Fix extra requirements

GSITK doesn't support python2

.gitlab-ci.yml

@@ -55,6 +55,7 @@ push-latest:
   only:
     - master
     - triggers
+    - fix-makefiles
 
 push-github:
   stage: deploy
@@ -69,6 +70,7 @@ deploy_pypi:
   stage: deploy
   script:   # Configure the PyPI credentials, then push the package, and cleanup the creds.
     - echo "[server-login]" >> ~/.pypirc
+    - echo "repository=https://upload.pypi.org/legacy/" >> ~/.pypirc
     - echo "username=" ${PYPI_USER} >> ~/.pypirc
     - echo "password=" ${PYPI_PASSWORD} >> ~/.pypirc
     - make pip_upload

.makefiles/base.mk

@@ -2,18 +2,16 @@ export
 NAME ?= $(shell basename $(CURDIR))
 VERSION ?= $(shell git describe --tags --dirty 2>/dev/null)
 
+ifeq ($(VERSION),)
+	VERSION:=unknown
+endif
+
 # Get the location of this makefile.
 MK_DIR := $(dir $(abspath $(lastword $(MAKEFILE_LIST))))
 
 -include .env
 -include ../.env
 
-.FORCE:
-
-version: .FORCE
-	@echo $(VERSION) > $(NAME)/VERSION
-	@echo $(VERSION)
-
 help:           ## Show this help.
 	@fgrep -h "##" $(MAKEFILE_LIST) | fgrep -v fgrep | sed -e 's/\\$$//' | sed -e 's/\(.*:\)[^#]*##\s*\(.*\)/\1\t\2/' | column -t -s "	"
 
@@ -35,4 +33,4 @@ include $(MK_DIR)/git.mk
 info:: ## List all variables
 	env
 
-.PHONY:: config help ci version .FORCE
+.PHONY:: config help ci

.makefiles/docker.mk

@@ -1,4 +1,5 @@
-IMAGEWTAG ?= $(IMAGENAME):$(VERSION)
+IMAGENAME?=$(NAME)
+IMAGEWTAG?=$(IMAGENAME):$(VERSION)
 
 docker-login: ## Log in to the registry. It will only be used in the server, or when running a CI task locally (if CI_BUILD_TOKEN is set).
 ifeq ($(CI_BUILD_TOKEN),)
@@ -22,4 +23,7 @@ login:: docker-login
 
 clean:: docker-clean
 
+docker-info:
+	@echo IMAGEWTAG=${IMAGEWTAG}
+
 .PHONY:: docker-login docker-clean login clean

.makefiles/python.mk

@@ -1,9 +1,17 @@
-PYVERSIONS ?= 2.7
+PYVERSIONS ?= 3.5
 PYMAIN ?= $(firstword $(PYVERSIONS))
 TARNAME ?= $(NAME)-$(VERSION).tar.gz 
+VERSIONFILE ?= $(NAME)/VERSION
 
 DEVPORT ?= 6000
 
+
+.FORCE:
+
+version: .FORCE
+	@echo $(VERSION) > $(VERSIONFILE)
+	@echo $(VERSION)
+
 yapf: ## Format python code
 	yapf -i -r $(NAME)
 	yapf -i -r tests
@@ -20,7 +28,7 @@ quick_build: $(addprefix build-, $(PYMAIN))
 build: $(addprefix build-, $(PYVERSIONS)) ## Build all images / python versions
 
 build-%: version Dockerfile-%  ## Build a specific version (e.g. build-2.7)
-	docker build -t '$(IMAGEWTAG)-python$*' --cache-from $(IMAGENAME):python$* -f Dockerfile-$* .;
+	docker build -t '$(IMAGEWTAG)-python$*' -f Dockerfile-$* .;
 
 dev-%: ## Launch a specific development environment using docker (e.g. dev-2.7)
 	@docker start $(NAME)-dev$* || (\
@@ -34,10 +42,10 @@ dev: dev-$(PYMAIN) ## Launch a development environment using docker, using the d
 
 quick_test: test-$(PYMAIN)
 
-test-%: ## Run setup.py from in an isolated container, built from the base image. (e.g. test-2.7)
+test-%: build-% ## Run setup.py from in an isolated container, built from the base image. (e.g. test-2.7)
 # This speeds tests up because the image has most (if not all) of the dependencies already.
 	docker rm $(NAME)-test-$* || true
-	docker create -ti --name $(NAME)-test-$* --entrypoint="" -w /usr/src/app/ $(IMAGENAME):python$* python setup.py test
+	docker create -ti --name $(NAME)-test-$* --entrypoint="" -w /usr/src/app/ $(IMAGEWTAG)-python$* python setup.py test
 	docker cp . $(NAME)-test-$*:/usr/src/app
 	docker start -a $(NAME)-test-$*
 
@@ -89,4 +97,4 @@ clean:: ## Clean older docker images and containers related to this project and
 	@docker ps -a | grep $(IMAGENAME) | awk '{ split($$2, vers, "-"); if(vers[0] != "${VERSION}"){ print $$1;}}' | xargs docker rm -v 2>/dev/null|| true
 	@docker images | grep $(IMAGENAME) | awk '{ split($$2, vers, "-"); if(vers[0] != "${VERSION}"){ print $$1":"$$2;}}' | xargs docker rmi 2>/dev/null|| true
 
-.PHONY:: yapf dockerfiles Dockerfile-% quick_build build build-% dev-% quick-dev test quick_test push-latest push-latest-% push-% push
+.PHONY:: yapf dockerfiles Dockerfile-% quick_build build build-% dev-% quick-dev test quick_test push-latest push-latest-% push-% push version .FORCE

Dockerfile.template

@@ -2,6 +2,12 @@ from python:{{PYVERSION}}
 
 MAINTAINER J. Fernando Sánchez <jf.sanchez@upm.es>
 
+RUN apt-get update && apt-get install -y \
+libblas-dev liblapack-dev liblapacke-dev gfortran \
+ && rm -rf /var/lib/apt/lists/*
+
+RUN pip install --no-cache-dir --upgrade numpy scipy scikit-learn
+
 RUN mkdir /cache/ /senpy-plugins /data/
 
 VOLUME /data/
@@ -14,9 +20,9 @@ ONBUILD WORKDIR /senpy-plugins/
 
 
 WORKDIR /usr/src/app
-COPY test-requirements.txt requirements.txt /usr/src/app/
+COPY test-requirements.txt requirements.txt extra-requirements.txt /usr/src/app/
-RUN pip install --use-wheel -r test-requirements.txt -r requirements.txt
+RUN pip install --no-cache-dir -r test-requirements.txt -r requirements.txt -r extra-requirements.txt
 COPY . /usr/src/app/
-RUN pip install --no-index --no-deps --editable .
+RUN pip install --no-cache-dir --no-index --no-deps --editable .
 
 ENTRYPOINT ["python", "-m", "senpy", "-f", "/senpy-plugins/", "--host", "0.0.0.0"]

README.rst

@@ -1,5 +1,5 @@
 .. image:: img/header.png
-   :height: 6em
+   :width: 100%
    :target: http://demos.gsi.dit.upm.es/senpy
 
 .. image:: https://travis-ci.org/gsi-upm/senpy.svg?branch=master

docs/SenpyClientUse.rst

@@ -0,0 +1,106 @@
+
+Client
+======
+
+Demo Endpoint
+-------------
+
+Import Client and send a request
+
+.. code:: python
+
+    from senpy.client import Client
+    
+    c = Client('http://latest.senpy.cluster.gsi.dit.upm.es/api')
+    r = c.analyse('I like Pizza', algorithm='sentiment140')
+
+Print response
+
+.. code:: python
+
+    for entry in r.entries:
+          print('{} -> {}'.format(entry['text'], entry['sentiments'][0]['marl:hasPolarity']))
+
+
+.. parsed-literal::
+
+    I like Pizza -> marl:Positive
+
+
+Obtain a list of available plugins
+
+.. code:: python
+
+    for plugin in c.request('/plugins')['plugins']:
+        print(plugin['name'])
+
+
+.. parsed-literal::
+
+    emoRand
+    rand
+    sentiment140
+
+
+Local Endpoint
+--------------
+
+Run a docker container with Senpy image and default plugins
+
+.. code::
+
+    docker run -ti --name 'SenpyEndpoint' -d -p 5000:5000 gsiupm/senpy:0.8.6 --host 0.0.0.0 --default-plugins
+
+
+.. parsed-literal::
+
+    a0157cd98057072388bfebeed78a830da7cf0a796f4f1a3fd9188f9f2e5fe562
+
+
+Import client and send a request to localhost
+
+.. code:: python
+
+    c_local = Client('http://127.0.0.1:5000/api')
+    r = c_local.analyse('Hello world', algorithm='sentiment140')
+
+Print response
+
+.. code:: python
+
+    for entry in r.entries:
+          print('{} -> {}'.format(entry['text'], entry['sentiments'][0]['marl:hasPolarity']))
+
+
+.. parsed-literal::
+
+    Hello world -> marl:Neutral
+
+
+Obtain a list of available plugins deployed locally
+
+.. code:: python
+
+    c_local.plugins().keys()
+
+
+.. parsed-literal::
+
+    rand
+    sentiment140
+    emoRand
+
+
+Stop the docker container
+
+.. code:: python
+
+    !docker stop SenpyEndpoint
+    !docker rm SenpyEndpoint
+
+
+.. parsed-literal::
+
+    SenpyEndpoint
+    SenpyEndpoint
+

docs/plugins-definition.rst

@@ -0,0 +1,113 @@
+Advanced plugin definition
+--------------------------
+In addition to finding plugins defined in source code files, senpy can also load a special type of definition file (`.senpy` files).
+This used to be the only mechanism for loading in earlier versions of senpy.
+
+The definition file contains basic information 
+
+Lastly, it is also possible to add new plugins programmatically.
+
+.. contents:: :local:
+
+What is a plugin?
+=================
+
+A plugin is a program that, given a text, will add annotations to it.
+In practice, a plugin consists of at least two files:
+
+- Definition file: a `.senpy` file that describes the plugin (e.g. what input parameters it accepts, what emotion model it uses).
+- Python module: the actual code that will add annotations to each input.
+
+This separation allows us to deploy plugins that use the same code but employ different parameters.
+For instance, one could use the same classifier and processing in several plugins, but train with different datasets.
+This scenario is particularly useful for evaluation purposes.
+
+The only limitation is that the name of each plugin needs to be unique.
+
+Definition files
+================
+
+The definition file complements and overrides the attributes provided by the plugin.
+It can be written in YAML or JSON.
+The most important attributes are:
+
+* **name**: unique name that senpy will use internally to identify the plugin.
+* **module**: indicates the module that contains the plugin code, which will be automatically loaded by senpy.
+* **version**
+* extra_params: to add parameters to the senpy API when this plugin is requested. Those parameters may be required, and have aliased names. For instance:
+
+  .. code:: yaml
+
+            extra_params:
+                hello_param:
+                    aliases: # required
+                        - hello_param
+                        - hello
+                    required: true
+                    default: Hi you
+                    values:
+                        - Hi you
+                        - Hello y'all
+                        - Howdy
+
+A complete example:
+
+.. code:: yaml
+          
+          name: <Name of the plugin>
+          module: <Python file>
+          version: 0.1
+
+And the json equivalent:
+
+.. code:: json
+
+          {
+            "name": "<Name of the plugin>",
+            "module": "<Python file>",
+            "version": "0.1"
+          }
+
+
+Example plugin with a definition file
+=====================================
+
+In this section, we will implement a basic sentiment analysis plugin.
+To determine the polarity of each entry, the plugin will compare the length of the string to a threshold.
+This threshold will be included in the definition file.
+
+The definition file would look like this:
+
+.. code:: yaml
+
+          name: helloworld
+          module: helloworld
+          version: 0.0
+          threshold: 10
+          description: Hello World
+
+Now, in a file named ``helloworld.py``:
+
+.. code:: python
+
+          #!/bin/env python
+          #helloworld.py
+
+          from senpy import AnalysisPlugin
+          from senpy import Sentiment
+
+
+          class HelloWorld(AnalysisPlugin):
+
+              def analyse_entry(entry, params):
+                  '''Basically do nothing with each entry'''
+
+                  sentiment = Sentiment()
+                  if len(entry.text) < self.threshold:
+                      sentiment['marl:hasPolarity'] = 'marl:Positive'
+                  else:
+                      sentiment['marl:hasPolarity'] = 'marl:Negative'
+                  entry.sentiments.append(sentiment)
+                  yield entry
+
+The complete code of the example plugin is available `here <https://lab.cluster.gsi.dit.upm.es/senpy/plugin-prueba>`__.

docs/plugins.rst

@@ -1,6 +1,8 @@
 Developing new plugins
 ----------------------
-This document describes how to develop a new analysis plugin. For an example of conversion plugins, see :doc:`conversion`.
+This document contains the minimum to get you started with developing new analysis plugin.
+For an example of conversion plugins, see :doc:`conversion`.
+For a description of definition files, see :doc:`plugins-definition`.
 
 A more step-by-step tutorial with slides is available `here <https://lab.cluster.gsi.dit.upm.es/senpy/senpy-tutorial>`__ 
 
@@ -9,83 +11,29 @@ A more step-by-step tutorial with slides is available `here <https://lab.cluster
 What is a plugin?
 =================
 
-A plugin is a program that, given a text, will add annotations to it.
+A plugin is a python object that can process entries. Given an entry, it will modify it, add annotations to it, or generate new entries.
-In practice, a plugin consists of at least two files:
-
-- Definition file: a `.senpy` file that describes the plugin (e.g. what input parameters it accepts, what emotion model it uses).
-- Python module: the actual code that will add annotations to each input.
-
-This separation allows us to deploy plugins that use the same code but employ different parameters.
-For instance, one could use the same classifier and processing in several plugins, but train with different datasets.
-This scenario is particularly useful for evaluation purposes.
-
-The only limitation is that the name of each plugin needs to be unique.
-
-Plugin Definition files
-=======================
-
-The definition file contains all the attributes of the plugin, and can be written in YAML or JSON.
-When the server is launched, it will recursively search for definition files in the plugin folder (the current folder, by default).
-The most important attributes are:
-
-* **name**: unique name that senpy will use internally to identify the plugin.
-* **module**: indicates the module that contains the plugin code, which will be automatically loaded by senpy.
-* **version**
-* extra_params: to add parameters to the senpy API when this plugin is requested. Those parameters may be required, and have aliased names. For instance:
-
-  .. code:: yaml
-
-            extra_params:
-                hello_param:
-                    aliases: # required
-                        - hello_param
-                        - hello
-                    required: true
-                    default: Hi you
-                    values:
-                        - Hi you
-                        - Hello y'all
-                        - Howdy
-
-  Parameter validation will fail if a required parameter without a default has not been provided, or if the definition includes a set of values and the provided one does not match one of them.
 
 
-A complete example:
+What is an entry?
-
+=================
-.. code:: yaml
-          
-          name: <Name of the plugin>
-          module: <Python file>
-          version: 0.1
-
-And the json equivalent:
-
-.. code:: json
-
-          {
-            "name": "<Name of the plugin>",
-            "module": "<Python file>",
-            "version": "0.1"
-          }
-
-
-Plugins Code
-============
-
-The basic methods in a plugin are:
-
-* __init__
-* activate: used to load memory-hungry resources
-* deactivate: used to free up resources
-* analyse_entry: called in every user requests. It takes two parameters: ``Entry``, the entry object, and ``params``, the parameters supplied by the user. It should yield one or more ``Entry`` objects.
-
-Plugins are loaded asynchronously, so don't worry if the activate method takes too long. The plugin will be marked as activated once it is finished executing the method.
-
-Entries
-=======
 
 Entries are objects that can be annotated.
+In general, they will be a piece of text.
 By default, entries are `NIF contexts <http://persistence.uni-leipzig.org/nlp2rdf/ontologies/nif-core/nif-core.html>`_ represented in JSON-LD format.
+It is a dictionary/JSON object that looks like this:
+
+  .. code:: python
+
+            {
+               "@id": "<unique identifier or blank node name>",
+               "nif:isString": "input text",
+               "sentiments": [ {
+                     ...
+               }
+               ],
+               ...
+            }
+
 Annotations are added to the object like this:
 
 .. code:: python
@@ -100,96 +48,111 @@ The value may be any valid JSON-LD dictionary.
 For simplicity, senpy includes a series of models by default in the ``senpy.models`` module.
 
 
-Example plugin
+What are annotations?
-==============
+=====================
+They are objects just like entries.
+Senpy ships with several default annotations, including: ``Sentiment``, ``Emotion``, ``EmotionSet``...jk bb
 
-In this section, we will implement a basic sentiment analysis plugin.
-To determine the polarity of each entry, the plugin will compare the length of the string to a threshold.
-This threshold will be included in the definition file.
 
-The definition file would look like this:
+What's a plugin made of?
+========================
 
-.. code:: yaml
+When receiving a query, senpy selects what plugin or plugins should process each entry, and in what order.
+It also makes sure the every entry and the parameters provided by the user meet the plugin requirements.
 
-          name: helloworld
+Hence, two parts are necessary: 1) the code that will process the entry, and 2) some attributes and metadata that will tell senpy how to interact with the plugin.
-          module: helloworld
-          version: 0.0
-          threshold: 10
-          description: Hello World
 
-Now, in a file named ``helloworld.py``:
+In practice, this is what a plugin looks like, tests included:
+
+
+.. literalinclude:: ../senpy/plugins/example/rand_plugin.py
+   :emphasize-lines: 5-11
+   :language: python
+
+
+The lines highlighted contain some information about the plugin.
+In particular, the following information is mandatory:
+
+* A unique name for the class. In our example, Rand.
+* The subclass/type of plugin. This is typically either `SentimentPlugin` or `EmotionPlugin`. However, new types of plugin can be created for different annotations. The only requirement is that these new types inherit from `senpy.Analysis`
+* A description of the plugin. This can be done simply by adding a doc to the class.
+* A version, which should get updated.
+* An author name.
+
+
+Plugins Code
+============
+
+The basic methods in a plugin are:
+
+* analyse_entry: called in every user requests. It takes two parameters: ``Entry``, the entry object, and ``params``, the parameters supplied by the user. It should yield one or more ``Entry`` objects.
+* activate: used to load memory-hungry resources. For instance, to train a classifier.
+* deactivate: used to free up resources when the plugin is no longer needed.
+
+Plugins are loaded asynchronously, so don't worry if the activate method takes too long. The plugin will be marked as activated once it is finished executing the method.
+
+
+How does senpy find modules?
+============================
+
+Senpy looks for files of two types:
+
+* Python files of the form `senpy_<NAME>.py` or `<NAME>_plugin.py`. In these files, it will look for: 1) Instances that inherit from `senpy.Plugin`, or subclasses of `senpy.Plugin` that can be initialized without a configuration file. i.e. classes that contain all the required attributes for a plugin.
+* Plugin definition files (see :doc:`advanced-plugins`)
+
+Defining additional parameters
+==============================
+
+Your plugin may ask for additional parameters from the users of the service by using the attribute ``extra_params`` in your plugin definition.
+It takes a dictionary, where the keys are the name of the argument/parameter, and the value has the following fields:
+
+* aliases: the different names which can be used in the request to use the parameter.
+* required: if set to true, users need to provide this parameter unless a default is set.
+* options: the different acceptable values of the parameter (i.e. an enum). If set, the value provided must match one of the options.
+* default: the default value of the parameter, if none is provided in the request.
 
 .. code:: python
 
-          #!/bin/env python
+          "extra_params":{
-          #helloworld.py
+             "language": {
-
+                "aliases": ["language", "lang", "l"],
-          from senpy.plugins import AnalysisPlugin
+                "required": True,
-          from senpy.models import Sentiment
+                "options": ["es", "en"],
+                "default": "es"
+                }
+             }
 
 
-          class HelloWorld(AnalysisPlugin):
-
-              def analyse_entry(entry, params):
-                  '''Basically do nothing with each entry'''
-
-                  sentiment = Sentiment()
-                  if len(entry.text) < self.threshold:
-                      sentiment['marl:hasPolarity'] = 'marl:Positive'
-                  else:
-                      sentiment['marl:hasPolarity'] = 'marl:Negative'
-                  entry.sentiments.append(sentiment)
-                  yield entry
-
-The complete code of the example plugin is available `here <https://lab.cluster.gsi.dit.upm.es/senpy/plugin-prueba>`__.
 
 Loading data and files
 ======================
 
 Most plugins will need access to files (dictionaries, lexicons, etc.).
-It is good practice to specify the paths of these files in the plugin configuration, so the same code can be reused with different resources.
+These files are usually heavy or under a license that does not allow redistribution.
+For this reason, senpy has a `data_folder` that is separated from the source files.
+The location of this folder is controlled programmatically or by setting the `SENPY_DATA` environment variable.
 
+Plugins have a convenience function `self.open` which will automatically prepend the data folder to relative paths:
 
-.. code:: yaml
-
-          name: dictworld
-          module: dictworld
-          dictionary_path: <PATH OF THE FILE>
-         
-The path can be either absolute, or relative.
-
-From absolute paths
-???????????????????
-
-Absolute paths (such as ``/data/dictionary.csv`` are straightfoward:
-
-.. code:: python
-
-   with open(os.path.join(self.dictionary_path) as f:
-      ...
-
-From relative paths
-???????????????????
-Since plugins are loading dynamically, relative paths will refer to the current working directory.
-Instead, what you usually want is to load files *relative to the plugin source folder*, like so:
-
-
-:: 
-
-   .
-   ..
-   plugin.senpy
-   plugin.py
-   dictionary.csv
-
-For this, we need to first get the path of your source folder first, like so:
 
 .. code:: python
 
           import os
-   root = os.path.realpath(__file__)
+
-   with open(os.path.join(root, self.dictionary_path) as f:
+
-       ...
+          class PluginWithResources(AnalysisPlugin):
+              file_in_data = <FILE PATH>
+              file_in_sources = <FILE PATH>
+
+              def activate(self):
+                  with self.open(self.file_in_data) as f:
+                      self._classifier = train_from_file(f)
+                  file_in_source = os.path.join(self.get_folder(), self.file_in_sources)
+                  with self.open(file_in_source) as f:
+                      pass
+
+         
+It is good practice to specify the paths of these files in the plugin configuration, so the same code can be reused with different resources.
 
