__init__.py 17.1 KB
Newer Older
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
1 2
from future import standard_library
standard_library.install_aliases()
3 4


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
5
from future.utils import with_metaclass
6

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
7
import os.path
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
8
import os
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
9
import re
10
import pickle
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
11
import logging
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
12
import copy
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
13
import pprint
14 15 16 17 18 19

import inspect
import sys
import subprocess
import importlib
import yaml
20
import threading
21

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
22
from .. import models, utils
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
23
from .. import api
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
24 25 26

logger = logging.getLogger(__name__)

27

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
28
class PluginMeta(models.BaseMeta):
29
    _classes = {}
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
30 31 32 33 34 35

    def __new__(mcs, name, bases, attrs, **kwargs):
        plugin_type = []
        if hasattr(bases[0], 'plugin_type'):
            plugin_type += bases[0].plugin_type
        plugin_type.append(name)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
36
        alias = attrs.get('name', name)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
37
        attrs['plugin_type'] = plugin_type
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
38 39 40 41 42 43 44 45
        attrs['name'] = alias
        if 'description' not in attrs:
            doc = attrs.get('__doc__', None)
            if not doc:
                raise Exception(('Please, add a description or '
                                 'documentation to class {}').format(name))
            attrs['description'] = doc
            attrs['name'] = alias
46
        cls = super(PluginMeta, mcs).__new__(mcs, name, bases, attrs)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
47 48 49 50 51 52 53 54

        if alias in mcs._classes:
            if os.environ.get('SENPY_TESTING', ""):
                raise Exception(('The type of plugin {} already exists. '
                                'Please, choose a different name').format(name))
            else:
                logger.warn('Overloading plugin class: {}'.format(alias))
        mcs._classes[alias] = cls
55 56 57 58 59
        return cls

    @classmethod
    def for_type(cls, ptype):
        return cls._classes[ptype]
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
60 61 62


class Plugin(with_metaclass(PluginMeta, models.Plugin)):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
63 64 65 66 67 68 69 70 71 72 73
    '''
    Base class for all plugins in senpy.
    A plugin must provide at least these attributes:

        - version
        - description (or docstring)
        - author

    Additionally, they may provide a URL (url) of a repository or website.

    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
74 75

    def __init__(self, info=None, data_folder=None, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
76 77 78 79
        """
        Provides a canonical name for plugins and serves as base for other
        kinds of plugins.
        """
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
80
        logger.debug("Initialising {}".format(info))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
81
        super(Plugin, self).__init__(**kwargs)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
82 83
        if info:
            self.update(info)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
84
        self.validate()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
85
        self.id = 'plugins/{}_{}'.format(self['name'], self['version'])
86
        self.is_activated = False
87
        self._lock = threading.Lock()
88
        self.data_folder = data_folder or os.getcwd()
89

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
90
    def validate(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
91 92 93 94 95 96
        missing = []
        for x in ['name', 'description', 'version']:
            if x not in self:
                missing.append(x)
        if missing:
            raise models.Error('Missing configuration parameters: {}'.format(missing))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
97

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
98 99 100
    def get_folder(self):
        return os.path.dirname(inspect.getfile(self.__class__))

101 102 103 104 105 106
    def activate(self):
        pass

    def deactivate(self):
        pass

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
107 108 109 110 111 112 113 114
    def test(self, test_cases=None):
        if not test_cases:
            if not hasattr(self, 'test_cases'):
                raise AttributeError(('Plugin {} [{}] does not have any defined '
                                      'test cases').format(self.id,
                                                           inspect.getfile(self.__class__)))
            test_cases = self.test_cases
        for case in test_cases:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
115
            try:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
116
                self.test_case(case)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
117
                logger.debug('Test case passed:\n{}'.format(pprint.pformat(case)))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
118 119
            except Exception as ex:
                logger.warn('Test case failed:\n{}'.format(pprint.pformat(case)))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
120 121
                raise

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
122 123 124
    def test_case(self, case):
        entry = models.Entry(case['entry'])
        given_parameters = case.get('params', case.get('parameters', {}))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
125
        expected = case.get('expected', None)
126
        should_fail = case.get('should_fail', False)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
127 128
        try:
            params = api.parse_params(given_parameters, self.extra_params)
129 130 131 132 133 134 135
            res = list(self.analyse_entries([entry, ], params))

            if not isinstance(expected, list):
                expected = [expected]
            utils.check_template(res, expected)
            for r in res:
                r.validate()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
136
        except models.Error:
137
            if should_fail:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
138
                return
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
139
            raise
140
        assert not should_fail
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
141

