '{:03d}'.format(int(''.join(time_list[4:7])) + 1)
    return datetime.datetime.strptime(
        "%s%s%s" % (''.join(time_list[0:4]), doy, ''.join(time_list[7:])),
        "%Y%j%H%M%S%f"
    )
9390ec89   Goutte   Initial experimen...
280

ce8af118   Goutte   Fix the favicon.
281

927c69c3   Goutte   Make the local ca...
282
283
284
285
286
287
288
289
290
291
292
293
294
295
def get_local_filename(url):
    """
    Build the local cache filename for the distant file
    :param url: string
    :return: string
    """
    from slugify import slugify
    n = len('http://')
    if url.startswith('https'):
        n += 1
    s = url[n:]
    return slugify(s)


180d7d97   Goutte   Refactor heavily.
296
def get_target_config(slug):
2fedd73b   Goutte   Initial implement...
297
    for s in config['targets']:  # dumb
8644387c   Goutte   Use real data.
298
299
        if s['slug'] == slug:
            return s
180d7d97   Goutte   Refactor heavily.
300
    raise Exception("No target found in configuration for '%s'." % slug)
8644387c   Goutte   Use real data.
301
302


180d7d97   Goutte   Refactor heavily.
303
304
305
306
307
def check_target_config(slug):
    get_target_config(slug)


def retrieve_amda_netcdf(orbiter, what, started_at, stopped_at):
8644387c   Goutte   Use real data.
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
    """
    Handles remote querying Myriam's API, downloading, extracting and caching
    the netCDF files.
    :param orbiter: key of the source in the YAML config
    :param what: either 'model' or 'orbit', a key in the config of the source
    :param started_at:
    :param stopped_at:
    :return: a list of local file paths to netCDF (.nc) files
    """

    url = config['amda'].format(
        dataSet=what,
        startTime=started_at.isoformat(),
        stopTime=stopped_at.isoformat()
    )
c50cc9d8   Goutte   Continue fixing.
323
    log.info("Fetching remote gzip files list at '%s'." % url)
b2837a08   Goutte   Add three retries...
324
325
    retries = 0
    success = False
92abc15b   Goutte   Mistrust the API ...
326
    errors = []
b2837a08   Goutte   Add three retries...
327
328
329
330
331
332
333
334
    remote_gzip_files = []
    while not success and retries < 3:
        try:
            response = urllib.urlopen(url)
            remote_gzip_files = json.loads(response.read())
            if not remote_gzip_files:
                raise Exception("Failed to fetch data at '%s'." % url)
            if remote_gzip_files == 'NODATASET':
92abc15b   Goutte   Mistrust the API ...
335
336
337
                raise Exception("API says there's no dataset at '%s'." % url)
            if remote_gzip_files == 'ERROR':
                raise Exception("API returned an error at '%s'." % url)
077980eb   Goutte   Improve availabil...
338
            if remote_gzip_files == ['OUTOFTIME']:  # it happens
80352490   Goutte   Multi model suppo...
339
340
                return []
                # raise Exception("API says it's out of time at '%s'." % url)
b2837a08   Goutte   Add three retries...
341
342
343
            success = True
        except Exception as e:
            log.warn("Failed (%d/3) '%s' : %s" % (retries+1, url, e.message))
92abc15b   Goutte   Mistrust the API ...
344
345
            remote_gzip_files = []
            errors.append(e)
b2837a08   Goutte   Add three retries...
346
347
348
        finally:
            retries += 1
    if not remote_gzip_files:
08abc2d4   Goutte   Remove duplicate ...
349
350
351
352
        abort(400, "Failed to fetch gzip files list for %s at '%s' : %s" %
                   (orbiter, url, errors))
    else:
        remote_gzip_files = list(set(remote_gzip_files))
9bfa6c42   Goutte   More bug hunting.
353
354

    log.debug("Fetched remote gzip files list : %s." % str(remote_gzip_files))
8644387c   Goutte   Use real data.
355

8644387c   Goutte   Use real data.
356
357
    local_gzip_files = []
    for remote_gzip_file in remote_gzip_files:
077980eb   Goutte   Improve availabil...
358
359
360
        # hotfixes to remove when fixed upstream @Myriam
        if remote_gzip_file in ['OUTOFTIME', 'ERROR']:
            continue  # sometimes half the response is okay, the other not
8644387c   Goutte   Use real data.
361
        if remote_gzip_file.endswith('/.gz'):
80352490   Goutte   Multi model suppo...
362
            continue  # this is just a plain bug
8644387c   Goutte   Use real data.
363
        remote_gzip_file = remote_gzip_file.replace('cdpp1', 'cdpp', 1)
077980eb   Goutte   Improve availabil...
364
        ################################################
e18701b6   Goutte   Cache clear (remo...
365
        local_gzip_file = join(CACHE_DIR, get_local_filename(remote_gzip_file))
8644387c   Goutte   Use real data.
366
367
        local_gzip_files.append(local_gzip_file)
        if not isfile(local_gzip_file):
9bfa6c42   Goutte   More bug hunting.
368
            log.debug("Retrieving '%s'..." % local_gzip_file)
8644387c   Goutte   Use real data.
369
            urllib.urlretrieve(remote_gzip_file, local_gzip_file)
9bfa6c42   Goutte   More bug hunting.
370
            log.debug("Retrieved '%s'." % local_gzip_file)
dc0be992   Goutte   Support having no...
371
372
        else:
            log.debug("Found '%s' in the cache." % local_gzip_file)
8644387c   Goutte   Use real data.
373
374
375
376

