1
0
mirror of https://github.com/gsi-upm/senpy synced 2025-10-19 09:48:26 +00:00

Compare commits

...

24 Commits

Author SHA1 Message Date
J. Fernando Sánchez
d145a852e7 remove pytest warning errors
In the previous commit I naively configured pytest to
treat any warning as an error, excep for numpy
pendingdeprecationwarning errors,
This causes the build in py2.7 to fail, due to some
warnings from nltk.
2018-12-07 18:02:01 +01:00
J. Fernando Sánchez
c090501534 New schema for parameters
* Add parameters as an entity in the schema
* Update examples to include parameters
* Change the API for processing plugins, params is a parameter again, instead of
only adding the request.
* Update tests
2018-12-07 16:46:07 +01:00
J. Fernando Sánchez
6a1069780b Draft merge 51-improve parameters
There are some unsolved issues, like representing the mix of analysis+parameters
in a sensible way.
I think we should somehow represent each of the analysis tasks with a unique ID,
and it should contain the specific parameters used.

Right now results.parameters is a mix of a dict with global parameters and a
list with a dict of parameters per plugin.
2018-11-22 18:56:30 +01:00
J. Fernando Sánchez
41aa142ce0 Refactored conversion and postprocessing 2018-11-22 17:27:43 +01:00
J. Fernando Sánchez
b48730137d Remove makefiles from auto push/pull 2018-11-06 17:12:54 +01:00
J. Fernando Sánchez
f1ec057b16 Add fetch to makefiles push 2018-11-06 17:02:59 +01:00
J. Fernando Sánchez
f6ca82cac8 Merge branch '56-exception-when-using-post' into 'master'
Replace algorithm list with a tuple

Closes #56

See merge request senpy/senpy!25
2018-11-06 14:56:00 +00:00
J. Fernando Sánchez
318acd5a71 Replace algorithm list with a tuple 2018-11-06 15:52:05 +01:00
J. Fernando Sánchez
c8f6f5613d Change CI to include make push
This replaces the makes for each python version with a simple `make push`.
It will also add a "main image" for each version, i.e. `gsiupm/senpy:1.0.0` in
addition to `gsiupm/senpy:1.0.0-python2.7` and `gsiupm/senpy:1.0.0-python3.5`.
2018-10-30 17:45:44 +01:00
J. Fernando Sánchez
748d1a00bd Fix bug in POST 2018-10-30 16:35:17 +01:00
J. Fernando Sánchez
a82e4ed440 Fix bug in py3.5 2018-10-30 16:14:06 +01:00
J. Fernando Sánchez
c939b095de Fix POST. Closes senpy/senpy#56 2018-10-30 15:15:37 +01:00
J. Fernando Sánchez
ca69bddc17 Improve extra requirement handling
This commit adds a new mechanism to handle parameters beforehand in chained
calls, and the ability to get help on available parameters in chained
calls (through `?help`).
It also includes tests for this feature.

Closes #51
2018-08-20 15:44:54 +02:00
J. Fernando Sánchez
aa35e62a27 Avoid duplication in split plugin 2018-08-20 14:07:33 +02:00
J. Fernando Sánchez
6dd4a44924 Make algorithm part of the URI
This also includes a couple of changes URIs to pass the tests with python 3.7

Closes #50
2018-08-17 11:01:56 +02:00
J. Fernando Sánchez
4291c5eabf Fix typo in requirements 2018-07-23 19:19:05 +02:00
J. Fernando Sánchez
7c7a815d1a Add *responses* to improve mocking 2018-07-23 19:07:57 +02:00
J. Fernando Sánchez
a3eb8f196c Several changes
* Add flag to run tests (and exit, or run the server)
* Add ntriples outformat
* Modify dependency installation logic to avoid installing several times
* Add encoded URLs as base/prefix
* Allow plugin activation to fail
2018-07-04 16:24:42 +02:00
J. Fernando Sánchez
00ffbb3804 Several changes
* Add flag to run tests
* Add ntriples outformat
2018-07-04 16:14:09 +02:00
J. Fernando Sánchez
13cf0c71c5 WIP
* Modify dependency installation logic (avoid installing several times)
* Add encoded URLs for as base/prefix
2018-06-28 18:24:18 +02:00
J. Fernando Sánchez
e5662d482e Allow activation fails 2018-06-20 11:51:06 +02:00
J. Fernando Sánchez
61181db199 Fix sentiment140 plugin 2018-06-18 17:43:10 +02:00
J. Fernando Sánchez
a1663a3f31 Upload latest with version 2018-06-18 17:36:30 +02:00
J. Fernando Sánchez
83b23dbdf4 UI improvements
* Add option to add multiple plugins
* Improve UI hints for collapsed parameters
* Refactored plugins without requirements
* Hide evaluation tab for the moment. You can see it by adding "?evaluation" to
  the URL.
2018-06-18 16:46:49 +02:00
60 changed files with 1806 additions and 810 deletions

View File

@@ -31,29 +31,19 @@ test-2.7:
variables:
PYTHON_VERSION: "2.7"
.image: &image_definition
push:
stage: push
script:
- make -e push-$PYTHON_VERSION
- make -e push
only:
- tags
- triggers
- fix-makefiles
push-3.5:
<<: *image_definition
variables:
PYTHON_VERSION: "3.5"
push-2.7:
<<: *image_definition
variables:
PYTHON_VERSION: "2.7"
push-latest:
<<: *image_definition
variables:
PYTHON_VERSION: latest
stage: push
script:
- make -e push-latest
only:
- master
- triggers

View File

@@ -22,7 +22,4 @@ else
rm $(KEY_FILE)
endif
push:: git-push
pull:: git-pull
.PHONY:: commit tag push git-push git-pull push-github
.PHONY:: commit tag git-push git-pull push-github

View File

@@ -1,17 +1,15 @@
makefiles-remote:
@git remote add makefiles ssh://git@lab.cluster.gsi.dit.upm.es:2200/docs/templates/makefiles.git 2>/dev/null || true
git ls-remote --exit-code makefiles 2> /dev/null || git remote add makefiles ssh://git@lab.cluster.gsi.dit.upm.es:2200/docs/templates/makefiles.git
makefiles-commit: makefiles-remote
git add -f .makefiles
git commit -em "Updated makefiles from ${NAME}"
makefiles-push:
git fetch makefiles $(NAME)
git subtree push --prefix=.makefiles/ makefiles $(NAME)
makefiles-pull: makefiles-remote
git subtree pull --prefix=.makefiles/ makefiles master --squash
pull:: makefiles-pull
push:: makefiles-push
.PHONY:: makefiles-remote makefiles-commit makefiles-push makefiles-pull pull push
.PHONY:: makefiles-remote makefiles-commit makefiles-push makefiles-pull

View File

@@ -76,8 +76,9 @@ pip_upload: pip_test ## Upload package to pip
push-latest: $(addprefix push-latest-,$(PYVERSIONS)) ## Push the "latest" tag to dockerhub
docker tag '$(IMAGEWTAG)-python$(PYMAIN)' '$(IMAGEWTAG)'
docker tag '$(IMAGEWTAG)-python$(PYMAIN)' '$(IMAGENAME)'
docker tag '$(IMAGEWTAG)-python$(PYMAIN)' '$(IMAGENAME):latest'
docker push '$(IMAGENAME):latest'
docker push '$(IMAGEWTAG)'
push-latest-%: build-% ## Push the latest image for a specific python version
docker tag $(IMAGENAME):$(VERSION)-python$* $(IMAGENAME):python$*

10
docker-compose.dev.yml Normal file
View File

@@ -0,0 +1,10 @@
version: '3'
services:
senpy:
image: "${IMAGENAME-gsiupm/senpy}:${VERSION-latest}"
entrypoint: ["/bin/bash"]
working_dir: "/senpy-plugins"
ports:
- 5000:5000
volumes:
- ".:/usr/src/app/"

9
docker-compose.test.yml Normal file
View File

@@ -0,0 +1,9 @@
version: '3'
services:
test:
image: "${IMAGENAME-gsiupm/senpy}:${VERSION-dev}"
entrypoint: ["py.test"]
volumes:
- ".:/usr/src/app/"
command:
[]

11
docker-compose.yml Normal file
View File

@@ -0,0 +1,11 @@
version: '3'
services:
senpy:
image: "${IMAGENAME-gsiupm/senpy}:${VERSION-dev}"
build:
context: .
dockerfile: Dockerfile${PYVERSION--2.7}
ports:
- 5001:5000
volumes:
- "./data:/data"

View File

@@ -6,13 +6,9 @@
],
"entries": [
{
"@type": [
"nif:RFC5147String",
"nif:Context"
],
"nif:beginIndex": 0,
"nif:endIndex": 40,
"nif:isString": "My favourite actress is Natalie Portman"
"text": "An entry should have a nif:isString key"
}
]
}

View File

@@ -3,10 +3,21 @@
"@id": "me:Result1",
"@type": "results",
"analysis": [
"me:SAnalysis1",
"me:SgAnalysis1",
"me:EmotionAnalysis1",
"me:NER1"
{
"@id": "_:SAnalysis1_Activity",
"@type": "marl:SentimentAnalysis",
"prov:wasAssociatedWith": "me:SAnalysis1"
},
{
"@id": "_:EmotionAnalysis1_Activity",
"@type": "onyx:EmotionAnalysis",
"prov:wasAssociatedWith": "me:EmotionAnalysis1"
},
{
"@id": "_:NER1_Activity",
"@type": "me:NER",
"prov:wasAssociatedWith": "me:NER1"
}
],
"entries": [
{
@@ -23,7 +34,7 @@
"nif:endIndex": 13,
"nif:anchorOf": "Microsoft",
"me:references": "http://dbpedia.org/page/Microsoft",
"prov:wasGeneratedBy": "me:NER1"
"prov:wasGeneratedBy": "_:NER1_Activity"
},
{
"@id": "http://micro.blog/status1#char=25,37",
@@ -31,7 +42,7 @@
"nif:endIndex": 37,
"nif:anchorOf": "Windows Phone",
"me:references": "http://dbpedia.org/page/Windows_Phone",
"prov:wasGeneratedBy": "me:NER1"
"prov:wasGeneratedBy": "_:NER1_Activity"
}
],
"suggestions": [
@@ -40,7 +51,7 @@
"nif:beginIndex": 16,
"nif:endIndex": 77,
"nif:anchorOf": "put your Windows Phone on your newest #open technology program",
"prov:wasGeneratedBy": "me:SgAnalysis1"
"prov:wasGeneratedBy": "_:SgAnalysis1_Activity"
}
],
"sentiments": [
@@ -51,14 +62,14 @@
"nif:anchorOf": "You'll be awesome.",
"marl:hasPolarity": "marl:Positive",
"marl:polarityValue": 0.9,
"prov:wasGeneratedBy": "me:SAnalysis1"
"prov:wasGeneratedBy": "_:SgAnalysis1_Activity"
}
],
"emotions": [
{
"@id": "http://micro.blog/status1#char=0,109",
"nif:anchorOf": "Dear Microsoft, put your Windows Phone on your newest #open technology program. You'll be awesome. #opensource",
"prov:wasGeneratedBy": "me:EAnalysis1",
"prov:wasGeneratedBy": "_:EmotionAnalysis1_Activity",
"onyx:hasEmotion": [
{
"onyx:hasEmotionCategory": "wna:liking"

View File

@@ -1,78 +0,0 @@
{
"@context": "http://mixedemotions-project.eu/ns/context.jsonld",
"@id": "me:Result1",
"@type": "results",
"analysis": [
"me:SAnalysis1",
"me:SgAnalysis1",
"me:EmotionAnalysis1",
"me:NER1",
{
"@type": "analysis",
"@id": "anonymous"
}
],
"entries": [
{
"@id": "http://micro.blog/status1",
"@type": [
"nif:RFC5147String",
"nif:Context"
],
"nif:isString": "Dear Microsoft, put your Windows Phone on your newest #open technology program. You'll be awesome. #opensource",
"entities": [
{
"@id": "http://micro.blog/status1#char=5,13",
"nif:beginIndex": 5,
"nif:endIndex": 13,
"nif:anchorOf": "Microsoft",
"me:references": "http://dbpedia.org/page/Microsoft",
"prov:wasGeneratedBy": "me:NER1"
},
{
"@id": "http://micro.blog/status1#char=25,37",
"nif:beginIndex": 25,
"nif:endIndex": 37,
"nif:anchorOf": "Windows Phone",
"me:references": "http://dbpedia.org/page/Windows_Phone",
"prov:wasGeneratedBy": "me:NER1"
}
],
"suggestions": [
{
"@id": "http://micro.blog/status1#char=16,77",
"nif:beginIndex": 16,
"nif:endIndex": 77,
"nif:anchorOf": "put your Windows Phone on your newest #open technology program",
"prov:wasGeneratedBy": "me:SgAnalysis1"
}
],
"sentiments": [
{
"@id": "http://micro.blog/status1#char=80,97",
"nif:beginIndex": 80,
"nif:endIndex": 97,
"nif:anchorOf": "You'll be awesome.",
"marl:hasPolarity": "marl:Positive",
"marl:polarityValue": 0.9,
"prov:wasGeneratedBy": "me:SAnalysis1"
}
],
"emotions": [
{
"@id": "http://micro.blog/status1#char=0,109",
"nif:anchorOf": "Dear Microsoft, put your Windows Phone on your newest #open technology program. You'll be awesome. #opensource",
"prov:wasGeneratedBy": "me:EAnalysis1",
"onyx:hasEmotion": [
{
"onyx:hasEmotionCategory": "wna:liking"
},
{
"onyx:hasEmotionCategory": "wna:excitement"
}
]
}
]
}
]
}

View File

@@ -1,9 +1,8 @@
{
"@context": "http://mixedemotions-project.eu/ns/context.jsonld",
"@id": "http://example.com#NIFExample",
"@id": "me:Result1",
"@type": "results",
"analysis": [
],
"analysis": [ ],
"entries": [
{
"@id": "http://example.org#char=0,40",

