Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Node.js Best Practices - Decimals and Big Integers #1010

Draft
wants to merge 1 commit into
base: main
Choose a base branch
from
Draft
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 40 additions & 0 deletions node.js/best-practices.md
Original file line number Diff line number Diff line change
Expand Up @@ -367,6 +367,46 @@ The following articles might be of interest:



## Decimals and Big Integers

JSON numbers cannot reliably capture 64-bit binary format IEEE 754 values.
That is, they loose precision for numbers with more than 16 digits and/or when converting to binary, for example, for a calculation:
```
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
```
```txt
> 12345678901234567
12345678901234568
> 0.1+0.2
0.30000000000000004
```

Please add a language tag. For plain text add txt as language tag.

> 12345678901234567
12345678901234568
> 0.1+0.2
0.30000000000000004
```

By default (= new odata adapter + @cap-js database), `cds.Int64` are selected as strings but `cds.Decimal` are selected as numbers.
The rationale is that `cds.Int64` are more commonly "big numbers", whereas decimals are mostly in the safe range.
However, you can instruct the @cap-js databases to select decimals as strings as well via `cds.env.features.string_decimals = true`.
TODO: change name?

Inbound, the new (and default) protocol adapters accept numbers in decimal notation (`1.23`), numbers in exponential notation (`1.23e1`), as well as strings in decimal notation (`'1.23'`).
The are taken as is and __not__ normalized.
Hence, your custom code must be able to handle both numbers and strings:
```
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
```
```txt
> 1 + '1'
'11'
> Number(1) + Number('1')
2
```

Please add a language tag. For plain text add txt as language tag.

> 1 + '1'
'11'
> Number(1) + Number('1')
2
```

Behavior of remote systems is inconsistent:
- OData v2 always sends strings
- OData v4 may ignore a IEEE754Compatible instruction
→ behavior in production may differ from behavior of local mocking
→ data from remote mixed into local data may result in inconsistencies

Summary:
- you must be able to deal with numbers and strings
- `cds.env.features.string_decimals = true` to select decimals as strings
- `cds.env.features.ensure_ieee754 = true` ensures that a `cds.ApplicationService` returns strings for decimals and big integers
+ TODO: promote? was meant as compat for data returned by custom handlers while we always select as string...



## Timestamps

When using [timestamps](events#timestamp) (for example for managed dates) the Node.js runtime offers a way to easily deal with that without knowing the format of the time string. The `req` object contains a property `timestamp` that holds the current time (specifically `new Date()`, which is comparable to `CURRENT_TIMESTAMP` in SQL). It also stays the same until the request finished, so if it is used in multiple places in the same transaction or request it will always be the same.
Expand Down
Loading