    local_netc_files = []
    for local_gzip_file in local_gzip_files:
        local_netc_file = local_gzip_file[0:-3]
9bfa6c42   Goutte   More bug hunting.
377
        log.debug("Unzipping '%s'..." % local_gzip_file)
3c064b17   Goutte   Ignore failures w...
378
379
380
381
382
383
384
385
        success = True
        try:
            with gzip.open(local_gzip_file, 'rb') as f:
                file_content = f.read()
                with open(local_netc_file, 'w+b') as g:
                    g.write(file_content)
        except Exception as e:
            success = False
dc0be992   Goutte   Support having no...
386
387
388
389
390
            log.error("Cannot process gz file '%s' from '%s' : %s" %
                      (local_gzip_file, url, e))
            # Sometimes, the downloaded gz is corrupted, and CRC checks fail.
            # We want to delete the local gz file and try again next time.
            removefile(local_gzip_file)
3c064b17   Goutte   Ignore failures w...
391
        if success:
dc0be992   Goutte   Support having no...
392
            local_netc_files.append(local_netc_file)
3c064b17   Goutte   Ignore failures w...
393
            log.debug("Unzipped '%s'." % local_gzip_file)
8644387c   Goutte   Use real data.
394

ea6c8d5d   Goutte   Add interval cons...
395
    return list(set(local_netc_files))  # remove possible dupes
8644387c   Goutte   Use real data.
396
397


180d7d97   Goutte   Refactor heavily.
398
399
400
401
402
def get_data_for_target(target_config, started_at, stopped_at):
    """
    :return: dict whose keys are datetime as str, values tuples of data
    """
    log.debug("Grabbing data for '%s'..." % target_config['slug'])
80352490   Goutte   Multi model suppo...
403

8644387c   Goutte   Use real data.
404
    try:
80352490   Goutte   Multi model suppo...
405
        models = target_config['models']
077980eb   Goutte   Improve availabil...
406
407
    except Exception as e:
        abort(500, "Invalid model configuration for '%s' : %s"
180d7d97   Goutte   Refactor heavily.
408
409
              % (target_config['slug'], str(e)))
    try:
80352490   Goutte   Multi model suppo...
410
        orbits = target_config['orbit']['models']
180d7d97   Goutte   Refactor heavily.
411
412
413
    except Exception as e:
        abort(500, "Invalid orbit configuration for '%s' : %s"
              % (target_config['slug'], str(e)))
28ef3790   Goutte   Clean up.
414

ea6c8d5d   Goutte   Add interval cons...
415
416
417
418
419
420
421
422
423
    # try:
    #     started_at = datetime.datetime.strptime(started_at, FILE_DATE_FMT)
    # except:
    #     abort(400, "Invalid started_at config : '%s'." % started_at)
    # try:
    #     stopped_at = datetime.datetime.strptime(stopped_at, FILE_DATE_FMT)
    # except:
    #     abort(400, "Invalid stopped_at config : '%s'." % stopped_at)

28ef3790   Goutte   Clean up.
424
    # Grab the list of netCDF files from Myriam's API
80352490   Goutte   Multi model suppo...
425
    model_files = []
80352490   Goutte   Multi model suppo...
426
427
428
429
    for model in models:
        model_files = model_files + retrieve_amda_netcdf(
            target_config['slug'], model['slug'], started_at, stopped_at
        )
80352490   Goutte   Multi model suppo...
430
431
    # Remove possible duplicates
    model_files = set(model_files)
80352490   Goutte   Multi model suppo...
432

ea6c8d5d   Goutte   Add interval cons...
433
    # Let's return an empty dict when there's no data instead of crashing
80352490   Goutte   Multi model suppo...
434
435
436
437
438
439
    # if not len(model_files):
    #     abort(500, "No model files found for '%s'." % target_config['slug'])
    # if not len(orbit_files):
    #     abort(500, "No orbit files found for '%s'." % target_config['slug'])

    precision = "%Y-%m-%dT%H"  # model and orbits times are only equal-ish
180d7d97   Goutte   Refactor heavily.
440
    orbit_data = {}  # keys are datetime as str, values arrays of XY
ea6c8d5d   Goutte   Add interval cons...
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