 
 Docker image
@@ -199,7 +162,16 @@ Add the following dockerfile to your project to generate a docker image with you
 
 .. code:: dockerfile
 
-   FROM gsiupm/senpy:0.8.8
+   FROM gsiupm/senpy
+
+Once you make sure your plugin works with a specific version of senpy, modify that file to make sure your build will work even if senpy gets updated.
+e.g.:
+
+
+.. code:: dockerfile
+
+   FROM gsiupm/senpy:1.0.1
+
      
 This will copy your source folder to the image, and install all dependencies.
 Now, to build an image:
@@ -215,7 +187,7 @@ And you can run it with:
    docker run -p 5000:5000 gsiupm/exampleplugin
 
 
-If the plugin non-source files (:ref:`loading data and files`), the recommended way is to use absolute paths.
+If the plugin uses non-source files (:ref:`loading data and files`), the recommended way is to use `SENPY_DATA` folder.
 Data can then be mounted in the container or added to the image.
 The former is recommended for open source plugins with licensed resources, whereas the latter is the most convenient and can be used for private images.
 
@@ -229,7 +201,7 @@ Adding data to the image:
 
 .. code:: dockerfile
 
-   FROM gsiupm/senpy:0.8.8
+   FROM gsiupm/senpy:1.0.1
    COPY data /
 
 F.A.Q.
@@ -245,7 +217,7 @@ Why does the analyse function yield instead of return?
 ??????????????????????????????????????????????????????
 
 This is so that plugins may add new entries to the response or filter some of them.
-For instance, a `context detection` plugin may add a new entry for each context in the original entry.
+For instance, a chunker may split one entry into several.
 On the other hand, a conversion plugin may leave out those entries that do not contain relevant information.
 
 
@@ -275,11 +247,13 @@ Training a classifier can be time time consuming. To avoid running the training
               def deactivate(self):
                   self.close()
 
-You can specify a 'shelf_file' in your .senpy file. By default the ShelfMixin creates a file based on the plugin name and stores it in that plugin's folder.
+
+By default the ShelfMixin creates a file based on the plugin name and stores it in that plugin's folder.
+However, you can manually specify a 'shelf_file' in your .senpy file.
 
 Shelves may get corrupted if the plugin exists unexpectedly.
 A corrupt shelf prevents the plugin from loading.
-If you do not care about the pickle, you can force your plugin to remove the corrupted file and load anyway, set the  'force_shelf' to True in your .senpy file.
+If you do not care about the data in the shelf, you can force your plugin to remove the corrupted file and load anyway, set the  'force_shelf' to True in your plugin and start it again.
 
 How can I turn an external service into a plugin?
 ?????????????????????????????????????????????????
@@ -313,50 +287,11 @@ This example ilustrate how to implement a plugin that accesses the Sentiment140
                                       prefix=p,
                                       marl__hasPolarity=polarity,
                                       marl__polarityValue=polarity_value)
-                  sentiment.prov__wasGeneratedBy = self.id
+                  sentiment.prov(self)
                   entry.sentiments.append(sentiment)
                   yield entry
 
 
-Can my plugin require additional parameters from the user?
-??????????????????????????????????????????????????????????
-
-You can add extra parameters in the definition file under the attribute ``extra_params``.
-It takes a dictionary, where the keys are the name of the argument/parameter, and the value has the following fields:
-
-* aliases: the different names which can be used in the request to use the parameter.
-* required: if set to true, users need to provide this parameter unless a default is set.
-* options: the different acceptable values of the parameter (i.e. an enum). If set, the value provided must match one of the options.
-* default: the default value of the parameter, if none is provided in the request.
-
-.. code:: python
-
-          extra_params
-             language:
-                aliases:
-                    - language
-                    - lang
-                    - l
-                required: true,
-                options:
-                    - es
-                    - en
-                default: es
-
-This example shows how to introduce a parameter associated with language.
-The extraction of this paremeter is used in the analyse method of the Plugin interface.
-
-.. code:: python
-
-          lang = params.get("language")
-
-Where can I set up variables for using them in my plugin?
-?????????????????????????????????????????????????????????
-
-You can add these variables in the definition file with the structure of attribute-value pairs.
-
-Every field added to the definition file is available to the plugin instance.
-
 Can I activate a DEBUG mode for my plugin?
 ???????????????????????????????????????????
 
@@ -371,7 +306,7 @@ Additionally, with the ``--pdb`` option you will be dropped into a pdb post mort
 
 .. code:: bash
 
-   senpy --pdb
+   python -m pdb yourplugin.py
 
 Where can I find more code examples?
 ????????????????????????????????????

docs/senpy.rst

@@ -1,8 +1,11 @@
 What is Senpy?
 --------------
 
-Web services can get really complex: data validation, user interaction, formatting, logging., etc. 
+Senpy is a framework for text analysis using Linked Data. There are three main applications of Senpy so far: sentiment and emotion analysis, user profiling and entity recoginition. Annotations and Services are compliant with NIF (NLP Interchange Format).
-The figure below summarizes the typical features in an analysis service.
+
+Senpy aims at providing a framework where analysis modules can be integrated easily as plugins, and providing a core functionality for managing tasks such as data validation, user interaction, formatting, logging, translation to linked data, etc. 
+
+The figure below summarizes the typical features in a text analysis service.
 Senpy implements all the common blocks, so developers can focus on what really matters: great analysis algorithms that solve real problems.
 
 .. image:: senpy-framework.png

docs/server.rst

@@ -7,7 +7,8 @@ The senpy server is launched via the `senpy` command:
 
     usage: senpy [-h] [--level logging_level] [--debug] [--default-plugins]
                 [--host HOST] [--port PORT] [--plugins-folder PLUGINS_FOLDER]
-                [--only-install]
+                [--only-install] [--only-list] [--data-folder DATA_FOLDER]
+                [--threaded] [--version]
 
     Run a Senpy server
 
@@ -22,6 +23,13 @@ The senpy server is launched via the `senpy` command:
       --plugins-folder PLUGINS_FOLDER, -f PLUGINS_FOLDER
                             Where to look for plugins.
       --only-install, -i    Do not run a server, only install plugin dependencies
+      --only-list, --list   Do not run a server, only list plugins found
+      --data-folder DATA_FOLDER, --data DATA_FOLDER
+                            Where to look for data. It be set with the SENPY_DATA
+                            environment variable as well.
+      --threaded            Run a threaded server
+      --version, -v         Output the senpy version and exit
+
 
 
 When launched, the server will recursively look for plugins in the specified plugins folder (the current working directory by default).

docs/vocabularies.rst

@@ -1,8 +1,24 @@
 Vocabularies and model
 ======================
 
-The model used in Senpy is based on the following vocabularies:
+The model used in Senpy is based on NIF 2.0 [1], which defines a semantic format and API for improving interoperability among natural language processing services.
 
-* Marl, a vocabulary designed to annotate and describe subjetive opinions expressed on the web or in information systems.
+Senpy has been applied to sentiment and emotion analysis services using the following vocabularies:
-* Onyx, which is built one the same principles as Marl to annotate and describe emotions, and provides interoperability with Emotion Markup Language.
+
-* NIF 2.0, which defines a semantic format and APO for improving interoperability among natural language processing services
+* Marl [2,6], a vocabulary designed to annotate and describe subjetive opinions expressed on the web or in information systems.
+* Onyx [3,5], which is built one the same principles as Marl to annotate and describe emotions, and provides interoperability with Emotion Markup Language.
+
+An overview of the vocabularies and their use can be found in [4].
+
+
+[1] Guidelines for developing NIF-based NLP services, Final Community Group Report 22 December 2015 Available at: https://www.w3.org/2015/09/bpmlod-reports/nif-based-nlp-webservices/
+
+[2] Marl Ontology Specification, available at http://www.gsi.dit.upm.es/ontologies/marl/
+
+[3] Onyx Ontology Specification, available at http://www.gsi.dit.upm.es/ontologies/onyx/
+
+[4] Iglesias, C. A., Sánchez-Rada, J. F., Vulcu, G., & Buitelaar, P. (2017). Linked Data Models for Sentiment and Emotion Analysis in Social Networks. In Sentiment Analysis in Social Networks (pp. 49-69).
+
+[5] Sánchez-Rada, J. F., & Iglesias, C. A. (2016). Onyx: A linked data approach to emotion representation. Information Processing & Management, 52(1), 99-114.
+
+[6] Westerski, A., Iglesias Fernandez, C. A., & Tapia Rico, F. (2011). Linked opinions: Describing sentiments on the structured web of data.

example-plugins/README.md

@@ -0,0 +1,23 @@
+This is a collection of plugins that exemplify certain aspects of plugin development with senpy.
+
+The first series of plugins the `basic` ones.
+Their starting point is a classification function defined in `basic.py`.
+They all include testing and running them as a script will run all tests.
+In ascending order of customization, the plugins are:
+
+* Basic is the simplest plugin of all. It leverages the `SentimentBox` Plugin class to create a plugin out of a classification method, and `MappingMixin` to convert the labels from (`pos`, `neg`) to (`marl:Positive`, `marl:Negative`
+* Basic_box is just like the previous one, but replaces the mixin with a custom function.
+* Basic_configurable is a version of `basic` with a configurable map of emojis for each sentiment.
+* Basic_parameterized like `basic_info`, but users set the map in each query (via `extra_parameters`).
+* Basic_analyse\_entry uses the more general `analyse_entry` method and adds the annotations individually.
+
+
+In rest of the plugins show advanced topics:
+
+* mynoop: shows how to add a definition file with external requirements for a plugin. Doing this with a python-only module would require moving all imports of the requirements to their functions, which is considered bad practice.
+* Async: a barebones example of training a plugin and analyzing data in parallel.
+
+All of the plugins in this folder include a set of test cases and they are periodically tested with the latest version of senpy.
+
+Additioanlly, for an example of stand-alone plugin that can be tested and deployed with docker, take a look at: lab.cluster.gsi.dit.upm.es/senpy/plugin-example
+ bbm

example-plugins/async_plugin.py

@@ -1,4 +1,4 @@
-from senpy.plugins import AnalysisPlugin
+from senpy import AnalysisPlugin
 
 import multiprocessing
 
@@ -7,10 +7,15 @@ def _train(process_number):
     return process_number
 
 
-class AsyncPlugin(AnalysisPlugin):
+class Async(AnalysisPlugin):
+    '''An example of an asynchronous module'''
+    author = '@balkian'
+    version = '0.2'
+    async = True
+
     def _do_async(self, num_processes):
         pool = multiprocessing.Pool(processes=num_processes)
-        values = pool.map(_train, range(num_processes))
+        values = sorted(pool.map(_train, range(num_processes)))
 
         return values
 
@@ -22,5 +27,11 @@ class AsyncPlugin(AnalysisPlugin):
         entry.async_values = values
         yield entry
 
-    def test(self):
+    test_cases = [
-        pass
+        {
+            'input': 'any',
+            'expected': {
+                'async_values': [0, 1]
+            }
+        }
+    ]

example-plugins/basic.py

@@ -0,0 +1,23 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+emoticons = {
+    'pos': [':)', ':]', '=)', ':D'],
+    'neg': [':(', ':[', '=(']
+}
+
+emojis = {
+    'pos': ['😁', '😂', '😃', '😄', '😆', '😅', '😄' '😍'],
+    'neg': ['😢', '😡', '😠', '😞', '😖', '😔', '😓', '😒']
+}
+
+
+def get_polarity(text, dictionaries=[emoticons, emojis]):
+    polarity = 'marl:Neutral'
+    for dictionary in dictionaries:
+        for label, values in dictionary.items():
+            for emoticon in values:
+                if emoticon and emoticon in text:
+                    polarity = label
+                    break
+    return polarity

example-plugins/basic_analyse_entry_plugin.py

@@ -0,0 +1,47 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+from senpy import easy_test, models, plugins
+
+import basic
+
+
+class BasicAnalyseEntry(plugins.SentimentPlugin):
+    '''Equivalent to Basic, implementing the analyse_entry method'''
+
+    author = '@balkian'
+    version = '0.1'
+
+    mappings = {
+        'pos': 'marl:Positive',
+        'neg': 'marl:Negative',
+        'default': 'marl:Neutral'
+    }
+
+    def analyse_entry(self, entry, params):
+        polarity = basic.get_polarity(entry.text)
+
+        polarity = self.mappings.get(polarity, self.mappings['default'])
+
+        s = models.Sentiment(marl__hasPolarity=polarity)
+        s.prov(self)
+        entry.sentiments.append(s)
+        yield entry
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Negative'
+    }]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/basic_box_plugin.py

@@ -0,0 +1,41 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+from senpy import easy_test, SentimentBox
+
+import basic
+
+
+class BasicBox(SentimentBox):
+    ''' A modified version of Basic that also does converts annotations manually'''
+
+    author = '@balkian'
+    version = '0.1'
+
+    mappings = {
+        'pos': 'marl:Positive',
+        'neg': 'marl:Negative',
+        'default': 'marl:Neutral'
+    }
+
+    def predict_one(self, input):
+        output = basic.get_polarity(input)
+        return self.mappings.get(output, self.mappings['default'])
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Negative'
+    }]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/basic_plugin.py

@@ -0,0 +1,40 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+from senpy import easy_test, SentimentBox, MappingMixin
+
+import basic
+
+
+class Basic(MappingMixin, SentimentBox):
+    '''Provides sentiment annotation using a lexicon'''
+
+    author = '@balkian'
+    version = '0.1'
+
+    mappings = {
+        'pos': 'marl:Positive',
+        'neg': 'marl:Negative',
+        'default': 'marl:Neutral'
+    }
+
+    def predict_one(self, input):
+        return basic.get_polarity(input)
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Negative'
+    }]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/configurable_plugin.py

@@ -0,0 +1,106 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+from senpy import easy_test, models, plugins
+
+import basic
+
+
+class Dictionary(plugins.SentimentPlugin):
+    '''Sentiment annotation using a configurable lexicon'''
+
+    author = '@balkian'
+    version = '0.2'
+
+    dictionaries = [basic.emojis, basic.emoticons]
+
+    mappings = {'pos': 'marl:Positive', 'neg': 'marl:Negative'}
+
+    def analyse_entry(self, entry, params):
+        polarity = basic.get_polarity(entry.text, self.dictionaries)
+        if polarity in self.mappings:
+            polarity = self.mappings[polarity]
+
+        s = models.Sentiment(marl__hasPolarity=polarity)
+        s.prov(self)
+        entry.sentiments.append(s)
+        yield entry
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Negative'
+    }]
+
+
+class EmojiOnly(Dictionary):
+    '''Sentiment annotation with a basic lexicon of emojis'''
+    description = 'A plugin'
+    dictionaries = [basic.emojis]
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Neutral'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Neutral'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Negative'
+    }]
+
+
+class EmoticonsOnly(Dictionary):
+    '''Sentiment annotation with a basic lexicon of emoticons'''
+    dictionaries = [basic.emoticons]
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'So sad :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Neutral'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Neutral'
+    }]
+
+
+class Salutes(Dictionary):
+    '''Sentiment annotation with a custom lexicon, for illustration purposes'''
+    dictionaries = [{
+        'marl:Positive': ['Hello', '!'],
+        'marl:Negative': ['Good bye', ]
+    }]
+
+    test_cases = [{
+        'input': 'Hello :)',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'Good bye :(',
+        'polarity': 'marl:Negative'
+    }, {
+        'input': 'Yay! Emojis  😁',
+        'polarity': 'marl:Positive'
+    }, {
+        'input': 'But no emoticons 😢',
+        'polarity': 'marl:Neutral'
+    }]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/dummy_plugin.py

@@ -0,0 +1,25 @@
+from senpy import AnalysisPlugin, easy
+
+
+class Dummy(AnalysisPlugin):
+    '''This is a dummy self-contained plugin'''
+    author = '@balkian'
+    version = '0.1'
+
+    def analyse_entry(self, entry, params):
+        entry['nif:isString'] = entry['nif:isString'][::-1]
+        entry.reversed = entry.get('reversed', 0) + 1
+        yield entry
+
+    test_cases = [{
+        'entry': {
+            'nif:isString': 'Hello',
+        },
+        'expected': {
+            'nif:isString': 'olleH'
+        }
+    }]
+
+
+if __name__ == '__main__':
+    easy()

example-plugins/dummy_required_plugin.py

@@ -0,0 +1,40 @@
+from senpy import AnalysisPlugin, easy
+
+
+class DummyRequired(AnalysisPlugin):
+    '''This is a dummy self-contained plugin'''
+    author = '@balkian'
+    version = '0.1'
+    extra_params = {
+        'example': {
+            'description': 'An example parameter',
+            'required': True,
+            'options': ['a', 'b']
+        }
+    }
+
+    def analyse_entry(self, entry, params):
+        entry['nif:isString'] = entry['nif:isString'][::-1]
+        entry.reversed = entry.get('reversed', 0) + 1
+        yield entry
+
+    test_cases = [{
+        'entry': {
+            'nif:isString': 'Hello',
+        },
+        'should_fail': True
+    }, {
+        'entry': {
+            'nif:isString': 'Hello',
+        },
+        'params': {
+            'example': 'a'
+        },
+        'expected': {
+            'nif:isString': 'olleH'
+        }
+    }]
+
+
+if __name__ == '__main__':
+    easy()

example-plugins/mynoop.py

@@ -0,0 +1,24 @@
+import noop
+from senpy.plugins import SentimentPlugin
+
+
+class NoOp(SentimentPlugin):
+    '''This plugin does nothing. Literally nothing.'''
+
+    version = 0
+
+    def analyse_entry(self, entry, *args, **kwargs):
+        yield entry
+
+    def test(self):
+        print(dir(noop))
+        super(NoOp, self).test()
+
+    test_cases = [{
+        'entry': {
+            'nif:isString': 'hello'
+        },
+        'expected': {
+            'nif:isString': 'hello'
+        }
+    }]

example-plugins/mynoop.senpy

@@ -0,0 +1,3 @@
+module: mynoop
+requirements:
+ - noop

example-plugins/parameterized_plugin.py

@@ -0,0 +1,63 @@
+#!/usr/local/bin/python
+# coding: utf-8
+
+from senpy import easy_test, models, plugins
+
+import basic
+
+
+class ParameterizedDictionary(plugins.SentimentPlugin):
+    '''This is a basic self-contained plugin'''
+
+    author = '@balkian'
+    version = '0.2'
+
+    extra_params = {
+        'positive-words': {
+            'description': 'Comma-separated list of words that are considered positive',
+            'aliases': ['positive'],
+            'required': True
+        },
+        'negative-words': {
+            'description': 'Comma-separated list of words that are considered negative',
+            'aliases': ['negative'],
+            'required': False
+        }
+    }
+
+    def analyse_entry(self, entry, params):
+        positive_words = params['positive-words'].split(',')
+        negative_words = params['negative-words'].split(',')
+        dictionary = {
+            'marl:Positive': positive_words,
+            'marl:Negative': negative_words,
+        }
+        polarity = basic.get_polarity(entry.text, [dictionary])
+
+        s = models.Sentiment(marl__hasPolarity=polarity)
+        s.prov(self)
+        entry.sentiments.append(s)
+        yield entry
+
+    test_cases = [
+        {
+            'input': 'Hello :)',
+            'polarity': 'marl:Positive',
+            'parameters': {
+                'positive': "Hello,:)",
+                'negative': "sad,:()"
+            }
+        },
+        {
+            'input': 'Hello :)',
+            'polarity': 'marl:Negative',
+            'parameters': {
+                'positive': "",
+                'negative': "Hello"
+            }
+        }
+    ]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/sklearn/mydata.py

@@ -0,0 +1,33 @@
+'''
+Create a dummy dataset.
+Messages with a happy emoticon are labelled positive
+Messages with a sad emoticon are labelled negative
+'''
+import random
+
+dataset = []
+
+vocabulary = ['hello', 'world', 'senpy', 'cool', 'goodbye', 'random', 'text']
+
+emojimap = {
+    1: [':)', ],
+    -1: [':(', ]
+}
+
+
+for tag, values in emojimap.items():
+    for i in range(1000):
+        msg = ''
+        for j in range(3):
+            msg += random.choice(vocabulary)
+            msg += " "
+        msg += random.choice(values)
+        dataset.append([msg, tag])
+
+
+text = []
+labels = []
+
+for i in dataset:
+    text.append(i[0])
+    labels.append(i[1])

example-plugins/sklearn/mypipeline.py

@@ -0,0 +1,30 @@
+from sklearn.pipeline import Pipeline
+from sklearn.feature_extraction.text import CountVectorizer
+from sklearn.model_selection import train_test_split
+
+from mydata import text, labels
+
+X_train, X_test, y_train, y_test = train_test_split(text, labels, test_size=0.12, random_state=42)
+
+from sklearn.naive_bayes import MultinomialNB
+
+
+count_vec = CountVectorizer(tokenizer=lambda x: x.split())
+clf3 = MultinomialNB()
+pipeline = Pipeline([('cv', count_vec),
+                    ('clf', clf3)])
+
+pipeline.fit(X_train, y_train)
+print('Feature names: {}'.format(count_vec.get_feature_names()))
+print('Class count: {}'.format(clf3.class_count_))
+
+
+if __name__ == '__main__':
+    print('--Results--')
+    tests = [
+        (['The sentiment for senpy should be positive :)', ], 1),
+        (['The sentiment for anything else should be negative :()', ], -1)
+    ]
+    for features, expected in tests:
+        result = pipeline.predict(features)
+        print('Input: {}\nExpected: {}\nGot: {}'.format(features[0], expected, result))

example-plugins/sklearn/pipeline_plugin.py

@@ -0,0 +1,37 @@
+from senpy import SentimentBox, MappingMixin, easy_test
+
+from mypipeline import pipeline
+
+
+class PipelineSentiment(MappingMixin, SentimentBox):
+    '''
+    This is a pipeline plugin that wraps a classifier defined in another module
+    (mypipeline).
+    '''
+    author = '@balkian'
+    version = 0.1
+    maxPolarityValue = 1
+    minPolarityValue = -1
+
+    mappings = {
+        1: 'marl:Positive',
+        -1: 'marl:Negative'
+    }
+
+    def predict_one(self, input):
+        return pipeline.predict([input, ])[0]
+
+    test_cases = [
+        {
+            'input': 'The sentiment for senpy should be positive :)',
+            'polarity': 'marl:Positive'
+        },
+        {
+            'input': 'The sentiment for senpy should be negative :(',
+            'polarity': 'marl:Negative'
+        }
+    ]
+
+
+if __name__ == '__main__':
+    easy_test()

example-plugins/sleep_plugin.py

@@ -0,0 +1,27 @@
+from senpy.plugins import AnalysisPlugin
+from time import sleep
+
+
+class Sleep(AnalysisPlugin):
+    '''Dummy plugin to test async'''
+    author = "@balkian"
+    version = "0.2"
+    timeout = 0.05
+    extra_params = {
+        "timeout": {
+            "@id": "timeout_sleep",
+            "aliases": ["timeout", "to"],
+            "required": False,
+            "default": 0
+        }
+    }
+
+    def activate(self, *args, **kwargs):
+        sleep(self.timeout)
+
+    def analyse_entry(self, entry, params):
+        sleep(float(params.get("timeout", self.timeout)))
+        yield entry
+
+    def test(self):
+        pass

extra-requirements.txt

@@ -0,0 +1 @@
+gsitk

k8s/senpy-deployment.yaml

@@ -13,7 +13,7 @@ spec:
     spec:
       containers:
       - name: senpy-latest
-        image: gsiupm/senpy:latest
+        image: $IMAGEWTAG
         imagePullPolicy: Always
         args:
           - "--default-plugins"

requirements.txt

@@ -3,7 +3,6 @@ requests>=2.4.1
 tornado>=4.4.3
 PyLD>=0.6.5
 nltk
-six
 future
 jsonschema
 jsonref

senpy/__init__.py

@@ -25,4 +25,10 @@ logger = logging.getLogger(__name__)
 
 logger.info('Using senpy version: {}'.format(__version__))
 
