Purpose:

The script checks if the “IB Last Updated” date is two or more days old and flags it for alert if certain conditions are met.

To Elaborate

The Python script is designed to monitor the timeliness of inventory bidding feed updates. It identifies instances where the “IB Last Updated” date is two or more days old, indicating a potential issue with feed processing. The script flags these instances by setting a “Stale Feed Alert” to “TRUE” under specific conditions: both the Campaign Status and Group Status must be active, the Campaign must contain “RBA-Pro-SEM” or “RBA-Pro-BSEM”, and Inventory Bidding must be set to TRUE. This alert is intended to trigger further reporting and serve as a notification for necessary action. The script filters data based on these criteria to ensure only relevant alerts are generated.

Walking Through the Code

1. Data Preparation
   • The script begins by copying the input DataFrame (inputDf) to a new DataFrame (outputDf) to preserve the original data while making modifications.
2. Time Difference Calculation
   • It calculates the time difference between the “IB Last Updated” date and the current date to determine if the feed is stale.
3. Boolean Mask Creation
   • A boolean mask is created to identify rows where the time difference is two or more days, indicating a stale feed.
4. Stale Feed Alert Assignment
   • The script sets the “Stale Feed Alert” column to “TRUE” for rows identified by the mask, otherwise leaving it blank.
5. Filter Modified Rows
   • Finally, the script filters the DataFrame to include only those rows where the “Stale Feed Alert” has been set, ensuring that only relevant alerts are processed further.

Vitals

• Script ID : 157
• Client ID / Customer ID: 1306913045 / 60268001
• Action Type: Bulk Upload
• Item Changed: AdGroup
• Output Columns: Account, Campaign, Group, Stale Feed Alert
• Linked Datasource: M1 Report
• Reference Datasource: None
• Owner: Jeff Sands (jsands@marinsoftware.com)
• Created by Jeff Sands on 2023-05-31 20:20
• Last Updated by Jeff Sands on 2023-12-06 04:01

> See it in Action

Python Code

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
# Column Names
RPT_COL_GROUP = 'Group'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_CAMPAIGN_STATUS = 'Campaign Status'
RPT_COL_GROUP_STATUS = 'Group Status'
RPT_COL_INVENTORY_BIDDING = 'Inventory Bidding'
RPT_COL_IB_LASTUPDATED = 'IB Last Updated'
RPT_COL_AUCTIONBOOST = 'AuctionBoost'
RPT_COL_STALE_FEEDALERT = 'Stale Feed Alert'

BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_GROUP = 'Group'
BULK_COL_STATUS = 'Status'
BULK_COL_STALE_FEEDALERT = 'Stale Feed Alert'

# Copy inputDf to outputDf
outputDf = inputDf.copy()

# Get the current date
today = datetime.datetime.now(CLIENT_TIMEZONE).date()

# Calculate time difference between RPT_COL_IB_LASTUPDATED and today
time_difference = today - inputDf[RPT_COL_IB_LASTUPDATED].dt.date

# Create a boolean mask for rows with time difference >= 2 days
mask = time_difference >= datetime.timedelta(days=2)

# Set BULK_COL_STALE_FEEDALERT to "TRUE" where mask is True, otherwise keep it blank
outputDf[BULK_COL_STALE_FEEDALERT] = np.where(mask, "TRUE", "")

# Create outputDf with only modified rows based on the mask
outputDf = outputDf[mask].copy()

Post generated on 2025-03-11 01:25:51 GMT