142 143 144
    def open(self, fpath, *args, **kwargs):
        if not os.path.isabs(fpath):
            fpath = os.path.join(self.data_folder, fpath)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
145 146
        return open(fpath, *args, **kwargs)

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
147 148
    def serve(self, debug=True, **kwargs):
        utils.easy(plugin_list=[self, ], plugin_folder=None, debug=debug, **kwargs)
149

150

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
151
# For backwards compatibility
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
152 153 154
SenpyPlugin = Plugin


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
155 156 157 158
class Analysis(Plugin):
    '''
    A subclass of Plugin that analyses text and provides an annotation.
    '''
159

160
    def analyse(self, *args, **kwargs):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
161
        raise NotImplementedError(
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
162
            'Your plugin should implement either analyse or analyse_entry')
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
163 164 165 166 167 168 169 170

    def analyse_entry(self, entry, parameters):
        """ An implemented plugin should override this method.
        This base method is here to adapt old style plugins which only
        implement the *analyse* function.
        Note that this method may yield an annotated entry or a list of
        entries (e.g. in a tokenizer)
        """
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
171
        text = entry['nif:isString']
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
172 173 174 175 176
        params = copy.copy(parameters)
        params['input'] = text
        results = self.analyse(**params)
        for i in results.entries:
            yield i
177

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
178 179 180
    def analyse_entries(self, entries, parameters):
        for entry in entries:
            logger.debug('Analysing entry with plugin {}: {}'.format(self, entry))
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
181 182 183 184 185 186 187 188 189 190 191 192 193
            results = self.analyse_entry(entry, parameters)
            if inspect.isgenerator(results):
                for result in results:
                    yield result
            else:
                yield results

    def test_case(self, case):
        if 'entry' not in case and 'input' in case:
            entry = models.Entry(_auto_id=False)
            entry.nif__isString = case['input']
            case['entry'] = entry
        super(Analysis, self).test_case(case)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
194

195

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
196 197 198 199 200 201 202 203
AnalysisPlugin = Analysis


class Conversion(Plugin):
    '''
    A subclass of Plugins that convert between different annotation models.
    e.g. a conversion of emotion models, or normalization of sentiment values.
    '''
204
    pass
205

206

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
207 208 209 210 211 212 213
ConversionPlugin = Conversion


class SentimentPlugin(Analysis, models.SentimentPlugin):
    '''
    Sentiment plugins provide sentiment annotation (using Marl)
    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
214 215
    minPolarityValue = 0
    maxPolarityValue = 1
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
216

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
217 218
    def test_case(self, case):
        if 'polarity' in case:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
219
            expected = case.get('expected', {})
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
220 221 222 223 224 225 226 227 228 229 230 231 232
            s = models.Sentiment(_auto_id=False)
            s.marl__hasPolarity = case['polarity']
            if 'sentiments' not in expected:
                expected['sentiments'] = []
            expected['sentiments'].append(s)
            case['expected'] = expected
        super(SentimentPlugin, self).test_case(case)


class EmotionPlugin(Analysis, models.EmotionPlugin):
    '''
    Emotion plugins provide emotion annotation (using Onyx)
    '''
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
233 234
    minEmotionValue = 0
    maxEmotionValue = 1
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
235 236


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
237 238 239 240
class EmotionConversion(Conversion):
    '''
    A subclass of Conversion that converts emotion annotations using different models
    '''
241
    pass
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
242 243


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
244 245 246
EmotionConversionPlugin = EmotionConversion


247 248 249 250 251 252 253
class Box(AnalysisPlugin):
    '''
    Black box plugins delegate analysis to a function.
    The flow is like so:

    .. code-block::

254
                   entry --> input() --> predict() --> output() --> entry'
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269


    In other words: their ``input`` method convers a query (entry and a set of parameters) into
    the input to the box method. The ``output`` method convers the results given by the box into
    an entry that senpy can handle.
    '''

    def input(self, entry, params=None):
        '''Transforms a query (entry+param) into an input for the black box'''
        return entry

    def output(self, output, entry=None, params=None):
        '''Transforms the results of the black box into an entry'''
        return output