+from .utils import easy, easy_load, easy_test  # noqa: F401
+
+from .models import *  # noqa: F401,F403
+from .plugins import *  # noqa: F401,F403
+from .extensions import *  # noqa: F401,F403
+
 __all__ = ['api', 'blueprints', 'cli', 'extensions', 'models', 'plugins']

senpy/__main__.py

@@ -22,9 +22,11 @@ the server.
 
 from flask import Flask
 from senpy.extensions import Senpy
+from senpy.utils import easy_test
 
 import logging
 import os
+import sys
 import argparse
 import senpy
 
@@ -38,7 +40,7 @@ def main():
         '-l',
         metavar='logging_level',
         type=str,
-        default="INFO",
+        default="WARN",
         help='Logging level')
     parser.add_argument(
         '--debug',
@@ -66,7 +68,7 @@ def main():
         '--plugins-folder',
         '-f',
         type=str,
-        default='plugins',
+        default='.',
         help='Where to look for plugins.')
     parser.add_argument(
         '--only-install',
@@ -74,6 +76,24 @@ def main():
         action='store_true',
         default=False,
         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(
+        '--only-list',
+        '--list',
+        action='store_true',
+        default=False,
+        help='Do not run a server, only list plugins found')
+    parser.add_argument(
+        '--data-folder',
+        '--data',
+        type=str,
+        default=None,
+        help='Where to look for data. It be set with the SENPY_DATA environment variable as well.')
     parser.add_argument(
         '--threaded',
         action='store_false',
@@ -88,17 +108,30 @@ def main():
     args = parser.parse_args()
     if args.version:
         print('Senpy version {}'.format(senpy.__version__))
+        print(sys.version)
         exit(1)
-    logging.basicConfig()
     rl = logging.getLogger()
     rl.setLevel(getattr(logging, args.level))
     app = Flask(__name__)
     app.debug = args.debug
-    sp = Senpy(app, args.plugins_folder, default_plugins=args.default_plugins)
+    sp = Senpy(app, args.plugins_folder,
+               default_plugins=args.default_plugins,
+               data_folder=args.data_folder)
+    if args.only_list:
+        plugins = sp.plugins()
+        maxwidth = max(len(x.id) for x in plugins)
+        for plugin in plugins:
+            import inspect
+            fpath = inspect.getfile(plugin.__class__)
+            print('{: <{width}} @ {}'.format(plugin.id, fpath, width=maxwidth))
+        return
     sp.install_deps()
     if args.only_install:
         return
     sp.activate_all()
+    if args.only_test:
+        easy_test(sp.plugins())
+        return
     print('Senpy version {}'.format(senpy.__version__))
     print('Server running on port %s:%d. Ctrl+C to quit' % (args.host,
                                                             args.port))

senpy/api.py

@@ -13,8 +13,9 @@ API_PARAMS = {
     "expanded-jsonld": {
         "@id": "expanded-jsonld",
         "aliases": ["expanded"],
+        "options": "boolean",
         "required": True,
-        "default": 0
+        "default": False
     },
     "with_parameters": {
         "aliases": ['withparameters',
@@ -23,13 +24,6 @@ API_PARAMS = {
         "default": False,
         "required": True
     },
-    "plugin_type": {
-        "@id": "pluginType",
-        "description": 'What kind of plugins to list',
-        "aliases": ["pluginType"],
-        "required": True,
-        "default": "analysisPlugin"
-    },
     "outformat": {
         "@id": "outformat",
         "aliases": ["o"],
@@ -59,6 +53,31 @@ API_PARAMS = {
     }
 }
 
+EVAL_PARAMS = {
+    "algorithm": {
+        "aliases": ["plug", "p", "plugins", "algorithms", 'algo', 'a', 'plugin'],
+        "description": "Plugins to be evaluated",
+        "required": True,
+        "help": "See activated plugins in /plugins"
+    },
+    "dataset": {
+        "aliases": ["datasets", "data", "d"],
+        "description": "Datasets to be evaluated",
+        "required": True,
+        "help": "See avalaible datasets in /datasets"
+    }
+}
+
+PLUGINS_PARAMS = {
+    "plugin_type": {
+        "@id": "pluginType",
+        "description": 'What kind of plugins to list',
+        "aliases": ["pluginType"],
+        "required": True,
+        "default": 'analysisPlugin'
+    }
+}
+
 WEB_PARAMS = {
     "inHeaders": {
         "aliases": ["headers"],
@@ -95,7 +114,7 @@ NIF_PARAMS = {
         "aliases": ["f"],
         "required": False,
         "default": "text",
-        "options": ["turtle", "text", "json-ld"],
+        "options": ["text", "json-ld"],
     },
     "language": {
         "@id": "language",
@@ -126,39 +145,38 @@ def parse_params(indict, *specs):
     wrong_params = {}
     for spec in specs:
         for param, options in iteritems(spec):
-            if param[0] != "@":  # Exclude json-ld properties
             for alias in options.get("aliases", []):
                 # Replace each alias with the correct name of the parameter
                 if alias in indict and alias is not param:
                     outdict[param] = indict[alias]
-                        del indict[alias]
+                    del outdict[alias]
                     continue
             if param not in outdict:
-                    if options.get("required", False) and "default" not in options:
-                        wrong_params[param] = spec[param]
-                    else:
                 if "default" in options:
+                    # We assume the default is correct
                     outdict[param] = options["default"]
-                elif "options" in spec[param]:
+                elif options.get("required", False):
-                    if spec[param]["options"] == "boolean":
+                    wrong_params[param] = spec[param]
+                continue
+            if "options" in options:
+                if options["options"] == "boolean":
                     outdict[param] = outdict[param] in [None, True, 'true', '1']
-                    elif outdict[param] not in spec[param]["options"]:
+                elif outdict[param] not in options["options"]:
                     wrong_params[param] = spec[param]
     if wrong_params:
         logger.debug("Error parsing: %s", wrong_params)
         message = Error(
             status=400,
-            message="Missing or invalid parameters",
+            message='Missing or invalid parameters',
             parameters=outdict,
-            errors={param: error
+            errors=wrong_params)
-                    for param, error in iteritems(wrong_params)})
         raise message
     if 'algorithm' in outdict and not isinstance(outdict['algorithm'], list):
         outdict['algorithm'] = outdict['algorithm'].split(',')
     return outdict
 
 
-def get_extra_params(request, plugin=None):
+def parse_extra_params(request, plugin=None):
     params = request.parameters.copy()
     if plugin:
         extra_params = parse_params(params, plugin.get('extra_params', {}))
@@ -176,7 +194,7 @@ def parse_call(params):
         results.entries.append(entry)
     elif params['informat'] == 'json-ld':
         results = from_string(params['input'], cls=Results)
-    else:
+    else:  # pragma: no cover
-        raise NotImplemented('Informat {} is not implemented'.format(params['informat']))
+        raise NotImplementedError('Informat {} is not implemented'.format(params['informat']))
     results.parameters = params
     return results

senpy/blueprints.py

@@ -19,18 +19,19 @@ Blueprints for Senpy
 """
 from flask import (Blueprint, request, current_app, render_template, url_for,
                    jsonify)
-from .models import Error, Response, Help, Plugins, read_schema
+from .models import Error, Response, Help, Plugins, read_schema, Datasets
 from . import api
 from .version import __version__
 from functools import wraps
 
 import logging
+import traceback
 import json
 
 logger = logging.getLogger(__name__)
 
 api_blueprint = Blueprint("api", __name__)
-demo_blueprint = Blueprint("demo", __name__)
+demo_blueprint = Blueprint("demo", __name__, template_folder='templates')
 ns_blueprint = Blueprint("ns", __name__)
 
 
@@ -72,41 +73,52 @@ def schema(schema="definitions"):
 
 
 def basic_api(f):
+    default_params = {
+        'inHeaders': False,
+        'expanded-jsonld': False,
+        'outformat': 'json-ld',
+        'with_parameters': True,
+    }
+
     @wraps(f)
     def decorated_function(*args, **kwargs):
         raw_params = get_params(request)
+        logger.info('Getting request: {}'.format(raw_params))
         headers = {'X-ORIGINAL-PARAMS': json.dumps(raw_params)}
+        params = default_params
 
-        outformat = 'json-ld'
         try:
-            print('Getting request:')
-            print(request)
             params = api.parse_params(raw_params, api.WEB_PARAMS, api.API_PARAMS)
             if hasattr(request, 'parameters'):
                 request.parameters.update(params)
             else:
                 request.parameters = params
             response = f(*args, **kwargs)
-        except Error as ex:
+        except (Exception) as ex:
-            response = ex
-            response.parameters = params
-            logger.error(ex)
             if current_app.debug:
                 raise
+            if not isinstance(ex, Error):
+                msg = "{}:\n\t{}".format(ex,
+                                         traceback.format_exc())
+                ex = Error(message=msg, status=500)
+            logger.exception('Error returning analysis result')
+            response = ex
+            response.parameters = raw_params
+            logger.error(ex)
 
-        in_headers = params['inHeaders']
+        if 'parameters' in response and not params['with_parameters']:
-        expanded = params['expanded-jsonld']
+            del response.parameters
-        outformat = params['outformat']
 
+        logger.info('Response: {}'.format(response))
         return response.flask(
-            in_headers=in_headers,
+            in_headers=params['inHeaders'],
             headers=headers,
             prefix=url_for('.api_root', _external=True),
             context_uri=url_for('api.context',
                                 entity=type(response).__name__,
                                 _external=True),
-            outformat=outformat,
+            outformat=params['outformat'],
-            expanded=expanded)
+            expanded=params['expanded-jsonld'])
 
     return decorated_function
 
@@ -116,21 +128,31 @@ def basic_api(f):
 def api_root():
     if request.parameters['help']:
         dic = dict(api.API_PARAMS, **api.NIF_PARAMS)
+        response = Help(valid_parameters=dic)
+        return response
+    req = api.parse_call(request.parameters)
+    return current_app.senpy.analyse(req)
+
+@api_blueprint.route('/evaluate/', methods=['POST', 'GET'])
+@basic_api
+def evaluate():
+    if request.parameters['help']:
+        dic = dict(api.EVAL_PARAMS)
         response = Help(parameters=dic)
         return response
     else:
-        req = api.parse_call(request.parameters)
+        params = api.parse_params(request.parameters, api.EVAL_PARAMS)
-        response = current_app.senpy.analyse(req)
+        response = current_app.senpy.evaluate(params)
         return response
 
-
 @api_blueprint.route('/plugins/', methods=['POST', 'GET'])
 @basic_api
 def plugins():
     sp = current_app.senpy
-    ptype = request.parameters.get('plugin_type')
+    params = api.parse_params(request.parameters, api.PLUGINS_PARAMS)
-    plugins = sp.filter_plugins(plugin_type=ptype)
+    ptype = params.get('plugin_type')
-    dic = Plugins(plugins=list(plugins.values()))
+    plugins = list(sp.plugins(plugin_type=ptype))
+    dic = Plugins(plugins=plugins)
     return dic
 
 
@@ -138,12 +160,13 @@ def plugins():
 @basic_api
 def plugin(plugin=None):
     sp = current_app.senpy
-    if plugin == 'default' and sp.default_plugin:
+    return sp.get_plugin(plugin)
-        return sp.default_plugin
+
-    plugins = sp.filter_plugins(
+
-        id='plugins/{}'.format(plugin)) or sp.filter_plugins(name=plugin)
+@api_blueprint.route('/datasets/', methods=['POST','GET'])
-    if plugins:
+@basic_api
-        response = list(plugins.values())[0]
+def datasets():
-    else:
+    sp = current_app.senpy
-        return Error(message="Plugin not found", status=404)
+    datasets = sp.datasets
-    return response
+    dic = Datasets(datasets = list(datasets.values()))
+    return dic

senpy/cli.py

@@ -28,11 +28,15 @@ def main_function(argv):
                               api.API_PARAMS,
                               api.NIF_PARAMS)
     plugin_folder = params['plugin_folder']
-    sp = Senpy(default_plugins=False, plugin_folder=plugin_folder)
+    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', sp.plugins.keys())
+    algos = request.parameters.get('algorithm', None)
+    if algos:
         for algo in algos:
             sp.activate_plugin(algo)
+    else:
+        sp.activate_all()
     res = sp.analyse(request)
     return res
 
@@ -42,9 +46,9 @@ def main():
     '''
     try:
         res = main_function(sys.argv[1:])
-        print(res.to_JSON())
+        print(res.serialize())
     except Error as err:
-        print(err.to_JSON())
+        print(err.serialize())
         sys.exit(2)
 
 

senpy/client.py

@@ -1,7 +1,6 @@
 import requests
 import logging
 from . import models
-from .plugins import default_plugin_type
 
 logger = logging.getLogger(__name__)
 
@@ -13,10 +12,17 @@ class Client(object):
     def analyse(self, input, method='GET', **kwargs):
         return self.request('/', method=method, input=input, **kwargs)
 
-    def plugins(self, ptype=default_plugin_type):
+    def evaluate(self, input, method='GET', **kwargs):
-        resp = self.request(path='/plugins', plugin_type=ptype).plugins
+        return self.request('/evaluate', method = method, input=input, **kwargs)
+
+    def plugins(self, *args, **kwargs):
+        resp = self.request(path='/plugins').plugins
         return {p.name: p for p in resp}
 
+    def datasets(self):
+        resp = self.request(path='/datasets').datasets
+        return {d.name: d for d in resp}
+
     def request(self, path=None, method='GET', **params):
         url = '{}{}'.format(self.endpoint, path)
         response = requests.request(method=method, url=url, params=params)

senpy/extensions.py

@@ -6,42 +6,58 @@ from future import standard_library
 standard_library.install_aliases()
 
 from . import plugins, api
-from .plugins import SenpyPlugin
+from .plugins import Plugin, evaluate
-from .models import Error
+from .models import Error, AggregatedEvaluation
 from .blueprints import api_blueprint, demo_blueprint, ns_blueprint
 
 from threading import Thread
 from functools import partial
-
 import os
 import copy
+import errno
 import logging
-import traceback
+
 
 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=".",
+                 data_folder=None,
                  default_plugins=False):
-        self.app = app
-        self._search_folders = set()
-        self._plugin_list = []
-        self._outdated = True
-        self._default = None
 
+        default_data = os.path.join(os.getcwd(), 'senpy_data')
+        self.data_folder = data_folder or os.environ.get('SENPY_DATA', default_data)
+        try:
+            os.makedirs(self.data_folder)
+        except OSError as e:
+            if e.errno == errno.EEXIST:
+                logger.debug('Data folder exists: {}'.format(self.data_folder))
+            else:  # pragma: no cover
+                raise
+
+        self._default = None
+        self._plugins = {}
+        if plugin_folder:
             self.add_folder(plugin_folder)
+
         if default_plugins:
             self.add_folder('plugins', from_root=True)
         else:
             # Add only conversion plugins
             self.add_folder(os.path.join('plugins', 'conversion'),
                             from_root=True)
-
+        self.app = app
         if app is not None:
             self.init_app(app)
 
@@ -56,21 +72,52 @@ class Senpy(object):
         # otherwise fall back to the request context
         if hasattr(app, 'teardown_appcontext'):
             app.teardown_appcontext(self.teardown)
-        else:
+        else:  # pragma: no cover
             app.teardown_request(self.teardown)
         app.register_blueprint(api_blueprint, url_prefix="/api")
         app.register_blueprint(ns_blueprint, url_prefix="/ns")
         app.register_blueprint(demo_blueprint, url_prefix="/")
 
+    def add_plugin(self, plugin):
+        self._plugins[plugin.name.lower()] = plugin
+
+    def delete_plugin(self, plugin):
+        del self._plugins[plugin.name.lower()]
+
+    def plugins(self, **kwargs):
+        """ Return the plugins registered for a given application. Filtered by criteria  """
+        return list(plugins.pfilter(self._plugins, **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]
+
+        results = self.plugins(id='plugins/{}'.format(name))
+
+        if not results:
+            return Error(message="Plugin not found", status=404)
+        return results[0]
+
+    @property
+    def analysis_plugins(self):
+        """ Return only the analysis plugins """
+        return self.plugins(plugin_type='analysisPlugin')
+
     def add_folder(self, folder, from_root=False):
+        """ Find plugins in this folder and add them to this instance """
         if from_root:
             folder = os.path.join(os.path.dirname(__file__), folder)
         logger.debug("Adding folder: %s", folder)
         if os.path.isdir(folder):
-            self._search_folders.add(folder)
+            new_plugins = plugins.from_folder([folder],
-            self._outdated = True
+                                              data_folder=self.data_folder)
+            for plugin in new_plugins:
+                self.add_plugin(plugin)
         else:
-            logger.debug("Not a folder: %s", folder)
+            raise AttributeError("Not a folder or does not exist: %s", folder)
 
     def _get_plugins(self, request):
         if not self.analysis_plugins:
@@ -89,14 +136,16 @@ class Senpy(object):
 
         plugins = list()
         for algo in algos:
-            if algo not in self.plugins:
+            algo = algo.lower()
-                logger.debug(("The algorithm '{}' is not valid\n"
+            if algo not in self._plugins:
+                msg = ("The algorithm '{}' is not valid\n"
                        "Valid algorithms: {}").format(algo,
-                                                             self.plugins.keys()))
+                                                      self._plugins.keys())
+                logger.debug(msg)
                 raise Error(
                     status=404,
-                    message="The algorithm '{}' is not valid".format(algo))
+                    message=msg)
-            plugins.append(self.plugins[algo])
+            plugins.append(self._plugins[algo])
         return plugins
 
     def _process_entries(self, entries, req, plugins):
@@ -110,7 +159,7 @@ class Senpy(object):
             return
         plugin = plugins[0]
         self._activate(plugin)  # Make sure the plugin is activated
-        specific_params = api.get_extra_params(req, plugin)
+        specific_params = api.parse_extra_params(req, plugin)
         req.analysis.append({'plugin': plugin,
                              'parameters': specific_params})
         results = plugin.analyse_entries(entries, specific_params)
@@ -118,7 +167,7 @@ class Senpy(object):
             yield i
 
     def install_deps(self):
-        for plugin in self.filter_plugins(is_activated=True):
+        for plugin in self.plugins(is_activated=True):
             plugins.install_deps(plugin)
 
     def analyse(self, request):
@@ -128,7 +177,6 @@ class Senpy(object):
         by api.parse_call().
         """
         logger.debug("analysing request: {}".format(request))
-        try:
         entries = request.entries
         request.entries = []
         plugins = self._get_plugins(request)
@@ -136,28 +184,67 @@ class Senpy(object):
         for i in self._process_entries(entries, results, plugins):
             results.entries.append(i)
         self.convert_emotions(results)
-            if 'with_parameters' not in results.parameters:
-                del results.parameters
         logger.debug("Returning analysis result: {}".format(results))
-        except (Error, Exception) as ex:
-            if not isinstance(ex, Error):
-                msg = "Error during analysis: {} \n\t{}".format(ex,
-                                                                traceback.format_exc())
-                ex = Error(message=msg, status=500)
-            logger.exception('Error returning analysis result')
-            raise ex
         results.analysis = [i['plugin'].id for i in results.analysis]
         return results
 
+    def _get_datasets(self, request):
+        if not self.datasets:
+            raise Error(
+                status=404,
+                message=("No datasets found."
+                         " Please verify DatasetManager"))
+        datasets_name = request.parameters.get('dataset', None).split(',')
+        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()))
+                raise Error(
+                    status=404,
+                    message="The dataset '{}' is not valid".format(dataset))
+        dm = 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()
+        for item in dm.get_datasets():
+            for key in item:
+                if key in self._dataset_list:
+                    continue
+                properties = item[key]
+                properties['@id'] = key
+                self._dataset_list[key] = properties
+        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):
+            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.filter_plugins(plugin_type='emotionConversionPlugin')
+        candidates = self.plugins(plugin_type='emotionConversionPlugin')
-        for name, candidate in candidates.items():
+        for candidate in candidates:
             for pair in candidate.onyx__doesConversion:
                 logging.debug(pair)
 
                 if pair['onyx:conversionFrom'] == fromModel \
                    and pair['onyx:conversionTo'] == toModel:
-                    # logging.debug('Found candidate: {}'.format(candidate))
                     yield candidate
 
     def convert_emotions(self, resp):
@@ -184,7 +271,8 @@ class Senpy(object):
                 logger.debug('Analysis plugin {} uses model: {}'.format(plugin.id, fromModel))
             except StopIteration:
                 e = Error(('No conversion plugin found for: '
-                           '{} -> {}'.format(fromModel, toModel)))
+                           '{} -> {}'.format(fromModel, toModel)),
+                          status=404)
                 e.original_response = resp
                 e.parameters = params
                 raise e
@@ -210,36 +298,40 @@ class Senpy(object):
 