View File

@@ -4,22 +4,34 @@
"@type": "results",
"analysis": [
{
"@id": "me:SAnalysis1",
"@id": "_:SAnalysis1_Activity",
"@type": "marl:SentimentAnalysis",
"marl:maxPolarityValue": 1,
"marl:minPolarityValue": 0
"prov:wasAssociatedWith": "me:SentimentAnalysis",
"prov:used": [
{
"name": "marl:maxPolarityValue",
"prov:value": "1"
},
{
"@id": "me:SgAnalysis1",
"name": "marl:minPolarityValue",
"prov:value": "0"
}
]
},
{
"@id": "_:SgAnalysis1_Activity",
"prov:wasAssociatedWith": "me:SgAnalysis1",
"@type": "me:SuggestionAnalysis"
},
{
"@id": "me:EmotionAnalysis1",
"@type": "me:EmotionAnalysis"
"@id": "_:EmotionAnalysis1_Activity",
"@type": "me:EmotionAnalysis",
"prov:wasAssociatedWith": "me:EmotionAnalysis1"
},
{
"@id": "me:NER1",
"@type": "me:NER"
"@id": "_:NER1_Activity",
"@type": "me:NER",
"prov:wasAssociatedWith": "me:EmotionNER1"
}
],
"entries": [

View File

@@ -4,8 +4,9 @@
"@type": "results",
"analysis": [
{
"@id": "me:EmotionAnalysis1",
"@type": "onyx:EmotionAnalysis"
"@id": "me:EmotionAnalysis1_Activity",
"@type": "me:EmotionAnalysis1",
"prov:wasAssociatedWith": "me:EmotionAnalysis1"
}
],
"entries": [
@@ -26,7 +27,7 @@
{
"@id": "http://micro.blog/status1#char=0,109",
"nif:anchorOf": "Dear Microsoft, put your Windows Phone on your newest #open technology program. You'll be awesome. #opensource",
"prov:wasGeneratedBy": "me:EmotionAnalysis1",
"prov:wasGeneratedBy": "_:EmotionAnalysis1_Activity",
"onyx:hasEmotion": [
{
"onyx:hasEmotionCategory": "wna:liking"

View File

@@ -4,8 +4,9 @@
"@type": "results",
"analysis": [
{
"@id": "me:NER1",
"@type": "me:NERAnalysis"
"@id": "_:NER1_Activity",
"@type": "me:NERAnalysis",
"prov:wasAssociatedWith": "me:NER1"
}
],
"entries": [

View File

@@ -9,9 +9,15 @@
"@type": "results",
"analysis": [
{
"@id": "me:HesamsAnalysis",
"@id": "me:HesamsAnalysis_Activity",
"@type": "onyx:EmotionAnalysis",
"onyx:usesEmotionModel": "emovoc:pad-dimensions"
"prov:wasAssociatedWith": "me:HesamsAnalysis",
"prov:used": [
{
"name": "emotion-model",
"prov:value": "emovoc:pad-dimensions"
}
]
}
],
"entries": [
@@ -32,7 +38,7 @@
{
"@id": "Entry1#char=0,21",
"nif:anchorOf": "This is a test string",
"prov:wasGeneratedBy": "me:HesamAnalysis",
"prov:wasGeneratedBy": "_:HesamAnalysis_Activity",
"onyx:hasEmotion": [
{
"emovoc:pleasure": 0.5,

View File

@@ -4,10 +4,9 @@
"@type": "results",
"analysis": [
{
"@id": "me:SAnalysis1",
"@id": "_:SAnalysis1_Activity",
"@type": "marl:SentimentAnalysis",
"marl:maxPolarityValue": 1,
"marl:minPolarityValue": 0
"prov:wasAssociatedWith": "me:SAnalysis1"
}
],
"entries": [
@@ -30,7 +29,7 @@
"nif:anchorOf": "You'll be awesome.",
"marl:hasPolarity": "marl:Positive",
"marl:polarityValue": 0.9,
"prov:wasGeneratedBy": "me:SAnalysis1"
"prov:wasGeneratedBy": "_:SAnalysis1_Activity"
}
],
"emotionSets": [

View File

@@ -3,7 +3,11 @@
"@id": "me:Result1",
"@type": "results",
"analysis": [
"me:SgAnalysis1"
{
"@id": "_:SgAnalysis1_Activity",
"@type": "me:SuggestionAnalysis",
"prov:wasAssociatedWith": "me:SgAnalysis1"
}
],
"entries": [
{
@@ -12,7 +16,6 @@
"nif:RFC5147String",
"nif:Context"
],
"prov:wasGeneratedBy": "me:SAnalysis1",
"nif:isString": "Dear Microsoft, put your Windows Phone on your newest #open technology program. You'll be awesome. #opensource",
"entities": [
],
@@ -22,7 +25,7 @@
"nif:beginIndex": 16,
"nif:endIndex": 77,
"nif:anchorOf": "put your Windows Phone on your newest #open technology program",
"prov:wasGeneratedBy": "me:SgAnalysis1"
"prov:wasGeneratedBy": "_:SgAnalysis1_Activity"
}
],
"sentiments": [

View File

@@ -11,7 +11,7 @@ class Async(AnalysisPlugin):
'''An example of an asynchronous module'''
author = '@balkian'
version = '0.2'
async = True
sync = False
def _do_async(self, num_processes):
pool = multiprocessing.Pool(processes=num_processes)

View File

@@ -43,7 +43,6 @@ class Dictionary(plugins.SentimentPlugin):
class EmojiOnly(Dictionary):
'''Sentiment annotation with a basic lexicon of emojis'''
description = 'A plugin'
dictionaries = [basic.emojis]
test_cases = [{

View File

@@ -12,3 +12,4 @@ rdflib-jsonld
numpy
scipy
scikit-learn
responses

View File

@@ -78,10 +78,15 @@ def main():
help='Do not run a server, only install plugin dependencies')
parser.add_argument(
'--only-test',
'-t',
action='store_true',
default=False,
help='Do not run a server, just test all plugins')
parser.add_argument(
'--test',
'-t',
action='store_true',
default=False,
help='Test all plugins before launching the server')
parser.add_argument(
'--only-list',
'--list',
@@ -99,12 +104,24 @@ def main():
action='store_false',
default=True,
help='Run a threaded server')
parser.add_argument(
'--no-deps',
'-n',
action='store_true',
default=False,
help='Skip installing dependencies')
parser.add_argument(
'--version',
'-v',
action='store_true',
default=False,
help='Output the senpy version and exit')
parser.add_argument(
'--allow-fail',
'--fail',
action='store_true',
default=False,
help='Do not exit if some plugins fail to activate')
args = parser.parse_args()
if args.version:
print('Senpy version {}'.format(senpy.__version__))
@@ -119,18 +136,26 @@ def main():
data_folder=args.data_folder)
if args.only_list:
plugins = sp.plugins()
maxwidth = max(len(x.id) for x in plugins)
maxname = max(len(x.name) for x in plugins)
maxversion = max(len(x.version) for x in plugins)
print('Found {} plugins:'.format(len(plugins)))
for plugin in plugins:
import inspect
fpath = inspect.getfile(plugin.__class__)
print('{: <{width}} @ {}'.format(plugin.id, fpath, width=maxwidth))
print('\t{: <{maxname}} @ {: <{maxversion}} -> {}'.format(plugin.name,
plugin.version,
fpath,
maxname=maxname,
maxversion=maxversion))
return
if not args.no_deps:
sp.install_deps()
if args.only_install:
return
sp.activate_all()
if args.only_test:
sp.activate_all(allow_fail=args.allow_fail)
if args.test or args.only_test:
easy_test(sp.plugins(), debug=args.debug)
if args.only_test:
return
print('Senpy version {}'.format(senpy.__version__))
print('Server running on port %s:%d. Ctrl+C to quit' % (args.host,

View File

@@ -1,26 +1,29 @@
from future.utils import iteritems
from .models import Error, Results, Entry, from_string
from .models import Analysis, Error, Results, Entry, from_string
import logging
logger = logging.getLogger(__name__)
boolean = [True, False]
API_PARAMS = {
"algorithm": {
"aliases": ["algorithms", "a", "algo"],
"required": False,
"required": True,
"default": 'default',
"description": ("Algorithms that will be used to process the request."
"It may be a list of comma-separated names."),
},
"expanded-jsonld": {
"@id": "expanded-jsonld",
"aliases": ["expanded"],
"options": "boolean",
"options": boolean,
"required": True,
"default": False
},
"with_parameters": {
"aliases": ['withparameters',
'with-parameters'],
"options": "boolean",
"options": boolean,
"default": False,
"required": True
},
@@ -29,16 +32,24 @@ API_PARAMS = {
"aliases": ["o"],
"default": "json-ld",
"required": True,
"options": ["json-ld", "turtle"],
"options": ["json-ld", "turtle", "ntriples"],
},
"help": {
"@id": "help",
"description": "Show additional help to know more about the possible parameters",
"aliases": ["h"],
"required": True,
"options": "boolean",
"options": boolean,
"default": False
},
"verbose": {
"@id": "verbose",
"description": "Show all help, including the common API parameters, or only plugin-related info",
"aliases": ["v"],
"required": True,
"options": boolean,
"default": True
},
"emotionModel": {
"@id": "emotionModel",
"aliases": ["emoModel"],
@@ -83,7 +94,7 @@ WEB_PARAMS = {
"aliases": ["headers"],
"required": True,
"default": False,
"options": "boolean"
"options": boolean
},
}
@@ -132,10 +143,19 @@ NIF_PARAMS = {
"aliases": ["u"],
"required": False,
"default": "RFC5147String",
"options": "RFC5147String"
"options": ["RFC5147String", ]
}
}
BUILTIN_PARAMS = {}
for d in [
NIF_PARAMS, CLI_PARAMS, WEB_PARAMS, PLUGINS_PARAMS, EVAL_PARAMS,
API_PARAMS
]:
for k, v in d.items():
BUILTIN_PARAMS[k] = v
def parse_params(indict, *specs):
if not specs:
@@ -157,10 +177,9 @@ def parse_params(indict, *specs):
outdict[param] = options["default"]
elif options.get("required", False):
wrong_params[param] = spec[param]
continue
if "options" in options:
if options["options"] == "boolean":
outdict[param] = outdict[param] in [None, True, 'true', '1']
elif "options" in options:
if options["options"] == boolean:
outdict[param] = str(outdict[param]).lower() in ['true', '1']
elif outdict[param] not in options["options"]:
wrong_params[param] = spec[param]
if wrong_params:
@@ -171,30 +190,126 @@ def parse_params(indict, *specs):
parameters=outdict,
errors=wrong_params)
raise message
if 'algorithm' in outdict and not isinstance(outdict['algorithm'], list):
outdict['algorithm'] = outdict['algorithm'].split(',')
return outdict
def parse_extra_params(request, plugin=None):
params = request.parameters.copy()
if plugin:
extra_params = parse_params(params, plugin.get('extra_params', {}))
params.update(extra_params)
def get_all_params(plugins, *specs):
'''Return a list of parameters for a given set of specifications and plugins.'''
dic = {}
for s in specs:
dic.update(s)
dic.update(get_extra_params(plugins))
return dic
def get_extra_params(plugins):
'''Get a list of possible parameters given a list of plugins'''
params = {}
extra_params = {}
for plugin in plugins:
this_params = plugin.get('extra_params', {})
for k, v in this_params.items():
if k not in extra_params:
extra_params[k] = {}
extra_params[k][plugin.name] = v
for k, v in extra_params.items(): # Resolve conflicts
if len(v) == 1: # Add the extra options that do not collide
params[k] = list(v.values())[0]
else:
required = False
aliases = None
options = None
default = None
nodefault = False # Set when defaults are not compatible
for plugin, opt in v.items():
params['{}.{}'.format(plugin, k)] = opt
required = required or opt.get('required', False)
newaliases = set(opt.get('aliases', []))
if aliases is None:
aliases = newaliases
else:
aliases = aliases & newaliases
if 'options' in opt:
newoptions = set(opt['options'])
options = newoptions if options is None else options & newoptions
if 'default' in opt:
newdefault = opt['default']
if newdefault:
if default is None and not nodefault:
default = newdefault
elif newdefault != default:
nodefault = True
default = None
# Check for incompatibilities
if options != set():
params[k] = {
'default': default,
'aliases': list(aliases),
'required': required,
'options': list(options)
}
return params
def parse_analysis(params, plugins):
'''
Parse the given parameters individually for each plugin, and get a list of the parameters that
belong to each of the plugins. Each item can then be used in the plugin.analyse_entries method.
'''
analysis_list = []
for i, plugin in enumerate(plugins):
if not plugin:
continue
this_params = filter_params(params, plugin, i)
parsed = parse_params(this_params, plugin.get('extra_params', {}))
analysis = plugin.activity(parsed)
analysis_list.append(analysis)
return analysis_list
def filter_params(params, plugin, ith=-1):
'''
Get the values within params that apply to a plugin.
More specific names override more general names, in this order:
<index_order>.parameter > <plugin.name>.parameter > parameter
Example:
>>> filter_params({'0.hello': True, 'hello': False}, Plugin(), 0)
{ '0.hello': True, 'hello': True}
'''
thisparams = {}
if ith >= 0:
ith = '{}.'.format(ith)
else:
ith = ""
for k, v in params.items():
if ith and k.startswith(str(ith)):
thisparams[k[len(ith):]] = v
elif k.startswith(plugin.name):
thisparams[k[len(plugin.name) + 1:]] = v
elif k not in thisparams:
thisparams[k] = v
return thisparams
def parse_call(params):
'''Return a results object based on the parameters used in a call/request.
'''
Return a results object based on the parameters used in a call/request.
'''
params = parse_params(params, NIF_PARAMS)
if params['informat'] == 'text':
results = Results()
entry = Entry(nif__isString=params['input'])
entry = Entry(nif__isString=params['input'], id='#') # Use @base
results.entries.append(entry)
elif params['informat'] == 'json-ld':
results = from_string(params['input'], cls=Results)
else: # pragma: no cover
raise NotImplementedError('Informat {} is not implemented'.format(params['informat']))
raise NotImplementedError('Informat {} is not implemented'.format(
params['informat']))
results.parameters = params
return results