    for orbit in orbits:
        if 'started_at' in orbit:
            s0 = datetime.datetime.strptime(orbit['started_at'], FILE_DATE_FMT)
        else:
            s0 = started_at
        if 'stopped_at' in orbit:
            s1 = datetime.datetime.strptime(orbit['stopped_at'], FILE_DATE_FMT)
        else:
            s1 = stopped_at

        orbit_files = retrieve_amda_netcdf(
            target_config['slug'], orbit['slug'], s0, s1
        )
        for orbit_file in orbit_files:
            log.debug("%s: opening orbit NETCDF4 '%s'..." %
                      (target_config['name'], orbit_file))
            cdf_handle = Dataset(orbit_file, "r", format="NETCDF4")
            times = cdf_handle.variables['Time']  # YYYY DOY HH MM SS .ms
            try:
                data_hee = cdf_handle.variables['HEE']
            except KeyError:
                data_hee = cdf_handle.variables['XYZ_HEE']  # p67 uses this

            log.debug("%s: aggregating data from '%s'..." %
                      (target_config['name'], orbit_file))
            for time, datum_hee in zip(times, data_hee):
                dtime = datetime_from_list(time)
                if s0 <= dtime <= s1:
                    dkey = dtime.strftime(precision)
                    orbit_data[dkey] = datum_hee
            cdf_handle.close()
180d7d97   Goutte   Refactor heavily.
473

8644387c   Goutte   Use real data.
474
475
476
    all_data = {}  # keys are datetime as str, values tuples of data
    for model_file in model_files:
        # Time, StartTime, StopTime, V, B, N, T, Delta_angle, P_dyn
a7ef1487   Goutte   More logs !
477
        log.debug("%s: opening model NETCDF4 '%s'..." %
180d7d97   Goutte   Refactor heavily.
478
                  (target_config['name'], model_file))
8644387c   Goutte   Use real data.
479
480
481
482
483
484
485
486
        cdf_handle = Dataset(model_file, "r", format="NETCDF4")
        times = cdf_handle.variables['Time']  # YYYY DOY HH MM SS .ms
        data_v = cdf_handle.variables['V']
        data_b = cdf_handle.variables['B']
        data_t = cdf_handle.variables['T']
        data_n = cdf_handle.variables['N']
        data_p = cdf_handle.variables['P_dyn']
        data_d = cdf_handle.variables['Delta_angle']
2fedd73b   Goutte   Initial implement...
487
        log.debug("%s: aggregating data from '%s'..." %
180d7d97   Goutte   Refactor heavily.
488
                  (target_config['name'], model_file))
8380e043   Goutte   Fix an awful bug ...
489
        for time, datum_v, datum_b, datum_t, datum_n, datum_p, datum_d \
8644387c   Goutte   Use real data.
490
491
492
493
494
495
                in zip(times, data_v, data_b, data_t, data_n, data_p, data_d):
            vrad = datum_v[0]
            vtan = datum_v[1]
            dtime = datetime_from_list(time)
            if started_at <= dtime <= stopped_at:
                dkey = dtime.strftime(precision)
6491a1f1   Goutte   Fix up the bugs l...
496
497
                x_hee = None
                y_hee = None
180d7d97   Goutte   Refactor heavily.
498
                if dkey in orbit_data:
6491a1f1   Goutte   Fix up the bugs l...
499
500
                    x_hee = orbit_data[dkey][0]
                    y_hee = orbit_data[dkey][1]
8644387c   Goutte   Use real data.
501
502
503
504
                all_data[dkey] = (
                    dtime.strftime("%Y-%m-%dT%H:%M:%S+00:00"),
                    vrad, vtan, sqrt(vrad * vrad + vtan * vtan),
                    datum_b, datum_t, datum_n, datum_p, datum_d,
6491a1f1   Goutte   Fix up the bugs l...
505
                    x_hee, y_hee
8644387c   Goutte   Use real data.
506
507
508
                )
        cdf_handle.close()

180d7d97   Goutte   Refactor heavily.
509
510
511
512
513
514
515
516
517
518
519
520
521
    return all_data


def generate_csv_contents(target_slug, started_at, stopped_at):
    target_config = get_target_config(target_slug)
    log.debug("Crunching CSV contents for '%s'..." % target_config['name'])
    si = StringIO.StringIO()
    cw = csv_writer(si)
    cw.writerow(PROPERTIES)

    all_data = get_data_for_target(target_config, started_at, stopped_at)

    log.debug("Writing and sorting CSV for '%s'..." % target_config['slug'])
8644387c   Goutte   Use real data.
522
523
    for dkey in sorted(all_data):
        cw.writerow(all_data[dkey])
2d2af24b   Goutte   Add a basic orbit...
524

180d7d97   Goutte   Refactor heavily.
525
    log.info("Generated CSV contents for '%s'." % target_config['slug'])
2d2af24b   Goutte   Add a basic orbit...
526
527
    return si.getvalue()