     @property
     def default_plugin(self):
-        candidate = self._default
+        if not self._default or not self._default.is_activated:
-        if not candidate:
+            candidates = self.plugins(plugin_type='analysisPlugin',
-            candidates = self.filter_plugins(plugin_type='analysisPlugin',
                                       is_activated=True)
             if len(candidates) > 0:
-                candidate = list(candidates.values())[0]
+                self._default = candidates[0]
-        logger.debug("Default: {}".format(candidate))
+            else:
-        return candidate
+                self._default = None
+            logger.debug("Default: {}".format(self._default))
+        return self._default
 
     @default_plugin.setter
     def default_plugin(self, value):
-        if isinstance(value, SenpyPlugin):
+        if isinstance(value, Plugin):
+            if not value.is_activated:
+                raise AttributeError('The default plugin has to be activated.')
             self._default = value
+
         else:
-            self._default = self.plugins[value]
+            self._default = self._plugins[value.lower()]
 
     def activate_all(self, sync=True):
         ps = []
-        for plug in self.plugins.keys():
+        for plug in self._plugins.keys():
             ps.append(self.activate_plugin(plug, sync=sync))
         return ps
 
     def deactivate_all(self, sync=True):
         ps = []
-        for plug in self.plugins.keys():
+        for plug in self._plugins.keys():
             ps.append(self.deactivate_plugin(plug, sync=sync))
         return ps
 
     def _set_active(self, plugin, active=True, *args, **kwargs):
-        ''' We're using a variable in the plugin itself to activate/deactive plugins.\
+        ''' We're using a variable in the plugin itself to activate/deactivate plugins.\
         Note that plugins may activate themselves by setting this variable.
         '''
         plugin.is_activated = active
@@ -249,24 +341,18 @@ class Senpy(object):
         with plugin._lock:
             if plugin.is_activated:
                 return
-            try:
             plugin.activate()
             msg = "Plugin activated: {}".format(plugin.name)
             logger.info(msg)
             success = True
             self._set_active(plugin, success)
-            except Exception as ex:
-                msg = "Error activating plugin {} - {} : \n\t{}".format(
-                    plugin.name, ex, traceback.format_exc())
-                logger.error(msg)
-                raise Error(msg)
 
     def activate_plugin(self, plugin_name, sync=True):
-        try:
+        plugin_name = plugin_name.lower()
-            plugin = self.plugins[plugin_name]
+        if plugin_name not in self._plugins:
-        except KeyError:
             raise Error(
                 message="Plugin not found: {}".format(plugin_name), status=404)
+        plugin = self._plugins[plugin_name]
 
         logger.info("Activating plugin: {}".format(plugin.name))
 
@@ -281,20 +367,15 @@ class Senpy(object):
         with plugin._lock:
             if not plugin.is_activated:
                 return
-            try:
             plugin.deactivate()
             logger.info("Plugin deactivated: {}".format(plugin.name))
-            except Exception as ex:
-                logger.error(
-                    "Error deactivating plugin {}: {}".format(plugin.name, ex))
-                logger.error("Trace: {}".format(traceback.format_exc()))
 
     def deactivate_plugin(self, plugin_name, sync=True):
-        try:
+        plugin_name = plugin_name.lower()
-            plugin = self.plugins[plugin_name]
+        if plugin_name not in self._plugins:
-        except KeyError:
             raise Error(
                 message="Plugin not found: {}".format(plugin_name), status=404)
+        plugin = self._plugins[plugin_name]
 
         self._set_active(plugin, False)
 
@@ -307,19 +388,3 @@ class Senpy(object):
 
     def teardown(self, exception):
         pass
-
-    @property
-    def plugins(self):
-        """ Return the plugins registered for a given application.  """
-        if self._outdated:
-            self._plugin_list = plugins.load_plugins(self._search_folders)
-            self._outdated = False
-        return self._plugin_list
-
-    def filter_plugins(self, **kwargs):
-        return plugins.pfilter(self.plugins, **kwargs)
-
-    @property
-    def analysis_plugins(self):
-        """ Return only the analysis plugins """
-        return self.filter_plugins(plugin_type='analysisPlugin')

senpy/meta.py

@@ -0,0 +1,257 @@
+'''
+Meta-programming for the models.
+'''
+import os
+import json
+import jsonschema
+import inspect
+import copy
+
+from abc import ABCMeta
+from collections import MutableMapping, namedtuple
+
+
+class BaseMeta(ABCMeta):
+    '''
+    Metaclass for models. It extracts the default values for the fields in
+    the model.
+
+    For instance, instances of the following class wouldn't need to mark
+    their version or description on initialization:
+
+    .. code-block:: python
+
+       class MyPlugin(Plugin):
+           version=0.3
+           description='A dull plugin'
+
+
+    Note that these operations could be included in the __init__ of the
+    class, but it would be very inefficient.
+    '''
+    _subtypes = {}
+
+    def __new__(mcs, name, bases, attrs, **kwargs):
+        register_afterwards = False
+        defaults = {}
+
+        attrs = mcs.expand_with_schema(name, attrs)
+        if 'schema' in attrs:
+            register_afterwards = True
+        for base in bases:
+            if hasattr(base, '_defaults'):
+                defaults.update(getattr(base, '_defaults'))
+
+        info, rest = mcs.split_attrs(attrs)
+
+        for i in list(info.keys()):
+            if isinstance(info[i], _Alias):
+                fget, fset, fdel = make_property(info[i].indict)
+                rest[i] = property(fget=fget, fset=fset, fdel=fdel)
+            else:
+                defaults[i] = info[i]
+
+        rest['_defaults'] = defaults
+
+        cls = super(BaseMeta, mcs).__new__(mcs, name, tuple(bases), rest)
+
+        if register_afterwards:
+            mcs.register(cls, defaults['@type'])
+        return cls
+
+    @classmethod
+    def register(mcs, rsubclass, rtype=None):
+        mcs._subtypes[rtype or rsubclass.__name__] = rsubclass
+
+    @staticmethod
+    def expand_with_schema(name, attrs):
+        if 'schema' in attrs:  # Schema specified by name
+            schema_file = '{}.json'.format(attrs['schema'])
+        elif 'schema_file' in attrs:
+            schema_file = attrs['schema_file']
+            del attrs['schema_file']
+        else:
+            return attrs
+
+        if '/' not in 'schema_file':
+            thisdir = os.path.dirname(os.path.realpath(__file__))
+            schema_file = os.path.join(thisdir,
+                                       'schemas',
+                                       schema_file)
+
+        schema_path = 'file://' + schema_file
+
+        with open(schema_file) as f:
+            schema = json.load(f)
+
+        resolver = jsonschema.RefResolver(schema_path, schema)
+        attrs['@type'] = "".join((name[0].lower(), name[1:]))
+        attrs['_schema_file'] = schema_file
+        attrs['schema'] = schema
+        attrs['_validator'] = jsonschema.Draft4Validator(schema, resolver=resolver)
+
+        schema_defaults = BaseMeta.get_defaults(attrs['schema'])
+        attrs.update(schema_defaults)
+
+        return attrs
+
+    @staticmethod
+    def is_func(v):
+        return inspect.isroutine(v) or inspect.ismethod(v) or \
+            inspect.ismodule(v) or isinstance(v, property)
+
+    @staticmethod
+    def is_internal(k):
+        return k[0] == '_' or k == 'schema' or k == 'data'
+
+    @staticmethod
+    def get_key(key):
+        if key[0] != '_':
+            key = key.replace("__", ":", 1)
+        return key
+
+    @staticmethod
+    def split_attrs(attrs):
+        '''
+        Extract the attributes of the class.
+
+        This allows adding default values in the class definition.
+        e.g.:
+        '''
+        isattr = {}
+        rest = {}
+        for key, value in attrs.items():
+            if not (BaseMeta.is_internal(key)) and (not BaseMeta.is_func(value)):
+                isattr[key] = value
+            else:
+                rest[key] = value
+        return isattr, rest
+
+    @staticmethod
+    def get_defaults(schema):
+        temp = {}
+        for obj in [
+                schema,
+        ] + schema.get('allOf', []):
+            for k, v in obj.get('properties', {}).items():
+                if 'default' in v and k not in temp:
+                    temp[k] = v['default']
+        return temp
+
+
+def make_property(key):
+
+    def fget(self):
+        return self[key]
+
+    def fdel(self):
+        del self[key]
+
+    def fset(self, value):
+        self[key] = value
+
+    return fget, fset, fdel
+
+
+class CustomDict(MutableMapping, object):
+    '''
+    A dictionary whose elements can also be accessed as attributes. Since some
+    characters are not valid in the dot-notation, the attribute names also
+    converted. e.g.:
+
+    > d = CustomDict()
+    > d.key = d['ns:name'] = 1
+    > d.key == d['key']
+    True
+    > d.ns__name == d['ns:name']
+    '''
+
+    _defaults = {}
+    _map_attr_key = {'id': '@id'}
+
+    def __init__(self, *args, **kwargs):
+        super(CustomDict, self).__init__()
+        for k, v in self._defaults.items():
+            self[k] = copy.copy(v)
+        for arg in args:
+            self.update(arg)
+        for k, v in kwargs.items():
+            self[self._attr_to_key(k)] = v
+        return self
+
+    def serializable(self):
+        def ser_or_down(item):
+            if hasattr(item, 'serializable'):
+                return item.serializable()
+            elif isinstance(item, dict):
+                temp = dict()
+                for kp in item:
+                    vp = item[kp]
+                    temp[kp] = ser_or_down(vp)
+                return temp
+            elif isinstance(item, list) or isinstance(item, set):
+                return list(ser_or_down(i) for i in item)
+            else:
+                return item
+
+        return ser_or_down(self.as_dict())
+
+    def __getitem__(self, key):
+        key = self._key_to_attr(key)
+        return self.__dict__[key]
+
+    def __setitem__(self, key, value):
+        '''Do not insert data directly, there might be a property in that key. '''
+        key = self._key_to_attr(key)
+        return setattr(self, key, value)
+
+    def as_dict(self):
+        return {self._attr_to_key(k): v for k, v in self.__dict__.items()
+                if not self._internal_key(k)}
+
+    def __iter__(self):
+        return (k for k in self.__dict__ if not self._internal_key(k))
+
+    def __len__(self):
+        return len(self.__dict__)
+
+    def __delitem__(self, key):
+        del self.__dict__[key]
+
+    def update(self, other):
+        for k, v in other.items():
+            self[k] = v
+
+    def _attr_to_key(self, key):
+        key = key.replace("__", ":", 1)
+        key = self._map_attr_key.get(key, key)
+        return key
+
+    def _key_to_attr(self, key):
+        if self._internal_key(key):
+            return key
+        key = key.replace(":", "__", 1)
+        return key
+
+    def __getattr__(self, key):
+        try:
+            return self.__dict__[self._attr_to_key(key)]
+        except KeyError:
+            raise AttributeError
+
+    @staticmethod
+    def _internal_key(key):
+        return key[0] == '_'
+
+    def __str__(self):
+        return str(self.serializable())
+
+    def __repr__(self):
+        return str(self.serializable())
+
+
+_Alias = namedtuple('Alias', 'indict')
+
+
+def alias(key):
+    return _Alias(key)

senpy/models.py

@@ -6,24 +6,30 @@ For compatibility with Py3 and for easier debugging, this new version drops
 introspection and adds all arguments to the models.
 '''
 from __future__ import print_function
-from six import string_types
+from future import standard_library
+standard_library.install_aliases()
+
+from future.utils import with_metaclass
+from past.builtins import basestring
 
 import time
 import copy
 import json
 import os
 import jsonref
-import jsonschema
-
 from flask import Response as FlaskResponse
 from pyld import jsonld
 
-from rdflib import Graph
-
 import logging
 
+logging.getLogger('rdflib').setLevel(logging.WARN)
 logger = logging.getLogger(__name__)
 
+from rdflib import Graph
+
+
+from .meta import BaseMeta, CustomDict, alias
+
 DEFINITIONS_FILE = 'definitions.json'
 CONTEXT_PATH = os.path.join(
     os.path.dirname(os.path.realpath(__file__)), 'schemas', 'context.jsonld')
@@ -45,38 +51,96 @@ def read_schema(schema_file, absolute=False):
         return jsonref.load(f, base_uri=schema_uri)
 
 
-base_schema = read_schema(DEFINITIONS_FILE)
+def load_context(context):
-
-
-class Context(dict):
-    @staticmethod
-    def load(context):
     logging.debug('Loading context: {}'.format(context))
     if not context:
         return context
     elif isinstance(context, list):
         contexts = []
         for c in context:
-                contexts.append(Context.load(c))
+            contexts.append(load_context(c))
         return contexts
     elif isinstance(context, dict):
-            return Context(context)
+        return dict(context)
-        elif isinstance(context, string_types):
+    elif isinstance(context, basestring):
         try:
             with open(context) as f:
-                    return Context(json.loads(f.read()))
+                return dict(json.loads(f.read()))
         except IOError:
             return context
     else:
         raise AttributeError('Please, provide a valid context')
 
 
-base_context = Context.load(CONTEXT_PATH)
+base_context = load_context(CONTEXT_PATH)
 
 
-class SenpyMixin(object):
+def register(rsubclass, rtype=None):
+    BaseMeta.register(rsubclass, rtype)
+
+
+class BaseModel(with_metaclass(BaseMeta, CustomDict)):
+    '''
+    Entities of the base model are a special kind of dictionary that emulates
+    a JSON-LD object. The structure of the dictionary is checked via JSON-schema.
+    For convenience, the values can also be accessed as attributes
+    (a la Javascript). e.g.:
+
+    >>> myobject.key == myobject['key']
+    True
+    >>> myobject.ns__name == myobject['ns:name']
+    True
+
+    Additionally, subclasses of this class can specify default values for their
+    instances. These defaults are inherited by subclasses. e.g.:
+
+    >>> class NewModel(BaseModel):
+    ...     mydefault = 5
+    >>> n1 = NewModel()
+    >>> n1['mydefault'] == 5
+    True
+    >>> n1.mydefault = 3
+    >>> n1['mydefault'] = 3
+    True
+    >>> n2 = NewModel()
+    >>> n2 == 5
+    True
+    >>> class SubModel(NewModel):
+            pass
+    >>> subn = SubModel()
+    >>> subn.mydefault == 5
+    True
+
+    Lastly, every subclass that also specifies a schema will get registered, so it
+    is possible to deserialize JSON and get the right type.
+    i.e. to recover an instance of the original class from a plain JSON.
+
+    '''
+
+    schema_file = DEFINITIONS_FILE
     _context = base_context["@context"]
 
+    def __init__(self, *args, **kwargs):
+        auto_id = kwargs.pop('_auto_id', True)
+
+        super(BaseModel, self).__init__(*args, **kwargs)
+
+        if auto_id:
+            self.id
+
+        if '@type' not in self:
+            logger.warn('Created an instance of an unknown model')
+
+    @property
+    def id(self):
+        if '@id' not in self:
+            self['@id'] = ':{}_{}'.format(type(self).__name__, time.time())
+        return self['@id']
+
+    @id.setter
+    def id(self, value):
+        self['@id'] = value
+
     def flask(self,
               in_headers=True,
               headers=None,
@@ -102,7 +166,7 @@ class SenpyMixin(object):
             })
         return FlaskResponse(
             response=content,
-            status=getattr(self, "status", 200),
+            status=self.get('status', 200),
             headers=headers,
             mimetype=mimetype)
 
@@ -130,25 +194,8 @@ class SenpyMixin(object):
         else:
             return content
 
-    def serializable(self):
-        def ser_or_down(item):
-            if hasattr(item, 'serializable'):
-                return item.serializable()
-            elif isinstance(item, dict):
-                temp = dict()
-                for kp in item:
-                    vp = item[kp]
-                    temp[kp] = ser_or_down(vp)
-                return temp
-            elif isinstance(item, list) or isinstance(item, set):
-                return list(ser_or_down(i) for i in item)
-            else:
-                return item
-
-        return ser_or_down(self._plain_dict())
-
     def jsonld(self,
-               with_context=True,
+               with_context=False,
                context_uri=None,
                prefix=None,
                expanded=False):
@@ -172,10 +219,6 @@ class SenpyMixin(object):
             del result['@context']
         return result
 
-    def to_JSON(self, *args, **kwargs):
-        js = json.dumps(self.jsonld(*args, **kwargs), indent=4, sort_keys=True)
-        return js
-
     def validate(self, obj=None):
         if not obj:
             obj = self
@@ -183,86 +226,22 @@ class SenpyMixin(object):
             obj = obj.jsonld()
         self._validator.validate(obj)
 
-    def __str__(self):
+    def prov(self, another):
-        return str(self.serialize())
+        self['prov:wasGeneratedBy'] = another.id
 
 
-class BaseModel(SenpyMixin, dict):
+def subtypes():
-
+    return BaseMeta._subtypes
-    schema = base_schema
-
-    def __init__(self, *args, **kwargs):
-        if 'id' in kwargs:
-            self.id = kwargs.pop('id')
-        elif kwargs.pop('_auto_id', True):
-            self.id = '_:{}_{}'.format(type(self).__name__, time.time())
-        temp = dict(*args, **kwargs)
-
-        for obj in [
-                self.schema,
-        ] + self.schema.get('allOf', []):
-            for k, v in obj.get('properties', {}).items():
-                if 'default' in v and k not in temp:
-                    temp[k] = copy.deepcopy(v['default'])
-
-        for i in temp:
-            nk = self._get_key(i)
-            if nk != i:
-                temp[nk] = temp[i]
-                del temp[i]
-        try:
-            temp['@type'] = getattr(self, '@type')
-        except AttributeError:
-            logger.warn('Creating an instance of an unknown model')
-
-        super(BaseModel, self).__init__(temp)
-
-    def _get_key(self, key):
-        if key is 'id':
-            key = '@id'
-        key = key.replace("__", ":", 1)
-        return key
-
-    def __delitem__(self, key):
-        dict.__delitem__(self, key)
-
-    def __getattr__(self, key):
-        try:
-            return self.__getitem__(self._get_key(key))
-        except KeyError:
-            raise AttributeError(key)
-
-    def __setattr__(self, key, value):
-        self.__setitem__(self._get_key(key), value)
-
-    def __delattr__(self, key):
-        try:
-            object.__delattr__(self, key)
-        except AttributeError:
-            self.__delitem__(self._get_key(key))
-
-    def _plain_dict(self):
-        d = {k: v for (k, v) in self.items() if k[0] != "_"}
-        return d
-
-
-def register(rsubclass, rtype=None):
-    _subtypes[rtype or rsubclass.__name__] = rsubclass
-
-
-_subtypes = {}
 
 
 def from_dict(indict, cls=None):
     if not cls:
         target = indict.get('@type', None)
+        cls = BaseModel
         try:
-            if target and target in _subtypes:
+            cls = subtypes()[target]
-                cls = _subtypes[target]
+        except KeyError:
-            else:
+            pass
-                cls = BaseModel
-        except Exception:
-            cls = BaseModel
     outdict = dict()
     for k, v in indict.items():
         if k == '@context':
@@ -270,10 +249,11 @@ def from_dict(indict, cls=None):
         elif isinstance(v, dict):
             v = from_dict(indict[k])
         elif isinstance(v, list):
+            v = v[:]
             for ix, v2 in enumerate(v):
                 if isinstance(v2, dict):
                     v[ix] = from_dict(v2)
-        outdict[k] = v
+        outdict[k] = copy.copy(v)
     return cls(**outdict)
 
 
@@ -286,37 +266,54 @@ def from_json(injson):
     return from_dict(indict)
 
 
-def from_schema(name, schema=None, schema_file=None, base_classes=None):
+class Entry(BaseModel):
+    schema = 'entry'
+
+    text = alias('nif:isString')
+
+
+class Sentiment(BaseModel):
+    schema = 'sentiment'
+
+    polarity = alias('marl:hasPolarity')
+    polarityValue = alias('marl:hasPolarityValue')
+
+
+class Error(BaseModel, Exception):
+    schema = 'error'
+
+    def __init__(self, message='Generic senpy exception', *args, **kwargs):
+        Exception.__init__(self, message)
+        super(Error, self).__init__(*args, **kwargs)
+        self.message = message
+
+    def __str__(self):
+        if not hasattr(self, 'errors'):
+            return self.message
+        return '{}:\n\t{}'.format(self.message, self.errors)
+
+    def __hash__(self):
+        return Exception.__hash__(self)
+
+
+# Add the remaining schemas programmatically
+
+def _class_from_schema(name, schema=None, schema_file=None, base_classes=None):
     base_classes = base_classes or []
     base_classes.append(BaseModel)
-    schema_file = schema_file or '{}.json'.format(name)
+    attrs = {}
-    class_name = '{}{}'.format(name[0].upper(), name[1:])
+    if schema:
-    if '/' not in 'schema_file':
+        attrs['schema'] = schema
-        schema_file = os.path.join(os.path.dirname(os.path.realpath(__file__)),
+    elif schema_file:
-                                   'schemas',
+        attrs['schema_file'] = schema_file
-                                   schema_file)
+    else:
-
+        attrs['schema'] = name
-    schema_path = 'file://' + schema_file
+    name = "".join((name[0].upper(), name[1:]))
-
+    return BaseMeta(name, base_classes, attrs)
-    with open(schema_file) as f:
-        schema = json.load(f)
-
-    dct = {}
-
-    resolver = jsonschema.RefResolver(schema_path, schema)
-    dct['@type'] = name
-    dct['_schema_file'] = schema_file
-    dct['schema'] = schema
-    dct['_validator'] = jsonschema.Draft4Validator(schema, resolver=resolver)
-
-    newclass = type(class_name, tuple(base_classes), dct)
-
-    register(newclass, name)
-    return newclass
 
