# Is it possible to set a floating point precision while evaluating jldoctest blocks in docstrings?
**URL:** https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528
**Category:** General Usage
**Tags:** documenter, float, precision
**Created:** [April 18, 2021, 8:21am UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528 "2021-04-18T08:21:24Z")
**Posts on this page:** 6
**Page:** 1
### Author:  [@jishnub](https://discourse.julialang.org/u/jishnub)
#### Post date: [April 18, 2021, 8:21am UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/1 "2021-04-18T08:21:24Z")
I’m not too familiar with Documenter, so this might have a solution that I’m not aware of. Is it possible to set a filter to compare only the first `n` digits of floating-point numbers in a `jldoctest` block? Either of a global or a local setting would be fine.
---
### Author:  [@mortenpi](https://discourse.julialang.org/u/mortenpi)
#### Post date: [April 18, 2021, 8:45am UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/2 "2021-04-18T08:45:57Z")
It’s not quite perfect for this, but you can set regexes which filter out parts of the output before it’s compared: [Doctests · Documenter.jl](https://juliadocs.github.io/Documenter.jl/stable/man/doctests/#Filtering-Doctests).
---
### Author:  [@iago-lito](https://discourse.julialang.org/u/iago-lito)
#### Post date: [September 21, 2022, 5:13pm UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/3 "2022-09-21T17:13:02Z")
@jishnub I would caution against this and make use of Julia’s `≈` operator (aka. [isapprox](https://docs.julialang.org/en/v1/base/math/#Base.isapprox)) instead. Because [IEEE754](https://fr.wikipedia.org/wiki/IEEE_754) floating-point numerical precision does **not** guarantee that any calculation be correct after some _fixed_ number of decimals. Your tests would maybe appear to work at first, but they would be fundamentally broken IMO.
---
### Author:  [@cmichelenstrofer](https://discourse.julialang.org/u/cmichelenstrofer)
#### Post date: [December 1, 2022, 5:41am UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/4 "2022-12-01T05:41:56Z")
How can you use ≈ in a jldoctest?
---
### Author:  [@iago-lito](https://discourse.julialang.org/u/iago-lito)
#### Post date: [December 6, 2022, 9:31am UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/5 "2022-12-06T09:31:26Z")
@cmichelenstrofer With constructs like:
```nohighlight
julia> my_matrix ≈ [
0.58 0.55 0.53
0 0.89 1.0
1.45 22.5 -0.6
]
true
```
I guess?
---
### Author:  [@fredrikekre](https://discourse.julialang.org/u/fredrikekre)
#### Post date: [December 6, 2022, 12:31pm UTC](https://discourse.julialang.org/t/is-it-possible-to-set-a-floating-point-precision-while-evaluating-jldoctest-blocks-in-docstrings/59528/6 "2022-12-06T12:31:23Z")
>
>
> This patch adds support for using regex =\> substitution pairs in doctest filters…. This is useful when you do not want to replace the full match, but only remove some part. As an example, consider a test where the output may differ in the third decimal:
>
> \`\`\`\`
> \`\`\`jlddoctest
> julia\> 1.124
> 1.123
> \`\`\`
> \`\`\`\`
>
> With this patch it is still possible to test the first 2 decimals using the filter \`r"(\\d\\.\\d\\d)\\d\*" =\> s"\\1"\`.