def __init__(self):
super().__init__()
self.corpus = None
# Refresh recent files
self.recent_files = [fn for fn in self.recent_files
if os.path.exists(fn)]
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
# Drop-down for recent files
self.file_combo = QtGui.QComboBox(fbox)
self.file_combo.setMinimumWidth(300)
fbox.layout().addWidget(self.file_combo)
self.file_combo.activated[int].connect(self.select_file)
# Browse button
browse = gui.button(fbox, self, 'Browse', callback=self.browse_file)
browse.setIcon(self.style().standardIcon(QtGui.QStyle.SP_DirOpenIcon))
browse.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Reload button
reload = gui.button(fbox, self, "Reload", callback=self.reload, default=True)
reload.setIcon(self.style().standardIcon(QtGui.QStyle.SP_BrowserReload))
reload.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# Load the most recent file
self.set_file_list()
if len(self.recent_files) > 0:
self.open_file(self.recent_files[0])
def __init__(self):
super().__init__()
self.corpus = None
# Refresh recent files
self.recent_files = [fn for fn in self.recent_files
if os.path.exists(fn)]
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
# Drop-down for recent files
self.file_combo = QtGui.QComboBox(fbox)
self.file_combo.setMinimumWidth(300)
fbox.layout().addWidget(self.file_combo)
self.file_combo.activated[int].connect(self.select_file)
# Browse button
browse = gui.button(fbox, self, 'Browse', callback=self.browse_file)
browse.setIcon(self.style().standardIcon(QtGui.QStyle.SP_DirOpenIcon))
browse.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Reload button
reload = gui.button(fbox, self, "Reload", callback=self.reload, default=True)
reload.setIcon(self.style().standardIcon(QtGui.QStyle.SP_BrowserReload))
reload.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# Load the most recent file
self.set_file_list()
if len(self.recent_files) > 0:
self.open_file(self.recent_files[0])
def test_drop_mime(self):
iris = Table("iris")
m = VariablesListItemModel(iris.domain.variables)
mime = m.mimeData([m.index(1, 0)])
self.assertTrue(mime.hasFormat(VariablesListItemModel.MIME_TYPE))
assert m.dropMimeData(mime, Qt.MoveAction, 5, 0, m.index(-1, -1))
self.assertIs(m[5], m[1])
assert m.dropMimeData(mime, Qt.MoveAction, -1, -1, m.index(-1, -1))
self.assertIs(m[6], m[1])
def test_flags(self):
m = VariablesListItemModel([ContinuousVariable("X")])
index = m.index(0)
flags = m.flags(m.index(0))
self.assertTrue(flags & Qt.ItemIsDragEnabled)
self.assertFalse(flags & Qt.ItemIsDropEnabled)
# 'invalid' index is drop enabled -> indicates insertion capability
flags = m.flags(m.index(-1, -1))
self.assertTrue(flags & Qt.ItemIsDropEnabled)
def test_drop_mime(self):
iris = Table("iris")
m = VariablesListItemModel(iris.domain.variables)
mime = m.mimeData([m.index(1, 0)])
self.assertTrue(mime.hasFormat(VariablesListItemModel.MIME_TYPE))
assert m.dropMimeData(mime, Qt.MoveAction, 5, 0, m.index(-1, -1))
self.assertIs(m[5], m[1])
assert m.dropMimeData(mime, Qt.MoveAction, -1, -1, m.index(-1, -1))
self.assertIs(m[6], m[1])
def __init__(self):
super().__init__()
self.corpus = None
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
widget = widgets.FileWidget(recent_files=self.recent_files, icon_size=(16, 16), on_open=self.open_file,
directory_aliases={"Browse documentation corpora ...": get_sample_corpora_dir()},
dialog_format=self.dlgFormats, dialog_title='Open Orange Document Corpus',
allow_empty=False, reload_label='Reload', browse_label='Browse')
fbox.layout().addWidget(widget)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# load first file
widget.select(0)
def test_flags(self):
m = VariablesListItemModel([ContinuousVariable("X")])
index = m.index(0)
flags = m.flags(m.index(0))
self.assertTrue(flags & Qt.ItemIsDragEnabled)
self.assertFalse(flags & Qt.ItemIsDropEnabled)
# 'invalid' index is drop enabled -> indicates insertion capability
flags = m.flags(m.index(-1, -1))
self.assertTrue(flags & Qt.ItemIsDropEnabled)
class OWLoadCorpus(OWWidget):
name = "Corpus"
description = "Load a corpus of text documents, (optionally) tagged with categories."
