Hi, Anatoly. I don�t believe there is a mechanism to maintain two versions of a data package in a single Bioconductor release.
Consider refactoring your database to expose versioned views to the user rather than tables. All end-user data in the database are exposed as �versioned views� (create view as �) of the underlying database tables. The database can then store any tables you like and you just need to ensure that v1_�. and v2_� do not ever change the information returned to the user. If you add more tables or information in the future, ensure that the v1_� and v2_� return the same information as before, but you can move on to v3_�. In this way, the �data� are versioned by the package version (the information in the backing tables is versioned by the database file), but the schema are versioned inside the database. Of course, you can write your R code, then, to check that the database contains the correct version to ensure that a user doesn�t try to use an old version that doesn�t have the correctly versioned views. If there is R code meant for v2, it will work for all future versions of the database, but R code written for v4 will not work with a database (perhaps a historical copy) that doesn�t have v4 views; this would allow a user to run code using the �old� database and repeat the code using the �new� database with new data with the code being unchanged. There are, of course, other approaches to consider. Happy to discuss further�. Sean From: Bioc-devel <bioc-devel-boun...@r-project.org> on behalf of Anatoly Sorokin <lpto...@gmail.com> Date: Tuesday, August 19, 2025 at 10:46 To: bioc-devel@r-project.org <bioc-devel@r-project.org> Subject: [Bioc-devel] New database for AnnotationData package Hi, I have a question: we have developed the new version of the database for our synaptome.data/synaptome.db packages. This version has a different API and an extended set of data; however, we would like to preserve access to the old database because some results would look different between versions. Is it possible to keep the old version of the package in the new release, together with the new one? If not, what would be the Bioconductor policy about this? Cheers, Anatoly _______________________________________________ Bioc-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel [[alternative HTML version deleted]]
_______________________________________________ Bioc-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/bioc-devel
