## 💥 Breaking changes * `miom$compute_multiregional_multipliers()` output columns renamed: `destination_country`, `destination_sector`, and `destination_label` are now `shock_country`, `shock_sector`, and `shock_label` (they identify the origin of the final-demand shock). * `miom$get_regional_interdependence()` now uses block sums from the spillover matrix for `total_spillover_out` and `total_spillover_in`. Removed `interdependence_index`; added `spillover_balance` and `spillover_export_share`.

# fio 0.1.6 ## 🐞 Bug fixes * Fix R-devel (4.5) issues (#119). * Remove CARGO_LINKER environment variable to use the correct linker for Windows (#123). * Fix `makevars.in` and `makevars.win.in` to avoid race condition when building package in parallel make (#127). ## 💥 Breaking changes * Removed `threads` slot from `iom` object (#95). * Raise minimum supported Rust version to 1.77 due to pulp crate (faer dependency) (#127). ## 🛠️ Other improvements * Fix homebrew installation instructions in README and `tools/msrv.r` (#105). * Translate `br_2020` dataset to English (#113). * Fedora 36, 37, 38 and 39 from test workflow, since CRAN runners were updated to Fedora 40 (#126). **Full Changelog**: https://github.com/albersonmiranda/fio/compare/v0.1.2...v0.1.6

## 🐞 Bug fixes * Fix R-devel (4.5) issues (#119). * Remove CARGO_LINKER environment variable to use the correct linker for Windows (#123). * Fix `makevars.in` and `makevars.win.in` to avoid race condition when building package in parallel make (#127). ## 💥 Breaking changes * Removed `threads` slot from `iom` object (#95). * Raise minimum supported Rust version to 1.77 due to pulp crate (faer dependency) (#127). ## 🛠️ Other improvements * Fix homebrew installation instructions in README and `tools/msrv.r` (#105). * Translate `br_2020` dataset to English (#113). * Fedora 36, 37, 38 and 39 from test workflow, since CRAN runners were updated to Fedora 40 (#126).

## 💥 Breaking changes * Removed `threads` slot from `iom` object (#95). ## 🛠️ Other improvements * Fix homebrew installation instructions in README and `tools/msrv.r` (#105). * Translate `br_2020` dataset to English (#113). ## 🐞 Bug fixes * Fix R-devel (4.5) issues (#119).

## 🛠️ Other improvements * Fixes system checks scripts to correctly handle errors when Rust and Cargo are not found.

## 🛠️ Other improvements * Reduce minimum supported Rust version from 1.71 to 1.67. * Set minimum version of rustc >= 1.67.1 in `SystemRequirements`. * Add system checks to `cargo` and `rustc` and prompt users either install, if missing, or update when version is lower than specified in `DESCRIPTION`.

## ✨ Enhancements * New `import_element()` to programmatically import data from an Excel file * New data import addin makes easy to import data from clipboard or an Excel file * New `iom` uses the efficiency of R6 classes * New `add` and `remove` methods to add and remove elements from an `iom` object * New `br_2020` dataset with a 51 sector Brazilian input-output table for 2020 * New `compute_tech_coeff()` and `compute_leontief_inverse()` methods uses the power of Rust and `faer` crate to substantially increase performance * New `compute_multiplier_*()` methods to compute multipliers from a given input-output table: * `compute_multiplier_output()` for the output multiplier * `compute_multiplier_employment()` for the employment multiplier * `compute_multiplier_wages()` for the wages multiplier * `compute_multiplier_taxes()` for the taxes multiplier * New `compute_influence_field()` method to compute the field of influence of each sector * New `compute_key_sectors()` method to compute the key sectors of an input-output table, based on power of dispersion and sensitivity of dispersion, and their coefficients of variations * New `compute_allocation_coeff()` method to compute the allocation coefficients of an input-output table * New `compute_ghosh_inverse()` method to compute the Ghosh inverse of an input-output table * New `compute_hypothetical_extraction()` method to compute impact on demand, supply and both structures after hypothetical extraction of a sector * New `set_max_threads()` method to allow the user to control number of threads used by {fio} (required by CRAN). ## 🚀 Performance improvements * Use Rust instead of base R for `compute_*()` functions * Use `faer` crate instead of `nalgebra` for faster linear algebra calculations * Use R6 classes for a clean, object state-changes, memory-efficient object-oriented programming ## 🛠️ Other improvements * Added assertions to check if elements imported into slots are matrices at initialization of `iom` object and at `add()` method * Added assertions to check matrices dimensions at initialization of `iom` object and at `add()` method * More informative warnings and errors messages with {cli} package * Added assertions to check if number format of slots is `double` at initialization of `iom` object and at `add()` method **Full Changelog**: https://github.com/albersonmiranda/fio/commits/v0.1.0

## ✨ Enhancements * New `import_element()` to programmatically import data from an Excel file * New data import addin makes easy to import data from clipboard or an Excel file * New `iom` uses the efficiency of R6 classes to store data reducing memory footprint * New `add` and `remove` methods to add and remove elements from an `iom` object * New `br_2020` dataset with a 51 sector Brazilian input-output table for 2020 * New `compute_tech_coeff()` and `compute_leontief_inverse()` methods uses the power of Rust and `faer` crate to substantially increase performance * New `compute_multiplier_*()` methods to compute multipliers from a given input-output table: * `compute_multiplier_output()` for the output multiplier * New `compute_influence_field()` method to compute the field of influence of each sector * New `compute_key_sectors()` method to compute the key sectors of an input-output table, based on power of dispersion and sensitivity of dispersion, and their coefficients of variations * New `compute_allocation_coeff()` method to compute the allocation coefficients of an input-output table * New `compute_ghosh_inverse()` method to compute the Ghosh inverse of an input-output table * New `compute_hypothetical_extraction()` method to compute impact on demand and supply structures after hypothetical extraction of a sector ## 🚀 Performance improvements * Use Rust instead of base R for `compute_*()` functions * Use `faer` crate instead of `nalgebra` for faster linear algebra calculations * Use R6 classes for a clean, object state-changes, memory-efficient object-oriented programming **Full Changelog**: https://github.com/albersonmiranda/fio/commits/v0.0.1-dev