 
-def _add_from_schema(*args, **kwargs):
+def _add_class_from_schema(*args, **kwargs):
-    generatedClass = from_schema(*args, **kwargs)
+    generatedClass = _class_from_schema(*args, **kwargs)
     globals()[generatedClass.__name__] = generatedClass
     del generatedClass
 
@@ -330,55 +327,19 @@ for i in [
         'emotionModel',
         'emotionPlugin',
         'emotionSet',
-        'entry',
+        'entity',
         'help',
         'plugin',
         'plugins',
         'response',
         'results',
-        'sentiment',
         'sentimentPlugin',
         'suggestion',
+        'aggregatedEvaluation',
+        'evaluation',
+        'metric',
+        'dataset',
+        'datasets',
+
 ]:
-    _add_from_schema(i)
+    _add_class_from_schema(i)
-
-_ErrorModel = from_schema('error')
-
-
-class Error(SenpyMixin, Exception):
-    def __init__(self, message, *args, **kwargs):
-        super(Error, self).__init__(self, message, message)
-        self._error = _ErrorModel(message=message, *args, **kwargs)
-        self.message = message
-
-    def validate(self, obj=None):
-        self._error.validate()
-
-    def __getitem__(self, key):
-        return self._error[key]
-
-    def __setitem__(self, key, value):
-        self._error[key] = value
-
-    def __delitem__(self, key):
-        del self._error[key]
-
-    def __getattr__(self, key):
-        if key != '_error' and hasattr(self._error, key):
-            return getattr(self._error, key)
-        raise AttributeError(key)
-
-    def __setattr__(self, key, value):
-        if key != '_error':
-            return setattr(self._error, key, value)
-        else:
-            super(Error, self).__setattr__(key, value)
-
-    def __delattr__(self, key):
-        delattr(self._error, key)
-
-    def __str__(self):
-        return str(self.to_JSON(with_context=False))
-
-
-register(Error, 'error')

senpy/plugins/__init__.py

@@ -1,14 +1,17 @@
 from future import standard_library
 standard_library.install_aliases()
 
+
+from future.utils import with_metaclass
+
 import os.path
 import os
+import re
 import pickle
 import logging
-import tempfile
 import copy
+import pprint
 
-import fnmatch
 import inspect
 import sys
 import subprocess
@@ -16,26 +19,92 @@ import importlib
 import yaml
 import threading
 
+import numpy as np
+
 from .. import models, utils
-from ..api import API_PARAMS
+from .. import api
+
 
 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 Plugin(models.Plugin):
+
-    def __init__(self, info=None):
+class PluginMeta(models.BaseMeta):
+    _classes = {}
+
+    def __new__(mcs, name, bases, attrs, **kwargs):
+        plugin_type = []
+        if hasattr(bases[0], 'plugin_type'):
+            plugin_type += bases[0].plugin_type
+        plugin_type.append(name)
+        alias = attrs.get('name', name)
+        attrs['plugin_type'] = plugin_type
+        attrs['name'] = alias
+        if 'description' not in attrs:
+            doc = attrs.get('__doc__', None)
+            if not doc:
+                raise Exception(('Please, add a description or '
+                                 'documentation to class {}').format(name))
+            attrs['description'] = doc
+            attrs['name'] = alias
+        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. '
+                                'Please, choose a different name').format(name))
+            else:
+                logger.warn('Overloading plugin class: {}'.format(alias))
+        mcs._classes[alias] = cls
+        return cls
+
+    @classmethod
+    def for_type(cls, ptype):
+        return cls._classes[ptype]
+
+
+class Plugin(with_metaclass(PluginMeta, models.Plugin)):
+    '''
+    Base class for all plugins in senpy.
+    A plugin must provide at least these attributes:
+
+        - version
+        - description (or docstring)
+        - author
+
+    Additionally, they may provide a URL (url) of a repository or website.
+
+    '''
+
+    def __init__(self, info=None, data_folder=None, **kwargs):
         """
         Provides a canonical name for plugins and serves as base for other
         kinds of plugins.
         """
-        if not info:
-            raise models.Error(message=("You need to provide configuration"
-                                        "information for the plugin."))
         logger.debug("Initialising {}".format(info))
-        id = 'plugins/{}_{}'.format(info['name'], info['version'])
+        super(Plugin, self).__init__(**kwargs)
-        super(Plugin, self).__init__(id=id, **info)
+        if info:
+            self.update(info)
+        self.validate()
+        self.id = 'plugins/{}_{}'.format(self['name'], self['version'])
         self.is_activated = False
         self._lock = threading.Lock()
+        self.data_folder = data_folder or os.getcwd()
+
+    def validate(self):
+        missing = []
+        for x in ['name', 'description', 'version']:
+            if x not in self:
+                missing.append(x)
+        if missing:
+            raise models.Error('Missing configuration parameters: {}'.format(missing))
 
     def get_folder(self):
         return os.path.dirname(inspect.getfile(self.__class__))
@@ -46,30 +115,62 @@ class Plugin(models.Plugin):
     def deactivate(self):
         pass
 
-    def test(self):
+    def test(self, test_cases=None):
+        if not test_cases:
             if not hasattr(self, 'test_cases'):
-            import inspect
                 raise AttributeError(('Plugin {} [{}] does not have any defined '
-                                  'test cases').format(self.id, inspect.getfile(self.__class__)))
+                                      'test cases').format(self.id,
-        for case in self.test_cases:
+                                                           inspect.getfile(self.__class__)))
-            res = list(self.analyse_entry(models.Entry(case['entry']),
+            test_cases = self.test_cases
-                                          case['params']))
+        for case in test_cases:
-            exp = case['expected']
+            try:
-            if not isinstance(exp, list):
+                self.test_case(case)
-                exp = [exp]
+                logger.debug('Test case passed:\n{}'.format(pprint.pformat(case)))
-            utils.check_template(res, exp)
+            except Exception as ex:
+                logger.warn('Test case failed:\n{}'.format(pprint.pformat(case)))
+                raise
+
+    def test_case(self, case):
+        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)
+        try:
+            params = api.parse_params(given_parameters, self.extra_params)
+            res = list(self.analyse_entries([entry, ], params))
+
+            if not isinstance(expected, list):
+                expected = [expected]
+            utils.check_template(res, expected)
             for r in res:
                 r.validate()
+        except models.Error:
+            if should_fail:
+                return
+            raise
+        assert not should_fail
+
+    def open(self, fpath, *args, **kwargs):
+        if not os.path.isabs(fpath):
+            fpath = os.path.join(self.data_folder, fpath)
+        return open(fpath, *args, **kwargs)
+
+    def serve(self, debug=True, **kwargs):
+        utils.easy(plugin_list=[self, ], plugin_folder=None, debug=debug, **kwargs)
 
 
+# For backwards compatibility
 SenpyPlugin = Plugin
 
 
-class AnalysisPlugin(Plugin):
+class Analysis(Plugin):
+    '''
+    A subclass of Plugin that analyses text and provides an annotation.
+    '''
 
     def analyse(self, *args, **kwargs):
-        raise NotImplemented(
+        raise NotImplementedError(
-            'Your method should implement either analyse or analyse_entry')
+            'Your plugin should implement either analyse or analyse_entry')
 
     def analyse_entry(self, entry, parameters):
         """ An implemented plugin should override this method.
@@ -88,40 +189,193 @@ class AnalysisPlugin(Plugin):
     def analyse_entries(self, entries, parameters):
         for entry in entries:
             logger.debug('Analysing entry with plugin {}: {}'.format(self, entry))
-            for result in self.analyse_entry(entry, parameters):
+            results = self.analyse_entry(entry, parameters)
+            if inspect.isgenerator(results):
+                for result in results:
                     yield result
+            else:
+                yield results
+
+    def test_case(self, case):
+        if 'entry' not in case and 'input' in case:
+            entry = models.Entry(_auto_id=False)
+            entry.nif__isString = case['input']
+            case['entry'] = entry
+        super(Analysis, self).test_case(case)
 
 
-class ConversionPlugin(Plugin):
+AnalysisPlugin = Analysis
+
+
+class Conversion(Plugin):
+    '''
+    A subclass of Plugins that convert between different annotation models.
+    e.g. a conversion of emotion models, or normalization of sentiment values.
+    '''
     pass
 
 
-class SentimentPlugin(models.SentimentPlugin, AnalysisPlugin):
+ConversionPlugin = Conversion
-    def __init__(self, info, *args, **kwargs):
-        super(SentimentPlugin, self).__init__(info, *args, **kwargs)
-        self.minPolarityValue = float(info.get("minPolarityValue", 0))
-        self.maxPolarityValue = float(info.get("maxPolarityValue", 1))
 
 
-class EmotionPlugin(models.EmotionPlugin, AnalysisPlugin):
+class SentimentPlugin(Analysis, models.SentimentPlugin):
-    def __init__(self, info, *args, **kwargs):
+    '''
-        super(EmotionPlugin, self).__init__(info, *args, **kwargs)
+    Sentiment plugins provide sentiment annotation (using Marl)
-        self.minEmotionValue = float(info.get("minEmotionValue", -1))
+    '''
-        self.maxEmotionValue = float(info.get("maxEmotionValue", 1))
+    minPolarityValue = 0
+    maxPolarityValue = 1
+
+    def test_case(self, case):
+        if 'polarity' in case:
+            expected = case.get('expected', {})
+            s = models.Sentiment(_auto_id=False)
+            s.marl__hasPolarity = case['polarity']
+            if 'sentiments' not in expected:
+                expected['sentiments'] = []
+            expected['sentiments'].append(s)
+            case['expected'] = expected
+        super(SentimentPlugin, self).test_case(case)
 
 
-class EmotionConversionPlugin(models.EmotionConversionPlugin, ConversionPlugin):
+class EmotionPlugin(Analysis, models.EmotionPlugin):
+    '''
+    Emotion plugins provide emotion annotation (using Onyx)
+    '''
+    minEmotionValue = 0
+    maxEmotionValue = 1
+
+
+class EmotionConversion(Conversion):
+    '''
+    A subclass of Conversion that converts emotion annotations using different models
+    '''
     pass
 
 
+EmotionConversionPlugin = EmotionConversion
+
+
+class Box(AnalysisPlugin):
+    '''
+    Black box plugins delegate analysis to a function.
+    The flow is like so:
+
+    .. code-block::
+
+                   entry --> input() --> predict_one() --> output() --> entry'
+
+
+    In other words: their ``input`` method convers a query (entry and a set of parameters) into
+    the input to the box method. The ``output`` method convers the results given by the box into
+    an entry that senpy can handle.
+    '''
+
+    def input(self, entry, params=None):
+        '''Transforms a query (entry+param) into an input for the black box'''
+        return entry
+
+    def output(self, output, entry=None, params=None):
+        '''Transforms the results of the black box into an entry'''
+        return output
+
+    def predict_one(self, input):
+        raise NotImplementedError('You should define the behavior of this plugin')
+
+    def analyse_entries(self, entries, params):
+        for entry in entries:
+            input = self.input(entry=entry, params=params)
+            results = self.predict_one(input=input)
+            yield self.output(output=results, entry=entry, params=params)
+
+    def fit(self, X=None, y=None):
+        return self
+
+    def transform(self, X):
+        return np.array([self.predict_one(x) for x in X])
+
+    def predict(self, X):
+        return self.transform(X)
+
+    def fit_transform(self, X, y):
+        self.fit(X, y)
+        return self.transform(X)
+
+    def as_pipe(self):
+        pipe = Pipeline([('plugin', self)])
+        pipe.name = self.name
+        return pipe
+
+
+class TextBox(Box):
+    '''A black box plugin that takes only text as input'''
+
+    def input(self, entry, params):
+        entry = super(TextBox, self).input(entry, params)
+        return entry['nif:isString']
+
+
+class SentimentBox(TextBox, SentimentPlugin):
+    '''
+    A box plugin where the output is only a polarity label or a tuple (polarity, polarityValue)
+    '''
+
+    def output(self, output, entry, **kwargs):
+        s = models.Sentiment()
+        try:
+            label, value = output
+        except ValueError:
+            label, value = output, None
+        s.prov(self)
+        s.polarity = label
+        if value is not None:
+            s.polarityValue = value
+        entry.sentiments.append(s)
+        return entry
+
+
+class EmotionBox(TextBox, EmotionPlugin):
+    '''
+    A box plugin where the output is only an a tuple of emotion labels
+    '''
+
+    def output(self, output, entry, **kwargs):
+        if not isinstance(output, list):
+            output = [output]
+        s = models.EmotionSet()
+        entry.emotions.append(s)
+        for label in output:
+            e = models.Emotion(onyx__hasEmotionCategory=label)
+            s.append(e)
+        return entry
+
+
+class MappingMixin(object):
+
+    @property
+    def mappings(self):
+        return self._mappings
+
+    @mappings.setter
+    def mappings(self, value):
+        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)
+
+
 class ShelfMixin(object):
     @property
     def sh(self):
         if not hasattr(self, '_sh') or self._sh is None:
-            self.__dict__['_sh'] = {}
+            self._sh = {}
             if os.path.isfile(self.shelf_file):
                 try:
-                    self.__dict__['_sh'] = pickle.load(open(self.shelf_file, 'rb'))
+                    with self.open(self.shelf_file, 'rb') as p:
+                        self._sh = pickle.load(p)
                 except (IndexError, EOFError, pickle.UnpicklingError):
                     logger.warning('{} has a corrupted shelf file!'.format(self.id))
                     if not self.get('force_shelf', False):
@@ -132,36 +386,42 @@ class ShelfMixin(object):
     def sh(self):
         if os.path.isfile(self.shelf_file):
             os.remove(self.shelf_file)
-            del self.__dict__['_sh']
+            del self._sh
         self.save()
 
+    @sh.setter
+    def sh(self, value):
+        self._sh = value
+
     @property
     def shelf_file(self):
-        if 'shelf_file' not in self or not self['shelf_file']:
+        if not hasattr(self, '_shelf_file') or not self._shelf_file:
-            sd = os.environ.get('SENPY_DATA', tempfile.gettempdir())
+            self._shelf_file = os.path.join(self.data_folder, self.name + '.p')
-            self.shelf_file = os.path.join(sd, self.name + '.p')
+        return self._shelf_file
-        return self['shelf_file']
+
+    @shelf_file.setter
+    def shelf_file(self, value):
+        self._shelf_file = value
 
     def save(self):
         logger.debug('saving pickle')
         if hasattr(self, '_sh') and self._sh is not None:
-            with open(self.shelf_file, 'wb') as f:
+            with self.open(self.shelf_file, 'wb') as f:
                 pickle.dump(self._sh, f)
 
 
-default_plugin_type = API_PARAMS['plugin_type']['default']
-
-
 def pfilter(plugins, **kwargs):
     """ Filter plugins by different criteria """
     if isinstance(plugins, models.Plugins):
         plugins = plugins.plugins
     elif isinstance(plugins, dict):
         plugins = plugins.values()
-    ptype = kwargs.pop('plugin_type', default_plugin_type)
+    ptype = kwargs.pop('plugin_type', Plugin)
     logger.debug('#' * 100)
     logger.debug('ptype {}'.format(ptype))
     if ptype:
+        if isinstance(ptype, PluginMeta):
+            ptype = ptype.__name__
         try:
             ptype = ptype[0].upper() + ptype[1:]
             pclass = globals()[ptype]
@@ -183,11 +443,7 @@ def pfilter(plugins, **kwargs):
 
     if kwargs:
         candidates = filter(matches, candidates)
-    return {p.name: p for p in candidates}
+    return candidates
-
-
-def validate_info(info):
-    return all(x in info for x in ('name', 'module', 'description', 'version'))
 
 
 def load_module(name, root=None):
@@ -199,7 +455,7 @@ def load_module(name, root=None):
     return tmp
 
 
-def log_subprocess_output(process):
+def _log_subprocess_output(process):
     for line in iter(process.stdout.readline, b''):
         logger.info('%r', line)
     for line in iter(process.stderr.readline, b''):
@@ -207,77 +463,176 @@ def log_subprocess_output(process):
 
 
 def install_deps(*plugins):
+    installed = False
     for info in plugins:
         requirements = info.get('requirements', [])
         if requirements:
-            pip_args = ['pip']
+            pip_args = [sys.executable, '-m', 'pip', 'install']
-            pip_args.append('install')
-            pip_args.append('--use-wheel')
             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)
-            log_subprocess_output(process)
+            _log_subprocess_output(process)
             exitcode = process.wait()
+            installed = True
             if exitcode != 0:
                 raise models.Error("Dependencies not properly installed")
+    return installed
 
 
-def load_plugin_from_info(info, root=None, validator=validate_info, install=True):
+is_plugin_file = re.compile(r'.*\.senpy$|senpy_[a-zA-Z0-9_]+\.py$|'
-    if not root and '_path' in info:
+                            '^(?!test_)[a-zA-Z0-9_]+_plugin.py$')
-        root = os.path.dirname(info['_path'])
-    if not validator(info):
-        raise ValueError('Plugin info is not valid: {}'.format(info))
-    module = info["module"]
-
-    try:
-        tmp = load_module(module, root)
-    except ImportError:
-        if not install:
-            raise
-        install_deps(info)
-        tmp = load_module(module, root)
-    candidate = None
-    for _, obj in inspect.getmembers(tmp):
-        if inspect.isclass(obj) and inspect.getmodule(obj) == tmp:
-            logger.debug(("Found plugin class:"
-                          " {}@{}").format(obj, inspect.getmodule(obj)))
-            candidate = obj
-            break
-    if not candidate:
-        logger.debug("No valid plugin for: {}".format(module))
-        return
-    module = candidate(info=info)
-    return module
 
 
-def parse_plugin_info(fpath):
+def find_plugins(folders):
-    logger.debug("Loading plugin: {}".format(fpath))
-    with open(fpath, 'r') as f:
-        info = yaml.load(f)
-    info['_path'] = fpath
-    name = info['name']
-    return name, info
-
-
-def load_plugin(fpath):
-    name, info = parse_plugin_info(fpath)
-    logger.debug("Info: {}".format(info))
-    plugin = load_plugin_from_info(info)
-    return name, plugin
-
-
-def load_plugins(folders, loader=load_plugin):
-    plugins = {}
     for search_folder in folders:
         for root, dirnames, filenames in os.walk(search_folder):
             # Do not look for plugins in hidden or special folders
             dirnames[:] = [d for d in dirnames if d[0] not in ['.', '_']]
-            for filename in fnmatch.filter(filenames, '*.senpy'):
+            for filename in filter(is_plugin_file.match, filenames):
                 fpath = os.path.join(root, filename)
-                name, plugin = loader(fpath)
+                yield fpath
-                if plugin and name:
+
-                    plugins[name] = plugin
+
+def from_path(fpath, **kwargs):
+    logger.debug("Loading plugin from {}".format(fpath))
+    if fpath.endswith('.py'):
+        # We asume root is the dir of the file, and module is the name of the file
+        root = os.path.dirname(fpath)
+        module = os.path.basename(fpath)[:-3]
+        for instance in _from_module_name(module=module, root=root, **kwargs):
+            yield instance
+    else:
+        info = parse_plugin_info(fpath)
+        yield from_info(info, **kwargs)
+
+
+def from_folder(folders, loader=from_path, **kwargs):
+    plugins = []
+    for fpath in find_plugins(folders):
+        for plugin in loader(fpath, **kwargs):
+            plugins.append(plugin)
     return plugins
+
+
+def from_info(info, root=None, **kwargs):
+    if any(x not in info for x in ('module',)):
+        raise ValueError('Plugin info is not valid: {}'.format(info))
+    module = info["module"]
+
+    if not root and '_path' in info:
+        root = os.path.dirname(info['_path'])
+
+    return one_from_module(module, root=root, info=info, **kwargs)
+
+
+def parse_plugin_info(fpath):
+    logger.debug("Parsing plugin info: {}".format(fpath))
+    with open(fpath, 'r') as f:
+        info = yaml.load(f)
+    info['_path'] = fpath
+    return info
+
+
+def from_module(module, **kwargs):
+
+    if inspect.ismodule(module):
+        res = _from_loaded_module(module, **kwargs)
+    else:
+        res = _from_module_name(module, **kwargs)
+    for p in res:
+        yield p
+
+
+def one_from_module(module, root, info, **kwargs):
+    if '@type' in info:
+        cls = PluginMeta.from_type(info['@type'])
+        return cls(info=info, **kwargs)
+    instance = next(from_module(module=module, root=root, info=info, **kwargs), None)
+    if not instance:
+        raise Exception("No valid plugin for: {}".format(module))
+    return instance
+
+
+def _classes_in_module(module):
+    for _, obj in inspect.getmembers(module):
+        if inspect.isclass(obj) and inspect.getmodule(obj) == module:
+            logger.debug(("Found plugin class:"
+                          " {}@{}").format(obj, inspect.getmodule(obj)))
+            yield obj
+
+
+def _instances_in_module(module):
+    for _, obj in inspect.getmembers(module):
+        if isinstance(obj, Plugin) and inspect.getmodule(obj) == module:
+            logger.debug(("Found plugin instance:"
+                          " {}@{}").format(obj, inspect.getmodule(obj)))
+            yield obj
+
+
+def _from_module_name(module, root, info=None, install=True, **kwargs):
+    try:
+        module = load_module(module, root)
+    except ImportError:
+        if not install or not info:
+            raise
+        install_deps(info)
+        module = load_module(module, root)
+    for plugin in _from_loaded_module(module=module, root=root, info=info, **kwargs):
+        yield plugin
+
+
+def _from_loaded_module(module, info=None, **kwargs):
+    for cls in _classes_in_module(module):
+        yield cls(info=info, **kwargs)
+    for instance in _instances_in_module(module):
+        yield instance
+
+
+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,
+              datasets=datasets,
+              pipelines=[plugin.as_pipe() for plugin in plugins])
+    ev.evaluate()
+    results = ev.results
+    evaluations = evaluations_to_JSONLD(results, **kwargs)
+    return evaluations
+
+
+def evaluations_to_JSONLD(results, flatten=False):
+    '''
+    Map the evaluation results to a JSONLD scheme
+    '''
+
+    evaluations = list()
+    metric_names = ['accuracy', 'precision_macro', 'recall_macro',
+                    'f1_macro', 'f1_weighted', 'f1_micro', 'f1_macro']
+
+    for index, row in results.iterrows():
+        evaluation = models.Evaluation()
+        if row.get('CV', True):
+            evaluation['@type'] = ['StaticCV', 'Evaluation']
+        evaluation.evaluatesOn = row['Dataset']
+        evaluation.evaluates = row['Model']
+        i = 0
+        if flatten:
+            metric = models.Metric()
+            for name in metric_names:
+                metric[name] = row[name]
+            evaluation.metrics.append(metric)
+        else:
+            # We should probably discontinue this representation
+            for name in metric_names:
+                metric = models.Metric()
+                metric['@id'] = 'Metric' + str(i)
+                metric['@type'] = name.capitalize()
+                metric.value = row[name]
+                evaluation.metrics.append(metric)
+                i += 1
+        evaluations.append(evaluation)
+    return evaluations

senpy/plugins/conversion/emotion/centroids.py

@@ -6,7 +6,12 @@ logger = logging.getLogger(__name__)
 
 
 class CentroidConversion(EmotionConversionPlugin):
-    def __init__(self, info):
+    '''
+    This plugin converts emotion annotations from a dimensional model to a
+    categorical one, and vice versa. The centroids used in the conversion
+    are configurable and appear in the semantic description of the plugin.
+    '''
+    def __init__(self, info, *args, **kwargs):
         if 'centroids' not in info:
             raise Error('Centroid conversion plugins should provide '
                         'the centroids in their senpy file')
@@ -33,7 +38,7 @@ class CentroidConversion(EmotionConversionPlugin):
                 ncentroids[aliases.get(k1, k1)] = nv1
             info['centroids'] = ncentroids
 
