fgof-cache Public

Notifications Fork 0 Star 0
da83934
Branches trunk
1 Branches 0 Tags
Go to file T
Code
espadonne Build cache roots
da83934 5 Commits
.github Add cache CI
src Build cache roots
test Build cache roots
.gitignore Ignore local docs
README.md Polish cache readme
fpm.toml Scaffold cache package

fgof-cache

Disk cache helpers for modern Fortran.

fgof-cache is intended to be a small, standalone library for cache directories, cache entries, and repeatable on-disk storage flows that CLI tools and local developer tooling keep hand-rolling.

It is part of the FortranGoingOnForty lib-modules catalog, but it is intended to stand on its own as a normal fpm package.

Current v1 target:

  • stable cache option and entry types
  • predictable cache-root resolution and initialization
  • explicit cache key and path helpers
  • safe write-through and cleanup semantics built on fgof-temp

Future scope:

  • TTL and invalidation helpers
  • metadata-aware cache indexes
  • cache pruning and size-budget policies

Status

Initial scaffold is in place.

Tracked today:

  • public fgof_cache and fgof_cache_types modules
  • stable cache constants and placeholder types
  • scaffold coverage in fpm test
  • CI on macOS and Ubuntu

Why Use It

  • cache logic shows up in tools, build helpers, editors, and local services
  • most projects still re-solve naming, layout, and cleanup rules themselves
  • a focused cache library can build naturally on the released temp and fs layers

Public API Shape

Primary modules:

  • fgof_cache
  • fgof_cache_types

Public types:

  • cache_options
  • cache_entry

Public constants:

  • FGOF_CACHE_OK
  • FGOF_CACHE_ERR_INVALID_OPTIONS
  • FGOF_CACHE_ERR_NOT_FOUND
  • FGOF_CACHE_ERR_IO
  • FGOF_CACHE_ERR_INTERNAL

Current public procedures:

  • clear_cache_options
  • clear_cache_entry
  • cache_backend_name
  • cache_error_name

Build And Test

fpm test

That is the baseline verification command locally and in CI.

Supported Platforms

  • macOS
  • Linux

Boundaries

  • intended to stay independently versioned and releasable
  • focused on cache ergonomics, not full database or state management
  • should stay useful on its own even if future state-oriented packages build on top

License

MIT