Purpose

Python script that parses campaign names and extracts the campaign type into a dimension called “Type de campagne”.

To Elaborate

The Python script solves the problem of extracting the campaign type from campaign names. The campaign type is identified by a tag that appears after a hyphen in the campaign name. The script takes an input DataFrame that contains campaign data and outputs a modified DataFrame with the campaign type extracted into a new column called “Type de campagne”. The script follows the following business rules:

• The campaign type tag appears after a hyphen in the campaign name.
• If the campaign name does not contain a hyphen, it is skipped.
• If the extracted campaign type is different from the existing campaign type in the DataFrame, it is updated.
• The output DataFrame only includes rows with non-empty campaign types.
• Extra whitespace in the campaign name is removed to avoid issues with previewing the data.

The script uses the following user-changeable parameters:

• PLACEMENT_KEY: The character that indicates the placement of the campaign type tag in the campaign name.
• inputDf: The input DataFrame that contains the campaign data.
• RPT_COL_CAMPAIGN: The column name in the input DataFrame that contains the campaign names.
• RPT_COL_TYPE_CAMPAGNE: The column name in the input DataFrame where the campaign type will be extracted.
• BULK_COL_TYPE_CAMPAGNE: The column name in the output DataFrame where the campaign type will be stored.

Walking Through the Code

1. The script defines a function get_tag_from_campaign_name that extracts the campaign type tag from a campaign name using a regular expression pattern.
2. The script creates an output DataFrame by copying the input DataFrame.
3. The script loops through each row in the input DataFrame.
4. For each row, it retrieves the existing campaign type and campaign name.
5. If the campaign name does not contain the placement key, the row is skipped.
6. The script calls the get_tag_from_campaign_name function to extract the campaign type tag from the campaign name.
7. If the extracted tag is different from the existing campaign type, it is updated in the output DataFrame.
8. The script drops rows with empty campaign types from the output DataFrame.
9. Extra whitespace in the campaign name is removed in the output DataFrame.
10. If the output DataFrame is not empty, it is printed. Otherwise, a message indicating an empty output is printed.

Vitals

• Script ID : 1071
• Client ID / Customer ID: 1306927809 / 60270355
• Action Type: Bulk Upload (Preview)
• Item Changed: Campaign
• Output Columns: Account, Campaign, Type de Campagne
• Linked Datasource: M1 Report
• Reference Datasource: None
• Owner: Grégory Pantaine (gpantaine@marinsoftware.com)
• Created by Grégory Pantaine on 2024-05-10 16:50
• Last Updated by Grégory Pantaine on 2024-05-10 17:02

> 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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
## name: Script - Campaign - Type de Campagne
## description:
## Parse Campaign Name and pick out the campaign type into a dimension 'Type de campagne'.
## Tag appears after '-' in campaign name.
## 
## Copied by Grégory Pantaine
## created: 2024-05-10

########### Configurable Params - START ##########
PLACEMENT_KEY = '-'

# Primary data source and columns
inputDf = dataSourceDict["1"]

# Output columns and initial values
RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_ACCOUNT = 'Account'
RPT_COL_TYPE_CAMPAGNE = 'Type de Campagne'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_TYPE_CAMPAGNE = 'Type de Campagne'

# Function to extract tag from campaign name

def get_tag_from_campaign_name(campaign_name):
    regex_pattern = r"(?:[^-]+-){3}(\w+)"

    placement_match = re.search(regex_pattern, campaign_name)
    if placement_match:
        placement_value = placement_match.group(1)
        # print("Placement value:", placement_value)
        return placement_value.strip()
    else:
        print("Placement value not found: " + campaign_name)
    
    # Return the entire campaign name if 'pl:' is not present
    return campaign_name.strip()

# Copy all input rows to output
outputDf = inputDf.copy()

# Loop through all rows
for index, row in inputDf.iterrows():
    existing_tag = row[RPT_COL_TYPE_CAMPAGNE]
    campaign_name = row[RPT_COL_CAMPAIGN]
    
    # Skip processing if campaign name does not contain the placement key
    if PLACEMENT_KEY not in campaign_name:
        continue

    tag = get_tag_from_campaign_name(campaign_name)

    # Print campaign and tag information
    # print("Campaign [%s] => Tag [%s]" % (campaign_name, tag))

    # Only tag if it's different than the existing tag
    if (len(tag) > 0) & (tag != existing_tag):
        outputDf.at[index, BULK_COL_TYPE_CAMPAGNE] = tag
    else:
        outputDf.at[index, BULK_COL_TYPE_CAMPAGNE] = np.nan

# Only include non-empty tags in bulk
outputDf = outputDf.dropna(subset=[BULK_COL_TYPE_CAMPAGNE])

# Remove extra whitespace from campaign name that breaks Preview
outputDf[RPT_COL_CAMPAIGN] = outputDf[RPT_COL_CAMPAIGN].str.strip()

if not outputDf.empty:
    print("outputDf", tableize(outputDf.head()))
else:
    print("Empty outputDf")

Post generated on 2024-05-15 07:44:05 GMT