8644387c   Goutte   Use real data.
528

180d7d97   Goutte   Refactor heavily.
529
530
def generate_csv_file_if_needed(target_slug, started_at, stopped_at):
    filename = "%s_%s_%s.csv" % (target_slug,
c0df94bc   Goutte   Adding more logs.
531
532
                                 started_at.strftime(FILE_DATE_FMT),
                                 stopped_at.strftime(FILE_DATE_FMT))
e18701b6   Goutte   Cache clear (remo...
533
    local_csv_file = join(CACHE_DIR, filename)
80352490   Goutte   Multi model suppo...
534
535
536
537
538
539
540
541
542
543
544
545
546

    generate = True
    if isfile(local_csv_file):
        # It need to have more than one line to not be empty (headers)
        with open(local_csv_file) as f:
            cnt = 0
            for _ in f:
                cnt += 1
                if cnt > 1:
                    generate = False
                    break

    if generate:
c0df94bc   Goutte   Adding more logs.
547
548
549
        log.info("Generating CSV '%s'..." % local_csv_file)
        try:
            with open(local_csv_file, mode="w+") as f:
180d7d97   Goutte   Refactor heavily.
550
                f.write(generate_csv_contents(target_slug,
c0df94bc   Goutte   Adding more logs.
551
552
553
554
                                              started_at=started_at,
                                              stopped_at=stopped_at))
            log.info("Generation of '%s' done." % filename)
        except Exception as e:
dc0be992   Goutte   Support having no...
555
            log.error(e)
5ede388f   Goutte   Make sure failed ...
556
            if isfile(local_csv_file):
92abc15b   Goutte   Mistrust the API ...
557
                log.warn("Removing failed CSV '%s'..." % local_csv_file)
5ede388f   Goutte   Make sure failed ...
558
                removefile(local_csv_file)
9bfa6c42   Goutte   More bug hunting.
559
            abort(500, "Failed creating CSV '%s' : %s" % (filename, e))
c0df94bc   Goutte   Adding more logs.
560
561


e18701b6   Goutte   Cache clear (remo...
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
def remove_all_files(in_directory):
    """
    Will throw if something horrible happens.
    Does not remove recursively (could be done with os.walk if needed).
    Does not remove directories either.
    :param in_directory: absolute path to directory
    :return:
    """
    import os

    if not os.path.isdir(in_directory):
        raise ValueError("No directory to clean at '%s'.")

    removed_files = []
    for file_name in os.listdir(in_directory):
        file_path = os.path.join(in_directory, file_name)
        if os.path.isfile(file_path):
            os.remove(file_path)
            removed_files.append(file_path)

    return removed_files


28bb4b28   Goutte   API for the cache...
585
586
def remove_files_created_before(date, in_directory):
    """
077980eb   Goutte   Improve availabil...
587
588
589
    Will throw if something horrible happens.
    Does not remove recursively (could be done with os.walk if needed).
    Does not remove directories either.
28bb4b28   Goutte   API for the cache...
590
    :param date: datetime object
077980eb   Goutte   Improve availabil...
591
    :param in_directory: absolute path to directory
28bb4b28   Goutte   API for the cache...
592
593
594
595
596
597
598
    :return:
    """
    import os
    import time

    secs = time.mktime(date.timetuple())

077980eb   Goutte   Improve availabil...
599
600
    if not os.path.isdir(in_directory):
        raise ValueError("No directory to clean at '%s'.")
28bb4b28   Goutte   API for the cache...
601
602
603
604

    removed_files = []
    for file_name in os.listdir(in_directory):
        file_path = os.path.join(in_directory, file_name)
077980eb   Goutte   Improve availabil...
605
606
607
608
609
        if os.path.isfile(file_path):
            t = os.stat(file_path)
            if t.st_ctime < secs:
                os.remove(file_path)
                removed_files.append(file_path)
28bb4b28   Goutte   API for the cache...
610
611
612
613

    return removed_files


077980eb   Goutte   Improve availabil...
614
615
616
617
618
619
620
621
622
623
624
def get_hit_counter():
    hit_count_path = get_path("../VISITS")

    if isfile(hit_count_path):
        hit_count = int(open(hit_count_path).read())
    else:
        hit_count = 1

    return hit_count


a4a9ef03   Goutte   Cache generated C...
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
def increment_hit_counter():
    hit_count_path = get_path("../VISITS")

    if isfile(hit_count_path):
        hit_count = int(open(hit_count_path).read())
        hit_count += 1
    else:
        hit_count = 1

    hit_counter_file = open(hit_count_path, 'w')
    hit_counter_file.write(str(hit_count))
    hit_counter_file.close()

    return hit_count


077980eb   Goutte   Improve availabil...
641
642
643
tpl_global_vars['visits'] = get_hit_counter()


a4a9ef03   Goutte   Cache generated C...
644
645
646
# ROUTING #####################################################################

