thomasabishop/eolas
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Autosave: 2024-06-21 07:30:04

Browse source
This commit is contained in:
thomasabishop 2024-06-21 07:30:04 +01:00
parent cfca55cc11
commit 65a4efc2f2
2 changed files with 40 additions and 23 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

BIN
.zk/notebook.db
View file

Binary file not shown.

63
zk/DynamoDB.md → zk/DynamoDB_data_structure.md
View file

@ -25,8 +25,9 @@ non-relational meaning there cannot be JOIN operations via
Although the data is stored as a table, one of the attributes is a primary key Although the data is stored as a table, one of the attributes is a primary key
and the rest of the attributes are effectively the "value" associated with it. and the rest of the attributes are effectively the "value" associated with it.
Because DynamoDB is schemaless, neither the attributes or their data types need Because DynamoDB is schemaless, other than the primary key, neither the
to be defined beforehand and each item can have its own distinct attributes. attributes or their data types need to be defined beforehand and each item can
have its own distinct attributes.
Each item in the table is uniquely identifiable by its primary key. Each item in the table is uniquely identifiable by its primary key.
@ -40,7 +41,7 @@ There are two types of primary key available:
are attributes. In a table that has a partition key and a sort key, it's are attributes. In a table that has a partition key and a sort key, it's
possible for multiple items to have the same partition key value. However, possible for multiple items to have the same partition key value. However,
those items must have different sort key values. You could then query by those items must have different sort key values. You could then query by
either key or both. For instance using the `personId` along with `LastName` either key or both. For instance using the `PersonId` along with `LastName`
### Secondary index ### Secondary index
@ -64,39 +65,51 @@ project:
```json ```json
{ {
"TableName": "TimeEntries", "TableName": "TimeEntries",
"AttributeDefinitions": [ "KeyAttributes": {
{ "PartitionKey": {
"AttributeName": "activity_start_end", "AttributeName": "activity_start_end",
"AttributeType": "S" "AttributeType": "S"
}
},
"NonKeyAttributes": [
{
"AttributeName": "activity_type",
"AttributeType": "S"
}, },
{ {
"AttributeName": "start", "AttributeName": "start",
"AttributeType": "S" "AttributeType": "S"
}, },
{ {
"AttributeName": "activity_type", "AttributeName": "end",
"AttributeType": "S" "AttributeType": "S"
} },
],
"KeySchema": [
{ {
"AttributeName": "activity_start_end", "AttributeName": "duration",
"KeyType": "HASH" "AttributeType": "N"
},
{
"AttributeName": "description",
"AttributeType": "S"
},
{
"AttributeName": "year",
"AttributeType": "S"
} }
], ],
"GlobalSecondaryIndexes": [ "GlobalSecondaryIndexes": [
{ {
"IndexName": "StartIndex", "IndexName": "YearIndex",
"KeySchema": [ "KeyAttributes": {
{ "PartitionKey": {
"AttributeName": "start", "AttributeName": "year",
"KeyType": "HASH" "AttributeType": "S"
}, },
{ "SortKey": {
"AttributeName": "activity_type", "AttributeName": "start",
"KeyType": "RANGE" "AttributeType": "S"
} }
], },
"Projection": { "Projection": {
"ProjectionType": "ALL" "ProjectionType": "ALL"
} }
@ -109,9 +122,13 @@ This defines the attribute `activity_start_end` as the primary key. This string
(`S`) value is a concatenation of three attributes, which is a way of ensuring (`S`) value is a concatenation of three attributes, which is a way of ensuring
each entry for the attribute will be unique. each entry for the attribute will be unique.
I also define two GSIs. -- check that these are correct for getting entries by The `NonKeyAttributes` are all the other attributes in addition to the primary
range! key. As mentioned these do not actually need to be defined when setting up the
table but they are listed here for clarity.
## Usage I have also defined a GSI. This is derived from the `Year` attribute. This will
group all the items by their `Year`, allowing me to query directly by year but
also helping to chunk the entries which will make look-ups quicker and less
expensive.
## Related notes ## Related notes