Is it possible to involve a search bar on a DataFrame in Pandas? - python-3.x

Question

I have a DataFrame in Pandas which collects some data from an Excel document. I created a GUI with PyQt5 in order to make it look more interesting but here is the thing.
Is it possbile to make a dynamic search bar in order to search through that DataFrame? For example, my DataFrame has over 3k+ rows and I wanna search for John Doe, then the results will come up on the GUI. As far as I know, QLineEdit is used for this but I can't seem to implement it on my code.
Is it me that is doing wrong or it is not possible to do it on a DataFrame? And if anyone wanna help me, just let me know, I would be so grateful and thankful, I guess it'll only take 10-15 minutes. I can also post the code here, but talking on Discord and explaining you in detail and also sharing screens would be a lot easier.

Answer 1

This can be done by subclassing QAbstractTableModel to create a custom table model that uses the underlying dataframe for supplying data to a QTableView. This custom model can then be combined with a QProxyFilterSortModel to filter the data in the table. To create a custom non-editable model from QAbstractTableModel you need to implement rowCount, columnCount, data, and headerData at the very least. In this case, minimal implemetation could be something like this:
class DataFrameModel(QtCore.QAbstractTableModel):
def __init__(self, data_frame, parent = None):
super().__init__(parent)
self.data_frame = data_frame
def rowCount(self, index):
if index.isValid():
return 0
return self.data_frame.shape[0]
def columnCount(self, index):
if index.isValid():
return 0
return self.data_frame.shape[1]
def data(self, index, role):
if not index.isValid() or role != QtCore.Qt.DisplayRole:
return None
return str(self.data_frame.iloc[index.row(), index.column()])
def headerData(self, section, orientation, role=None):
if role != QtCore.Qt.DisplayRole:
return None
if orientation == QtCore.Qt.Vertical:
return self.data_frame.index[section]
else:
return self.data_frame.columns[section]
To show and filter the data in a table you could do something like this:
class MyWidget(QtWidgets.QWidget):
def __init__(self, parent = None):
super().__init__(parent)
self.table_view = QtWidgets.QTableView()
self.proxy_model = QtCore.QSortFilterProxyModel()
# by default, the QSortFilterProxyModel will search for keys in the first column only
# setting QSortFilterProxyModel.filterKeyColumn to -1 will match values in all columns
self.proxy_model.setFilterKeyColumn(-1)
self.table_view.setModel(self.proxy_model)
# line edit for entering (part of) the key that should be searched for
self.line_edit = QtWidgets.QLineEdit()
self.line_edit.textChanged.connect(self.filter_text)
vlayout = QtWidgets.QVBoxLayout(self)
vlayout.addWidget(self.line_edit)
vlayout.addWidget(self.table_view)
def filter_text(self, text):
self.proxy_model.setFilterFixedString(text)
def set_data(self, data):
self.model = DataFrameModel(pd.DataFrame(data))
self.proxy_model.setSourceModel(self.model)
if __name__ == "__main__":
app = QtWidgets.QApplication([])
win = MyWidget()
win.set_data({'a':['apple', 'banana', 'cherry'], 'b':[4,5,6], 'c':['green', 'yellow', 'red']})
win.show()
app.exec()
Of course this is a very basic implementation but I might help you get started.

Answer 2

I think the library pandas_ui is what you are looking for here.
It allows you to play with your dataframe interactively, i.e. in a microsoft excel fashion. Particularly, the column filtering option should do what you want.