@app.route('/favicon.ico')
bde97e4d   Goutte   Add more changes ...
647
def favicon():  # we want it served from the root, not from static/
a4a9ef03   Goutte   Cache generated C...
648
649
650
651
652
653
654
655
656
657
    return send_from_directory(
        join(app.root_path, 'static', 'img'),
        'favicon.ico', mimetype='image/vnd.microsoft.icon'
    )


@app.route("/")
@app.route("/home.html")
@app.route("/index.html")
def home():
077980eb   Goutte   Improve availabil...
658
    increment_hit_counter()
bde97e4d   Goutte   Add more changes ...
659
660
    parameters = PARAMETERS.values()
    parameters.sort(key=lambda x: x['position'])
a4a9ef03   Goutte   Cache generated C...
661
662
    return render_view('home.html.jinja2', {
        'targets': config['targets'],
bde97e4d   Goutte   Add more changes ...
663
        'parameters': parameters,
a4a9ef03   Goutte   Cache generated C...
664
665
666
        'planets': [s for s in config['targets'] if s['type'] == 'planet'],
        'probes':  [s for s in config['targets'] if s['type'] == 'probe'],
        'comets':  [s for s in config['targets'] if s['type'] == 'comet'],
077980eb   Goutte   Improve availabil...
667
        'visits':  get_hit_counter(),
a4a9ef03   Goutte   Cache generated C...
668
669
670
    })


180d7d97   Goutte   Refactor heavily.
671
672
@app.route("/<target>_<started_at>_<stopped_at>.csv")
def download_target_csv(target, started_at, stopped_at):
a4a9ef03   Goutte   Cache generated C...
673
674
675
676
677
    """
    Grab data and orbit data for the specified `target`,
    rearrange it and return it as a CSV file.
    `started_at` and `stopped_at` should be UTC.
    """
180d7d97   Goutte   Refactor heavily.
678
    check_target_config(target)
a4a9ef03   Goutte   Cache generated C...
679
    try:
c0df94bc   Goutte   Adding more logs.
680
        started_at = datetime.datetime.strptime(started_at, FILE_DATE_FMT)
a4a9ef03   Goutte   Cache generated C...
681
682
683
    except:
        abort(400, "Invalid started_at parameter : '%s'." % started_at)
    try:
c0df94bc   Goutte   Adding more logs.
684
        stopped_at = datetime.datetime.strptime(stopped_at, FILE_DATE_FMT)
a4a9ef03   Goutte   Cache generated C...
685
686
687
    except:
        abort(400, "Invalid stopped_at parameter : '%s'." % stopped_at)

180d7d97   Goutte   Refactor heavily.
688
    filename = "%s_%s_%s.csv" % (target,
c0df94bc   Goutte   Adding more logs.
689
690
                                 started_at.strftime(FILE_DATE_FMT),
                                 stopped_at.strftime(FILE_DATE_FMT))
e18701b6   Goutte   Cache clear (remo...
691
    local_csv_file = join(CACHE_DIR, filename)
180d7d97   Goutte   Refactor heavily.
692
    generate_csv_file_if_needed(target, started_at, stopped_at)
a4a9ef03   Goutte   Cache generated C...
693
694
695
    if not isfile(local_csv_file):
        abort(500, "Could not cache CSV file at '%s'." % local_csv_file)

e18701b6   Goutte   Cache clear (remo...
696
    return send_from_directory(CACHE_DIR, filename)
a4a9ef03   Goutte   Cache generated C...
697
698


0511eed7   Goutte   Tarball generatio...
699
700
@app.route("/<targets>_<started_at>_<stopped_at>.tar.gz")
def download_targets_tarball(targets, started_at, stopped_at):
b2837a08   Goutte   Add three retries...
701
    """
bc18b96c   Goutte   Implement first (...
702
703
704
    Grab data and orbit data for each of the specified `targets`,
    in their own CSV file, and make a tarball of them.
    `started_at` and `stopped_at` should be UTC strings.
b2837a08   Goutte   Add three retries...
705

ea6c8d5d   Goutte   Add interval cons...
706
707
    Note: we do not use this route anymore, but let's keep it shelved for now.