-        super(CentroidConversion, self).__init__(info)
+        super(CentroidConversion, self).__init__(info, *args, **kwargs)
 
         self.dimensions = set()
         for c in self.centroids.values():

senpy/plugins/example/emoRand/emoRand.py

@@ -4,7 +4,15 @@ from senpy.plugins import EmotionPlugin
 from senpy.models import EmotionSet, Emotion, Entry
 
 
-class RmoRandPlugin(EmotionPlugin):
+class EmoRand(EmotionPlugin):
+    name = "emoRand"
+    description = 'A sample plugin that returns a random emotion annotation'
+    author = '@balkian'
+    version = '0.1'
+    url = "https://github.com/gsi-upm/senpy-plugins-community"
+    requirements = {}
+    onyx__usesEmotionModel = "emoml:big6"
+
     def analyse_entry(self, entry, params):
         category = "emoml:big6happiness"
         number = max(-1, min(1, random.gauss(0, 0.5)))

senpy/plugins/example/emoRand/emoRand.senpy

@@ -1,9 +0,0 @@
----
-name: emoRand
-module: emoRand
-description: A sample plugin that returns a random emotion annotation
-author: "@balkian"
-version: '0.1'
-url: "https://github.com/gsi-upm/senpy-plugins-community"
-requirements: {}
-onyx:usesEmotionModel: "emoml:big6"

senpy/plugins/example/emorand_plugin.py

@@ -0,0 +1,32 @@
+import random
+
+from senpy.plugins import EmotionPlugin
+from senpy.models import EmotionSet, Emotion, Entry
+
+
+class EmoRand(EmotionPlugin):
+    '''A sample plugin that returns a random emotion annotation'''
+    author = '@balkian'
+    version = '0.1'
+    url = "https://github.com/gsi-upm/senpy-plugins-community"
+    onyx__usesEmotionModel = "emoml:big6"
+
+    def analyse_entry(self, entry, params):
+        category = "emoml:big6happiness"
+        number = max(-1, min(1, random.gauss(0, 0.5)))
+        if number > 0:
+            category = "emoml:big6anger"
+        emotionSet = EmotionSet()
+        emotion = Emotion({"onyx:hasEmotionCategory": category})
+        emotionSet.onyx__hasEmotion.append(emotion)
+        emotionSet.prov__wasGeneratedBy = self.id
+        entry.emotions.append(emotionSet)
+        yield entry
+
+    def test(self):
+        params = dict()
+        results = list()
+        for i in range(100):
+            res = next(self.analyse_entry(Entry(nif__isString="Hello"), params))
+            res.validate()
+            results.append(res.emotions[0]['onyx:hasEmotion'][0]['onyx:hasEmotionCategory'])

senpy/plugins/example/rand/rand.senpy

@@ -1,10 +0,0 @@
----
-name: rand
-module: rand
-description: A sample plugin that returns a random sentiment annotation
-author: "@balkian"
-version: '0.1'
-url: "https://github.com/gsi-upm/senpy-plugins-community"
-requirements: {}
-marl:maxPolarityValue: '1'
-marl:minPolarityValue: "-1"

senpy/plugins/example/rand_plugin.py

@@ -1,33 +1,35 @@
 import random
-
+from senpy import SentimentPlugin, Sentiment, Entry
-from senpy.plugins import SentimentPlugin
-from senpy.models import Sentiment, Entry
 
 
-class RandPlugin(SentimentPlugin):
+class Rand(SentimentPlugin):
+    '''A sample plugin that returns a random sentiment annotation'''
+    author = "@balkian"
+    version = '0.1'
+    url = "https://github.com/gsi-upm/senpy-plugins-community"
+    marl__maxPolarityValue = '1'
+    marl__minPolarityValue = "-1"
+
     def analyse_entry(self, entry, params):
-        lang = params.get("language", "auto")
-
         polarity_value = max(-1, min(1, random.gauss(0.2, 0.2)))
         polarity = "marl:Neutral"
         if polarity_value > 0:
             polarity = "marl:Positive"
         elif polarity_value < 0:
             polarity = "marl:Negative"
-        sentiment = Sentiment({
+        sentiment = Sentiment(marl__hasPolarity=polarity,
-            "marl:hasPolarity": polarity,
+                              marl__polarityValue=polarity_value)
-            "marl:polarityValue": polarity_value
+        sentiment.prov(self)
-        })
-        sentiment["prov:wasGeneratedBy"] = self.id
         entry.sentiments.append(sentiment)
-        entry.language = lang
         yield entry
 
     def test(self):
+        '''Run several random analyses.'''
         params = dict()
         results = list()
-        for i in range(100):
+        for i in range(50):
-            res = next(self.analyse_entry(Entry(nif__isString="Hello"), params))
+            res = next(self.analyse_entry(Entry(nif__isString="Hello"),
+                                          params))
             res.validate()
             results.append(res.sentiments[0]['marl:hasPolarity'])
         assert 'marl:Positive' in results

senpy/plugins/misc/split.py

@@ -6,19 +6,21 @@ import nltk
 
 
 class SplitPlugin(AnalysisPlugin):
+    '''description: A sample plugin that chunks input text'''
 
     def activate(self):
         nltk.download('punkt')
 
     def analyse_entry(self, entry, params):
-        chunker_type = params.get("delimiter", "sentence")
+        chunker_type = params["delimiter"]
-        original_text = entry.get('nif:isString', None)
+        original_text = entry['nif:isString']
         if chunker_type == "sentence":
             tokenizer = PunktSentenceTokenizer()
         if chunker_type == "paragraph":
             tokenizer = LineTokenizer()
-        chars = tokenizer.span_tokenize(original_text)
+        chars = list(tokenizer.span_tokenize(original_text))
         for i, chunk in enumerate(tokenizer.tokenize(original_text)):
+            print(chunk)
             e = Entry()
             e['nif:isString'] = chunk
             if entry.id:
@@ -44,20 +46,20 @@ class SplitPlugin(AnalysisPlugin):
         },
         {
             'entry': {
-                "id": ":test",
+                "@id": ":test",
-                'nif:isString': 'Hello. World.'
+                'nif:isString': 'Hello\nWorld'
             },
             'params': {
-                'delimiter': 'sentence',
+                'delimiter': 'paragraph',
             },
             'expected': [
                 {
-                    "@id": ":test#char=0,6",
+                    "@id": ":test#char=0,5",
-                    'nif:isString': 'Hello.'
+                    'nif:isString': 'Hello'
                 },
                 {
-                    "@id": ":test#char=7,13",
+                    "@id": ":test#char=6,11",
-                    'nif:isString': 'World.'
+                    'nif:isString': 'World'
                 }
             ]
         }

senpy/plugins/sentiment/sentiment140/sentiment140.py

@@ -6,8 +6,9 @@ from senpy.models import Sentiment
 
 
 class Sentiment140Plugin(SentimentPlugin):
+    '''Connects to the sentiment140 free API: http://sentiment140.com'''
     def analyse_entry(self, entry, params):
-        lang = params.get("language", "auto")
+        lang = params["language"]
         res = requests.post("http://www.sentiment140.com/api/bulkClassifyJson",
                             json.dumps({
                                 "language": lang,
@@ -35,6 +36,18 @@ class Sentiment140Plugin(SentimentPlugin):
         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.test 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': {

senpy/plugins/sentiment/sentiment140/sentiment140.senpy

@@ -16,6 +16,7 @@ extra_params:
     - es
     - en
     - auto
+    default: auto
 requirements: {}
 maxPolarityValue: 1
 minPolarityValue: 0

senpy/schemas/aggregatedEvaluation.json

@@ -0,0 +1,38 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "allOf": [
+    {"$ref": "response.json"},
+    {
+      "title": "AggregatedEvaluation",
+      "description": "The results of the evaluation",
+      "type": "object",
+      "properties": {
+        "@context": {
+          "$ref": "context.json"
+        },
+        "@type": {
+          "default": "AggregatedEvaluation"
+        },
+        "@id": {
+          "description": "ID of the aggregated evaluation",
+          "type": "string"
+        },
+        "evaluations": {
+          "default": [],
+          "type": "array",
+          "items": {
+            "anyOf": [
+              {
+                "$ref": "evaluation.json"
+              },{
+                "type": "string"
+              }
+            ]
+          }
+        }
+
+      },
+      "required": ["@id", "evaluations"]
+    }
+  ]
+}

senpy/schemas/dataset.json

@@ -0,0 +1,29 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "name": "Dataset",
+  "properties": {
+    "@id": {
+      "type": "string"
+    },
+    "name": {
+      "type": "string"
+    },
+    "compression": {
+      "type": "string"
+    },
+    "expected_bytes": {
+      "type": "int"
+    },
+    "filename": {
+      "description": "Name of the dataset",
+      "type": "string"
+    },
+    "url": {
+      "description": "Classifier or plugin evaluated",
+      "type": "string"
+    },
+    "stats": {        
+    }
+  },
+  "required": ["@id"]
+}

senpy/schemas/datasets.json

@@ -0,0 +1,18 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "allOf": [
+    {"$ref": "response.json"},
+    {
+     "required": ["datasets"],
+      "properties": {
+        "datasets": {
+          "type": "array",
+          "default": [],
+          "items": {
+            "$ref": "dataset.json"
+          }
+        }
+      }
+    }
+  ]
+}

senpy/schemas/definitions.json

@@ -41,5 +41,20 @@
   },
   "Response": {
     "$ref": "response.json"
+  },
+  "AggregatedEvaluation": {
+    "$ref": "aggregatedEvaluation.json"
+  },
+  "Evaluation": {
+    "$ref": "evaluation.json"
+  },
+  "Metric": {
+    "$ref": "metric.json"
+  },
+  "Dataset": {
+    "$ref": "dataset.json"
+  },
+  "Datasets": {
+    "$ref": "datasets.json"
   }
 }

senpy/schemas/evaluation.json

@@ -0,0 +1,28 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "name": "Evalation",
+  "properties": {
+    "@id": {
+      "type": "string"
+    },
+    "@type": {
+      "type": "array",
+      "default": "Evaluation"
+
+    },
+    "metrics": {
+      "type": "array",
+      "items": {"$ref": "metric.json" },
+      "default": []
+    },
+    "evaluatesOn": {
+      "description": "Name of the dataset evaluated ",
+      "type": "string"
+    },
+    "evaluates": {
+      "description": "Classifier or plugin evaluated",
+      "type": "string"
+    }
+  },
+  "required": ["@id", "metrics"]
+}

senpy/schemas/help.json

@@ -7,11 +7,11 @@
       "description": "Help containing accepted parameters",
       "type": "object",
       "properties": {
-        "parameters": {
+        "valid_parameters": {
           "type": "object"
           }
         },
-        "required": "parameters"
+        "required": "valid_parameters"
       }
     ]
 }

senpy/schemas/metric.json

@@ -0,0 +1,24 @@
+{
+  "$schema": "http://json-schema.org/draft-04/schema#",
+  "properties": {
+    "@id": {
+      "type": "string"
+    },
+    "@type": {
+      "type": "string"
+    },
+    "maxValue": {
+      "type": "number"
+    },
+    "minValue": {
+      "type": "number"
+    },
+    "value": {
+      "type": "number"
+    },
+    "deviation": {
+      "type": "number"
+    }
+  },
+  "required": ["@id"]
+}

senpy/schemas/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "http://json-schema.org/draft-04/schema#",
   "type": "object",
-  "required": ["@id", "extra_params"],
+  "required": ["@id", "name", "description", "version", "plugin_type"],
   "properties": {
     "@id": {
       "type": "string",
@@ -9,7 +9,19 @@
     },
     "name": {
       "type": "string",
-      "description": "The name of the plugin, which will be used in the algorithm detection phase"
+      "description": "The name of the plugin, which will be used in the algorithm detection phase."
+    },
+    "description": {
+      "type": "string",
+      "description": "A summary of what the plugin does, and pointers to further information."
+    },
+    "version": {
+      "type": "string",
+      "description": "The version of the plugin."
+    },
+    "plugin_type": {
+      "type": "string",
+      "description": "Sub-type of plugin. e.g. sentimentPlugin"
     },
     "extra_params": {
       "type": "object",

senpy/static/css/main.css

@@ -152,3 +152,18 @@ textarea{
     /* background: white; */
     display: none;
 }
+
+.deco-none {
+    color: inherit;
+    text-decoration: inherit;
+}
+
+.deco-none:link {
+    color: inherit;
+    text-decoration: inherit;
+}
+
+.deco-none:hover {
+    color: inherit;
+    text-decoration: inherit;
+}

senpy/static/js/main.js

@@ -1,7 +1,10 @@
 var ONYX = "http://www.gsi.dit.upm.es/ontologies/onyx/ns#";
 var RDF_TYPE =  "http://www.w3.org/1999/02/22-rdf-syntax-ns#type";
-var plugins_params={};
+var plugins_params = default_params = {};
-var default_params = JSON.parse($.ajax({type: "GET", url: "/api?help=True" , async: false}).responseText);
+var plugins = [];
+var defaultPlugin = {};
+var gplugins = {};
+
 function replaceURLWithHTMLLinks(text) {
     console.log('Text: ' + text);
     var exp = /(\b(https?|ftp|file):\/\/[-A-Z0-9+&@#\/%?=~_|!:,.;]*[-A-Z0-9+&@#\/%=~_|])/ig;
@@ -25,59 +28,71 @@ function hashchanged(){
     }
 }
 
-$(document).ready(function() {
+
-    var response = JSON.parse($.ajax({type: "GET", url: "/api/plugins/" , async: false}).responseText);
+function get_plugins(response){
-    var defaultPlugin= JSON.parse($.ajax({type: "GET", url: "/api/plugins/default" , async: false}).responseText);
-	html="";
-    var availablePlugins = document.getElementById('availablePlugins');
     plugins = response.plugins;
-	gplugins = {};
+}
+
+function get_datasets(response){
+    datasets = response.datasets
+}
+
+function group_plugins(){
     for (r in plugins){
         ptype = plugins[r]['@type'];
         if(gplugins[ptype] == undefined){
-		  gplugins[ptype] = [r]
+            gplugins[ptype] = [r];
         }else{
-		  gplugins[ptype].push(r)
+            gplugins[ptype].push(r);
 	      }
     }
+}
+
+function get_parameters(){
+    for (p in plugins){	
+        plugin = plugins[p];
+        if (plugin["extra_params"]){
+            plugins_params[plugin["name"]] = plugin["extra_params"];
+        }
+    }
+}
+
+function draw_plugins_selection(){
+    html="";
+    group_plugins();
     for (g in gplugins){	
         html += "<optgroup label=\""+g+"\">"
         for (r in gplugins[g]){
-		  plugin = plugins[r]
+            plugin = plugins[gplugins[g][r]]
-		  if (plugin["name"]){
+            if (!plugin["name"]){
+                console.log("No name for plugin ", plugin);
+                continue;
+
+            }
+            html+= "<option value=\""+plugin.name+"\" "
             if (plugin["name"] == defaultPlugin["name"]){
-			  if (plugin["is_activated"]){
+                html+= " selected=\"selected\""
-				html+= "<option value=\""+plugin["name"]+"\" selected=\"selected\">"+plugin["name"]+"</option>"
-			  }else{
-				html+= "<option value=\""+plugin["name"]+"\" selected=\"selected\" disabled=\"disabled\">"+plugin["name"]+"</option>"
-			  }
-			}
-			else{
-			  if (plugin["is_activated"]){
-				html+= "<option value=\""+plugin["name"]+"\">"+plugin["name"]+"</option>"
-			  }
-			  else{
-				html+= "<option value=\""+plugin["name"]+"\" disabled=\"disabled\">"+plugin["name"]+"</option>"
-			  }
             }
+            if (!plugin["is_activated"]){
+                html+= " disabled=\"disabled\" "
             }
+            html+=">"+plugin["name"]+"</option>"
 
-		 if (plugin["extra_params"]){
-			plugins_params[plugin["name"]]={};
-			for (param in plugin["extra_params"]){
-			  if (typeof plugin["extra_params"][param] !="string"){
-				var params = new Array();
-				var alias = plugin["extra_params"][param]["aliases"][0];
-				params[alias]=new Array();
-				for (option in plugin["extra_params"][param]["options"]){
-				  params[alias].push(plugin["extra_params"][param]["options"][option])
-				}
-				plugins_params[plugin["name"]][alias] = (params[alias])
         }
     }
-		  }
+    html += "</optgroup>"
-		  var pluginList = document.createElement('li');
+    // 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;
+}
 
+function draw_plugins_list(){
+    var availablePlugins = document.getElementById('availablePlugins');
+
+    for(p in plugins){
+        var pluginEntry = document.createElement('li');
+        plugin = plugins[p];
         newHtml = ""
         if(plugin.url) {
             newHtml= "<a href="+plugin.url+">" + plugin.name + "</a>";
@@ -85,13 +100,34 @@ $(document).ready(function() {
             newHtml= plugin["name"];
         }
         newHtml += ": " + replaceURLWithHTMLLinks(plugin.description);
-		  pluginList.innerHTML = newHtml;
+        pluginEntry.innerHTML = newHtml;
-		  availablePlugins.appendChild(pluginList)
+        availablePlugins.appendChild(pluginEntry)
     }
-	 html += "</optgroup>"
+}
+
+function draw_datasets(){
+    html = "";
+    repeated_html = "<input class=\"checks-datasets\" type=\"checkbox\" value=\"";
+    for (dataset in datasets){
+        html += repeated_html+datasets[dataset]["@id"]+"\">"+datasets[dataset]["@id"];
+        html += "<br>"
     }
-    document.getElementById('plugins').innerHTML = html;
+    document.getElementById("datasets").innerHTML = html;
-    change_params();
+}
+
+$(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_datasets();
 
     $(window).on('hashchange', hashchanged);
     hashchanged();
@@ -99,38 +135,132 @@ $(document).ready(function() {
 
 });
 
+function get_default_parameters(){
+    default_params = JSON.parse($.ajax({type: "GET", url: "/api?help=true" , async: false}).responseText).valid_parameters;
+    // Remove the parameters that are always added
+    delete default_params["input"];
+    delete default_params["algorithm"];
+    delete default_params["help"];
 
-function change_params(){
+}
-      var plugin = document.getElementById("plugins").options[document.getElementById("plugins").selectedIndex].value;
-        html=""
-		for (param in default_params){
-		  if ((default_params[param]['options']) && (['help','conversion'].indexOf(param) < 0)){  
-			html+= "<label> "+param+"</label>"
-            html+= "<select id=\""+param+"\" name=\""+param+"\">"
-			for (option in default_params[param]['options']){
-				if (default_params[param]['options'][option] == default_params[param]['default']){  
-					html+="<option value \""+default_params[param]['options'][option]+"\" selected >"+default_params[param]['options'][option]+"</option>"
-				}
-				else{
-					html+="<option value \""+default_params[param]['options'][option]+"\">"+default_params[param]['options'][option]+"</option>"
 
