How to read Excel data by column name in python using xlrd

Question

I am trying to read the data of large excel file(almost 100000 row). I am using 'xlrd Module' in python to fetch the data from excel. I want to fetch data by column name( Cascade,Schedule Name,Market ) instead of column number( 0,1,2 ). Because my excel columns are not fixed. i know how to fetch data in case of fixed column.

here is the code by which i am fetching data from the excel for fixed column

import xlrd

file_location =r"C:\Users\Desktop\Vision.xlsx"
workbook=xlrd.open_workbook(file_location)
sheet= workbook.sheet_by_index(0)
print(sheet.ncols,sheet.nrows,sheet.name,sheet.number)

for i in range(sheet.nrows):
   flag = 0
   for j in range(sheet.ncols):
      value=sheet.cell(i,j).value

If anyone has any solution of this, kindly let me know

Thanks

Answer 1

Alternatively you could also make use of pandas , which is a comprehensive data analysis library with built-in excel I/O capabilities .

import pandas as pd

file_location =r"C:\Users\esatnir\Desktop\Sprint Vision.xlsx"

# Read out first sheet of excel file and return as pandas dataframe
df = pd.read_excel(file_location)

# Reduce dataframe to target columns (by filtering on column names)
df = df[['Cascade', 'Schedule Name', 'Market']]

where a quick view of the resulting dataframe df will show:

In [1]: df
Out[1]:
   Cascade     Schedule Name                Market
0  SF05UB0  DO Macro Upgrade  Upper Central Valley
1  DE03HO0  DO Macro Upgrade                Toledo
2  SF73XC4  DO Macro Upgrade                SF Bay

Answer 2

Your column names are in the first row of the spreadsheet, right? So read the first row and construct a mapping from names to column indices.

column_pos = [ (sheet.cell(0, i).value, i) for i in range(sheet.ncols) ]
colidx = dict(column_pos)

Or as a one-liner:

colidx = dict( (sheet.cell(0, i).value, i) for i in range(sheet.ncols) )

You can then use the index to interpret column names, for example:

print(sheet.cell(5, colidx["Schedule Name"]).value)

To get an entire column, you can use a list comprehension:

schedule = [ sheet.cell(i, colidx["Schedule Name"]).value for i in range(1, sheet.nrows) ]

If you really wanted to, you could create a wrapper for the cell function that handles the interpretation. But I think this is simple enough.

Answer 3

Comment : still not working when header of
fieldnames = ['Cascade', 'Market', 'Schedule', 'Name] and
Sheet(['Cascade', 'Schedule', 'Name', 'Market']) are equal.

Keep order of fieldnames in col_idx , was not my initial goal.

---

Question : I want to fetch data by column name

The following OOP solution will work:

class OrderedByName():
    """
    Privides a generator method, to iterate in Column Name ordered sequence
    Provides subscription, to get columns index by name. using class[name]
    """
    def __init__(self, sheet, fieldnames, row=0):
        """
        Create a OrderedDict {name:index} from 'fieldnames'
        :param sheet: The Worksheet to use
        :param fieldnames: Ordered List of Column Names
        :param row: Default Row Index for the Header Row
        """
        from collections import OrderedDict
        self.columns = OrderedDict().fromkeys(fieldnames, None)
        for n in range(sheet.ncols):
            self.columns[sheet.cell(row, n).value] = n

    @property
    def ncols(self):
        """
        Generator, equal usage as range(xlrd.ncols), 
          to iterate columns in ordered sequence
        :return: yield Column index
        """
        for idx in self.columns.values():
            yield idx

    def __getitem__(self, item):
        """
        Make class object subscriptable
        :param item: Column Name
        :return: Columns index
        """
        return self.columns[item]

Usage :

# Worksheet Data
sheet([['Schedule', 'Cascade', 'Market'],
       ['SF05UB0', 'DO Macro Upgrade', 'Upper Cnetral Valley'],
       ['DE03HO0', 'DO Macro Upgrade', 'Toledo'],
       ['SF73XC4', 'DO Macro Upgrade', 'SF Bay']]
      )

# Instantiate with Ordered List of Column Names
# NOTE the different Order of Column Names
by_name = OrderedByName(sheet, ['Cascade', 'Market', 'Schedule'])

# Iterate all Rows and all Columns Ordered as instantiated
for row in range(sheet.nrows):
    for col in by_name.ncols:
        value = sheet.cell(row, col).value
        print("cell({}).value == {}".format((row,col), value))

Output :

 cell((0, 1)).value == Cascade cell((0, 2)).value == Market cell((0, 0)).value == Schedule cell((1, 1)).value == DO Macro Upgrade cell((1, 2)).value == Upper Cnetral Valley cell((1, 0)).value == SF05UB0 cell((2, 1)).value == DO Macro Upgrade cell((2, 2)).value == Toledo cell((2, 0)).value == DE03HO0 cell((3, 1)).value == DO Macro Upgrade cell((3, 2)).value == SF Bay cell((3, 0)).value == SF73XC4

---

Get Index of one Column by Name

print("cell{}.value == {}".format((1, by_name['Schedule']), sheet.cell(1, by_name['Schedule']).value)) #>>> cell(1, 0).value == SF05UB0

Tested with Python: 3.5

Answer 4

You can make use of pandas. Below is the sample code for identifying the columns and rows in an excel sheet.

import pandas as pd

file_location =r"Your_Excel_Path"

# Read out first sheet of excel file and return as pandas dataframe
df = pd.read_excel(file_location)


total_rows=len(df.axes[0])
total_cols=len(df.axes[1])

# Print total number of rows in an excel sheet
print("Number of Rows: "+str(total_rows))

# Print total number of columns in an excel sheet
print("Number of Columns: "+str(total_cols))

# Print column names in an excel sheet
print(df.columns.ravel())

Now once you have the column data, you can convert it into a list of values.