View File

@@ -18,15 +18,15 @@
Blueprints for Senpy
"""
from flask import (Blueprint, request, current_app, render_template, url_for,
jsonify)
jsonify, redirect)
from .models import Error, Response, Help, Plugins, read_schema, dump_schema, Datasets
from . import api
from .version import __version__
from functools import wraps
import logging
import traceback
import json
import base64
logger = logging.getLogger(__name__)
@@ -34,6 +34,24 @@ api_blueprint = Blueprint("api", __name__)
demo_blueprint = Blueprint("demo", __name__, template_folder='templates')
ns_blueprint = Blueprint("ns", __name__)
_mimetypes_r = {'json-ld': ['application/ld+json'],
'turtle': ['text/turtle'],
'ntriples': ['application/n-triples'],
'text': ['text/plain']}
MIMETYPES = {}
for k, vs in _mimetypes_r.items():
for v in vs:
if v in MIMETYPES:
raise Exception('MIMETYPE {} specified for two formats: {} and {}'.format(v,
v,
MIMETYPES[v]))
MIMETYPES[v] = k
DEFAULT_MIMETYPE = 'application/ld+json'
DEFAULT_FORMAT = 'json-ld'
def get_params(req):
if req.method == 'POST':
@@ -45,22 +63,60 @@ def get_params(req):
return indict
def encoded_url(url=None, base=None):
code = ''
if not url:
if request.method == 'GET':
url = request.full_path[1:] # Remove the first slash
else:
hash(frozenset(tuple(request.parameters.items())))
code = 'hash:{}'.format(hash)
code = code or base64.urlsafe_b64encode(url.encode()).decode()
if base:
return base + code
return url_for('api.decode', code=code, _external=True)
def decoded_url(code, base=None):
if code.startswith('hash:'):
raise Exception('Can not decode a URL for a POST request')
base = base or request.url_root
path = base64.urlsafe_b64decode(code.encode()).decode()
return base + path
@demo_blueprint.route('/')
def index():
return render_template("index.html", version=__version__)
ev = str(get_params(request).get('evaluation', False))
evaluation_enabled = ev.lower() not in ['false', 'no', 'none']
return render_template("index.html",
evaluation=evaluation_enabled,
version=__version__)
@api_blueprint.route('/contexts/<entity>.jsonld')
def context(entity="context"):
context = Response._context
context['@vocab'] = url_for('ns.index', _external=True)
context['endpoint'] = url_for('api.api_root', _external=True)
return jsonify({"@context": context})
@api_blueprint.route('/d/<code>')
def decode(code):
try:
return redirect(decoded_url(code))
except Exception:
return Error('invalid URL').flask()
@ns_blueprint.route('/') # noqa: F811
def index():
context = Response._context
context['@vocab'] = url_for('.ns', _external=True)
context = Response._context.copy()
context['endpoint'] = url_for('api.api_root', _external=True)
return jsonify({"@context": context})
@@ -76,7 +132,7 @@ def basic_api(f):
default_params = {
'inHeaders': False,
'expanded-jsonld': False,
'outformat': 'json-ld',
'outformat': None,
'with_parameters': True,
}
@@ -95,43 +151,66 @@ def basic_api(f):
request.parameters = params
response = f(*args, **kwargs)
except (Exception) as ex:
if current_app.debug:
if current_app.debug or current_app.config['TESTING']:
raise
if not isinstance(ex, Error):
msg = "{}:\n\t{}".format(ex,
traceback.format_exc())
msg = "{}".format(ex)
ex = Error(message=msg, status=500)
logger.exception('Error returning analysis result')
response = ex
response.parameters = raw_params
logger.error(ex)
logger.exception(ex)
if 'parameters' in response and not params['with_parameters']:
del response.parameters
logger.info('Response: {}'.format(response))
mime = request.accept_mimetypes\
.best_match(MIMETYPES.keys(),
DEFAULT_MIMETYPE)
mimeformat = MIMETYPES.get(mime, DEFAULT_FORMAT)
outformat = params['outformat'] or mimeformat
return response.flask(
in_headers=params['inHeaders'],
headers=headers,
prefix=url_for('.api_root', _external=True),
prefix=params.get('prefix', encoded_url()),
context_uri=url_for('api.context',
entity=type(response).__name__,
_external=True),
outformat=params['outformat'],
outformat=outformat,
expanded=params['expanded-jsonld'])
return decorated_function
@api_blueprint.route('/', methods=['POST', 'GET'])
@api_blueprint.route('/', defaults={'plugin': None}, methods=['POST', 'GET'])
@api_blueprint.route('/<path:plugin>', methods=['POST', 'GET'])
@basic_api
def api_root():
def api_root(plugin):
if plugin:
if request.parameters['algorithm'] != api.API_PARAMS['algorithm']['default']:
raise Error('You cannot specify the algorithm with a parameter and a URL variable.'
' Please, remove one of them')
request.parameters['algorithm'] = tuple(plugin.replace('+', '/').split('/'))
params = request.parameters
plugin = request.parameters['algorithm']
sp = current_app.senpy
plugins = sp.get_plugins(plugin)
if request.parameters['help']:
dic = dict(api.API_PARAMS, **api.NIF_PARAMS)
response = Help(valid_parameters=dic)
apis = []
if request.parameters['verbose']:
apis.append(api.BUILTIN_PARAMS)
allparameters = api.get_all_params(plugins, *apis)
response = Help(valid_parameters=allparameters)
return response
req = api.parse_call(request.parameters)
return current_app.senpy.analyse(req)
analysis = api.parse_analysis(req.parameters, plugins)
results = current_app.senpy.analyse(req, analysis)
return results
@api_blueprint.route('/evaluate/', methods=['POST', 'GET'])
@@ -160,7 +239,7 @@ def plugins():
@api_blueprint.route('/plugins/<plugin>/', methods=['POST', 'GET'])
@basic_api
def plugin(plugin=None):
def plugin(plugin):
sp = current_app.senpy
return sp.get_plugin(plugin)

View File

@@ -31,10 +31,10 @@ def main_function(argv):
default_plugins = params.get('default-plugins', False)
sp = Senpy(default_plugins=default_plugins, plugin_folder=plugin_folder)
request = api.parse_call(params)
algos = request.parameters.get('algorithm', None)
algos = sp.get_plugins(request.parameters.get('algorithm', None))
if algos:
for algo in algos:
sp.activate_plugin(algo)
sp.activate_plugin(algo.name)
else:
sp.activate_all()
res = sp.analyse(request)

View File

@@ -24,8 +24,12 @@ class Client(object):
return {d.name: d for d in resp}
def request(self, path=None, method='GET', **params):
url = '{}{}'.format(self.endpoint, path)
url = '{}{}'.format(self.endpoint.rstrip('/'), path)
if method == 'POST':
response = requests.post(url=url, data=params)
else:
response = requests.request(method=method, url=url, params=params)
try:
resp = models.from_dict(response.json())
except Exception as ex:

View File

@@ -6,7 +6,6 @@ from future import standard_library
standard_library.install_aliases()
from . import plugins, api
from .plugins import Plugin, evaluate
from .models import Error, AggregatedEvaluation
from .blueprints import api_blueprint, demo_blueprint, ns_blueprint
@@ -17,19 +16,14 @@ import copy
import errno
import logging
from . import gsitk_compat
logger = logging.getLogger(__name__)
try:
from gsitk.datasets.datasets import DatasetManager
GSITK_AVAILABLE = True
except ImportError:
logger.warn('GSITK is not installed. Some functions will be unavailable.')
GSITK_AVAILABLE = False
class Senpy(object):
""" Default Senpy extension for Flask """
def __init__(self,
app=None,
plugin_folder=".",
@@ -55,7 +49,7 @@ class Senpy(object):
self.add_folder('plugins', from_root=True)
else:
# Add only conversion plugins
self.add_folder(os.path.join('plugins', 'conversion'),
self.add_folder(os.path.join('plugins', 'postprocessing'),
from_root=True)
self.app = app
if app is not None:
@@ -84,27 +78,47 @@ class Senpy(object):
def delete_plugin(self, plugin):
del self._plugins[plugin.name.lower()]
def plugins(self, **kwargs):
def plugins(self, plugin_type=None, is_activated=True, **kwargs):
""" Return the plugins registered for a given application. Filtered by criteria """
return list(plugins.pfilter(self._plugins, **kwargs))
return list(plugins.pfilter(self._plugins, plugin_type=plugin_type,
is_activated=is_activated, **kwargs))
def get_plugin(self, name, default=None):
if name == 'default':
return self.default_plugin
plugin = name.lower()
if plugin in self._plugins:
return self._plugins[plugin]
elif name == 'conversion':
return None
results = self.plugins(id='plugins/{}'.format(name))
if name.lower() in self._plugins:
return self._plugins[name.lower()]
if not results:
return Error(message="Plugin not found", status=404)
results = self.plugins(id='endpoint:plugins/{}'.format(name.lower()),
plugin_type=None)
if results:
return results[0]
results = self.plugins(id=name,
plugin_type=None)
if results:
return results[0]
msg = ("Plugin not found: '{}'\n"
"Make sure it is ACTIVATED\n"
"Valid algorithms: {}").format(name,
self._plugins.keys())
raise Error(message=msg, status=404)
def get_plugins(self, name):
try:
name = name.split(',')
except AttributeError:
pass # Assume it is a tuple or a list
return tuple(self.get_plugin(n) for n in name)
@property
def analysis_plugins(self):
""" Return only the analysis plugins """
return self.plugins(plugin_type='analysisPlugin')
""" Return only the analysis plugins that are active"""
return self.plugins(plugin_type='analysisPlugin', is_activated=True)
def add_folder(self, folder, from_root=False):
""" Find plugins in this folder and add them to this instance """
@@ -119,75 +133,140 @@ class Senpy(object):
else:
raise AttributeError("Not a folder or does not exist: %s", folder)
def _get_plugins(self, request):
if not self.analysis_plugins:
raise Error(
status=404,
message=("No plugins found."
" Please install one."))
algos = request.parameters.get('algorithm', None)
if not algos:
if self.default_plugin:
algos = [self.default_plugin.name, ]
else:
raise Error(
status=404,
message="No default plugin found, and None provided")
# def check_analysis_request(self, analysis):
# '''Check if the analysis request can be fulfilled'''
# if not self.plugins():
# raise Error(
# status=404,
# message=("No plugins found."
# " Please install one."))
# for a in analysis:
# algo = a.algorithm
# if algo == 'default' and not self.default_plugin:
# raise Error(
# status=404,
# message="No default plugin found, and None provided")
# else:
# self.get_plugin(algo)
plugins = list()
for algo in algos:
algo = algo.lower()
if algo not in self._plugins:
msg = ("The algorithm '{}' is not valid\n"
"Valid algorithms: {}").format(algo,
self._plugins.keys())
logger.debug(msg)
raise Error(
status=404,
message=msg)
plugins.append(self._plugins[algo])
return plugins
def _process_entries(self, entries, req, plugins):
def _process(self, req, pending, done=None):
"""
Recursively process the entries with the first plugin in the list, and pass the results
to the rest of the plugins.
"""
if not plugins:
for i in entries:
yield i
return
plugin = plugins[0]
self._activate(plugin) # Make sure the plugin is activated
specific_params = api.parse_extra_params(req, plugin)
req.analysis.append({'plugin': plugin,
'parameters': specific_params})
results = plugin.analyse_entries(entries, specific_params)
for i in self._process_entries(results, req, plugins[1:]):
yield i
done = done or []
if not pending:
return req
analysis = pending[0]
results = analysis.run(req)
results.analysis.append(analysis)
done += analysis
return self._process(results, pending[1:], done)
def install_deps(self):
for plugin in self.plugins(is_activated=True):
plugins.install_deps(plugin)
plugins.install_deps(*self.plugins())
def analyse(self, request):
def analyse(self, request, analysis=None):
"""
Main method that analyses a request, either from CLI or HTTP.
It takes a processed request, provided by the user, as returned
by api.parse_call().
"""
if not self.plugins():
raise Error(
status=404,
message=("No plugins found."
" Please install one."))
if analysis is None:
params = str(request)
plugins = self.get_plugins(request.parameters['algorithm'])
analysis = api.parse_analysis(request.parameters, plugins)
logger.debug("analysing request: {}".format(request))
entries = request.entries
request.entries = []
plugins = self._get_plugins(request)
results = request
for i in self._process_entries(entries, results, plugins):
results.entries.append(i)
self.convert_emotions(results)
logger.debug("Returning analysis result: {}".format(results))
results.analysis = [i['plugin'].id for i in results.analysis]
results = self._process(request, analysis)
logger.debug("Got analysis result: {}".format(results))
results = self.postprocess(results)
logger.debug("Returning post-processed result: {}".format(results))
return results
def convert_emotions(self, resp):
"""
Conversion of all emotions in a response **in place**.
In addition to converting from one model to another, it has
to include the conversion plugin to the analysis list.
Needless to say, this is far from an elegant solution, but it works.
@todo refactor and clean up
"""
plugins = resp.analysis
if 'parameters' not in resp:
return resp
params = resp['parameters']
toModel = params.get('emotionModel', None)
if not toModel:
return resp
logger.debug('Asked for model: {}'.format(toModel))
output = params.get('conversion', None)
candidates = {}
for plugin in plugins:
try:
fromModel = plugin.get('onyx:usesEmotionModel', None)
candidates[plugin.id] = next(self._conversion_candidates(fromModel, toModel))
logger.debug('Analysis plugin {} uses model: {}'.format(
plugin.id, fromModel))
except StopIteration:
e = Error(('No conversion plugin found for: '
'{} -> {}'.format(fromModel, toModel)),
status=404)
e.original_response = resp
e.parameters = params
raise e
newentries = []
done = []
for i in resp.entries:
if output == "full":
newemotions = copy.deepcopy(i.emotions)
else:
newemotions = []
for j in i.emotions:
plugname = j['prov:wasGeneratedBy']
candidate = candidates[plugname]
done.append({'plugin': candidate, 'parameters': params})
for k in candidate.convert(j, fromModel, toModel, params):
k.prov__wasGeneratedBy = candidate.id
if output == 'nested':
k.prov__wasDerivedFrom = j
newemotions.append(k)
i.emotions = newemotions
newentries.append(i)
resp.entries = newentries
return resp
def _conversion_candidates(self, fromModel, toModel):
candidates = self.plugins(plugin_type=plugins.EmotionConversion)
for candidate in candidates:
for pair in candidate.onyx__doesConversion:
logging.debug(pair)
if candidate.can_convert(fromModel, toModel):
yield candidate
def postprocess(self, response):
'''
Transform the results from the analysis plugins.
It has some pre-defined post-processing like emotion conversion,
and it also allows plugins to auto-select themselves.
'''
response = self.convert_emotions(response)
for plug in self.plugins(plugin_type=plugins.PostProcessing):
if plug.check(response, response.analysis):
response = plug.process(response)
return response
def _get_datasets(self, request):
if not self.datasets:
raise Error(
@@ -198,21 +277,19 @@ class Senpy(object):
for dataset in datasets_name:
if dataset not in self.datasets:
logger.debug(("The dataset '{}' is not valid\n"
"Valid datasets: {}").format(dataset,
self.datasets.keys()))
"Valid datasets: {}").format(
dataset, self.datasets.keys()))
raise Error(
status=404,
message="The dataset '{}' is not valid".format(dataset))
dm = DatasetManager()
dm = gsitk_compat.DatasetManager()
datasets = dm.prepare_datasets(datasets_name)
return datasets
@property
def datasets(self):
if not GSITK_AVAILABLE:
raise Exception('GSITK is not available. Install it to use this function.')
self._dataset_list = {}
dm = DatasetManager()
dm = gsitk_compat.DatasetManager()
for item in dm.get_datasets():
for key in item:
if key in self._dataset_list:
@@ -223,84 +300,26 @@ class Senpy(object):
return self._dataset_list
def evaluate(self, params):
if not GSITK_AVAILABLE:
raise Exception('GSITK is not available. Install it to use this function.')
logger.debug("evaluating request: {}".format(params))
results = AggregatedEvaluation()
results.parameters = params
datasets = self._get_datasets(results)
plugins = self._get_plugins(results)
for eval in evaluate(plugins, datasets):
plugins = []
for plugname in params.algorithm:
plugins = self.get_plugin(plugname)
for eval in plugins.evaluate(plugins, datasets):
results.evaluations.append(eval)
if 'with_parameters' not in results.parameters:
del results.parameters
logger.debug("Returning evaluation result: {}".format(results))
return results
def _conversion_candidates(self, fromModel, toModel):
candidates = self.plugins(plugin_type='emotionConversionPlugin')
for candidate in candidates:
for pair in candidate.onyx__doesConversion:
logging.debug(pair)
if pair['onyx:conversionFrom'] == fromModel \
and pair['onyx:conversionTo'] == toModel:
yield candidate
def convert_emotions(self, resp):
"""
Conversion of all emotions in a response **in place**.
In addition to converting from one model to another, it has
to include the conversion plugin to the analysis list.
Needless to say, this is far from an elegant solution, but it works.
@todo refactor and clean up
"""
plugins = [i['plugin'] for i in resp.analysis]
params = resp.parameters
toModel = params.get('emotionModel', None)
if not toModel:
return
logger.debug('Asked for model: {}'.format(toModel))
output = params.get('conversion', None)
candidates = {}
for plugin in plugins:
try:
fromModel = plugin.get('onyx:usesEmotionModel', None)
candidates[plugin.id] = next(self._conversion_candidates(fromModel, toModel))
logger.debug('Analysis plugin {} uses model: {}'.format(plugin.id, fromModel))
except StopIteration:
e = Error(('No conversion plugin found for: '
'{} -> {}'.format(fromModel, toModel)),
status=404)
e.original_response = resp
e.parameters = params
raise e
newentries = []
for i in resp.entries:
if output == "full":
newemotions = copy.deepcopy(i.emotions)
else:
newemotions = []
for j in i.emotions:
plugname = j['prov:wasGeneratedBy']
candidate = candidates[plugname]
resp.analysis.append({'plugin': candidate,
'parameters': params})
for k in candidate.convert(j, fromModel, toModel, params):
k.prov__wasGeneratedBy = candidate.id
if output == 'nested':
k.prov__wasDerivedFrom = j
newemotions.append(k)
i.emotions = newemotions
newentries.append(i)
resp.entries = newentries
@property
def default_plugin(self):
if not self._default or not self._default.is_activated:
candidates = self.plugins(plugin_type='analysisPlugin',
is_activated=True)
candidates = self.plugins(
plugin_type='analysisPlugin', is_activated=True)
if len(candidates) > 0:
self._default = candidates[0]
else:
@@ -310,7 +329,7 @@ class Senpy(object):
@default_plugin.setter
def default_plugin(self, value):
if isinstance(value, Plugin):
if isinstance(value, plugins.Plugin):
if not value.is_activated:
raise AttributeError('The default plugin has to be activated.')
self._default = value
@@ -318,10 +337,15 @@ class Senpy(object):
else:
self._default = self._plugins[value.lower()]
def activate_all(self, sync=True):
def activate_all(self, sync=True, allow_fail=False):
ps = []
for plug in self._plugins.keys():
ps.append(self.activate_plugin(plug, sync=sync))
try:
self.activate_plugin(plug, sync=sync)
except Exception as ex:
if not allow_fail:
raise
logger.error('Could not activate {}: {}'.format(plug, ex))
return ps
def deactivate_all(self, sync=True):
@@ -346,6 +370,7 @@ class Senpy(object):
logger.info(msg)
success = True
self._set_active(plugin, success)
return success
def activate_plugin(self, plugin_name, sync=True):
plugin_name = plugin_name.lower()
@@ -356,8 +381,9 @@ class Senpy(object):
logger.info("Activating plugin: {}".format(plugin.name))
if sync or 'async' in plugin and not plugin.async:
self._activate(plugin)
if sync or not getattr(plugin, 'async', True) or getattr(
plugin, 'sync', False):
return self._activate(plugin)
else:
th = Thread(target=partial(self._activate, plugin))
th.start()
@@ -379,7 +405,8 @@ class Senpy(object):
self._set_active(plugin, False)
if sync or 'async' in plugin and not plugin.async:
if sync or not getattr(plugin, 'async', True) or not getattr(
plugin, 'sync', False):
self._deactivate(plugin)
else:
th = Thread(target=partial(self._deactivate, plugin))

