Change job model.

app/services/forms.py

@@ -3,10 +3,50 @@ from wtforms import MultipleFileField, SelectField, StringField, SubmitField, Va
 from wtforms.validators import DataRequired, Length
 
 
+class NewNLPJobForm(FlaskForm):
+    description = StringField(
+        'Description',
+        validators=[DataRequired(), Length(1, 255)]
+    )
+    files = MultipleFileField('Files', validators=[DataRequired()])
+    language = SelectField(
+        'Language',
+        choices=[('', 'Choose your option'),
+                 ('en', 'English'),
+                 ('fr', 'French'),
+                 ('de', 'German'),
+                 ('it', 'Italian'),
+                 ('pt', 'Portuguese'),
+                 ('es', 'Spanish')
+                 ],
+        validators=[DataRequired()]
+    )
+    submit = SubmitField('Submit')
+    title = StringField(
+        'Title',
+        validators=[DataRequired(), Length(1, 32)]
+    )
+    version = SelectField(
+        'Version',
+        choices=[('', 'Choose your option'),
+                 ('latest', 'Latest'),
+                 ],
+        validators=[DataRequired()]
+    )
+
+    def validate_files(form, field):
+        for file in field.data:
+            if not file.filename.lower().endswith('.txt'):
+                raise ValidationError(
+                    'File does not have an approved extension: '
+                    '.txt'
+                )
+
+
 class NewOCRJobForm(FlaskForm):
     description = StringField(
         'Description',
-        validators=[DataRequired(), Length(1, 64)]
+        validators=[DataRequired(), Length(1, 255)]
     )
     files = MultipleFileField('Files', validators=[DataRequired()])
     language = SelectField(
@@ -44,43 +84,3 @@ class NewOCRJobForm(FlaskForm):
                     'File does not have an approved extension: '
                     '.pdf | .tif | .tiff'
                 )
-
-
-class NewNLPJobForm(FlaskForm):
-    description = StringField(
-        'Description',
-        validators=[DataRequired(), Length(1, 64)]
-    )
-    files = MultipleFileField('Files', validators=[DataRequired()])
-    language = SelectField(
-        'Language',
-        choices=[('', 'Choose your option'),
-                 ('en', 'English'),
-                 ('fr', 'French'),
-                 ('de', 'German'),
-                 ('it', 'Italian'),
-                 ('pt', 'Portuguese'),
-                 ('es', 'Spanish')
-                 ],
-        validators=[DataRequired()]
-    )
-    submit = SubmitField('Submit')
-    title = StringField(
-        'Title',
-        validators=[DataRequired(), Length(1, 32)]
-    )
-    version = SelectField(
-        'Version',
-        choices=[('', 'Choose your option'),
-                 ('latest', 'Latest'),
-                 ],
-        validators=[DataRequired()]
-    )
-
-    def validate_files(form, field):
-        for file in field.data:
-            if not file.filename.lower().endswith('.txt'):
-                raise ValidationError(
-                    'File does not have an approved extension: '
-                    '.txt'
-                )