+function draw_default_parameters(){
+    var basic_params = document.getElementById("basic_params");
+    basic_params.innerHTML = params_div(default_params);
+}
+
+function draw_extra_parameters(){
+    var plugin = document.getElementsByClassName('plugin')[0].options[document.getElementsByClassName('plugin')[0].selectedIndex].value;
+    get_parameters();
+
+    var extra_params = document.getElementById("extra_params");
+    extra_params.innerHTML = params_div(plugins_params[plugin]);
+}
+
+function draw_parameters(){
+    draw_default_parameters();
+    draw_extra_parameters();
+}
+
+
+function add_default_params(){
+    var html = "";
+    // html += '<a href="#basic_params" class="btn btn-info" data-toggle="collapse">Basic API parameters</a>';
+    html += '<span id="basic_params" class="panel-collapse collapse">';
+    html += '<ul class="list-group">'
+    html += params_div(default_params);
+    html += '</span>';
+    return html;
+}
+
+function params_div(params){
+    var html = '<div class="container-fluid">';
+    if (Object.keys(params).length === 0) {
+        html += '<p class="text text-muted text-center">This plugin does not take any extra parameters</p>';
     }
+    // Iterate over the keys in order
+    pnames = Object.keys(params).sort()
+    for (ix in pnames){
+        pname = pnames[ix];
+        param = params[pname];
+        html+='<div class="form-group">';
+        html += '<div class="row">'
+        html+= '<label class="col-sm-4" for="'+pname+'">'+pname+'</label>'
+        if (param.options){
+            opts = param.options;
+            if(param.options.length == 1 && param.options[0] == 'boolean') {
+                opts = [true, false];
             }
-			html+="</select><br>"
+            html+= '<select class="col-sm-8" id="'+pname+"\" name=\""+pname+"\">"
-		  }
+            var defaultopt = param.default;
-		}
+            for (option in opts){
-        for (param in plugins_params[plugin]){
+                isselected = "";
-          if (param || plugins_params[plugin][param].length > 1){
+                if (defaultopt != undefined && opts[option] == defaultopt ){
-            html+= "<label> Parameter "+param+"</label>"
+                    isselected = ' selected="selected"'
-            html+= "<select id=\""+param+"\" name=\""+param+"\">"
-            for (option in plugins_params[plugin][param]){
-                  html+="<option value \""+plugins_params[plugin][param][option]+"\">"+plugins_params[plugin][param][option]+"</option>"
                 }
+                html+="<option value=\""+opts[option]+'"' + isselected +
+                '>'+opts[option]+"</option>"
             }
             html+="</select>"
         }
-        document.getElementById("params").innerHTML = html
+        else {
-};
+            default_value = "";
+            if(param.default != undefined){
+                default_value = param.default;
+            };
+            html +='<input class="col-sm-8" id="'+pname+'" name="'+pname+'" value="' + default_value + '"></input>';
+        }
+        html+='</div>';
+        html+='<div class="row">';
+        if ('description' in param){
+            html += '<p class="form-text sm-sm-12 text-muted text-center">' + param.description + '</p>';
+
+        }
+        html+='</div>';
+        html+='</div>';
+    }
+    html+='</div>';
+    return html;
+}
+
+function _get_form_parameters(id){
+    var element = document.getElementById(id);
+    params = {};
+    var selects = element.getElementsByTagName('select');
+    var inputs = element.getElementsByTagName('input');
+
+    Array.prototype.forEach.call(selects, function (sel) {
+        key = sel.name;
+        value = sel.options[sel.selectedIndex].value
+        params[key] = value;
+    });
+
+    Array.prototype.forEach.call(inputs, function (el) {
+        params[el.name] = el.value;
+    });
+
+    for (k in params){
+        value = params[k];
+        if (value == "" || value === "undefined"){
+            delete params[k];
+
+        }
+    }
+
+    return params;
+}
+
+function get_form_parameters(){
+    var p1 = _get_form_parameters("basic_params");
+    var p2 = _get_form_parameters("extra_params");
+    return Object.assign(p1, p2);
+}
+
+function add_param(key, value){
+    value = encodeURIComponent(value);
+    return "&"+key+"="+value;
+}
+
 
 function load_JSON(){
     url = "/api";
@@ -138,29 +268,20 @@ function load_JSON(){
     var rawcontainer = document.getElementById("jsonraw");
     rawcontainer.innerHTML = '';
     container.innerHTML = '';
-      var plugin = document.getElementById("plugins").options[document.getElementById("plugins").selectedIndex].value;
+
+    var plugin = document.getElementsByClassName("plugin")[0].options[document.getElementsByClassName("plugin")[0].selectedIndex].value;
+
     var input = encodeURIComponent(document.getElementById("input").value);
     url += "?algo="+plugin+"&i="+input
-      for (param in plugins_params[plugin]){
-        if (param != null){
-            var param_value = encodeURIComponent(document.getElementById(param).options[document.getElementById(param).selectedIndex].text);
-            if (param_value){
-              url+="&"+param+"="+param_value
-            }
-        }
-      }
 
-      for (param in default_params){
+    params = get_form_parameters();
-        if ((param != null) && (default_params[param]['options']) && (['help','conversion'].indexOf(param) < 0)){
+
-            var param_value = encodeURIComponent(document.getElementById(param).options[document.getElementById(param).selectedIndex].text);
+    for (key in params){
-            if (param_value){
+        url += add_param(key, params[key]);
-              url+="&"+param+"="+param_value
-            }
-        }
     }
 
     var response =  $.ajax({type: "GET", url: url , async: false}).responseText;
-  rawcontainer.innerHTML = replaceURLWithHTMLLinks(response)
+    rawcontainer.innerHTML = replaceURLWithHTMLLinks(response);
 
     document.getElementById("input_request").innerHTML = "<a href='"+url+"'>"+url+"</a>"
     document.getElementById("results-div").style.display = 'block';
@@ -171,12 +292,95 @@ function load_JSON(){
         };
         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?)");
+        $('#results-div a[href="#raw"]').click();
+        // location.hash = 'raw';
     }
+}
 
+function get_datasets_from_checkbox(){
+    var checks = document.getElementsByClassName("checks-datasets");
 
+    datasets = "";
+    for (var i = 0; i < checks.length; i++){
+        if (checks[i].checked){
+            datasets += checks[i].value + ",";
+        }
+    }
+    datasets = datasets.slice(0, -1);
 }
 
 
+function create_body_metrics(evaluations){
+	var new_tbody = document.createElement('tbody')
+	var metric_html = ""
+    for (var eval in evaluations){
+        metric_html += "<tr><th>"+evaluations[eval].evaluates+"</th><th>"+evaluations[eval].evaluatesOn+"</th>";
+        for (var metric in evaluations[eval].metrics){
+            metric_html +=  "<th>"+parseFloat(evaluations[eval].metrics[metric].value.toFixed(4))+"</th>";
+        }
+        metric_html += "</tr>";
+    }
+    new_tbody.innerHTML = metric_html
+    return new_tbody
+}
+
+function evaluate_JSON(){
+
+    url = "/api/evaluate";
+
+    var container = document.getElementById('results_eval');
+    var rawcontainer = document.getElementById('jsonraw_eval');
+    var table = document.getElementById("eval_table");
+
+    rawcontainer.innerHTML = "";
+    container.innerHTML = "";
+
+    var plugin = document.getElementsByClassName("plugin")[0].options[document.getElementsByClassName("plugin")[0].selectedIndex].value;
+
+    get_datasets_from_checkbox();
+
+    url += "?algo="+plugin+"&dataset="+datasets
+
+    var response =  $.ajax({type: "GET", url: url , async: false, dataType: 'json'}).responseText;
+    rawcontainer.innerHTML = replaceURLWithHTMLLinks(response);
+
+    document.getElementById("input_request_eval").innerHTML = "<a href='"+url+"'>"+url+"</a>"
+    document.getElementById("evaluate-div").style.display = 'block';
+    
+    try {
+        response = JSON.parse(response);
+        var options = {
+            mode: 'view'
+        };
+
+        //Control the single response results
+        if (!(Array.isArray(response.evaluations))){
+            response.evaluations = [response.evaluations]
+        }
+
+        new_tbody = create_body_metrics(response.evaluations)
+        table.replaceChild(new_tbody, table.lastElementChild)
+
+        var editor = new JSONEditor(container, options, response);
+        editor.expandAll();
+        // $('#results-div a[href="#viewer"]').tab('show');
+        $('#evaluate-div a[href="#evaluate-table"]').click();
+        // location.hash = 'raw';
+        
+        
+    }
+    catch(err){
+        console.log("Error decoding JSON (got turtle?)");
+        $('#evaluate-div a[href="#evaluate-raw"]').click();
+        // location.hash = 'raw';
+    }
+
+    
+
+}

senpy/templates/index.html

@@ -32,6 +32,8 @@
       <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>
+        <li role="presentation"><a class="active" href="#evaluate">Evaluate Plugins</a></li>
+
       </ul>
 
       <div class="tab-content">
@@ -54,6 +56,7 @@
                     <ul>
                       <li>List all available plugins: <a href="/api/plugins">/api/plugins</a></li>
                       <li>Get information about the default plugin: <a href="/api/plugins/default">/api/plugins/default</a></li>
+                      <li>List all available datasets: <a href="/api/datasets">/api/datasets</a></li>
                       <li>Download the JSON-LD context used: <a href="/api/contexts/Results.jsonld">/api/contexts/Results.jsonld</a></li>
                     </ul>
 
@@ -88,28 +91,52 @@
 
         <div class="tab-pane active" id="test">
           <div class="well">
-            <form id="form" onsubmit="return getPlugins();" accept-charset="utf-8">
+            <form id="form" class="container" onsubmit="return getPlugins();" accept-charset="utf-8">
-              <div id="inputswrapper">
                 <div><textarea id="input" class="boxsizingBorder" rows="5" name="i">This text makes me sad.
 whilst this text makes me happy and surprised at the same time.
-I cannot believe it!</textarea></div>
+I cannot believe it!</textarea>
-                <label>Select the plugin:</label>
-                <select id="plugins" name="plugins" onchange="change_params()">
-                </select>
-                        </br>
-                        <div id ="params">
                 </div>
-                </br>
+                <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">
+                                    Basic API parameters
+                                </h4>
+                            </div>
+                        </a>
+                        <div id="basic_params" class="panel-collapse collapse panel-body">
+                        </div>
+                    </div>
+                    <div class="panel panel-default">
+                        <a data-toggle="collapse" class="deco-none" href="#extra_params">
+                            <div class="panel-heading">
+                                <h4 class="panel-title">
+                                    Plugin extra parameters
+                                </h4>
+                            </div>
+                        </a>
+                        <div id="extra_params" class="panel-collapse collapse in panel-body">
+                        </div>
+                    </div>
+                </div> 
+                <!-- END PARAMETERS -->
+ 
                 <a id="preview" class="btn btn-lg btn-primary" onclick="load_JSON()">Analyse!</a>
                 <!--<button id="visualise" name="type" type="button">Visualise!</button>-->
-              </div>
             </form>
           </div>
             <span id="input_request"></span>
             <div id="results-div">
               <ul class="nav nav-tabs" role="tablist">
-                <li role="presentation" class="active"><a class="active" href="#viewer">Viewer</a></li>
+                <li role="presentation" class="active"><a data-toggle="tab" class="active" href="#viewer">Viewer</a></li>
-                <li role="presentation"><a class="active" href="#raw">Raw</a></li>
+                <li role="presentation"><a data-toggle="tab" class="active" href="#raw">Raw</a></li>
               </ul>
               <div class="tab-content" id="results-container">
                 
@@ -127,6 +154,70 @@ I cannot believe it!</textarea></div>
               </div>
             </div>
         </div>
+
+        <div class="tab-pane" id="evaluate">
+          <div class="well">
+            <form id="form" class="container" onsubmit="return getPlugins();" accept-charset="utf-8">
+                <div>
+                    <label>Select the plugin:</label>
+                    <select id="plugins-eval" name="plugins-eval" class=plugin onchange="draw_extra_parameters()">
+                    </select>
+                </div>
+                <div>
+                    <label>Select the datasets:</label>
+                    <div id="datasets" name="datasets" >
+                    </select>
+                </div>
+ 
+                <a id="preview" class="btn btn-lg btn-primary" onclick="evaluate_JSON()">Evaluate Plugin!</a>
+                <!--<button id="visualise" name="type" type="button">Visualise!</button>-->
+            </form>
+          </div>
+            <span id="input_request_eval"></span>
+            <div id="evaluate-div">
+              <ul class="nav nav-tabs" role="tablist">
+                <li role="presentation" class="active"><a data-toggle="tab" class="active" href="#evaluate-viewer">Viewer</a></li>
+                <li role="presentation"><a data-toggle="tab" class="active" href="#evaluate-raw">Raw</a></li>
+                <li role="presentation"><a data-toggle="tab" class="active" href="#evaluate-table">Table</a></li>
+              </ul>
+              <div class="tab-content" id="evaluate-container">
+                
+                <div class="tab-pane active" id="evaluate-viewer">
+                  <div id="content">
+                    <pre id="results_eval" class="results_eval"></pre>
+                  </div>
+                </div>
+                
+                <div class="tab-pane"  id="evaluate-raw">
+                  <div id="content">
+                    <pre id="jsonraw_eval" class="results_eval"></pre>
+                  </div>
+                </div>
+                <div class="tab-pane"  id="evaluate-table">
+                    <table id="eval_table" class="table table-condensed">
+                      <thead>
+                        <tr>
+                        <th>Plugin</th>
+                        <th>Dataset</th>
+                        <th>Accuracy</th>
+                        <th>Precision_macro</th>
+                        <th>Recall_macro</th>
+                        <th>F1_macro</th>
+                        <th>F1_weighted</th>
+                        <th>F1_micro</th>
+                        <th>F1</th>
+                        </tr>
+                      </thead>
+                      <tbody>
+                      </tbody>
+                    </table>
+                  </div>
+                </div>
+                </div>
+              </div>
+            </div>
+        </div>
+
         <a href="http://www.gsi.dit.upm.es" target="_blank"><img class="center-block" src="static/img/gsi.png"/>  </a> 
 
       </div>

senpy/test.py

@@ -0,0 +1,36 @@
+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
+
+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)
+    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

senpy/utils.py

@@ -1,25 +1,101 @@
-from . import models
+from . import models, __version__
+from collections import MutableMapping
+import pprint
+
+import logging
+logger = logging.getLogger(__name__)
+
+# MutableMapping should be enough, but it causes problems with py2
+DICTCLASSES = (MutableMapping, dict, models.BaseModel)
 
 
 def check_template(indict, template):
-    if isinstance(template, dict) and isinstance(indict, dict):
+    if isinstance(template, DICTCLASSES) and isinstance(indict, DICTCLASSES):
         for k, v in template.items():
             if k not in indict:
-                return '{} not in {}'.format(k, indict)
+                raise models.Error('{} not in {}'.format(k, indict))
             check_template(indict[k], v)
     elif isinstance(template, list) and isinstance(indict, list):
-        if len(indict) != len(template):
-            raise models.Error('Different size for {} and {}'.format(indict, template))
         for e in template:
-            found = False
             for i in indict:
                 try:
                     check_template(i, e)
-                    found = True
+                    break
                 except models.Error as ex:
+                    # raise
                     continue
-            if not found:
+            else:
-                raise models.Error('{} not found in {}'.format(e, indict))
+                raise models.Error(('Element not found.'
+                                   '\nExpected: {}\nIn: {}').format(pprint.pformat(e),
+                                                                    pprint.pformat(indict)))
     else:
         if indict != template:
-            raise models.Error('{} and {} are different'.format(indict, template))
+            raise models.Error(('Differences found.\n'
+                                '\tExpected: {}\n'
+                                '\tFound: {}').format(pprint.pformat(indict),
+                                                      pprint.pformat(template)))
+
+
+def convert_dictionary(original, mappings):
+    result = {}
+    for key, value in original.items():
+        if key in mappings:
+            key = mappings[key]
+        result[key] = value
+    return result
+
+
+def easy_load(app=None, plugin_list=None, plugin_folder=None, **kwargs):
+    '''
+    Run a server with a specific plugin.
+    '''
+
+    from flask import Flask
+    from .extensions import Senpy
+
+    if not app:
+        app = Flask(__name__)
+    sp = Senpy(app, plugin_folder=plugin_folder, **kwargs)
+    if not plugin_list:
+        from . import plugins
+        import __main__
+        plugin_list = plugins.from_module(__main__)
+    for plugin in plugin_list:
+        sp.add_plugin(plugin)
+    sp.install_deps()
+    sp.activate_all()
+    return sp, app
+
+
+def easy_test(plugin_list=None):
+    logger.setLevel(logging.DEBUG)
+    logging.getLogger().setLevel(logging.INFO)
+    if not plugin_list:
+        import __main__
+        logger.info('Loading classes from {}'.format(__main__))
+        from . import plugins
+        plugin_list = plugins.from_module(__main__)
+    for plug in plugin_list:
+        plug.test()
+        logger.info('The tests for {} passed!'.format(plug.name))
+    logger.info('All tests passed!')
+
+
+def easy(host='0.0.0.0', port=5000, debug=True, **kwargs):
+    '''
+    Run a server with a specific plugin.
+    '''
+    logging.getLogger().setLevel(logging.DEBUG)
+    logging.getLogger('senpy').setLevel(logging.INFO)
+    sp, app = easy_load(**kwargs)
+    easy_test(sp.plugins())
+    app.debug = debug
+    import time
+    logger.info(time.time())
+    logger.info('Senpy version {}'.format(__version__))
+    logger.info('Server running on port %s:%d. Ctrl+C to quit' % (host,
+                                                                  port))
+    app.debug = debug
+    app.run(host,
+            port,
+            debug=app.debug)

setup.py

@@ -1,20 +1,20 @@
-import pip
 from setuptools import setup
-# parse_requirements() returns generator of pip.req.InstallRequirement objects
-from pip.req import parse_requirements
-from senpy import __version__
 
-try:
+with open('senpy/VERSION') as f:
-    install_reqs = parse_requirements(
+    __version__ = f.read().strip()
-        "requirements.txt", session=pip.download.PipSession())
+    assert __version__
-    test_reqs = parse_requirements(
-        "test-requirements.txt", session=pip.download.PipSession())
-except AttributeError:
-    install_reqs = parse_requirements("requirements.txt")
-    test_reqs = parse_requirements("test-requirements.txt")
 
-install_reqs = [str(ir.req) for ir in install_reqs]
+
-test_reqs = [str(ir.req) for ir in test_reqs]
+def parse_requirements(filename):
+    """ load requirements from a pip requirements file """
+    with open(filename, 'r') as f:
+        lineiter = list(line.strip() for line in f)
+    return [line for line in lineiter if line and not line.startswith("#")]
+
+
+install_reqs = parse_requirements("requirements.txt")
+test_reqs = parse_requirements("test-requirements.txt")
+extra_reqs = parse_requirements("extra-requirements.txt")
 
 
 setup(
@@ -34,6 +34,9 @@ setup(
     install_requires=install_reqs,
     tests_require=test_reqs,
     setup_requires=['pytest-runner', ],
+    extras_require={
+        'evaluation': extra_reqs
+    },
     include_package_data=True,
     entry_points={
         'console_scripts':

test-requirements.txt

@@ -1,3 +1,5 @@
 mock
 pytest-cov
 pytest
+scikit-learn
+numpy

tests/plugins/async_plugin/asyncplugin.senpy

@@ -1,8 +0,0 @@
----
-name: Async
-module: asyncplugin
-description: I am async
-author: "@balkian"
-version: '0.1'
-async: true
-extra_params: {}

tests/plugins/dummy_plugin/dummy.py

@@ -1,11 +0,0 @@
-from senpy.plugins import SentimentPlugin
-
-
-class DummyPlugin(SentimentPlugin):
-    def analyse_entry(self, entry, params):
-        entry['nif:iString'] = entry['nif:isString'][::-1]
-        entry.reversed = entry.get('reversed', 0) + 1
-        yield entry
-
-    def test(self):
-        pass

tests/plugins/dummy_plugin/dummy.senpy

@@ -1,15 +0,0 @@
-{
-    "name": "Dummy",
-    "module": "dummy",
-    "description": "I am dummy",
-    "author": "@balkian",
-    "version": "0.1",
-    "extra_params": {
-        "example": {
-            "@id": "example_parameter",
-            "aliases": ["example", "ex"],
-            "required": false,
-            "default": 0
-        }
-    }
-}

tests/plugins/dummy_plugin/dummy_required.senpy

@@ -1,14 +0,0 @@
-{
-    "name": "DummyRequired",
-    "module": "dummy",
-    "description": "I am dummy",
-    "author": "@balkian",
-    "version": "0.1",
-    "extra_params": {
-        "example": {
-            "@id": "example_parameter",
-            "aliases": ["example", "ex"],
-            "required": true
-        }
-    }
-}

tests/plugins/sleep_plugin/sleep.py

@@ -1,14 +0,0 @@
-from senpy.plugins import AnalysisPlugin
-from time import sleep
-
-
-class SleepPlugin(AnalysisPlugin):
-    def activate(self, *args, **kwargs):
-        sleep(self.timeout)
-
-    def analyse_entry(self, entry, params):
-        sleep(float(params.get("timeout", self.timeout)))
-        yield entry
-
-    def test(self):
-        pass

tests/plugins/sleep_plugin/sleep.senpy

@@ -1,16 +0,0 @@
-{
-    "name": "Sleep",
-    "module": "sleep",
-    "description": "I am dummy",
-    "author": "@balkian",
-    "version": "0.1",
-    "timeout": 0.05,
-    "extra_params": {
-        "timeout": {
-            "@id": "timeout_sleep",
-            "aliases": ["timeout", "to"],
-            "required": false,
-            "default": 0
-        }
-     }
-}

tests/test_api.py

@@ -19,11 +19,8 @@ class APITest(TestCase):
 
     def test_basic(self):
         a = {}
-        try:
+        self.assertRaises(Error, parse_params, a)
-            parse_params(a, NIF_PARAMS)
+        self.assertRaises(Error, parse_params, a, NIF_PARAMS)
-            raise AssertionError()
-        except Error:
-            pass
         a = {'input': 'hello'}
         p = parse_params(a, NIF_PARAMS)
         assert 'input' in p
@@ -39,11 +36,7 @@ class APITest(TestCase):
                 'required': True
             }
         }
-        try:
+        self.assertRaises(Error, parse_params, plug_params)
-            parse_params(query, plug_params)
-            raise AssertionError()
-        except Error:
-            pass
         query['hello'] = 'world'
         p = parse_params(query, plug_params)
         assert 'hello' in p
@@ -53,7 +46,6 @@ class APITest(TestCase):
         query['hiya'] = 'dlrow'
         p = parse_params(query, plug_params)
         assert 'hello' in p
-        assert 'hiya' in p
         assert p['hello'] == 'dlrow'
 
     def test_default(self):

tests/test_blueprints.py

@@ -23,7 +23,7 @@ class BlueprintsTest(TestCase):
         cls.app = Flask("test_extensions")
         cls.app.debug = False
         cls.client = cls.app.test_client()
-        cls.senpy = Senpy()
+        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)
@@ -34,11 +34,14 @@ class BlueprintsTest(TestCase):
     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()
+
     def test_home(self):
         """
         Calling with no arguments should ask the user for more arguments
         """
-        self.app.debug = False
         resp = self.client.get("/api/")
         self.assertCode(resp, 400)
         js = parse_resp(resp)
@@ -84,6 +87,10 @@ class BlueprintsTest(TestCase):
         js = parse_resp(resp)
         logging.debug("Got response: %s", js)
         assert isinstance(js, models.Error)
+        resp = self.client.get("/api/?i=My aloha mohame&algo=DummyRequired&example=notvalid")
+        self.assertCode(resp, 400)
+        resp = self.client.get("/api/?i=My aloha mohame&algo=DummyRequired&example=a")
+        self.assertCode(resp, 200)
 
     def test_error(self):
         """
@@ -155,12 +162,15 @@ class BlueprintsTest(TestCase):
     def test_schema(self):
         resp = self.client.get("/api/schemas/definitions.json")
         self.assertCode(resp, 200)
-        js = parse_resp(resp)
+        assert "$schema" in resp.data.decode()
-        assert "$schema" in js
 
     def test_help(self):
         resp = self.client.get("/api/?help=true")
         self.assertCode(resp, 200)
         js = parse_resp(resp)
-        assert "parameters" in js
+        assert "valid_parameters" in js
-        assert "help" in js["parameters"]
+        assert "help" in js["valid_parameters"]
+
+    def test_conversion(self):
+        resp = self.client.get("/api/?input=hello&algo=emoRand&emotionModel=DOES NOT EXIST")
+        self.assertCode(resp, 404)

tests/test_cli.py

@@ -12,7 +12,8 @@ class CLITest(TestCase):
     def test_basic(self):
         self.assertRaises(Error, partial(main_function, []))
 