31
senpy/gsitk_compat.py Normal file
View File

@@ -0,0 +1,31 @@
import logging
from pkg_resources import parse_version, get_distribution, DistributionNotFound
logger = logging.getLogger(__name__)
MSG = 'GSITK is not (properly) installed.'
IMPORTMSG = '{} Some functions will be unavailable.'.format(MSG)
RUNMSG = '{} Install it to use this function.'.format(MSG)
def raise_exception(*args, **kwargs):
raise Exception(RUNMSG)
try:
gsitk_distro = get_distribution("gsitk")
GSITK_VERSION = parse_version(gsitk_distro.version)
GSITK_AVAILABLE = GSITK_VERSION > parse_version("0.1.9.1") # Earlier versions have a bug
except DistributionNotFound:
GSITK_AVAILABLE = False
GSITK_VERSION = ()
if GSITK_AVAILABLE:
from gsitk.datasets.datasets import DatasetManager
from gsitk.evaluation.evaluation import Evaluation as Eval
from sklearn.pipeline import Pipeline
modules = locals()
else:
logger.warning(IMPORTMSG)
DatasetManager = Eval = Pipeline = raise_exception

View File

@@ -85,6 +85,7 @@ class BaseMeta(ABCMeta):
schema = json.load(f)
resolver = jsonschema.RefResolver(schema_path, schema)
if '@type' not in attrs:
attrs['@type'] = "".join((name[0].lower(), name[1:]))
attrs['_schema_file'] = schema_file
attrs['schema'] = schema
@@ -244,10 +245,10 @@ class CustomDict(MutableMapping, object):
return key[0] == '_'
def __str__(self):
return str(self.serializable())
return json.dumps(self.serializable(), sort_keys=True, indent=4)
def __repr__(self):
return str(self.serializable())
return json.dumps(self.serializable(), sort_keys=True, indent=4)
_Alias = namedtuple('Alias', 'indict')

View File

@@ -121,11 +121,11 @@ class BaseModel(with_metaclass(BaseMeta, CustomDict)):
'''
schema_file = DEFINITIONS_FILE
# schema_file = DEFINITIONS_FILE
_context = base_context["@context"]
def __init__(self, *args, **kwargs):
auto_id = kwargs.pop('_auto_id', True)
auto_id = kwargs.pop('_auto_id', False)
super(BaseModel, self).__init__(*args, **kwargs)
@@ -133,12 +133,12 @@ class BaseModel(with_metaclass(BaseMeta, CustomDict)):
self.id
if '@type' not in self:
logger.warn('Created an instance of an unknown model')
logger.warning('Created an instance of an unknown model')
@property
def id(self):
if '@id' not in self:
self['@id'] = ':{}_{}'.format(type(self).__name__, time.time())
self['@id'] = '_:{}_{}'.format(type(self).__name__, time.time())
return self['@id']
@id.setter
@@ -146,7 +146,7 @@ class BaseModel(with_metaclass(BaseMeta, CustomDict)):
self['@id'] = value
def flask(self,
in_headers=True,
in_headers=False,
headers=None,
outformat='json-ld',
**kwargs):
@@ -176,20 +176,22 @@ class BaseModel(with_metaclass(BaseMeta, CustomDict)):
def serialize(self, format='json-ld', with_mime=False, **kwargs):
js = self.jsonld(**kwargs)
content = json.dumps(js, indent=2, sort_keys=True)
if format == 'json-ld':
content = json.dumps(js, indent=2, sort_keys=True)
mimetype = "application/json"
elif format in ['turtle', ]:
elif format in ['turtle', 'ntriples']:
logger.debug(js)
content = json.dumps(js, indent=2, sort_keys=True)
base = kwargs.get('prefix')
g = Graph().parse(
data=content,
format='json-ld',
base=kwargs.get('prefix'),
context=self._context)
base=base,
context=[self._context,
{'@base': base}])
logger.debug(
'Parsing with prefix: {}'.format(kwargs.get('prefix')))
content = g.serialize(format='turtle').decode('utf-8')
content = g.serialize(format=format,
base=base).decode('utf-8')
mimetype = 'text/{}'.format(format)
else:
raise Error('Unknown outformat: {}'.format(format))
@@ -205,25 +207,21 @@ class BaseModel(with_metaclass(BaseMeta, CustomDict)):
expanded=False):
result = self.serializable()
if context_uri or with_context:
result['@context'] = context_uri or self._context
# result = jsonld.compact(result,
# self._context,
# options={
# 'base': prefix,
# 'expandContext': self._context,
# 'senpy': prefix
# })
if expanded:
result = jsonld.expand(
result, options={'base': prefix,
'expandContext': self._context})
'expandContext': self._context})[0]
if not with_context:
try:
del result['@context']
except KeyError:
pass
elif context_uri:
result['@context'] = context_uri
else:
result['@context'] = self._context
return result
def validate(self, obj=None):
@@ -327,7 +325,6 @@ def _add_class_from_schema(*args, **kwargs):
for i in [
'aggregatedEvaluation',
'analysis',
'dataset',
'datasets',
'emotion',
@@ -341,7 +338,7 @@ for i in [
'entity',
'help',
'metric',
'plugin',
'parameter',
'plugins',
'response',
'results',
@@ -351,3 +348,55 @@ for i in [
]:
_add_class_from_schema(i)
class Analysis(BaseModel):
schema = 'analysis'
parameters = alias('prov:used')
@property
def params(self):
outdict = {}
outdict['algorithm'] = self.algorithm
for param in self.parameters:
outdict[param['name']] = param['value']
return outdict
@params.setter
def params(self, value):
for k, v in value.items():
for param in self.parameters:
if param.name == k:
param.value = v
break
else:
self.parameters.append(Parameter(name=k, value=v))
@property
def algorithm(self):
return self['prov:wasAssociatedWith']
@property
def plugin(self):
return self._plugin
@plugin.setter
def plugin(self, value):
self._plugin = value
self['prov:wasAssociatedWith'] = value.id
def run(self, request):
return self.plugin.process(request, self.params)
class Plugin(BaseModel):
schema = 'plugin'
def activity(self, parameters):
'''Generate a prov:Activity from this plugin and the '''
a = Analysis()
a.plugin = self
a.params = parameters
return a