icon = "icons/TextFile.svg"
priority = 10
outputs = [(Output.CORPUS, Corpus)]
want_main_area = False
resizing_enabled = False
dlgFormats = "Only tab files (*.tab)"
recent_files = Setting(["(none)"])
def __init__(self):
super().__init__()
self.corpus = None
# Refresh recent files
self.recent_files = [fn for fn in self.recent_files
if os.path.exists(fn)]
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
# Drop-down for recent files
self.file_combo = QtGui.QComboBox(fbox)
self.file_combo.setMinimumWidth(300)
fbox.layout().addWidget(self.file_combo)
self.file_combo.activated[int].connect(self.select_file)
# Browse button
browse = gui.button(fbox, self, 'Browse', callback=self.browse_file)
browse.setIcon(self.style().standardIcon(QtGui.QStyle.SP_DirOpenIcon))
browse.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Reload button
reload = gui.button(fbox, self, "Reload", callback=self.reload, default=True)
reload.setIcon(self.style().standardIcon(QtGui.QStyle.SP_BrowserReload))
reload.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# Load the most recent file
self.set_file_list()
if len(self.recent_files) > 0:
self.open_file(self.recent_files[0])
def set_file_list(self):
self.file_combo.clear()
if not self.recent_files:
self.file_combo.addItem("(none)")
for file in self.recent_files:
if file == "(none)":
self.file_combo.addItem("(none)")
else:
self.file_combo.addItem(os.path.split(file)[1])
self.file_combo.addItem("Browse documentation corpora ...")
def reload(self):
if self.recent_files:
return self.open_file(self.recent_files[0])
def select_file(self, n):
if n < len(self.recent_files) :
name = self.recent_files[n]
del self.recent_files[n]
self.recent_files.insert(0, name)
elif n:
self.browse_file(True)
if len(self.recent_files) > 0:
self.set_file_list()
self.open_file(self.recent_files[0])
def browse_file(self, demos_loc=False):
start_file = os.path.expanduser("~/")
if demos_loc:
start_file = get_sample_corpora_dir()
filename = QtGui.QFileDialog.getOpenFileName(
self, 'Open Orange Document Corpus', start_file, self.dlgFormats)
if not filename:
return
if filename in self.recent_files:
self.recent_files.remove(filename)
self.recent_files.insert(0, filename)
self.set_file_list()
self.open_file(filename)
def open_file(self, path):
self.error(1, '')
self.used_attrs[:] = []
self.unused_attrs[:] = []
try:
self.corpus = Corpus.from_file(path)
for f in self.corpus.domain.metas:
if f in self.corpus.text_features:
self.used_attrs.append(f)
else:
self.unused_attrs.append(f)
self.info_label.setText("Corpus of {} documents.".format(len(self.corpus)))
self.send(Output.CORPUS, self.corpus)
except BaseException as err:
self.error(1, str(err))
def update_feature_selection(self):
if self.corpus is not None:
self.corpus.set_text_features(self.used_attrs)
self.send(Output.CORPUS, self.corpus)
class OWLoadCorpus(OWWidget):
name = "Corpus"
description = "Load a corpus of text documents, (optionally) tagged with categories."
icon = "icons/TextFile.svg"
priority = 10
outputs = [(Output.CORPUS, Corpus)]
want_main_area = False
resizing_enabled = False
dlgFormats = (
"All readable files ({});;".format(
'*' + ' *'.join(FileFormat.readers.keys())) +
";;".join("{} (*{})".format(f.DESCRIPTION, ' *'.join(f.EXTENSIONS))
for f in sorted(set(FileFormat.readers.values()),
key=list(FileFormat.readers.values()).index)))
recent_files = Setting([])
class Error(OWWidget.Error):
read_file = Msg("Can't read file {} ({})")
def __init__(self):
super().__init__()
self.corpus = None
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
widget = widgets.FileWidget(recent_files=self.recent_files, icon_size=(16, 16), on_open=self.open_file,
directory_aliases={"Browse documentation corpora ...": get_sample_corpora_dir()},
dialog_format=self.dlgFormats, dialog_title='Open Orange Document Corpus',
allow_empty=False, reload_label='Reload', browse_label='Browse')
fbox.layout().addWidget(widget)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# load first file
widget.select(0)
def open_file(self, path):
self.Error.read_file.clear()
self.used_attrs[:] = []
self.unused_attrs[:] = []
if path:
try:
self.corpus = Corpus.from_file(path)
self.corpus.name = os.path.splitext(os.path.basename(path))[0]
self.info_label.setText("Corpus of {} documents.".format(len(self.corpus)))
self.used_attrs.extend(self.corpus.text_features)
self.unused_attrs.extend([f for f in self.corpus.domain.metas
if f.is_string and f not in self.corpus.text_features])
except BaseException as err:
self.Error.read_file(path, str(err))
def update_feature_selection(self):
# TODO fix VariablesListItemView so it does not emit
# duplicated data when reordering inside a single window
def remove_duplicates(l):
unique = []
for i in l:
if i not in unique:
unique.append(i)
return unique
if self.corpus is not None:
self.corpus.set_text_features(remove_duplicates(self.used_attrs))
self.send(Output.CORPUS, self.corpus)
class OWLoadCorpus(OWWidget):
name = "Corpus"
description = "Load a corpus of text documents, (optionally) tagged with categories."
