# Working with UTF-8 encoded data

**URL:** <https://fortran-lang.discourse.group/t/working-with-utf-8-encoded-data/10414>\
**Category:** Announcements\
**Created:** [September 23, 2025, 12:12pm UTC](https://fortran-lang.discourse.group/t/working-with-utf-8-encoded-data/10414 "2025-09-23T12:12:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![urbanjost](https://avatars.discourse-cdn.com/v4/letter/u/0ea827/32.png) [@urbanjost](https://fortran-lang.discourse.group/u/urbanjost)\
**Post date:** [September 23, 2025, 12:12pm UTC](https://fortran-lang.discourse.group/t/working-with-utf-8-encoded-data/10414/1 "2025-09-23T12:12:39Z")

</div>

An alpha release of [GitHub - urbanjost/M\_unicode: Unicode support when ISO\_10646 is not supported](https://github.com/urbanjost/M_unicode) provides a user-defined type that can be used with ragged arrays of ASCII or UTF-8 encoded data. It does not require the compiler to support the Fortran Unicode extension but provides overloading for all the basic operators and character intrinsics with both a procedural and OOP interface. The UPPER() and LOWER() funtions support the concept of case for the Unicode Latin characters not just the ASCII subset, and a basic SORT() function provides for ordering the data by Unicode codepoint values. Documentation and examples are still a WIP but complete enough to guide usage for anyone interesting in trying it. Only tested with ifx and gfortran so far but until proven otherwise I think it should work with any environment where UTF-8 files are supported. So far that includes allowing what-you=see-is-what-you-get string constants on Linux and Cygwin at a minimum as well. It needs a DT and more CD/CI unit testing but builds easily with `fpm`. The intrinsic overloads include TOKENIZE and SPLIT. It is just an alpha release but should be useful for anyone working with UTF-8 data, particularly if the compiler does not support the UCS-4 extensions of Fortran.