View File

@@ -1,15 +1,14 @@
from future import standard_library
standard_library.install_aliases()
from future.utils import with_metaclass
from functools import partial
import os.path
import os
import re
import pickle
import logging
import copy
import pprint
import inspect
@@ -18,22 +17,15 @@ import subprocess
import importlib
import yaml
import threading
import nltk
from nltk import download
from .. import models, utils
from .. import api
from .. import gsitk_compat
from .. import testing
logger = logging.getLogger(__name__)
try:
from gsitk.evaluation.evaluation import Evaluation as Eval
from sklearn.pipeline import Pipeline
GSITK_AVAILABLE = True
except ImportError:
logger.warn('GSITK is not installed. Some functions will be unavailable.')
GSITK_AVAILABLE = False
class PluginMeta(models.BaseMeta):
_classes = {}
@@ -51,16 +43,19 @@ class PluginMeta(models.BaseMeta):
if doc:
attrs['description'] = doc
else:
logger.warn(('Plugin {} does not have a description. '
'Please, add a short summary to help other developers').format(name))
logger.warning(
('Plugin {} does not have a description. '
'Please, add a short summary to help other developers'
).format(name))
cls = super(PluginMeta, mcs).__new__(mcs, name, bases, attrs)
if alias in mcs._classes:
if os.environ.get('SENPY_TESTING', ""):
raise Exception(('The type of plugin {} already exists. '
raise Exception(
('The type of plugin {} already exists. '
'Please, choose a different name').format(name))
else:
logger.warn('Overloading plugin class: {}'.format(alias))
logger.warning('Overloading plugin class: {}'.format(alias))
mcs._classes[alias] = cls
return cls
@@ -92,10 +87,12 @@ class Plugin(with_metaclass(PluginMeta, models.Plugin)):
if info:
self.update(info)
self.validate()
self.id = 'plugins/{}_{}'.format(self['name'], self['version'])
self.id = 'endpoint:plugins/{}_{}'.format(self['name'],
self['version'])
self.is_activated = False
self._lock = threading.Lock()
self._directory = os.path.abspath(os.path.dirname(inspect.getfile(self.__class__)))
self._directory = os.path.abspath(
os.path.dirname(inspect.getfile(self.__class__)))
data_folder = data_folder or os.getcwd()
subdir = os.path.join(data_folder, self.name)
@@ -123,7 +120,8 @@ class Plugin(with_metaclass(PluginMeta, models.Plugin)):
if x not in self:
missing.append(x)
if missing:
raise models.Error('Missing configuration parameters: {}'.format(missing))
raise models.Error(
'Missing configuration parameters: {}'.format(missing))
def get_folder(self):
return os.path.dirname(inspect.getfile(self.__class__))
@@ -134,35 +132,86 @@ class Plugin(with_metaclass(PluginMeta, models.Plugin)):
def deactivate(self):
pass
def process(self, request, parameters, **kwargs):
"""
An implemented plugin should override this method.
Here, we assume that a process_entries method exists."""
newentries = list(
self.process_entries(request.entries, parameters))
request.entries = newentries
return request
def process_entries(self, entries, parameters):
for entry in entries:
self.log.debug('Processing entry with plugin {}: {}'.format(
self, entry))
results = self.process_entry(entry, parameters)
if inspect.isgenerator(results):
for result in results:
yield result
else:
yield results
def process_entry(self, entry, parameters):
"""
This base method is here to adapt plugins which only
implement the *process* function.
Note that this method may yield an annotated entry or a list of
entries (e.g. in a tokenizer)
"""
raise NotImplementedError(
'You need to implement process, process_entries or process_entry in your plugin'
)
def test(self, test_cases=None):
if not test_cases:
if not hasattr(self, 'test_cases'):
raise AttributeError(('Plugin {} [{}] does not have any defined '
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:
try:
self.test_case(case)
self.log.debug('Test case passed:\n{}'.format(pprint.pformat(case)))
self.log.debug('Test case passed:\n{}'.format(
pprint.pformat(case)))
except Exception as ex:
self.log.warn('Test case failed:\n{}'.format(pprint.pformat(case)))
self.log.warning('Test case failed:\n{}'.format(
pprint.pformat(case)))
raise
def test_case(self, case):
def test_case(self, case, mock=testing.MOCK_REQUESTS):
if 'entry' not in case and 'input' in case:
entry = models.Entry(_auto_id=False)
entry.nif__isString = case['input']
case['entry'] = entry
entry = models.Entry(case['entry'])
given_parameters = case.get('params', case.get('parameters', {}))
expected = case.get('expected', None)
should_fail = case.get('should_fail', False)
responses = case.get('responses', [])
try:
params = api.parse_params(given_parameters, self.extra_params)
res = list(self.analyse_entries([entry, ], params))
request = models.Response()
parameters = api.parse_params(given_parameters,
self.extra_params)
request.entries = [
entry,
]
method = partial(self.process, request, parameters)
if mock:
res = method()
else:
with testing.patch_all_requests(responses):
res = method()
if not isinstance(expected, list):
expected = [expected]
utils.check_template(res, expected)
for r in res:
r.validate()
utils.check_template(res.entries, expected)
res.validate()
except models.Error:
if should_fail:
return
@@ -199,40 +248,26 @@ class Analysis(Plugin):
A subclass of Plugin that analyses text and provides an annotation.
'''
def analyse(self, *args, **kwargs):
raise NotImplementedError(
'Your plugin should implement either analyse or analyse_entry')
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)
"""
text = entry['nif:isString']
params = copy.copy(parameters)
params['input'] = text
results = self.analyse(**params)
for i in results.entries:
yield i
def analyse(self, request, parameters):
return super(Analysis, self).process(request, parameters)
def analyse_entries(self, entries, parameters):
for entry in entries:
self.log.debug('Analysing entry with plugin {}: {}'.format(self, entry))
results = self.analyse_entry(entry, parameters)
if inspect.isgenerator(results):
for result in results:
yield result
else:
yield results
for i in super(Analysis, self).process_entries(entries, parameters):
yield i
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)
def process(self, request, parameters, **kwargs):
return self.analyse(request, parameters)
def process_entries(self, entries, parameters):
for i in self.analyse_entries(entries, parameters):
yield i
def process_entry(self, entry, parameters, **kwargs):
if hasattr(self, 'analyse_entry'):
for i in self.analyse_entry(entry, parameters):
yield i
else:
super(Analysis, self).process_entry(entry, parameters, **kwargs)
AnalysisPlugin = Analysis
@@ -243,7 +278,20 @@ 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.
'''
pass
def process(self, response, parameters, plugins=None, **kwargs):
plugins = plugins or []
newentries = []
for entry in response.entries:
newentries.append(
self.convert_entry(entry, parameters, plugins))
response.entries = newentries
return response
def convert_entry(self, entry, parameters, conversions_applied):
raise NotImplementedError(
'You should implement a way to convert each entry, or a custom process method'
)
ConversionPlugin = Conversion
@@ -280,12 +328,28 @@ class EmotionConversion(Conversion):
'''
A subclass of Conversion that converts emotion annotations using different models
'''
pass
def can_convert(self, fromModel, toModel):
'''
Whether this plugin can convert from fromModel to toModel.
If fromModel is None, it is interpreted as "any Model"
'''
for pair in self.onyx__doesConversion:
if (pair['onyx:conversionTo'] == toModel) and \
((fromModel is None) or (pair['onyx:conversionFrom'] == fromModel)):
return True
return False
EmotionConversionPlugin = EmotionConversion
class PostProcessing(Plugin):
def check(self, request, plugins):
'''Should this plugin be run for this request?'''
return False
class Box(AnalysisPlugin):
'''
Black box plugins delegate analysis to a function.
@@ -310,9 +374,10 @@ class Box(AnalysisPlugin):
return output
def predict_one(self, input):
raise NotImplementedError('You should define the behavior of this plugin')
raise NotImplementedError(
'You should define the behavior of this plugin')
def analyse_entries(self, entries, params):
def process_entries(self, entries, params):
for entry in entries:
input = self.input(entry=entry, params=params)
results = self.predict_one(input=input)
@@ -332,7 +397,7 @@ class Box(AnalysisPlugin):
return self.transform(X)
def as_pipe(self):
pipe = Pipeline([('plugin', self)])
pipe = gsitk_compat.Pipeline([('plugin', self)])
pipe.name = self.name
return pipe
@@ -381,7 +446,6 @@ class EmotionBox(TextBox, EmotionPlugin):
class MappingMixin(object):
@property
def mappings(self):
return self._mappings
@@ -391,11 +455,10 @@ class MappingMixin(object):
self._mappings = value
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)
output = self.mappings.get(output, self.mappings.get(
'default', output))
return super(MappingMixin, self).output(
output=output, entry=entry, params=params)
class ShelfMixin(object):
@@ -408,7 +471,8 @@ class ShelfMixin(object):
with self.open(self.shelf_file, 'rb') as p:
self._sh = pickle.load(p)
except (IndexError, EOFError, pickle.UnpicklingError):
self.log.warning('Corrupted shelf file: {}'.format(self.shelf_file))
self.log.warning('Corrupted shelf file: {}'.format(
self.shelf_file))
if not self.get('force_shelf', False):
raise
return self._sh
@@ -456,8 +520,7 @@ def pfilter(plugins, plugin_type=Analysis, **kwargs):
plugin_type = plugin_type[0].upper() + plugin_type[1:]
pclass = globals()[plugin_type]
logger.debug('Class: {}'.format(pclass))
candidates = filter(lambda x: isinstance(x, pclass),
plugins)
candidates = filter(lambda x: isinstance(x, pclass), plugins)
except KeyError:
raise models.Error('{} is not a valid type'.format(plugin_type))
else:
@@ -467,8 +530,7 @@ def pfilter(plugins, plugin_type=Analysis, **kwargs):
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))
logger.debug("matching {} with {}: {}".format(plug.name, kwargs, res))
return res
if kwargs:
@@ -502,17 +564,17 @@ def install_deps(*plugins):
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)
process = subprocess.Popen(
pip_args, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
_log_subprocess_output(process)
exitcode = process.wait()
installed = True
if exitcode != 0:
raise models.Error("Dependencies not properly installed")
raise models.Error(
"Dependencies not properly installed: {}".format(pip_args))
nltk_resources |= set(info.get('nltk_resources', []))
installed |= nltk.download(list(nltk_resources))
installed |= download(list(nltk_resources))
return installed
@@ -530,7 +592,7 @@ def find_plugins(folders):
yield fpath
def from_path(fpath, **kwargs):
def from_path(fpath, install_on_fail=False, **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
@@ -540,7 +602,7 @@ def from_path(fpath, **kwargs):
yield instance
else:
info = parse_plugin_info(fpath)
yield from_info(info, **kwargs)
yield from_info(info, install_on_fail=install_on_fail, **kwargs)
def from_folder(folders, loader=from_path, **kwargs):
@@ -551,7 +613,7 @@ def from_folder(folders, loader=from_path, **kwargs):
return plugins
def from_info(info, root=None, **kwargs):
def from_info(info, root=None, install_on_fail=True, **kwargs):
if any(x not in info for x in ('module', )):
raise ValueError('Plugin info is not valid: {}'.format(info))
module = info["module"]
@@ -559,7 +621,12 @@ def from_info(info, root=None, **kwargs):
if not root and '_path' in info:
root = os.path.dirname(info['_path'])
return one_from_module(module, root=root, info=info, **kwargs)
fun = partial(one_from_module, module, root=root, info=info, **kwargs)
try:
return fun()
except (ImportError, LookupError):
install_deps(info)
return fun()
def parse_plugin_info(fpath):
@@ -584,7 +651,8 @@ 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)
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
@@ -606,17 +674,10 @@ def _instances_in_module(module):
yield obj
def _from_module_name(module, root, info=None, install=True, **kwargs):
try:
def _from_module_name(module, root, info=None, **kwargs):
module = load_module(module, root)
except (ImportError, LookupError):
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):
if install:
install_deps(plugin)
for plugin in _from_loaded_module(
module=module, root=root, info=info, **kwargs):
yield plugin
@@ -628,10 +689,8 @@ def _from_loaded_module(module, info=None, **kwargs):
def evaluate(plugins, datasets, **kwargs):
if not GSITK_AVAILABLE:
raise Exception('GSITK is not available. Install it to use this function.')
ev = Eval(tuples=None,
ev = gsitk_compat.Eval(
tuples=None,
datasets=datasets,
pipelines=[plugin.as_pipe() for plugin in plugins])
ev.evaluate()

View File

@@ -1,19 +0,0 @@
---
name: split
module: senpy.plugins.misc.split
description: A sample plugin that chunks input text
author: "@militarpancho"
version: '0.2'
url: "https://github.com/gsi-upm/senpy"
requirements:
- nltk
extra_params:
delimiter:
aliases:
- type
- t
required: false
default: sentence
options:
- sentence
- paragraph

View File

@@ -5,13 +5,27 @@ from nltk.tokenize.simple import LineTokenizer
import nltk
class SplitPlugin(AnalysisPlugin):
class Split(AnalysisPlugin):
'''description: A sample plugin that chunks input text'''
author = ["@militarpancho", '@balkian']
version = '0.3'
url = "https://github.com/gsi-upm/senpy"
extra_params = {
'delimiter': {
'aliases': ['type', 't'],
'required': False,
'default': 'sentence',
'options': ['sentence', 'paragraph']
},
}
def activate(self):
nltk.download('punkt')
def analyse_entry(self, entry, params):
yield entry
chunker_type = params["delimiter"]
original_text = entry['nif:isString']
if chunker_type == "sentence":
@@ -19,12 +33,15 @@ class SplitPlugin(AnalysisPlugin):
if chunker_type == "paragraph":
tokenizer = LineTokenizer()
chars = list(tokenizer.span_tokenize(original_text))
for i, chunk in enumerate(tokenizer.tokenize(original_text)):
print(chunk)
if len(chars) == 1:
# This sentence was already split
return
for i, chunk in enumerate(chars):
start, end = chunk
e = Entry()
e['nif:isString'] = chunk
e['nif:isString'] = original_text[start:end]
if entry.id:
e.id = entry.id + "#char={},{}".format(chars[i][0], chars[i][1])
e.id = entry.id + "#char={},{}".format(start, end)
yield e
test_cases = [

View File

@@ -1,6 +1,6 @@
---
name: Ekman2FSRE
module: senpy.plugins.conversion.emotion.centroids
module: senpy.plugins.postprocessing.emotion.centroids
description: Plugin to convert emotion sets from Ekman to VAD
version: 0.2
# No need to specify onyx:doesConversion because centroids.py adds it automatically from centroids_direction

View File

@@ -1,6 +1,6 @@
---
name: Ekman2PAD
module: senpy.plugins.conversion.emotion.centroids
module: senpy.plugins.postprocessing.emotion.centroids
description: Plugin to convert emotion sets from Ekman to VAD
version: 0.2
# No need to specify onyx:doesConversion because centroids.py adds it automatically from centroids_direction

View File

@@ -0,0 +1,196 @@
from senpy import PostProcessing, easy_test
class MaxEmotion(PostProcessing):
'''Plugin to extract the emotion with highest value from an EmotionSet'''
author = '@dsuarezsouto'
version = '0.1'
def process_entry(self, entry, params):
if len(entry.emotions) < 1:
yield entry
return
set_emotions = entry.emotions[0]['onyx:hasEmotion']
# If there is only one emotion, do not modify it
if len(set_emotions) < 2:
yield entry
return
max_emotion = set_emotions[0]
# Extract max emotion from the set emotions (emotion with highest intensity)
for tmp_emotion in set_emotions:
if tmp_emotion['onyx:hasEmotionIntensity'] > max_emotion[
'onyx:hasEmotionIntensity']:
max_emotion = tmp_emotion
if max_emotion['onyx:hasEmotionIntensity'] == 0:
max_emotion['onyx:hasEmotionCategory'] = "neutral"
max_emotion['onyx:hasEmotionIntensity'] = 1.0
entry.emotions[0]['onyx:hasEmotion'] = [max_emotion]
entry.emotions[0]['prov:wasGeneratedBy'] = "maxSentiment"
yield entry
def check(self, request, plugins):
return 'maxemotion' in request.parameters and self not in plugins
# Test Cases:
# 1 Normal Situation.
# 2 Case to return a Neutral Emotion.
test_cases = [
{
"name":
"If there are several emotions within an emotion set, reduce it to one.",
"entry": {
"@type":
"entry",
"emotions": [
{
"@id":
"Emotions0",
"@type":
"emotionSet",
"onyx:hasEmotion": [
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "anger",
"onyx:hasEmotionIntensity": 0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "joy",
"onyx:hasEmotionIntensity": 0.3333333333333333
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "negative-fear",
"onyx:hasEmotionIntensity": 0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "sadness",
"onyx:hasEmotionIntensity": 0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "disgust",
"onyx:hasEmotionIntensity": 0
}
]
}
],
"nif:isString":
"Test"
},
'expected': {
"@type":
"entry",
"emotions": [
{
"@id":
"Emotions0",
"@type":
"emotionSet",
"onyx:hasEmotion": [
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "joy",
"onyx:hasEmotionIntensity": 0.3333333333333333
}
],
"prov:wasGeneratedBy":
'maxSentiment'
}
],
"nif:isString":
"Test"
}
},
{
"name":
"If the maximum emotion has an intensity of 0, return a neutral emotion.",
"entry": {
"@type":
"entry",
"emotions": [{
"@id":
"Emotions0",
"@type":
"emotionSet",
"onyx:hasEmotion": [
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "anger",
"onyx:hasEmotionIntensity": 0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "joy",
"onyx:hasEmotionIntensity": 0
},
{
"@id":
"_:Emotion_1538121033.74",
"@type":
"emotion",
"onyx:hasEmotionCategory":
"negative-fear",
"onyx:hasEmotionIntensity":
0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory":
"sadness",
"onyx:hasEmotionIntensity": 0
},
{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory":
"disgust",
"onyx:hasEmotionIntensity": 0
}]
}],
"nif:isString":
"Test"
},
'expected': {
"@type":
"entry",
"emotions": [{
"@id":
"Emotions0",
"@type":
"emotionSet",
"onyx:hasEmotion": [{
"@id": "_:Emotion_1538121033.74",
"@type": "emotion",
"onyx:hasEmotionCategory": "neutral",
"onyx:hasEmotionIntensity": 1
}],
"prov:wasGeneratedBy":
'maxSentiment'
}],
"nif:isString":
"Test"
}
}
]
if __name__ == '__main__':
easy_test()

