Unique Active Addresses

Goal

The goal of this calculation is to present the count of unique addresses active on the blockchain, either daily, weekly or monthly.

Data used

As source we use data collected in the impacted_addresses collection. For every transaction, we find which addresses (account, contract or public key) are impacted by this transaction and add this to the collection.

Calculation steps

Calculation happens separately for each grouping (daily, weekly, monthly). All three groupings use the same pipeline, only the height_for_first_block and height_for_last_block are selected differently (but all taken from the collection blocks_per_day)

Daily

For daily, it's an entry every day. The timestamp on the entry is the day itself.

Weekly

For weekly, we take the starting block on Monday and end with the last block the following Sunday. The timestamp on the entry is the Monday of the week. During the current week, the week is marked as incomplete, meaning that for every day of this week, the data will be recalculated. It is marked complete when the entire week is in the past.

Monthly

For monthly, we take the starting block on the first day of the month and end with the last block of the last day of the month. The timestamp on the entry is the first day of the month. During the current month, the month is marked as incomplete, meaning that for every day of this month, the data will be recalculated. It is marked complete when the entire month is in the past.

The following pipeline is used to extract the data:

pipeline = [
            {
                "$match": {
                    "block_height": {
                        "$gte": height_for_first_block,
                        "$lte": height_for_last_block,
                    },
                    "effect_type": {"$ne": "Account Reward"},
                }
            },
            {
                "$project": {
                    "address_length": {"$strLenCP": "$impacted_address_canonical"},
                    "impacted_address_canonical": 1,
                }
            },
            {
                "$group": {
                    "_id": {
                        "category": {
                            "$switch": {
                                "branches": [
                                    {
                                        "case": {"$lt": ["$address_length", 29]},
                                        "then": "contract",
                                    },
                                    {
                                        "case": {
                                            "$and": [
                                                {"$gte": ["$address_length", 29]},
                                                {"$lt": ["$address_length", 64]},
                                            ]
                                        },
                                        "then": "address",
                                    },
                                    {
                                        "case": {"$eq": ["$address_length", 64]},
                                        "then": "public_key",
                                    },
                                ],
                                "default": "other",
                            }
                        }
                    },
                    "unique_addresses": {"$addToSet": "$impacted_address_canonical"},
                }
            },
            {
                "$project": {
                    "_id": 0,
                    "category": "$_id.category",
                    "count": {"$size": "$unique_addresses"},
                }
            },
        ]

!!! Explanation: impacted_address_canonical is the field that is filled with the canonical version of the address (only applicable for account addresses). The length of this field determines the category.

Note

Payday Rewards are not taken into account as this is an automated system and does not indicative of actual intent by users.

The output is a a dictionary (counts_by_category) with keys: address, contract, public_key and counts as value.

This is stored in the statistics collection as a dictionary:

{
"_id": _id,
"date": d_date,
"type": analysis.value,
"unique_impacted_address_count": counts_by_category,
}

Where analysis is:

Grouping	Analysis
Daily	statistics_unique_addresses_v2_daily
Weekly	statistics_unique_addresses_v2_weekly
Monthly	statistics_unique_addresses_v2_monthly