-        res = main_function(['--input', 'test', '--algo', 'rand', '--with-parameters'])
+        res = main_function(['--input', 'test', '--algo', 'rand',
+                             '--with-parameters', '--default-plugins'])
         assert res.parameters['input'] == 'test'
         assert 'rand' in res.parameters['algorithm']
         assert res.parameters['input'] == 'test'

tests/test_client.py

@@ -1,24 +1,9 @@
 from unittest import TestCase
-try:
-    from unittest.mock import patch
-except ImportError:
-    from mock import patch
-
-import json
 
+from senpy.test import patch_requests
 from senpy.client import Client
 from senpy.models import Results, Plugins, Error
-from senpy.plugins import AnalysisPlugin, default_plugin_type
+from senpy.plugins import AnalysisPlugin
-
-
-class Call(dict):
-    def __init__(self, obj):
-        self.obj = obj.serialize()
-        self.status_code = 200
-        self.content = self.json()
-
-    def json(self):
-        return json.loads(self.obj)
 
 
 class ModelsTest(TestCase):
@@ -29,20 +14,18 @@ class ModelsTest(TestCase):
     def test_client(self):
         endpoint = 'http://dummy/'
         client = Client(endpoint)
-        success = Call(Results())
+        with patch_requests(Results()) as (request, response):
-        with patch('requests.request', return_value=success) as patched:
             resp = client.analyse('hello')
             assert isinstance(resp, Results)
-        patched.assert_called_with(
+        request.assert_called_with(
             url=endpoint + '/', method='GET', params={'input': 'hello'})
-        error = Call(Error('Nothing'))
+        with patch_requests(Error('Nothing')) as (request, response):
-        with patch('requests.request', return_value=error) as patched:
             try:
                 client.analyse(input='hello', algorithm='NONEXISTENT')
                 raise Exception('Exceptions should be raised. This is not golang')
             except Error:
                 pass
-        patched.assert_called_with(
+        request.assert_called_with(
             url=endpoint + '/',
             method='GET',
             params={'input': 'hello',
@@ -54,12 +37,11 @@ class ModelsTest(TestCase):
         plugins = Plugins()
         p1 = AnalysisPlugin({'name': 'AnalysisP1', 'version': 0, 'description': 'No'})
         plugins.plugins = [p1, ]
-        success = Call(plugins)
+        with patch_requests(plugins) as (request, response):
-        with patch('requests.request', return_value=success) as patched:
             response = client.plugins()
             assert isinstance(response, dict)
             assert len(response) == 1
             assert "AnalysisP1" in response
-        patched.assert_called_with(
+        request.assert_called_with(
             url=endpoint + '/plugins', method='GET',
-            params={'plugin_type': default_plugin_type})
+            params={})

tests/test_extensions.py

@@ -25,10 +25,11 @@ def analyse(instance, **kwargs):
 class ExtensionsTest(TestCase):
     def setUp(self):
         self.app = Flask('test_extensions')
-        self.dir = os.path.dirname(__file__)
+        self.examples_dir = os.path.join(os.path.dirname(__file__), '..', 'example-plugins')
-        self.senpy = Senpy(plugin_folder=self.dir,
+        self.senpy = Senpy(plugin_folder=self.examples_dir,
                            app=self.app,
                            default_plugins=False)
+        self.senpy.deactivate_all()
         self.senpy.activate_plugin("Dummy", sync=True)
 
     def test_init(self):
@@ -41,22 +42,37 @@ class ExtensionsTest(TestCase):
     def test_discovery(self):
         """ Discovery of plugins in given folders.  """
         # noinspection PyProtectedMember
-        assert self.dir in self.senpy._search_folders
+        print(self.senpy.plugins())
-        print(self.senpy.plugins)
+        assert self.senpy.get_plugin("dummy")
-        assert "Dummy" in self.senpy.plugins
+
+    def test_add_delete(self):
+        '''Should be able to add and delete new plugins. '''
+        new = plugins.Plugin(name='new', description='new', version=0)
+        self.senpy.add_plugin(new)
+        assert new in self.senpy.plugins()
+        self.senpy.delete_plugin(new)
+        assert new not in self.senpy.plugins()
+
+    def test_adding_folder(self):
+        """ It should be possible for senpy to look for plugins in more folders. """
+        senpy = Senpy(plugin_folder=None,
+                      app=self.app,
+                      default_plugins=False)
+        assert not senpy.analysis_plugins
+        senpy.add_folder(self.examples_dir)
+        assert senpy.analysis_plugins
+        self.assertRaises(AttributeError, senpy.add_folder, 'DOES NOT EXIST')
 
     def test_installing(self):
         """ Installing a plugin """
         info = {
             'name': 'TestPip',
-            'module': 'dummy',
+            'module': 'mynoop',
             'description': None,
             'requirements': ['noop'],
             'version': 0
         }
-        root = os.path.join(self.dir, 'plugins', 'dummy_plugin')
+        module = plugins.from_info(info, root=self.examples_dir, install=True)
-        module = plugins.load_plugin_from_info(info, root=root)
-        plugins.install_deps(info)
         assert module.name == 'TestPip'
         assert module
         import noop
@@ -65,8 +81,8 @@ class ExtensionsTest(TestCase):
     def test_enabling(self):
         """ Enabling a plugin """
         self.senpy.activate_all(sync=True)
-        assert len(self.senpy.plugins) >= 3
+        assert len(self.senpy.plugins()) >= 3
-        assert self.senpy.plugins["Sleep"].is_activated
+        assert self.senpy.get_plugin("Sleep").is_activated
 
     def test_installing_nonexistent(self):
         """ Fail if the dependencies cannot be met """
@@ -83,8 +99,8 @@ class ExtensionsTest(TestCase):
     def test_disabling(self):
         """ Disabling a plugin """
         self.senpy.deactivate_all(sync=True)
-        assert not self.senpy.plugins["Dummy"].is_activated
+        assert not self.senpy.get_plugin("dummy").is_activated
-        assert not self.senpy.plugins["Sleep"].is_activated
+        assert not self.senpy.get_plugin("sleep").is_activated
 
     def test_default(self):
         """ Default plugin should be set """
@@ -107,7 +123,18 @@ 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.entries[0]['nif:iString'] == 'input'
+        assert r1.entries[0]['nif:isString'] == 'input'
+
+    def test_analyse_empty(self):
+        """ Trying to analyse when no plugins are installed should raise an error."""
+        senpy = Senpy(plugin_folder=None,
+                      app=self.app,
+                      default_plugins=False)
+        self.assertRaises(Error, senpy.analyse, Results())
+
+    def test_analyse_wrong(self):
+        """ Trying to analyse with a non-existent plugin should raise an error."""
+        self.assertRaises(Error, analyse, self.senpy, algorithm='DOES NOT EXIST', input='test')
 
     def test_analyse_jsonld(self):
         """ Using a plugin with JSON-LD input"""
@@ -131,14 +158,15 @@ class ExtensionsTest(TestCase):
                      output="tuptuo")
         assert r1.analysis[0] == "plugins/Dummy_0.1"
         assert r2.analysis[0] == "plugins/Dummy_0.1"
-        assert r1.entries[0]['nif:iString'] == 'input'
+        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.plugins['MOCK'] = mm
+        self.senpy.add_plugin(mm)
         try:
             analyse(self.senpy, input='nothing', algorithm='MOCK')
             assert False
@@ -146,29 +174,28 @@ class ExtensionsTest(TestCase):
             assert 'error in analysis' in ex['message']
             assert ex['status'] == 500
 
-        mm.analyse.side_effect = Exception('generic exception on analysis')
+        ex = Exception('generic exception on analysis')
-        mm.analyse_entries.side_effect = Exception(
+        mm.analyse.side_effect = ex
-            'generic exception on analysis')
+        mm.analyse_entries.side_effect = ex
 
         try:
             analyse(self.senpy, input='nothing', algorithm='MOCK')
             assert False
-        except Error as ex:
+        except Exception as ex:
-            assert 'generic exception on analysis' in ex['message']
+            assert 'generic exception on analysis' in str(ex)
-            assert ex['status'] == 500
 
     def test_filtering(self):
         """ Filtering plugins """
-        assert len(self.senpy.filter_plugins(name="Dummy")) > 0
+        assert len(self.senpy.plugins(name="Dummy")) > 0
-        assert not len(self.senpy.filter_plugins(name="notdummy"))
+        assert not len(self.senpy.plugins(name="NotDummy"))
-        assert self.senpy.filter_plugins(name="Dummy", is_activated=True)
+        assert self.senpy.plugins(name="Dummy", is_activated=True)
         self.senpy.deactivate_plugin("Dummy", sync=True)
-        assert not len(
+        assert not len(self.senpy.plugins(name="Dummy",
-            self.senpy.filter_plugins(name="Dummy", is_activated=True))
+                                          is_activated=True))
 
     def test_load_default_plugins(self):
-        senpy = Senpy(plugin_folder=self.dir, default_plugins=True)
+        senpy = Senpy(plugin_folder=self.examples_dir, default_plugins=True)
-        assert len(senpy.plugins) > 1
+        assert len(senpy.plugins()) > 1
 
     def test_convert_emotions(self):
         self.senpy.activate_all(sync=True)
@@ -186,7 +213,7 @@ class ExtensionsTest(TestCase):
         response = Results({
             'analysis': [{'plugin': plugin}],
             'entries': [Entry({
-                'nif:iString': 'much ado about nothing',
+                'nif:isString': 'much ado about nothing',
                 'emotions': [eSet1]
             })]
         })

tests/test_models.py

@@ -9,13 +9,15 @@ from senpy.models import (Emotion,
                           EmotionAnalysis,
                           EmotionSet,
                           Entry,
+                          Entity,
                           Error,
                           Results,
                           Sentiment,
+                          SentimentPlugin,
                           Plugins,
-                          Plugin,
                           from_string,
-                          from_dict)
+                          from_dict,
+                          subtypes)
 from senpy import plugins
 from pprint import pprint
 
@@ -99,7 +101,8 @@ class ModelsTest(TestCase):
 
     def test_plugins(self):
         self.assertRaises(Error, plugins.Plugin)
-        p = plugins.Plugin({"name": "dummy",
+        p = plugins.SentimentPlugin({"name": "dummy",
+                                     "description": "I do nothing",
                                      "version": 0,
                                      "extra_params": {
                                          "none": {
@@ -110,7 +113,7 @@ class ModelsTest(TestCase):
                                      }})
         c = p.jsonld()
         assert '@type' in c
-        assert c['@type'] == 'plugin'
+        assert c['@type'] == 'sentimentPlugin'
         assert 'info' not in c
         assert 'repo' not in c
         assert 'extra_params' in c
@@ -123,7 +126,9 @@ class ModelsTest(TestCase):
     def test_str(self):
         """The string representation shouldn't include private variables"""
         r = Results()
-        p = plugins.Plugin({"name": "STR test", "version": 0})
+        p = plugins.Plugin({"name": "STR test",
+                            "description": "Test of private variables.",
+                            "version": 0})
         p._testing = 0
         s = str(p)
         assert "_testing" not in s
@@ -131,6 +136,11 @@ class ModelsTest(TestCase):
         s = str(r)
         assert "_testing" not in s
 
+    def test_serialize(self):
+        for k, v in subtypes().items():
+            e = v()
+            e.serialize()
+
     def test_turtle(self):
         """Any model should be serializable as a turtle file"""
         ana = EmotionAnalysis()
@@ -170,13 +180,14 @@ class ModelsTest(TestCase):
     def test_single_plugin(self):
         """A response with a single plugin should still return a list"""
         plugs = Plugins()
-        p = Plugin({'id': str(1),
+        p = SentimentPlugin({'id': str(1),
                              'version': 0,
                              'description': 'dummy'})
         plugs.plugins.append(p)
         assert isinstance(plugs.plugins, list)
         js = plugs.jsonld()
         assert isinstance(js['plugins'], list)
+        assert js['plugins'][0]['@type'] == 'sentimentPlugin'
 
     def test_from_string(self):
         results = {
@@ -189,6 +200,7 @@ class ModelsTest(TestCase):
             }]
         }
         recovered = from_dict(results)
+        assert recovered.id == results['@id']
         assert isinstance(recovered, Results)
         assert isinstance(recovered.entries[0], Entry)
 
@@ -196,3 +208,14 @@ class ModelsTest(TestCase):
         recovered = from_string(string)
         assert isinstance(recovered, Results)
         assert isinstance(recovered.entries[0], Entry)
+
+    def test_serializable(self):
+        r = Results()
+        e = Entry()
+        ent = Entity()
+        e.entities.append(ent)
+        r.entries.append(e)
+        d = r.serializable()
+        assert d
+        assert d['entries']
+        assert d['entries'][0]['entities']

tests/test_plugins.py

@@ -1,17 +1,25 @@
 #!/bin/env python
 
 import os
+import sys
 import pickle
 import shutil
 import tempfile
 
-from unittest import TestCase
+from unittest import TestCase, skipIf
-from senpy.models import Results, Entry, EmotionSet, Emotion
+from senpy.models import Results, Entry, EmotionSet, Emotion, Plugins
 from senpy import plugins
 from senpy.plugins.conversion.emotion.centroids import CentroidConversion
 
+import pandas as pd
+
 
 class ShelfDummyPlugin(plugins.SentimentPlugin, plugins.ShelfMixin):
+    '''Dummy plugin for tests.'''
+    name = 'Shelf'
+    version = 0
+    author = 'the senpy community'
+
     def activate(self, *args, **kwargs):
         if 'counter' not in self.sh:
             self.sh['counter'] = 0
@@ -40,19 +48,50 @@ class PluginsTest(TestCase):
         self.shelf_dir = tempfile.mkdtemp()
         self.shelf_file = os.path.join(self.shelf_dir, "shelf")
 
+    def test_serialize(self):
+        '''A plugin should be serializable and de-serializable'''
+        dummy = ShelfDummyPlugin()
+        dummy.serialize()
+
+    def test_jsonld(self):
+        '''A plugin should be serializable and de-serializable'''
+        dummy = ShelfDummyPlugin()
+        dummy.jsonld()
+
     def test_shelf_file(self):
         a = ShelfDummyPlugin(
             info={'name': 'default_shelve_file',
+                  'description': 'Dummy plugin for tests',
                   'version': 'test'})
         a.activate()
         assert os.path.isfile(a.shelf_file)
         os.remove(a.shelf_file)
 
+    def test_plugin_filter(self):
+        ps = Plugins()
+        for i in (plugins.SentimentPlugin,
+                  plugins.EmotionPlugin,
+                  plugins.AnalysisPlugin):
+            p = i(name='Plugin_{}'.format(i.__name__),
+                  description='TEST',
+                  version=0,
+                  author='NOBODY')
+            ps.plugins.append(p)
+        assert len(ps.plugins) == 3
+        cases = [('AnalysisPlugin', 3),
+                 ('SentimentPlugin', 1),
+                 ('EmotionPlugin', 1)]
+
+        for name, num in cases:
+            res = list(plugins.pfilter(ps.plugins, plugin_type=name))
+            assert len(res) == num
+
     def test_shelf(self):
         ''' A shelf is created and the value is stored '''
         newfile = self.shelf_file + "new"
         a = ShelfDummyPlugin(info={
             'name': 'shelve',
+            'description': 'Shelf plugin for tests',
             'version': 'test',
             'shelf_file': newfile
         })
@@ -75,6 +114,7 @@ class PluginsTest(TestCase):
             pickle.dump({'counter': 99}, f)
         a = ShelfDummyPlugin(info={
             'name': 'DummyShelf',
+            'description': 'Dummy plugin for tests',
             'shelf_file': self.shelf_file,
             'version': 'test'
         })
@@ -105,7 +145,8 @@ class PluginsTest(TestCase):
             with open(fn, 'rb') as f:
                 msg, error = files[fn]
                 a = ShelfDummyPlugin(info={
-                    'name': 'shelve',
+                    'name': 'test_corrupt_shelf_{}'.format(msg),
+                    'description': 'Dummy plugin for tests',
                     'version': 'test',
                     'shelf_file': f.name
                 })
@@ -126,6 +167,7 @@ class PluginsTest(TestCase):
         ''' Reusing the values of a previous shelf '''
         a = ShelfDummyPlugin(info={
             'name': 'shelve',
+            'description': 'Dummy plugin for tests',
             'version': 'test',
             'shelf_file': self.shelf_file
         })
@@ -136,6 +178,7 @@ class PluginsTest(TestCase):
 
         b = ShelfDummyPlugin(info={
             'name': 'shelve',
+            'description': 'Dummy plugin for tests',
             'version': 'test',
             'shelf_file': self.shelf_file
         })
@@ -148,6 +191,7 @@ class PluginsTest(TestCase):
         ''' Should be able to set extra parameters'''
         a = ShelfDummyPlugin(info={
             'name': 'shelve',
+            'description': 'Dummy shelf plugin for tests',
             'version': 'test',
             'shelf_file': self.shelf_file,
             'extra_params': {
@@ -160,6 +204,61 @@ class PluginsTest(TestCase):
         })
         assert 'example' in a.extra_params
 
+    def test_box(self):
+
+        class MyBox(plugins.Box):
+            ''' Vague description'''
+
+            author = 'me'
+            version = 0
+
+            def input(self, entry, **kwargs):
+                return entry.text
+
+            def predict_one(self, input):
+                return 'SIGN' in input
+
+            def output(self, output, entry, **kwargs):
+                if output:
+                    entry.myAnnotation = 'DETECTED'
+                return entry
+
+            test_cases = [
+                {
+                    'input': "nothing here",
+                    'expected': {'myAnnotation': 'DETECTED'},
+                    'should_fail': True
+                }, {
+                    'input': "SIGN",
+                    'expected': {'myAnnotation': 'DETECTED'}
+                }]
+
+        MyBox().test()
+
+    def test_sentimentbox(self):
+
+        class SentimentBox(plugins.MappingMixin, plugins.SentimentBox):
+            ''' Vague description'''
+
+            author = 'me'
+            version = 0
+
+            mappings = {'happy': 'marl:Positive', 'sad': 'marl:Negative'}
+
+            def predict_one(self, input, **kwargs):
+                return 'happy' if ':)' in input else 'sad'
+
+            test_cases = [
+                {
+                    'input': 'a happy face :)',
+                    'polarity': 'marl:Positive'
+                }, {
+                    'input': "Nothing",
+                    'polarity': 'marl:Negative'
+                }]
+
+        SentimentBox().test()
+
     def test_conversion_centroids(self):
         info = {
             "name": "CentroidTest",
@@ -213,21 +312,57 @@ 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):
+        testdata = []
+        for i in range(50):
+            testdata.append(["good", 1])
+        for i in range(50):
+            testdata.append(["bad", 0])
+        dataset = pd.DataFrame(testdata, columns=['text', 'polarity'])
 
-def make_mini_test(plugin_info):
+        class DummyPlugin(plugins.TextBox):
+            description = 'Plugin to test evaluation'
+            version = 0
+
+            def predict_one(self, input):
+                return 0
+
+        class SmartPlugin(plugins.TextBox):
+            description = 'Plugin to test evaluation'
+            version = 0
+
+            def predict_one(self, input):
+                if input == 'good':
+                    return 1
+                return 0
+
+        dpipe = DummyPlugin()
+        results = plugins.evaluate(datasets={'testdata': dataset}, plugins=[dpipe], flatten=True)
+        dumb_metrics = results[0].metrics[0]
+        assert abs(dumb_metrics['accuracy'] - 0.5) < 0.01
+
+        spipe = SmartPlugin()
+        results = plugins.evaluate(datasets={'testdata': dataset}, plugins=[spipe], flatten=True)
+        smart_metrics = results[0].metrics[0]
+        assert abs(smart_metrics['accuracy'] - 1) < 0.01
+
+
+def make_mini_test(fpath):
     def mini_test(self):
-        plugin = plugins.load_plugin_from_info(plugin_info, install=True)
+        for plugin in plugins.from_path(fpath, install=True):
             plugin.test()
     return mini_test
 
 
 def _add_tests():
-    root = os.path.dirname(__file__)
+    root = os.path.join(os.path.dirname(__file__), '..')
-    plugs = plugins.load_plugins(os.path.join(root, ".."), loader=plugins.parse_plugin_info)
+    print(root)
-    for k, v in plugs.items():
+    for fpath in plugins.find_plugins([root, ]):
         pass
-        t_method = make_mini_test(v)
+        t_method = make_mini_test(fpath)
-        t_method.__name__ = 'test_plugin_{}'.format(k)
+        t_method.__name__ = 'test_plugin_{}'.format(fpath)
         setattr(PluginsTest, t_method.__name__, t_method)
         del t_method
 

tests/test_test.py

@@ -0,0 +1,32 @@
+from unittest import TestCase
+
+import requests
+import json
+from senpy.test import patch_requests
+from senpy.models import Results
+
+
+class TestTest(TestCase):
+    def test_patch_text(self):
+        with patch_requests('hello'):
+            r = requests.get('http://example.com')
+            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())
+            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)
+            js = res.json()
+            assert js
+            assert js['nothing'] == 'new'