View File

@@ -1,22 +0,0 @@
---
name: sentiment140
module: sentiment140
description: "Connects to the sentiment140 free API: http://sentiment140.com"
author: "@balkian"
version: '0.2'
url: "https://github.com/gsi-upm/senpy-plugins-community"
extra_params:
language:
"@id": lang_sentiment140
aliases:
- language
- l
required: false
options:
- es
- en
- auto
default: auto
requirements: {}
maxPolarityValue: 1
minPolarityValue: 0

View File

@@ -4,12 +4,31 @@ import json
from senpy.plugins import SentimentPlugin
from senpy.models import Sentiment
ENDPOINT = 'http://www.sentiment140.com/api/bulkClassifyJson'
class Sentiment140Plugin(SentimentPlugin):
class Sentiment140(SentimentPlugin):
'''Connects to the sentiment140 free API: http://sentiment140.com'''
author = "@balkian"
version = '0.2'
url = "https://github.com/gsi-upm/senpy-plugins-community"
extra_params = {
'language': {
"@id": 'lang_sentiment140',
'aliases': ['language', 'l'],
'required': False,
'default': 'auto',
'options': ['es', 'en', 'auto']
}
}
maxPolarityValue = 1
minPolarityValue = 0
def analyse_entry(self, entry, params):
lang = params["language"]
res = requests.post("http://www.sentiment140.com/api/bulkClassifyJson",
res = requests.post(ENDPOINT,
json.dumps({
"language": lang,
"data": [{
@@ -31,23 +50,10 @@ class Sentiment140Plugin(SentimentPlugin):
marl__hasPolarity=polarity,
marl__polarityValue=polarity_value)
sentiment.prov__wasGeneratedBy = self.id
entry.sentiments = []
entry.sentiments.append(sentiment)
entry.language = lang
yield entry
def test(self, *args, **kwargs):
'''
To avoid calling the sentiment140 API, we will mock the results
from requests.
'''
from senpy.testing import patch_requests
expected = {"data": [{"polarity": 4}]}
with patch_requests(expected) as (request, response):
super(Sentiment140Plugin, self).test(*args, **kwargs)
assert request.called
assert response.json.called
test_cases = [
{
'entry': {
@@ -61,6 +67,9 @@ class Sentiment140Plugin(SentimentPlugin):
'marl:hasPolarity': 'marl:Positive',
}
]
}
},
'responses': [{'url': ENDPOINT,
'method': 'POST',
'json': {'data': [{'polarity': 4}]}}]
}
]

View File

@@ -9,7 +9,20 @@
"@type": {
"type": "string",
"description": "Type of the analysis. e.g. marl:SentimentAnalysis"
},
"prov:wasAssociatedWith": {
"@type": "string",
"description": "Algorithm/plugin that was used"
},
"prov:used": {
"description": "Parameters of the algorithm",
"@type": "array",
"default": [],
"type": "array",
"items": {
"$ref": "parameter.json"
}
}
},
"required": ["@id", "@type"]
"required": ["@type", "prov:wasAssociatedWith"]
}

View File

@@ -41,7 +41,7 @@
"@container": "@set"
},
"analysis": {
"@id": "AnalysisInvolved",
"@id": "prov:wasInformedBy",
"@type": "@id",
"@container": "@set"
},

View File

@@ -20,5 +20,5 @@
"description": "The ID of the analysis that generated this Emotion. The full object should be included in the \"analysis\" property of the root object"
}
},
"required": ["@id", "prov:wasGeneratedBy", "onyx:hasEmotion"]
"required": ["prov:wasGeneratedBy", "onyx:hasEmotion"]
}

View File

@@ -35,5 +35,5 @@
"default": []
}
},
"required": ["@id", "nif:isString"]
"required": ["nif:isString"]
}

View File

@@ -0,0 +1,16 @@
{
"$schema": "http://json-schema.org/draft-04/schema#",
"description": "Parameters for a senpy analysis",
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Name of the parameter"
},
"prov:value": {
"@type": "any",
"description": "Value of the parameter"
}
},
"required": ["name", "prov:value"]
}

View File

@@ -21,13 +21,7 @@
"default": [],
"type": "array",
"items": {
"anyOf": [
{
"$ref": "analysis.json"
},{
"type": "string"
}
]
}
},
"entries": {

View File

@@ -19,5 +19,5 @@
"description": "The ID of the analysis that generated this Sentiment. The full object should be included in the \"analysis\" property of the root object"
}
},
"required": ["@id", "prov:wasGeneratedBy"]
"required": ["prov:wasGeneratedBy"]
}

View File

@@ -167,3 +167,36 @@ textarea{
color: inherit;
text-decoration: inherit;
}
.collapsed .collapseicon {
display: none !important;
}
.collapsed .expandicon {
display: inline-block !important;
}
.expandicon {
display: none !important;
}
.collapseicon {
display: inline-block !important;
}
.loader {
border: 6px solid #f3f3f3; /* Light grey */
border-top: 6px solid blue;
border-bottom: 6px solid blue;
border-radius: 50%;
width: 3em;
height: 3em;
animation: spin 2s linear infinite;
}
@keyframes spin {
0% { transform: rotate(0deg); }
100% { transform: rotate(360deg); }
}

View File

@@ -4,6 +4,7 @@ var plugins_params = default_params = {};
var plugins = [];
var defaultPlugin = {};
var gplugins = {};
var pipeline = [];
function replaceURLWithHTMLLinks(text) {
console.log('Text: ' + text);
@@ -30,7 +31,10 @@ function hashchanged(){
function get_plugins(response){
plugins = response.plugins;
for(ix in response.plugins){
plug = response.plugins[ix];
plugins[plug.name] = plug;
}
}
function get_datasets(response){
@@ -83,10 +87,32 @@ function draw_plugins_selection(){
html += "</optgroup>"
// Two elements with plugin class
// One from the evaluate tab and another one from the analyse tab
document.getElementsByClassName('plugin')[0].innerHTML = html;
document.getElementsByClassName('plugin')[1].innerHTML = html;
plugin_lists = document.getElementsByClassName('plugin')
for (element in plugin_lists){
plugin_lists[element].innerHTML = html;
}
draw_plugin_pipeline();
}
function draw_plugin_pipeline(){
var pipeHTML = "";
console.log("Drawing pipeline: ", pipeline);
for (ix in pipeline){
plug = pipeline[ix];
pipeHTML += '<span onclick="remove_plugin_pipeline(\'' + plug + '\')" class="btn btn-primary"><span ><i class="fa fa-minus"></i></span> ' + plug + '</span> <i class="fa fa-arrow-right"></i> ';
}
console.log(pipeHTML);
$("#pipeline").html(pipeHTML);
}
function remove_plugin_pipeline(name){
console.log("Removing plugin: ", name);
var index = pipeline.indexOf(name);
pipeline.splice(index, 1);
draw_plugin_pipeline();
}
function draw_plugins_list(){
var availablePlugins = document.getElementById('availablePlugins');
@@ -105,6 +131,13 @@ function draw_plugins_list(){
}
}
function add_plugin_pipeline(){
var selected = get_selected_plugin();
pipeline.push(selected);
console.log("Adding ", selected);
draw_plugin_pipeline();
}
function draw_datasets(){
html = "";
repeated_html = "<input class=\"checks-datasets\" type=\"checkbox\" value=\"";
@@ -118,16 +151,20 @@ function draw_datasets(){
$(document).ready(function() {
var response = JSON.parse($.ajax({type: "GET", url: "/api/plugins/" , async: false}).responseText);
defaultPlugin= JSON.parse($.ajax({type: "GET", url: "/api/plugins/default" , async: false}).responseText);
var response2 = JSON.parse($.ajax({type: "GET", url: "/api/datasets/" , async: false}).responseText);
get_plugins(response);
get_default_parameters();
get_datasets(response2);
draw_plugins_list();
draw_plugins_selection();
draw_parameters();
draw_plugin_description();
if (evaluation_enabled) {
var response2 = JSON.parse($.ajax({type: "GET", url: "/api/datasets/" , async: false}).responseText);
get_datasets(response2);
draw_datasets();
}
$(window).on('hashchange', hashchanged);
hashchanged();
@@ -144,17 +181,34 @@ function get_default_parameters(){
}
function get_selected_plugin(){
return document.getElementsByClassName('plugin')[0].options[document.getElementsByClassName('plugin')[0].selectedIndex].value;
}
function draw_default_parameters(){
var basic_params = document.getElementById("basic_params");
basic_params.innerHTML = params_div(default_params);
}
function update_params(params, plug){
ep = plugins_params[plug];
for(k in ep){
params[k] = ep[k];
}
return params
}
function draw_extra_parameters(){
var plugin = document.getElementsByClassName('plugin')[0].options[document.getElementsByClassName('plugin')[0].selectedIndex].value;
var plugin = get_selected_plugin();
get_parameters();
var extra_params = document.getElementById("extra_params");
extra_params.innerHTML = params_div(plugins_params[plugin]);
var params = {};
for (sel in pipeline){
update_params(params, pipeline[sel]);
}
update_params(params, plugin);
extra_params.innerHTML = params_div(params);
}
function draw_parameters(){
@@ -261,6 +315,15 @@ function add_param(key, value){
return "&"+key+"="+value;
}
function get_pipeline_arg(){
arg = "";
for (ix in pipeline){
arg = arg + pipeline[ix] + ",";
}
arg = arg + get_selected_plugin();
return arg;
}
function load_JSON(){
url = "/api";
@@ -269,7 +332,9 @@ function load_JSON(){
rawcontainer.innerHTML = '';
container.innerHTML = '';
var plugin = document.getElementsByClassName("plugin")[0].options[document.getElementsByClassName("plugin")[0].selectedIndex].value;
var plugin = get_pipeline_arg();
$(".loading").addClass("loader");
$("#preview").hide();
var input = encodeURIComponent(document.getElementById("input").value);
url += "?algo="+plugin+"&i="+input
@@ -280,27 +345,27 @@ function load_JSON(){
url += add_param(key, params[key]);
}
var response = $.ajax({type: "GET", url: url , async: false}).responseText;
rawcontainer.innerHTML = replaceURLWithHTMLLinks(response);
document.getElementById("input_request").innerHTML = "<a href='"+url+"'>"+url+"</a>"
$.ajax({type: "GET", url: url}).always(function(response){
document.getElementById("results-div").style.display = 'block';
try {
response = JSON.parse(response);
if(typeof response=="object") {
var options = {
mode: 'view'
};
var editor = new JSONEditor(container, options, response);
editor.expandAll();
// $('#results-div a[href="#viewer"]').tab('show');
$('#results-div a[href="#viewer"]').click();
// location.hash = 'raw';
}
catch(err){
console.log("Error decoding JSON (got turtle?)");
response = JSON.stringify(response, null, 4);
} else {
console.log("Got turtle?");
$('#results-div a[href="#raw"]').click();
// location.hash = 'raw';
}
rawcontainer.innerHTML = replaceURLWithHTMLLinks(response);
document.getElementById("input_request").innerHTML = "<a href='"+url+"'>"+url+"</a>"
$(".loading").removeClass("loader");
$("#preview").show();
});
}
function get_datasets_from_checkbox(){
@@ -347,7 +412,11 @@ function evaluate_JSON(){
url += "?algo="+plugin+"&dataset="+datasets
var response = $.ajax({type: "GET", url: url , async: false, dataType: 'json'}).responseText;
$('#doevaluate').attr("disabled", true);
$.ajax({type: "GET", url: url, dataType: 'json'}).always(function(resp) {
$('#doevaluate').attr("disabled", false);
response = resp.responseText;
rawcontainer.innerHTML = replaceURLWithHTMLLinks(response);
document.getElementById("input_request_eval").innerHTML = "<a href='"+url+"'>"+url+"</a>"
@@ -380,7 +449,16 @@ function evaluate_JSON(){
$('#evaluate-div a[href="#evaluate-raw"]').click();
// location.hash = 'raw';
}
})
}
function draw_plugin_description(){
var plugin = plugins[get_selected_plugin()];
$("#plugdescription").text(plugin.description);
console.log(plugin);
}
function plugin_selected(){
draw_extra_parameters();
draw_plugin_description();
}