2fedd73b   Goutte   Initial implement...
708
    targets: string list of targets' slugs, separated by `-`.
b2837a08   Goutte   Add three retries...
709
    """
2fedd73b   Goutte   Initial implement...
710
    separator = '-'
0511eed7   Goutte   Tarball generatio...
711
712
    targets = targets.split(separator)
    targets.sort()
2fedd73b   Goutte   Initial implement...
713
714
    targets_configs = []
    for target in targets:
b2837a08   Goutte   Add three retries...
715
716
        if not target:
            abort(400, "Invalid targets format : `%s`." % targets)
180d7d97   Goutte   Refactor heavily.
717
        targets_configs.append(get_target_config(target))
2fedd73b   Goutte   Initial implement...
718
    if 0 == len(targets_configs):
b2837a08   Goutte   Add three retries...
719
720
        abort(400, "No valid targets specified. What are you doing?")

b2837a08   Goutte   Add three retries...
721
    try:
ea6c8d5d   Goutte   Add interval cons...
722
        started_at = datetime.datetime.strptime(started_at, FILE_DATE_FMT)
b2837a08   Goutte   Add three retries...
723
724
725
    except:
        abort(400, "Invalid started_at parameter : '%s'." % started_at)
    try:
ea6c8d5d   Goutte   Add interval cons...
726
        stopped_at = datetime.datetime.strptime(stopped_at, FILE_DATE_FMT)
b2837a08   Goutte   Add three retries...
727
728
    except:
        abort(400, "Invalid stopped_at parameter : '%s'." % stopped_at)
ea6c8d5d   Goutte   Add interval cons...
729
730
    sta = started_at.strftime(FILE_DATE_FMT)
    sto = stopped_at.strftime(FILE_DATE_FMT)
b2837a08   Goutte   Add three retries...
731

0511eed7   Goutte   Tarball generatio...
732
    gzip_filename = "%s_%s_%s.tar.gz" % (separator.join(targets), sta, sto)
e18701b6   Goutte   Cache clear (remo...
733
    local_gzip_file = join(CACHE_DIR, gzip_filename)
2fedd73b   Goutte   Initial implement...
734
735

    if not isfile(local_gzip_file):
0511eed7   Goutte   Tarball generatio...
736
        log.debug("Creating the CSV files for the tarball...")
2fedd73b   Goutte   Initial implement...
737
        for target_config in targets_configs:
0511eed7   Goutte   Tarball generatio...
738
            filename = "%s_%s_%s.csv" % (target_config['slug'], sta, sto)
e18701b6   Goutte   Cache clear (remo...
739
            local_csv_file = join(CACHE_DIR, filename)
2fedd73b   Goutte   Initial implement...
740
741
            if not isfile(local_csv_file):
                with open(local_csv_file, mode="w+") as f:
180d7d97   Goutte   Refactor heavily.
742
                    f.write(generate_csv_contents(target_config['slug'],
2fedd73b   Goutte   Initial implement...
743
744
745
                                                  started_at=started_at,
                                                  stopped_at=stopped_at))

0511eed7   Goutte   Tarball generatio...
746
        log.debug("Creating the tarball '%s'..." % local_gzip_file)
2fedd73b   Goutte   Initial implement...
747
748
        with tarfile.open(local_gzip_file, "w:gz") as tar:
            for target_config in targets_configs:
0511eed7   Goutte   Tarball generatio...
749
                filename = "%s_%s_%s.csv" % (target_config['slug'], sta, sto)
e18701b6   Goutte   Cache clear (remo...
750
                local_csv_file = join(CACHE_DIR, filename)
2fedd73b   Goutte   Initial implement...
751
752
753
                tar.add(local_csv_file, arcname=filename)

    if not isfile(local_gzip_file):
0511eed7   Goutte   Tarball generatio...
754
        abort(500, "No tarball to serve. Looked at '%s'." % local_gzip_file)
2fedd73b   Goutte   Initial implement...
755

e18701b6   Goutte   Cache clear (remo...
756
    return send_from_directory(CACHE_DIR, gzip_filename)
b2837a08   Goutte   Add three retries...
757

28bb4b28   Goutte   API for the cache...
758

bc18b96c   Goutte   Implement first (...
759
760
761
762
763
@app.route("/<targets>_<params>_<started_at>_<stopped_at>.nc")
def download_targets_netcdf(targets, params, started_at, stopped_at):
    """
    Grab data and orbit data for the specified `target`,
    rearrange it and return it as a CSV file.
e18701b6   Goutte   Cache clear (remo...
764
    `started_at` and `stopped_at` are expected to be UTC.
bc18b96c   Goutte   Implement first (...
765
766
767
768

    targets: string list of targets' slugs, separated by `-`.
    params: string list of targets' parameters, separated by `-`.
    """
e18701b6   Goutte   Cache clear (remo...
769
    separator = '-'  # /!\ this char should never be in target's slugs
bc18b96c   Goutte   Implement first (...
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
    targets = targets.split(separator)
    targets.sort()
    targets_configs = []
    for target in targets:
        if not target:
            abort(400, "Invalid targets format : `%s`." % targets)
        targets_configs.append(get_target_config(target))
    if 0 == len(targets_configs):
        abort(400, "No valid targets specified. What are you doing?")
    params = params.split(separator)
    params.sort()
    if 0 == len(params):
        abort(400, "No valid parameters specified. What are you doing?")
    if not is_list_in_list(params, PARAMETERS.keys()):
        abort(400, "Some parameters are not recognized in '%s'." % str(params))

