NEWS

PlantTxDbHub 0.99.4

Added getTxDbPath() to retrieve the local file path of a cached TxDb database for a given species identifier. This simplifies loading with GenomicFeatures::loadDb().

The package no longer depends on AnnotationHub. Databases are now downloaded directly from the URLs specified in the metadata CSV file (inst/extdata/metadata.csv). Users should call downloadPlantTxDbs() instead of relying on AnnotationHub queries.

Added listPlantTxDbSpecies() to display available species identifiers and their corresponding SQLite file names.
downloadPlantTxDbs() gained a species argument, allowing users to download a subset of databases (e.g., species = "Arabidopsis_TAIR10").
Added a download parameter to downloadPlantTxDbs() (download = FALSE) to allow examination of the destination path without initiating a download.
The download function now automatically re‑fetches corrupted SQLite files, using a new internal helper is_valid_sqlite() that validates the database integrity.
The list of available species and their download URLs is now maintained in a single CSV file (inst/extdata/metadata.csv). This file can be easily extended by contributors without modifying R code.
A contributing guide is included in the vignette, explaining how to add new plant TxDb databases.

Downloads are now more robust: when curl is available it is used with a configurable timeout; otherwise download.file() is used with a raised timeout.
The default cache directory uses tools::R_user_dir("PlantTxDbHub", "data"), keeping files in a persistent but unobtrusive location.
Vignette updated to reflect the new workflow, including installation, species listing, selective download, and contribution instructions.

Fixed a timeout issue when downloading the large Glycine max database by increasing the default timeout to 30000 seconds and adding curl support.
Removed non‑ASCII characters from source files to pass R CMD check without warnings.
Fixed undocumented dependencies (GenomicFeatures, GenomeInfoDb) by adding them to Suggests.

First release of the package.
Provided TxDb annotation databases for three plant species via AnnotationHub (concept only; not yet functional in the current release).
Included a vignette demonstrating retrieval and use of TxDb objects.