View File

@@ -5,6 +5,9 @@
<title>Playground {{version}}</title>
</head>
<script>
this.evaluation_enabled = {% if evaluation %}true{%else %}false{%endif%};
</script>
<script src="static/js/jquery-2.1.1.min.js" ></script>
<!--<script src="jquery.autosize.min.js"></script>-->
<link rel="stylesheet" href="static/css/bootstrap.min.css">
@@ -32,7 +35,9 @@
<ul class="nav nav-tabs" role="tablist">
<li role="presentation" ><a class="active" href="#about">About</a></li>
<li role="presentation"class="active"><a class="active" href="#test">Test it</a></li>
{% if evaluation %}
<li role="presentation"><a class="active" href="#evaluate">Evaluate Plugins</a></li>
{% endif %}
</ul>
@@ -61,6 +66,14 @@
</ul>
</p>
<p>Senpy is a research project. If you use it in your research, please cite:
<pre>
Senpy: A Pragmatic Linked Sentiment Analysis Framework.
Sánchez-Rada, J. F., Iglesias, C. A., Corcuera, I., & Araque, Ó.
In Data Science and Advanced Analytics (DSAA),
2016 IEEE International Conference on (pp. 735-742). IEEE.
</pre>
</p>
</div>
<div class="col-lg-6 ">
@@ -70,8 +83,6 @@
</div>
<div class="panel-body"><ul id=availablePlugins></ul></div>
</div>
</div>
<div class="col-lg-6 ">
<a href="http://senpy.readthedocs.io">
<div class="panel panel-default">
<div class="panel-heading"><i class="fa fa-book"></i> If you are new to senpy, you might want to read senpy's documentation</div>
@@ -82,9 +93,6 @@
<div class="panel-heading"><i class="fa fa-sign-in"></i> Feel free to follow us on GitHub</div>
</div>
</a>
<div class="panel panel-default">
<div class="panel-heading"><i class="fa fa-child"></i> Enjoy.</div>
</div>
</div>
</div>
</div>
@@ -96,17 +104,28 @@
whilst this text makes me happy and surprised at the same time.
I cannot believe it!</textarea>
</div>
<div>
<label>Select the plugin:</label>
<select id="plugins" name="plugins" class=plugin onchange="draw_extra_parameters()">
</select>
</div>
<!-- PARAMETERS -->
<div class="panel-group" id="parameters">
<div class="panel panel-default">
<a data-toggle="collapse" class="deco-none" href="#basic_params">
<div class="panel-heading">
<h4 class="panel-title">
Select the plugin.
</h4>
</div>
<div id="plugin_selection" class="panel-collapse panel-body">
<span id="pipeline"></span>
<select name="plugins" class="plugin" onchange="plugin_selected()">
</select>
<span onclick="add_plugin_pipeline()"><span class="btn"><i class="fa fa-plus" title="Add more plugins to the pipeline. Processing order is left to right. i.e. the results of the leftmost plugin will be used as input for the second leftmost, and so on."></i></span></span>
<label class="help-block " id="plugdescription"></label>
</div>
</div>
<div class="panel panel-default">
<a data-toggle="collapse" class="deco-none collapsed" href="#basic_params">
<div class="panel-heading">
<h4 class="panel-title">
<i class="fa fa-chevron-right pull-left expandicon"></i>
<i class="fa fa-chevron-down pull-left collapseicon"></i>
Basic API parameters
</h4>
</div>
@@ -118,6 +137,8 @@ I cannot believe it!</textarea>
<a data-toggle="collapse" class="deco-none" href="#extra_params">
<div class="panel-heading">
<h4 class="panel-title">
<i class="fa fa-chevron-right pull-left expandicon"></i>
<i class="fa fa-chevron-down pull-left collapseicon"></i>
Plugin extra parameters
</h4>
</div>
@@ -129,6 +150,7 @@ I cannot believe it!</textarea>
<!-- END PARAMETERS -->
<a id="preview" class="btn btn-lg btn-primary" onclick="load_JSON()">Analyse!</a>
<div id="loading-results" class="loading"></div>
<!--<button id="visualise" name="type" type="button">Visualise!</button>-->
</form>
</div>
@@ -155,6 +177,8 @@ I cannot believe it!</textarea>
</div>
</div>
{% if evaluation %}
<div class="tab-pane" id="evaluate">
<div class="well">
<form id="form" class="container" onsubmit="return getPlugins();" accept-charset="utf-8">
@@ -169,7 +193,7 @@ I cannot believe it!</textarea>
</select>
</div>
<a id="preview" class="btn btn-lg btn-primary" onclick="evaluate_JSON()">Evaluate Plugin!</a>
<a id="doevaluate" class="btn btn-lg btn-primary" onclick="evaluate_JSON()">Evaluate Plugin!</a>
<!--<button id="visualise" name="type" type="button">Visualise!</button>-->
</form>
</div>
@@ -216,6 +240,7 @@ I cannot believe it!</textarea>
</div>
</div>
</div>
{% endif %}
</div>
<a href="http://www.gsi.dit.upm.es" target="_blank"><img class="center-block" src="static/img/gsi.png"/> </a>

View File

@@ -1,36 +1,31 @@
try:
from unittest.mock import patch, MagicMock
except ImportError:
from mock import patch, MagicMock
from past.builtins import basestring
import json
from contextlib import contextmanager
import os
import responses as requestmock
from .models import BaseModel
@contextmanager
def patch_requests(value, code=200):
success = MagicMock()
if isinstance(value, BaseModel):
value = value.jsonld()
if not isinstance(value, basestring):
data = json.dumps(value)
MOCK_REQUESTS = os.environ.get('MOCK_REQUESTS', '').lower() in ['no', 'false']
def patch_all_requests(responses):
patched = requestmock.RequestsMock()
for response in responses or []:
args = response.copy()
if 'json' in args and isinstance(args['json'], BaseModel):
args['json'] = args['json'].jsonld()
args['method'] = getattr(requestmock, args.get('method', 'GET'))
patched.add(**args)
return patched
def patch_requests(url, response, method='GET', status=200):
args = {'url': url, 'method': method, 'status': status}
if isinstance(response, basestring):
args['body'] = response
else:
data = value
success.json.return_value = value
success.status_code = code
success.content = data
success.text = data
method_mocker = MagicMock()
method_mocker.return_value = success
with patch.multiple('requests', request=method_mocker,
get=method_mocker, post=method_mocker):
yield method_mocker, success
assert method_mocker.called
args['json'] = response
return patch_all_requests([args])

View File

@@ -80,7 +80,7 @@ def easy_test(plugin_list=None, debug=True):
for plug in plugin_list:
plug.test()
plug.log.info('My tests passed!')
logger.info('All tests passed!')
logger.info('All tests passed for {} plugins!'.format(len(plugin_list)))
except Exception:
if not debug:
raise

View File

@@ -12,6 +12,7 @@ max-line-length = 100
universal=1
[tool:pytest]
addopts = --cov=senpy --cov-report term-missing
filterwarnings =
ignore:the matrix subclass:PendingDeprecationWarning
[coverage:report]
omit = senpy/__main__.py

View File

@@ -3,8 +3,9 @@ import logging
logger = logging.getLogger(__name__)
from unittest import TestCase
from senpy.api import parse_params, API_PARAMS, NIF_PARAMS, WEB_PARAMS
from senpy.models import Error
from senpy.api import (boolean, parse_params, get_extra_params, parse_analysis,
API_PARAMS, NIF_PARAMS, WEB_PARAMS)
from senpy.models import Error, Plugin
class APITest(TestCase):
@@ -89,3 +90,156 @@ class APITest(TestCase):
assert "Dummy" in p['algorithm']
assert 'input' in p
assert p['input'] == 'Aloha my friend'
def test_parse_analysis(self):
'''The API should parse user parameters and return them in a format that plugins can use'''
plugins = [
Plugin({
'name': 'plugin1',
'extra_params': {
# Incompatible parameter
'param0': {
'aliases': ['p1', 'parameter1'],
'options': ['option1', 'option2'],
'default': 'option1',
'required': True
},
'param1': {
'aliases': ['p1', 'parameter1'],
'options': ['en', 'es'],
'default': 'en',
'required': False
},
'param2': {
'aliases': ['p2', 'parameter2'],
'required': False,
'options': ['value2_1', 'value2_2', 'value3_3']
}
}
}), Plugin({
'name': 'plugin2',
'extra_params': {
'param0': {
'aliases': ['parameter1'],
'options': ['new option', 'new option2'],
'default': 'new option',
'required': False
},
'param1': {
'aliases': ['myparam1', 'p1'],
'options': ['en', 'de', 'auto'],
'default': 'de',
'required': True
},
'param3': {
'aliases': ['p3', 'parameter3'],
'options': boolean,
'default': True
}
}
})
]
call = {
'param1': 'en',
'0.param0': 'option1',
'0.param1': 'en',
'param2': 'value2_1',
'param0': 'new option',
'1.param1': 'de',
'param3': False,
}
expected = [
{
'param0': 'option1',
'param1': 'en',
'param2': 'value2_1',
}, {
'param0': 'new option',
'param1': 'de',
'param3': False,
}
]
p = parse_analysis(call, plugins)
for i, arg in enumerate(expected):
params = p[i].params
for k, v in arg.items():
assert params[k] == v
def test_get_extra_params(self):
'''The API should return the list of valid parameters for a set of plugins'''
plugins = [
Plugin({
'name': 'plugin1',
'extra_params': {
# Incompatible parameter
'param0': {
'aliases': ['p1', 'parameter1'],
'options': ['option1', 'option2'],
'default': 'option1',
'required': True
},
'param1': {
'aliases': ['p1', 'parameter1'],
'options': ['en', 'es'],
'default': 'en',
'required': False
},
'param2': {
'aliases': ['p2', 'parameter2'],
'required': False,
'options': ['value2_1', 'value2_2', 'value3_3']
}
}
}), Plugin({
'name': 'plugin2',
'extra_params': {
'param0': {
'aliases': ['parameter1'],
'options': ['new option', 'new option2'],
'default': 'new option',
'required': False
},
'param1': {
'aliases': ['myparam1', 'p1'],
'options': ['en', 'de', 'auto'],
'default': 'de',
'required': True
},
'param3': {
'aliases': ['p3', 'parameter3'],
'options': boolean,
'default': True
}
}
})
]
expected = {
# Overlapping parameters
'plugin1.param0': plugins[0]['extra_params']['param0'],
'plugin1.param1': plugins[0]['extra_params']['param1'],
'plugin2.param0': plugins[1]['extra_params']['param0'],
'plugin2.param1': plugins[1]['extra_params']['param1'],
# Non-overlapping parameters
'param2': plugins[0]['extra_params']['param2'],
'param3': plugins[1]['extra_params']['param3'],
# Intersection of overlapping parameters
'param1': {
'aliases': ['p1'],
'options': ['en'],
'default': None,
'required': True
}
}
result = get_extra_params(plugins)
for ik, iv in expected.items():
assert ik in result
for jk, jv in iv.items():
assert jk in result[ik]
assert expected[ik][jk] == result[ik][jk]

View File

@@ -21,27 +21,29 @@ class BlueprintsTest(TestCase):
def setUpClass(cls):
"""Set up only once, and re-use in every individual test"""
cls.app = Flask("test_extensions")
cls.app.debug = False
cls.client = cls.app.test_client()
cls.senpy = Senpy(default_plugins=True)
cls.senpy.init_app(cls.app)
cls.dir = os.path.join(os.path.dirname(__file__), "..")
cls.senpy.add_folder(cls.dir)
cls.senpy.activate_plugin("Dummy", sync=True)
cls.senpy.activate_plugin("DummyRequired", sync=True)
cls.senpy.activate_all()
cls.senpy.default_plugin = 'Dummy'
def setUp(self):
self.app.config['TESTING'] = True # Tell Flask not to catch Exceptions
def assertCode(self, resp, code):
self.assertEqual(resp.status_code, code)
def test_playground(self):
resp = self.client.get("/")
assert "main.js" in resp.data.decode()
assert "main.js" in resp.get_data(as_text=True)
def test_home(self):
"""
Calling with no arguments should ask the user for more arguments
"""
self.app.config['TESTING'] = False # Errors are expected in this case
resp = self.client.get("/api/")
self.assertCode(resp, 400)
js = parse_resp(resp)
@@ -64,10 +66,25 @@ class BlueprintsTest(TestCase):
logging.debug("Got response: %s", js)
assert "@context" in js
assert "entries" in js
assert len(js['analysis']) == 1
def test_analysis_post(self):
"""
The results for a POST request should be the same as for a GET request.
"""
resp = self.client.post("/api/", data={'i': 'My aloha mohame',
'algorithm': 'rand',
'with_parameters': True})
self.assertCode(resp, 200)
js = parse_resp(resp)
logging.debug("Got response: %s", js)
assert "@context" in js
assert "entries" in js
assert len(js['analysis']) == 1
def test_analysis_extra(self):
"""
Extra params that have a default should
Extra params that have a default should use it
"""
resp = self.client.get("/api/?i=My aloha mohame&algo=Dummy&with_parameters=true")
self.assertCode(resp, 200)
@@ -81,7 +98,7 @@ class BlueprintsTest(TestCase):
Extra params that have a required argument that does not
have a default should raise an error.
"""
self.app.debug = False
self.app.config['TESTING'] = False # Errors are expected in this case
resp = self.client.get("/api/?i=My aloha mohame&algo=DummyRequired")
self.assertCode(resp, 400)
js = parse_resp(resp)
@@ -89,15 +106,117 @@ class BlueprintsTest(TestCase):
assert isinstance(js, models.Error)
resp = self.client.get("/api/?i=My aloha mohame&algo=DummyRequired&example=notvalid")
self.assertCode(resp, 400)
self.app.config['TESTING'] = True
resp = self.client.get("/api/?i=My aloha mohame&algo=DummyRequired&example=a")
self.assertCode(resp, 200)
def test_analysis_url(self):
"""
The algorithm can also be specified as part of the URL
"""
self.app.config['TESTING'] = False # Errors are expected in this case
resp = self.client.get("/api/DummyRequired?i=My aloha mohame")
self.assertCode(resp, 400)
js = parse_resp(resp)
logging.debug("Got response: %s", js)
assert isinstance(js, models.Error)
resp = self.client.get("/api/DummyRequired?i=My aloha mohame&example=notvalid")
self.assertCode(resp, 400)
resp = self.client.get("/api/DummyRequired?i=My aloha mohame&example=a")
self.assertCode(resp, 200)
def test_analysis_chain(self):
"""
More than one algorithm can be specified. Plugins will then be chained
"""
resp = self.client.get("/api/Dummy?i=My aloha mohame")
js = parse_resp(resp)
assert len(js['analysis']) == 1
assert js['entries'][0]['nif:isString'] == 'My aloha mohame'[::-1]
resp = self.client.get("/api/Dummy/Dummy?i=My aloha mohame")
# Calling dummy twice, should return the same string
self.assertCode(resp, 200)
js = parse_resp(resp)
assert len(js['analysis']) == 2
assert js['entries'][0]['nif:isString'] == 'My aloha mohame'
resp = self.client.get("/api/Dummy+Dummy?i=My aloha mohame")
# Same with pluses instead of slashes
self.assertCode(resp, 200)
js = parse_resp(resp)
assert len(js['analysis']) == 2
assert js['entries'][0]['nif:isString'] == 'My aloha mohame'
def test_analysis_chain_required(self):
"""
If a parameter is required and duplicated (because two plugins require it), specifying
it once should suffice
"""
resp = self.client.get("/api/DummyRequired/DummyRequired?i=My aloha mohame&example=a")
js = parse_resp(resp)
assert len(js['analysis']) == 2
assert js['entries'][0]['nif:isString'] == 'My aloha mohame'
assert js['entries'][0]['reversed'] == 2
def test_requirements_chain_help(self):
'''The extra parameters of each plugin should be merged if they are in a chain '''
resp = self.client.get("/api/split/DummyRequired?help=true")
self.assertCode(resp, 200)
js = parse_resp(resp)
assert 'valid_parameters' in js
vp = js['valid_parameters']
assert 'example' in vp
assert 'delimiter' in vp
def test_requirements_chain_repeat_help(self):
'''
If a plugin appears several times in a chain, there should be a way to set different
parameters for each.
'''
resp = self.client.get("/api/split/split?help=true")
self.assertCode(resp, 200)
js = parse_resp(resp)
assert 'valid_parameters' in js
vp = js['valid_parameters']
assert 'delimiter' in vp
resp = self.client.get("/api/split/split?help=true&verbose=false")
js = parse_resp(resp)
vp = js['valid_parameters']
assert len(vp.keys()) == 1
def test_requirements_chain(self):
"""
It should be possible to specify different parameters for each step in the chain.
"""
# First, we split by sentence twice. Each call should generate 3 additional entries
# (one per sentence in the original).
resp = self.client.get('/api/split/split?i=The first sentence. The second sentence.'
'\nA new paragraph&delimiter=sentence')
js = parse_resp(resp)
assert len(js['analysis']) == 2
assert len(js['entries']) == 7
# Now, we split by sentence. This produces 3 additional entries.
# Then, we split by paragraph. This should create 2 additional entries (One per paragraph
# in the original text)
resp = self.client.get('/api/split/split?i=The first sentence. The second sentence.'
'\nA new paragraph&0.delimiter=sentence&1.delimiter=paragraph')
# Calling dummy twice, should return the same string
self.assertCode(resp, 200)
js = parse_resp(resp)
assert len(js['analysis']) == 2
assert len(js['entries']) == 6
def test_error(self):
"""
The dummy plugin returns an empty response,\
it should contain the context
"""
self.app.debug = False
self.app.config['TESTING'] = False # Errors are expected in this case
resp = self.client.get("/api/?i=My aloha mohame&algo=DOESNOTEXIST")
self.assertCode(resp, 404)
js = parse_resp(resp)
@@ -139,7 +258,7 @@ class BlueprintsTest(TestCase):
js = parse_resp(resp)
logging.debug(js)
assert "@id" in js
assert js["@id"] == "plugins/Dummy_0.1"
assert js["@id"] == "endpoint:plugins/Dummy_0.1"
def test_default(self):
""" Show only one plugin"""
@@ -148,7 +267,7 @@ class BlueprintsTest(TestCase):
js = parse_resp(resp)
logging.debug(js)
assert "@id" in js
assert js["@id"] == "plugins/Dummy_0.1"
assert js["@id"] == "endpoint:plugins/Dummy_0.1"
def test_context(self):
resp = self.client.get("/api/contexts/context.jsonld")
@@ -172,5 +291,6 @@ class BlueprintsTest(TestCase):
assert "help" in js["valid_parameters"]
def test_conversion(self):
self.app.config['TESTING'] = False # Errors are expected in this case
resp = self.client.get("/api/?input=hello&algo=emoRand&emotionModel=DOES NOT EXIST")
self.assertCode(resp, 404)