57493104   Goutte   Add the time to t...
786
    date_fmt = FILE_DATE_FMT
bc18b96c   Goutte   Implement first (...
787
788
789
790
791
792
793
794
795
796
797
798
799
    try:
        started_at = datetime.datetime.strptime(started_at, date_fmt)
    except:
        abort(400, "Invalid started_at parameter : '%s'." % started_at)
    try:
        stopped_at = datetime.datetime.strptime(stopped_at, date_fmt)
    except:
        abort(400, "Invalid stopped_at parameter : '%s'." % stopped_at)
    sta = started_at.strftime(date_fmt)
    sto = stopped_at.strftime(date_fmt)

    nc_filename = "%s_%s_%s_%s.nc" % \
                  (separator.join(targets), separator.join(params), sta, sto)
e18701b6   Goutte   Cache clear (remo...
800
    nc_path = join(CACHE_DIR, nc_filename)
bc18b96c   Goutte   Implement first (...
801
802
803
804
805

    if not isfile(nc_path):
        log.debug("Creating the NetCDF file '%s'..." % nc_filename)
        nc_handle = Dataset(nc_path, "w", format="NETCDF4")
        try:
ea6c8d5d   Goutte   Add interval cons...
806
            nc_handle.description = "Model and orbit data for targets"  # todo
bc18b96c   Goutte   Implement first (...
807
            nc_handle.history = "Created " + time.ctime(time.time())
ea6c8d5d   Goutte   Add interval cons...
808
            nc_handle.source = "Heliopropa (CDDP)"
bc18b96c   Goutte   Implement first (...
809
810
811
812
813
814
815
            available_params = list(PROPERTIES)
            for target in targets_configs:
                target_slug = target['slug']
                log.debug("Adding group '%s' to the NetCDF..." % target_slug)
                nc_group = nc_handle.createGroup(target_slug)
                data = get_data_for_target(target, started_at, stopped_at)
                dkeys = sorted(data)
ceeb2f4a   Goutte   Add the target co...
816
817
                dimension = 'dim_'+target_slug
                nc_handle.createDimension(dimension, len(dkeys))
57493104   Goutte   Add the time to t...
818
819

                # TIME #
ceeb2f4a   Goutte   Add the target co...
820
                nc_time = nc_group.createVariable('time', 'i8', (dimension,))
57493104   Goutte   Add the time to t...
821
822
823
824
825
826
827
828
829
830
831
832
                nc_time.units = "hours since 1970-01-01 00:00:00"
                nc_time.calendar = "standard"
                times = []
                for dkey in dkeys:
                    time_as_string = data[dkey][0][:-6]  # remove +00:00 tail
                    date = datetime.datetime.strptime(time_as_string, date_fmt)
                    times.append(date2num(
                        date, units=nc_time.units, calendar=nc_time.calendar
                    ))
                nc_time[:] = times

                # SELECTED PARAMETERS #
bc18b96c   Goutte   Implement first (...
833
834
835
836
                nc_vars = []
                indices = []
                for param in params:
                    indices.append(available_params.index(param))
ceeb2f4a   Goutte   Add the target co...
837
                    nc_var = nc_group.createVariable(param, 'f8', (dimension,))
5a6d4498   Goutte   Add a title to ea...
838
                    nc_var.units = PARAMETERS[param]['units']
bc18b96c   Goutte   Implement first (...
839
840
841
842
843
844
845
846
                    nc_vars.append(nc_var)
                for i, nc_var in enumerate(nc_vars):
                    index = indices[i]
                    values = []
                    for dkey in dkeys:
                        dval = data[dkey]
                        values.append(dval[index])
                    nc_var[:] = values
ceeb2f4a   Goutte   Add the target co...
847
848

                # ORBIT #
6491a1f1   Goutte   Fix up the bugs l...
849
                nc_x = nc_group.createVariable('xhee', 'f8', (dimension,))
ceeb2f4a   Goutte   Add the target co...
850
                nc_x.units = 'Au'
6491a1f1   Goutte   Fix up the bugs l...
851
                nc_y = nc_group.createVariable('yhee', 'f8', (dimension,))
ceeb2f4a   Goutte   Add the target co...
852
853
854
                nc_y.units = 'Au'
                values_x = []
                values_y = []
6491a1f1   Goutte   Fix up the bugs l...
855
856
                index_x = available_params.index('xhee')
                index_y = available_params.index('yhee')
ceeb2f4a   Goutte   Add the target co...
857
858
859
860
861
862
863
864
                for dkey in dkeys:
                    dval = data[dkey]
                    values_x.append(dval[index_x])
                    values_y.append(dval[index_y])
                nc_x[:] = values_x
                nc_y[:] = values_y
            log.debug("Writing NetCDF '%s'..." % nc_filename)

bc18b96c   Goutte   Implement first (...
865
        except Exception as e:
57493104   Goutte   Add the time to t...
866
            log.error("Failed to generate NetCDF '%s'." % nc_filename)
bc18b96c   Goutte   Implement first (...
867
868
869
870
871
872
873
            raise e
        finally:
            nc_handle.close()

    if not isfile(nc_path):
        abort(500, "No NetCDF to serve. Looked at '%s'." % nc_path)

e18701b6   Goutte   Cache clear (remo...
874
    return send_from_directory(CACHE_DIR, nc_filename)
bc18b96c   Goutte   Implement first (...
875
876