icon = "icons/TextFile.svg"
priority = 10
outputs = [(Output.CORPUS, Corpus)]
want_main_area = False
dlgFormats = "Only tab files (*.tab)"
recent_files = Setting(["(none)"])
def __init__(self):
super().__init__()
self.corpus = None
# Refresh recent files
self.recent_files = [
fn for fn in self.recent_files if os.path.exists(fn)
]
# Browse file box
fbox = gui.widgetBox(self.controlArea, "Corpus file", orientation=0)
# Drop-down for recent files
self.file_combo = QtGui.QComboBox(fbox)
self.file_combo.setMinimumWidth(300)
fbox.layout().addWidget(self.file_combo)
self.file_combo.activated[int].connect(self.select_file)
# Browse button
browse = gui.button(fbox, self, 'Browse', callback=self.browse_file)
browse.setIcon(self.style().standardIcon(QtGui.QStyle.SP_DirOpenIcon))
browse.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Reload button
reload = gui.button(fbox,
self,
"Reload",
callback=self.reload,
default=True)
reload.setIcon(self.style().standardIcon(
QtGui.QStyle.SP_BrowserReload))
reload.setSizePolicy(QtGui.QSizePolicy.Fixed, QtGui.QSizePolicy.Fixed)
# Corpus info
ibox = gui.widgetBox(self.controlArea, "Corpus info", addSpace=True)
corp_info = "Corpus of 0 documents."
self.info_label = gui.label(ibox, self, corp_info)
# Used Text Features
fbox = gui.widgetBox(self.controlArea, orientation=0)
ubox = gui.widgetBox(fbox, "Used text features", addSpace=True)
self.used_attrs = VariablesListItemModel()
self.used_attrs_view = VariablesListItemView()
self.used_attrs_view.setModel(self.used_attrs)
ubox.layout().addWidget(self.used_attrs_view)
aa = self.used_attrs
aa.dataChanged.connect(self.update_feature_selection)
aa.rowsInserted.connect(self.update_feature_selection)
aa.rowsRemoved.connect(self.update_feature_selection)
# Ignored Text Features
ibox = gui.widgetBox(fbox, "Ignored text features", addSpace=True)
self.unused_attrs = VariablesListItemModel()
self.unused_attrs_view = VariablesListItemView()
self.unused_attrs_view.setModel(self.unused_attrs)
ibox.layout().addWidget(self.unused_attrs_view)
# Load the most recent file
self.set_file_list()
if len(self.recent_files) > 0:
self.open_file(self.recent_files[0])
def set_file_list(self):
self.file_combo.clear()
if not self.recent_files:
self.file_combo.addItem("(none)")
for file in self.recent_files:
if file == "(none)":
self.file_combo.addItem("(none)")
else:
self.file_combo.addItem(os.path.split(file)[1])
self.file_combo.addItem("Browse documentation corpora ...")
def reload(self):
if self.recent_files:
return self.open_file(self.recent_files[0])
def select_file(self, n):
if n < len(self.recent_files):
name = self.recent_files[n]
del self.recent_files[n]
self.recent_files.insert(0, name)
elif n:
self.browse_file(True)
if len(self.recent_files) > 0:
self.set_file_list()
self.open_file(self.recent_files[0])
def browse_file(self, demos_loc=False):
start_file = os.path.expanduser("~/")
if demos_loc:
start_file = get_sample_corpora_dir()
filename = QtGui.QFileDialog.getOpenFileName(
self, 'Open Orange Document Corpus', start_file, self.dlgFormats)
if not filename:
return
if filename in self.recent_files:
self.recent_files.remove(filename)
self.recent_files.insert(0, filename)
self.set_file_list()
self.open_file(filename)
def open_file(self, path):
self.error(1, '')
self.used_attrs[:] = []
self.unused_attrs[:] = []
try:
self.corpus = Corpus.from_file(path)
for i in self.corpus.used_features:
self.used_attrs.append(i)
for i in self.corpus.domain.metas:
if i not in self.corpus.used_features:
self.unused_attrs.append(i)
self.info_label.setText("Corpus of {} documents.".format(
len(self.corpus)))
self.send(Output.CORPUS, self.corpus)
except BaseException as err:
self.error(1, str(err))
def update_feature_selection(self):
if self.corpus is not None:
self.corpus.regenerate_documents(self.used_attrs)
self.send(Output.CORPUS, self.corpus)