View File

@@ -14,22 +14,28 @@ class ModelsTest(TestCase):
def test_client(self):
endpoint = 'http://dummy/'
client = Client(endpoint)
with patch_requests(Results()) as (request, response):
with patch_requests('http://dummy/', Results()):
resp = client.analyse('hello')
assert isinstance(resp, Results)
request.assert_called_with(
url=endpoint + '/', method='GET', params={'input': 'hello'})
with patch_requests(Error('Nothing')) as (request, response):
with patch_requests('http://dummy/', Error('Nothing')):
try:
client.analyse(input='hello', algorithm='NONEXISTENT')
raise Exception('Exceptions should be raised. This is not golang')
except Error:
pass
request.assert_called_with(
url=endpoint + '/',
method='GET',
params={'input': 'hello',
'algorithm': 'NONEXISTENT'})
def test_client_post(self):
endpoint = 'http://dummy/'
client = Client(endpoint)
with patch_requests('http://dummy/', Results()):
resp = client.analyse('hello')
assert isinstance(resp, Results)
with patch_requests('http://dummy/', Error('Nothing'), method='POST'):
try:
client.analyse(input='hello', method='POST', algorithm='NONEXISTENT')
raise Exception('Exceptions should be raised. This is not golang')
except Error:
pass
def test_plugins(self):
endpoint = 'http://dummy/'
@@ -37,11 +43,8 @@ class ModelsTest(TestCase):
plugins = Plugins()
p1 = AnalysisPlugin({'name': 'AnalysisP1', 'version': 0, 'description': 'No'})
plugins.plugins = [p1, ]
with patch_requests(plugins) as (request, response):
with patch_requests('http://dummy/plugins', plugins):
response = client.plugins()
assert isinstance(response, dict)
assert len(response) == 1
assert "AnalysisP1" in response
request.assert_called_with(
url=endpoint + '/plugins', method='GET',
params={})

View File

@@ -11,14 +11,15 @@ except ImportError:
from functools import partial
from senpy.extensions import Senpy
from senpy import plugins
from senpy.models import Error, Results, Entry, EmotionSet, Emotion, Plugin
from senpy.models import Analysis, Error, Results, Entry, EmotionSet, Emotion, Plugin
from senpy import api
from flask import Flask
from unittest import TestCase
def analyse(instance, **kwargs):
request = api.parse_call(kwargs)
basic = api.parse_params(kwargs, api.API_PARAMS)
request = api.parse_call(basic)
return instance.analyse(request)
@@ -49,9 +50,9 @@ class ExtensionsTest(TestCase):
'''Should be able to add and delete new plugins. '''
new = plugins.Analysis(name='new', description='new', version=0)
self.senpy.add_plugin(new)
assert new in self.senpy.plugins()
assert new in self.senpy.plugins(is_activated=False)
self.senpy.delete_plugin(new)
assert new not in self.senpy.plugins()
assert new not in self.senpy.plugins(is_activated=False)
def test_adding_folder(self):
""" It should be possible for senpy to look for plugins in more folders. """
@@ -60,7 +61,7 @@ class ExtensionsTest(TestCase):
default_plugins=False)
assert not senpy.analysis_plugins
senpy.add_folder(self.examples_dir)
assert senpy.analysis_plugins
assert senpy.plugins(plugin_type=plugins.AnalysisPlugin, is_activated=False)
self.assertRaises(AttributeError, senpy.add_folder, 'DOES NOT EXIST')
def test_installing(self):
@@ -121,8 +122,8 @@ class ExtensionsTest(TestCase):
# Leaf (defaultdict with __setattr__ and __getattr__.
r1 = analyse(self.senpy, algorithm="Dummy", input="tupni", output="tuptuo")
r2 = analyse(self.senpy, input="tupni", output="tuptuo")
assert r1.analysis[0] == "plugins/Dummy_0.1"
assert r2.analysis[0] == "plugins/Dummy_0.1"
assert r1.analysis[0].algorithm == "endpoint:plugins/Dummy_0.1"
assert r2.analysis[0].algorithm == "endpoint:plugins/Dummy_0.1"
assert r1.entries[0]['nif:isString'] == 'input'
def test_analyse_empty(self):
@@ -130,7 +131,7 @@ class ExtensionsTest(TestCase):
senpy = Senpy(plugin_folder=None,
app=self.app,
default_plugins=False)
self.assertRaises(Error, senpy.analyse, Results())
self.assertRaises(Error, senpy.analyse, Results(), [])
def test_analyse_wrong(self):
""" Trying to analyse with a non-existent plugin should raise an error."""
@@ -156,30 +157,32 @@ class ExtensionsTest(TestCase):
r2 = analyse(self.senpy,
input="tupni",
output="tuptuo")
assert r1.analysis[0] == "plugins/Dummy_0.1"
assert r2.analysis[0] == "plugins/Dummy_0.1"
assert r1.analysis[0].algorithm == "endpoint:plugins/Dummy_0.1"
assert r2.analysis[0].algorithm == "endpoint:plugins/Dummy_0.1"
assert r1.entries[0]['nif:isString'] == 'input'
def test_analyse_error(self):
mm = mock.MagicMock()
mm.id = 'magic_mock'
mm.name = 'mock'
mm.is_activated = True
mm.analyse_entries.side_effect = Error('error in analysis', status=500)
self.senpy.add_plugin(mm)
class ErrorPlugin(plugins.Analysis):
author = 'nobody'
version = 0
ex = Error()
def process(self, *args, **kwargs):
raise self.ex
m = ErrorPlugin(ex=Error('error in analysis', status=500))
self.senpy.add_plugin(m)
try:
analyse(self.senpy, input='nothing', algorithm='MOCK')
analyse(self.senpy, input='nothing', algorithm='ErrorPlugin')
assert False
except Error as ex:
assert 'error in analysis' in ex['message']
assert ex['status'] == 500
ex = Exception('generic exception on analysis')
mm.analyse.side_effect = ex
mm.analyse_entries.side_effect = ex
m.ex = Exception('generic exception on analysis')
try:
analyse(self.senpy, input='nothing', algorithm='MOCK')
analyse(self.senpy, input='nothing', algorithm='ErrorPlugin')
assert False
except Exception as ex:
assert 'generic exception on analysis' in str(ex)
@@ -195,7 +198,7 @@ class ExtensionsTest(TestCase):
def test_load_default_plugins(self):
senpy = Senpy(plugin_folder=self.examples_dir, default_plugins=True)
assert len(senpy.plugins()) > 1
assert len(senpy.plugins(is_activated=False)) > 1
def test_convert_emotions(self):
self.senpy.activate_all(sync=True)
@@ -211,27 +214,28 @@ class ExtensionsTest(TestCase):
'emoml:valence': 0
}))
response = Results({
'analysis': [{'plugin': plugin}],
'analysis': [plugin],
'entries': [Entry({
'nif:isString': 'much ado about nothing',
'emotions': [eSet1]
})]
})
params = {'emotionModel': 'emoml:big6',
'algorithm': ['conversion'],
'conversion': 'full'}
r1 = deepcopy(response)
r1.parameters = params
self.senpy.convert_emotions(r1)
self.senpy.analyse(r1)
assert len(r1.entries[0].emotions) == 2
params['conversion'] = 'nested'
r2 = deepcopy(response)
r2.parameters = params
self.senpy.convert_emotions(r2)
self.senpy.analyse(r2)
assert len(r2.entries[0].emotions) == 1
assert r2.entries[0].emotions[0]['prov:wasDerivedFrom'] == eSet1
params['conversion'] = 'filtered'
r3 = deepcopy(response)
r3.parameters = params
self.senpy.convert_emotions(r3)
self.senpy.analyse(r3)
assert len(r3.entries[0].emotions) == 1
r3.jsonld()

View File

@@ -5,7 +5,8 @@ import jsonschema
import json
import rdflib
from unittest import TestCase
from senpy.models import (Emotion,
from senpy.models import (Analysis,
Emotion,
EmotionAnalysis,
EmotionSet,
Entry,
@@ -61,7 +62,7 @@ class ModelsTest(TestCase):
def test_id(self):
""" Adding the id after creation should overwrite the automatic ID
"""
r = Entry()
r = Entry(_auto_id=True)
j = r.jsonld()
assert '@id' in j
r.id = "test"
@@ -189,6 +190,19 @@ class ModelsTest(TestCase):
assert isinstance(js['plugins'], list)
assert js['plugins'][0]['@type'] == 'sentimentPlugin'
def test_parameters(self):
'''An Analysis should contain the algorithm and the list of parameters to be used'''
a = Analysis()
a.params = {'param1': 1, 'param2': 2}
assert len(a.parameters) == 2
for param in a.parameters:
if param.name == 'param1':
assert param.value == 1
elif param.name == 'param2':
assert param.value == 2
else:
raise Exception('Unknown value %s' % param)
def test_from_string(self):
results = {
'@type': 'results',

View File

@@ -1,7 +1,6 @@
#!/bin/env python
import os
import sys
import pickle
import shutil
import tempfile
@@ -9,7 +8,8 @@ import tempfile
from unittest import TestCase, skipIf
from senpy.models import Results, Entry, EmotionSet, Emotion, Plugins
from senpy import plugins
from senpy.plugins.conversion.emotion.centroids import CentroidConversion
from senpy.plugins.postprocessing.emotion.centroids import CentroidConversion
from senpy.gsitk_compat import GSITK_AVAILABLE
import pandas as pd
@@ -312,9 +312,7 @@ class PluginsTest(TestCase):
res = c._backwards_conversion(e)
assert res["onyx:hasEmotionCategory"] == "c2"
@skipIf(sys.version_info < (3, 0),
reason="requires Python3")
def test_evaluation(self):
def _test_evaluation(self):
testdata = []
for i in range(50):
testdata.append(["good", 1])
@@ -348,6 +346,16 @@ class PluginsTest(TestCase):
smart_metrics = results[0].metrics[0]
assert abs(smart_metrics['accuracy'] - 1) < 0.01
@skipIf(not GSITK_AVAILABLE, "GSITK is not available")
def test_evaluation(self):
self._test_evaluation()
@skipIf(GSITK_AVAILABLE, "GSITK is available")
def test_evaluation_unavailable(self):
with self.assertRaises(Exception) as context:
self._test_evaluation()
self.assertTrue('GSITK ' in str(context.exception))
def make_mini_test(fpath):
def mini_test(self):

View File

@@ -5,28 +5,29 @@ import json
from senpy.testing import patch_requests
from senpy.models import Results
ENDPOINT = 'http://example.com'
class TestTest(TestCase):
def test_patch_text(self):
with patch_requests('hello'):
r = requests.get('http://example.com')
with patch_requests(ENDPOINT, 'hello'):
r = requests.get(ENDPOINT)
assert r.text == 'hello'
assert r.content == 'hello'
def test_patch_json(self):
r = Results()
with patch_requests(r):
res = requests.get('http://example.com')
assert res.content == json.dumps(r.jsonld())
with patch_requests(ENDPOINT, r):
res = requests.get(ENDPOINT)
assert res.text == json.dumps(r.jsonld())
js = res.json()
assert js
assert js['@type'] == r['@type']
def test_patch_dict(self):
r = {'nothing': 'new'}
with patch_requests(r):
res = requests.get('http://example.com')
assert res.content == json.dumps(r)
with patch_requests(ENDPOINT, r):
res = requests.get(ENDPOINT)
assert res.text == json.dumps(r)
js = res.json()
assert js
assert js['nothing'] == 'new'