28bb4b28   Goutte   API for the cache...
877
878
# API #########################################################################

e18701b6   Goutte   Cache clear (remo...
879
880
881
882
883
884
885
886
887
888
889
890
@app.route("/cache/clear")
def cache_clear():
    """
    Removes all files from the cache.
    Note: It also removes the .gitkeep file. Not a problem for prod.
    """
    removed_files = remove_all_files(CACHE_DIR)
    count = len(removed_files)
    return "Cache cleared! Removed %d file%s." \
           % (count, 's' if count != 1 else '')


d9710a98   Goutte   Rename the cleanu...
891
892
@app.route("/cache/cleanup")
def cache_cleanup():
28bb4b28   Goutte   API for the cache...
893
894
    """
    Removes all files from the cache that are older than roughly one month.
e18701b6   Goutte   Cache clear (remo...
895
    Note: It also removes the .gitkeep file. Maybe it should not, but hey.
28bb4b28   Goutte   API for the cache...
896
897
    """
    a_month_ago = datetime.datetime.now() - datetime.timedelta(days=32)
e18701b6   Goutte   Cache clear (remo...
898
    removed_files = remove_files_created_before(a_month_ago, CACHE_DIR)
d9710a98   Goutte   Rename the cleanu...
899
900
901
    count = len(removed_files)
    return "Cache cleaned! Removed %d old file%s." \
           % (count, 's' if count != 1 else '')
28bb4b28   Goutte   API for the cache...
902
903


b500e561   Goutte   Invert the orbits...
904
905
906
907
@app.route("/cache/warmup")
def cache_warmup():
    """
    Warms up the cache for the current day.
927c69c3   Goutte   Make the local ca...
908
    Linked to SpaceWeather#edit in swapp.ls to get the default time interval.
e18701b6   Goutte   Cache clear (remo...
909
    If you edit this code you'll need to edit the other as well and vice versa.
b500e561   Goutte   Invert the orbits...
910
    """
b500e561   Goutte   Invert the orbits...
911
912
913
914
915
    # relativedelta(years=3)
    # startted_at = datetime.datetime.now() - relativedelta(years=3)
    return "To Do"


bde97e4d   Goutte   Add more changes ...
916
917
918
919
920
921
922
@app.route("/run.log")
def run_log():
    with open(get_path('run.log'), 'r') as f:
        contents = f.read()
    return contents


1754789b   Goutte   Decorate and clea...
923
924
925
926
# DEV TOOLS ###################################################################

# @app.route("/inspect")
# def analyze_cdf():
a4a9ef03   Goutte   Cache generated C...
927
#     """
1754789b   Goutte   Decorate and clea...
928
#     For debug purposes.
a4a9ef03   Goutte   Cache generated C...
929
#     """
1754789b   Goutte   Decorate and clea...
930
931
#     cdf_to_inspect = get_path("../res/dummy.nc")
#     cdf_to_inspect = get_path("../res/dummy_jupiter_coordinates.nc")
a4a9ef03   Goutte   Cache generated C...
932
933
#
#     si = StringIO.StringIO()
1754789b   Goutte   Decorate and clea...
934
935
#     cw = csv.DictWriter(si, fieldnames=['Name', 'Shape', 'Length'])
#     cw.writeheader()
a4a9ef03   Goutte   Cache generated C...
936
#
1754789b   Goutte   Decorate and clea...
937
938
939
940
941
942
943
944
945
946
#     # Time, StartTime, StopTime, V, B, N, T, Delta_angle, P_dyn, QualityFlag
#     cdf_handle = Dataset(cdf_to_inspect, "r", format="NETCDF4")
#     for variable in cdf_handle.variables:
#         v = cdf_handle.variables[variable]
#         cw.writerow({
#             'Name': variable,
#             'Shape': v.shape,
#             'Length': v.size,
#         })
#     cdf_handle.close()
a4a9ef03   Goutte   Cache generated C...
947
948
949
950
#
#     return si.getvalue()


9390ec89   Goutte   Initial experimen...
951
952
953
# MAIN ########################################################################

if __name__ == "__main__":
e18701b6   Goutte   Cache clear (remo...
954
    # Debug mode on, as the production server does not use this but run.wsgi
9390ec89   Goutte   Initial experimen...
955
956
    extra_files = [get_path('../config.yml')]
    app.run(debug=True, extra_files=extra_files)