Really cool review of the RPZ format and schema! You seem to have figured most everything out on your own, but I just thought I'd point to the developer documentation (I just added some information about the trace db vs config file after reading your remarks -- thanks!) which includes the database schema.
You were right to note that some information is only in the database, and some only in the config file. The configuration does not need to include all the details from the trace of course (we include the trace anyway, for tools like reprounzip-graph, but the packer and unpackers only need the config), however there is also information that is only in the config because it is inferred once the experiment is done (e.g. which package do files come from, whether a file is read or written by the overall experiment, OS information, and resolving symlinks).
Hi @tmcphillips,
Really cool review of the RPZ format and schema! You seem to have figured most everything out on your own, but I just thought I'd point to the developer documentation (I just added some information about the trace db vs config file after reading your remarks -- thanks!) which includes the database schema.
You were right to note that some information is only in the database, and some only in the config file. The configuration does not need to include all the details from the trace of course (we include the trace anyway, for tools like reprounzip-graph, but the packer and unpackers only need the config), however there is also information that is only in the config because it is inferred once the experiment is done (e.g. which package do files come from, whether a file is read or written by the overall experiment, OS information, and resolving symlinks).
Again, thanks for the review!