270
    def predict(self, input):
271 272 273 274 275
        raise NotImplementedError('You should define the behavior of this plugin')

    def analyse_entries(self, entries, params):
        for entry in entries:
            input = self.input(entry=entry, params=params)
276
            results = self.predict(input=input)
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
            yield self.output(output=results, entry=entry, params=params)


class TextBox(Box):
    '''A black box plugin that takes only text as input'''

    def input(self, entry, params):
        entry = super(TextBox, self).input(entry, params)
        return entry['nif:isString']


class SentimentBox(TextBox, SentimentPlugin):
    '''
    A box plugin where the output is only a polarity label or a tuple (polarity, polarityValue)
    '''

    def output(self, output, entry, **kwargs):
        s = models.Sentiment()
        try:
            label, value = output
        except ValueError:
            label, value = output, None
        s.prov(self)
        s.polarity = label
        if value is not None:
            s.polarityValue = value
        entry.sentiments.append(s)
        return entry


class EmotionBox(TextBox, EmotionPlugin):
    '''
    A box plugin where the output is only an a tuple of emotion labels
    '''

    def output(self, output, entry, **kwargs):
        if not isinstance(output, list):
            output = [output]
        s = models.EmotionSet()
        entry.emotions.append(s)
        for label in output:
            e = models.Emotion(onyx__hasEmotionCategory=label)
            s.append(e)
        return entry


class MappingMixin(object):

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
325 326 327 328 329 330 331 332
    @property
    def mappings(self):
        return self._mappings

    @mappings.setter
    def mappings(self, value):
        self._mappings = value

333 334 335 336 337 338 339 340
    def output(self, output, entry, params):
        output = self.mappings.get(output,
                                   self.mappings.get('default', output))
        return super(MappingMixin, self).output(output=output,
                                                entry=entry,
                                                params=params)


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
341 342 343
class ShelfMixin(object):
    @property
    def sh(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
344
        if not hasattr(self, '_sh') or self._sh is None:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
345
            self._sh = {}
346
            if os.path.isfile(self.shelf_file):
347
                try:
348
                    with self.open(self.shelf_file, 'rb') as p:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
349
                        self._sh = pickle.load(p)
350
                except (IndexError, EOFError, pickle.UnpicklingError):
351 352 353
                    logger.warning('{} has a corrupted shelf file!'.format(self.id))
                    if not self.get('force_shelf', False):
                        raise
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
354 355 356 357
        return self._sh

    @sh.deleter
    def sh(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
358 359
        if os.path.isfile(self.shelf_file):
            os.remove(self.shelf_file)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
360
            del self._sh
361
        self.save()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
362

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
363 364 365 366
    @sh.setter
    def sh(self, value):
        self._sh = value

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
367 368
    @property
    def shelf_file(self):
369 370 371 372 373 374 375
        if not hasattr(self, '_shelf_file') or not self._shelf_file:
            self._shelf_file = os.path.join(self.data_folder, self.name + '.p')
        return self._shelf_file

    @shelf_file.setter
    def shelf_file(self, value):
        self._shelf_file = value
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
376

377
    def save(self):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
378
        logger.debug('saving pickle')
379
        if hasattr(self, '_sh') and self._sh is not None:
380
            with self.open(self.shelf_file, 'wb') as f:
381
                pickle.dump(self._sh, f)
382 383 384 385 386 387 388 389


def pfilter(plugins, **kwargs):
    """ Filter plugins by different criteria """
    if isinstance(plugins, models.Plugins):
        plugins = plugins.plugins
    elif isinstance(plugins, dict):
        plugins = plugins.values()
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
390
    ptype = kwargs.pop('plugin_type', Plugin)
391 392 393
    logger.debug('#' * 100)
    logger.debug('ptype {}'.format(ptype))
    if ptype:
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
394 395
        if isinstance(ptype, PluginMeta):
            ptype = ptype.__name__
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416
        try:
            ptype = ptype[0].upper() + ptype[1:]
            pclass = globals()[ptype]
            logger.debug('Class: {}'.format(pclass))
            candidates = filter(lambda x: isinstance(x, pclass),
                                plugins)
        except KeyError:
            raise models.Error('{} is not a valid type'.format(ptype))
    else:
        candidates = plugins

    logger.debug(candidates)

    def matches(plug):
        res = all(getattr(plug, k, None) == v for (k, v) in kwargs.items())
        logger.debug(
            "matching {} with {}: {}".format(plug.name, kwargs, res))
        return res

    if kwargs:
        candidates = filter(matches, candidates)
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
417
    return candidates
418 419


420 421 422
def load_module(name, root=None):
    if root:
        sys.path.append(root)
423
    tmp = importlib.import_module(name)
424 425
    if root:
        sys.path.remove(root)
426 427 428
    return tmp


