# DebugDataWriter.jl: simple formatted writer of debug information

**URL:** <https://discourse.julialang.org/t/debugdatawriter-jl-simple-formatted-writer-of-debug-information/96800>\
**Category:** Package Announcements\
**Created:** [March 29, 2023, 7:36pm UTC](https://discourse.julialang.org/t/debugdatawriter-jl-simple-formatted-writer-of-debug-information/96800 "2023-03-29T19:36:00Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![rssdev10](https://avatars.discourse-cdn.com/v4/letter/r/e9a140/32.png) [@rssdev10](https://discourse.julialang.org/u/rssdev10)\
**Post date:** [March 29, 2023, 7:36pm UTC](https://discourse.julialang.org/t/debugdatawriter-jl-simple-formatted-writer-of-debug-information/96800/1 "2023-03-29T19:36:00Z")

</div>

The package [DebugDataWriter.jl](https://github.com/rssdev10/DebugDataWriter.jl) provides writing of debug information into external files in a human readable format.

```julia
@debug_output path_id file_title data_to_be_formatted

```

If the output disabled, there is no overhead in a program execution. Only when `DebugDataWriter.config().enable_dump = true`, the module generates output.

Example:

```julia
    id = get_debug_id("Some query")
    @debug_output id "some complex structure" ones(2, 3)
    @debug_output id "some structure as lambda" begin
        zeros(5, 2) # here we can put some code for preparing data for output
    end

```

`id` is used as a name of further output sub-directory of `debug_out`. That name will be generated based on current date/time and the provided string in file name allowed format. All non alpha-num symbols are translated into the `_` character. The strings as a second argument of the macro `@debug_output` gives a name of the file for data output. Default data formatter is JSON.

The data for debug output might be provided as a literal, a variable or a lambda function. The lambda-function will be activated if only `enable_dump` is `true`.

```julia
    matrix = ones(2, 3)
    title = "table from variable from title variable"
    @debug_output id "some complex structure" matrix
    @debug_output id title matrix

    @debug_output id "some structure as lambda" begin
        zeros(5, 2)
    end

```

Implemented formats are JSON with JSON.jl, HTML and TXT with PrettyTables.jl, and SVG, XML as raw data output.

```julia
    @debug_output id "text table" ones(2, 3) :JSON
    @debug_output id "text table" ones(2, 3) :TXT
    @debug_output id "HTML table" ones(2, 3) :HTML
    @debug_output id "HTML table from variable" matrix :HTML
    @debug_output id title matrix :HTML  

```

The list of serializers/formatters might be modified.

Output files are present in the path `debug_out`. Default prefix name contains full date-time. If the parameter `DebugDataWriter.config().path_format_fulltime = false`, then HEX-time in seconds will be used as a prefix. So, directories based on the `id` might be naturally alphabetically ordered accordingly to the time when the `id` was generated. Debug output with the same `id` means creation of multiple files in the same directory.

```plaintext
    ├── debug_out
    │ ├── 1872ecb296f_Another_query
    │ │ └── with_details.json
    │ ├── 20230329-221437-037
    │ │ └── array.json
    │ ├── 20230329-221437-037_Another_query
    │ │ ├── HTML_table.html
    │ │ ├── HTML_table_from_variable.html
    │ │ ├── some_SVG.svg
    │ │ ├── some_XML.xml
    │ │ ├── some_complex_structure.json
    │ │ ├── some_structure_as_lambda.json
    │ │ ├── table_from_variable_from_title_variable.html
    │ │ ├── table_from_variable_from_title_variable.json
    │ │ ├── text_as_a_text.txt
    │ │ ├── text_table.json
    │ │ └── text_table.txt
    │ └── 20230329-221437-037_query_with_something_interesting
    │ └── some_dict.json

```