app/services/views.py

@@ -6,61 +6,8 @@ from ..models import Job
 from ..import swarm
 from .. import db
 from threading import Thread
-import os
 import json
-
-
-@services.route('/ocr', methods=['GET', 'POST'])
-@login_required
-def ocr():
-    new_ocr_job_form = NewOCRJobForm()
-    if new_ocr_job_form.validate_on_submit():
-        ocr_job = Job(creator=current_user._get_current_object(),
-                      description=new_ocr_job_form.description.data,
-                      service="ocr",
-                      ressources=json.dumps({"n_cores": 4,
-                                             "mem_mb": 8192}),
-                      service_args=json.dumps({"args": ["--keep-intermediates",
-                                                        "--skip-binarisation"],
-                                               "lang": new_ocr_job_form.language.data,
-                                               "version": new_ocr_job_form.version.data}),
-                      status="pending",
-                      title=new_ocr_job_form.title.data)
-
-        db.session.add(ocr_job)
-        db.session.commit()
-
-        dir = os.path.join(current_app.config['OPAQUE_STORAGE'],
-                           str(ocr_job.user_id),
-                           'jobs',
-                           str(ocr_job.id))
-
-        try:
-            os.makedirs(dir)
-        except OSError:
-            flash('OSError!')
-            db.session.remove(ocr_job)
-            db.session.commit()
-        else:
-            for file in new_ocr_job_form.files.data:
-                file.save(os.path.join(dir, file.filename))
-            '''
-            ' TODO: Let the scheduler run this job in the background.
-            '
-            ' NOTE: Using self created threads is just for testing purpose as
-            '       there is no scheduler available.
-            '''
-            db.session.expunge(ocr_job)
-            thread = Thread(target=swarm.run, args=(ocr_job,))
-            thread.start()
-            flash('Job created!')
-        return redirect(url_for('services.ocr'))
-
-    return render_template(
-        'services/ocr.html.j2',
-        title='Optical Character Recognition',
-        new_ocr_job_form=new_ocr_job_form
-    )
+import os
 
 
 @services.route('/nlp', methods=['GET', 'POST'])
@@ -68,15 +15,14 @@ def ocr():
 def nlp():
     new_nlp_job_form = NewNLPJobForm()
     if new_nlp_job_form.validate_on_submit():
-        nlp_job = Job(creator=current_user._get_current_object(),
+        nlp_job = Job(creator=current_user,
                       description=new_nlp_job_form.description.data,
-                      service="nlp",
-                      ressources=json.dumps({"n_cores": 2,
-                                             "mem_mb": 4096}),
-                      service_args=json.dumps({"args": [],
-                                               "lang": new_nlp_job_form.language.data,
-                                               "version": new_nlp_job_form.version.data}),
-                      status="pending",
+                      mem_mb=4096,
+                      n_cores=2,
+                      service='nlp',
+                      service_args=json.dumps(['-l {}'.format(new_nlp_job_form.language.data)]),
+                      service_version=new_nlp_job_form.version.data,
+                      status='submitted',
                       title=new_nlp_job_form.title.data)
 
         db.session.add(nlp_job)
@@ -113,3 +59,57 @@ def nlp():
         title='Natrual Language Processing',
         new_nlp_job_form=new_nlp_job_form
     )
+
+
+@services.route('/ocr', methods=['GET', 'POST'])
+@login_required
+def ocr():
+    new_ocr_job_form = NewOCRJobForm()
+    if new_ocr_job_form.validate_on_submit():
+        ocr_job = Job(creator=current_user,
+                      description=new_ocr_job_form.description.data,
+                      mem_mb=8192,
+                      n_cores=4,
+                      service='ocr',
+                      service_args=json.dumps([
+                        '-l {}'.format(new_ocr_job_form.language.data),
+                        '--keep-intermediates',
+                        '--skip-binarisation']),
+                      service_version=new_ocr_job_form.version.data,
+                      status='submitted',
+                      title=new_ocr_job_form.title.data)
+
+        db.session.add(ocr_job)
+        db.session.commit()
+
+        dir = os.path.join(current_app.config['OPAQUE_STORAGE'],
+                           str(ocr_job.user_id),
+                           'jobs',
+                           str(ocr_job.id))
+
+        try:
+            os.makedirs(dir)
+        except OSError:
+            flash('OSError!')
+            db.session.remove(ocr_job)
+            db.session.commit()
+        else:
+            for file in new_ocr_job_form.files.data:
+                file.save(os.path.join(dir, file.filename))
+            '''
+            ' TODO: Let the scheduler run this job in the background.
+            '
+            ' NOTE: Using self created threads is just for testing purpose as
+            '       there is no scheduler available.
+            '''
+            db.session.expunge(ocr_job)
+            thread = Thread(target=swarm.run, args=(ocr_job,))
+            thread.start()
+            flash('Job created!')
+        return redirect(url_for('services.ocr'))
+
+    return render_template(
+        'services/ocr.html.j2',
+        title='Optical Character Recognition',
+        new_ocr_job_form=new_ocr_job_form
+    )