429
def _log_subprocess_output(process):
430 431 432 433 434 435 436
    for line in iter(process.stdout.readline, b''):
        logger.info('%r', line)
    for line in iter(process.stderr.readline, b''):
        logger.error('%r', line)


def install_deps(*plugins):
437
    installed = False
438 439 440
    for info in plugins:
        requirements = info.get('requirements', [])
        if requirements:
441
            pip_args = [sys.executable, '-m', 'pip', 'install', '--use-wheel']
442 443 444 445 446 447
            for req in requirements:
                pip_args.append(req)
            logger.info('Installing requirements: ' + str(requirements))
            process = subprocess.Popen(pip_args,
                                       stdout=subprocess.PIPE,
                                       stderr=subprocess.PIPE)
448
            _log_subprocess_output(process)
449
            exitcode = process.wait()
450
            installed = True
451 452
            if exitcode != 0:
                raise models.Error("Dependencies not properly installed")
453
    return installed
454 455


456 457
is_plugin_file = re.compile(r'.*\.senpy$|senpy_[a-zA-Z0-9_]+\.py$|'
                            '^(?!test_)[a-zA-Z0-9_]+_plugin.py$')
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
458 459


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
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
def find_plugins(folders):
    for search_folder in folders:
        for root, dirnames, filenames in os.walk(search_folder):
            # Do not look for plugins in hidden or special folders
            dirnames[:] = [d for d in dirnames if d[0] not in ['.', '_']]
            for filename in filter(is_plugin_file.match, filenames):
                fpath = os.path.join(root, filename)
                yield fpath


def from_path(fpath, **kwargs):
    logger.debug("Loading plugin from {}".format(fpath))
    if fpath.endswith('.py'):
        # We asume root is the dir of the file, and module is the name of the file
        root = os.path.dirname(fpath)
        module = os.path.basename(fpath)[:-3]
        for instance in _from_module_name(module=module, root=root, **kwargs):
            yield instance
    else:
        info = parse_plugin_info(fpath)
        yield from_info(info, **kwargs)


def from_folder(folders, loader=from_path, **kwargs):
    plugins = []
    for fpath in find_plugins(folders):
        for plugin in loader(fpath, **kwargs):
            plugins.append(plugin)
    return plugins


def from_info(info, root=None, **kwargs):
    if any(x not in info for x in ('module',)):
493
        raise ValueError('Plugin info is not valid: {}'.format(info))
494 495
    module = info["module"]

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
496 497 498 499
    if not root and '_path' in info:
        root = os.path.dirname(info['_path'])

    return one_from_module(module, root=root, info=info, **kwargs)
500 501


502
def parse_plugin_info(fpath):
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
503
    logger.debug("Parsing plugin info: {}".format(fpath))
504 505
    with open(fpath, 'r') as f:
        info = yaml.load(f)
506
    info['_path'] = fpath
J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
507
    return info
508 509


J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
510
def from_module(module, **kwargs):
511

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
512 513 514 515 516 517
    if inspect.ismodule(module):
        res = _from_loaded_module(module, **kwargs)
    else:
        res = _from_module_name(module, **kwargs)
    for p in res:
        yield p
518

J. Fernando Sánchez's avatar
J. Fernando Sánchez committed
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

def one_from_module(module, root, info, **kwargs):
    if '@type' in info:
        cls = PluginMeta.from_type(info['@type'])
        return cls(info=info, **kwargs)
    instance = next(from_module(module=module, root=root, info=info, **kwargs), None)
    if not instance:
        raise Exception("No valid plugin for: {}".format(module))
    return instance


def _classes_in_module(module):
    for _, obj in inspect.getmembers(module):
        if inspect.isclass(obj) and inspect.getmodule(obj) == module:
            logger.debug(("Found plugin class:"
                          " {}@{}").format(obj, inspect.getmodule(obj)))
            yield obj


def _instances_in_module(module):
    for _, obj in inspect.getmembers(module):
        if isinstance(obj, Plugin) and inspect.getmodule(obj) == module:
            logger.debug(("Found plugin instance:"
                          " {}@{}").format(obj, inspect.getmodule(obj)))
            yield obj


def _from_module_name(module, root, info=None, install=True, **kwargs):
    try:
        module = load_module(module, root)
    except ImportError:
        if not install or not info:
            raise
        install_deps(info)
        module = load_module(module, root)
    for plugin in _from_loaded_module(module=module, root=root, info=info, **kwargs):
        yield plugin


def _from_loaded_module(module, info=None, **kwargs):
    for cls in _classes_in_module(module):
        yield cls(info=info, **kwargs)
    for instance in _instances_in_module(module):
        yield instance