Extracting a pattern from a string using python

Question

I'm trying to read data from a column in an Excel file and then extracint user id's used in that row. So far I was able to extract user id's using the following code and then write the results to an Excel file.

import xlrd
import pandas as pd


#Input File Path
file='file1.xlsx'
workbook = xlrd.open_workbook(file)

#open first worksheet
sheet=workbook.sheet_by_index(0)

#extract details from 4th column
description = sheet.col_values(4)

my_series = pd.Series(description)
numbers = my_series.str.findall('\d+')
All_Ids = pd.to_numeric(numbers, errors='ignore')
All_Ids_mapped = [map(int, x) for x in All_Ids]
df = pd.DataFrame(All_Ids_mapped)

# Create a Pandas Excel writer using XlsxWriter as the engine.
writer = pd.ExcelWriter('extracted_ids.xlsx', engine='xlsxwriter')

# Convert the dataframe to an XlsxWriter Excel object.
df.to_excel(writer, sheet_name='Sheet1')

# Close the Pandas Excel writer and output the Excel file.
writer.save()

But now my problem is that in the column there are many id's. So I want to extract the id's which start with the string 'user with id' For instance a string in the column looks like the following:

The user with id '123' started discussion with the user with id '456' in the discussion thread with id '5000'.

Since I'm interested only in user id's I want to update my search string to incorporate text. I tried the following but it doesn't give me the output.

  numbers=my_series.str.findall('^user with id.+\d+')

Please help me write the correct expression in str.findall .

Thank you.

Answer 1

using re module, I got the following result:

series = "The user with id '123' started discussion with the user with id '456' in the discussion thread with id '5000'."
>>>re.findall("user with id '\d+'", series)
["user with id '123'", "user with id '456'"]

Are these the expected matches? Since the resulting matches are ordered, it wouldn't be too hard to select one by index and extract the id.