I have enabled LFS with git and Bitbucket. I am now able to push a ZIP file with more than 100MB which is fantastic, but I am a little worried I will blot Bitbucket over time. The ZIP file is an export of an application component that is deployed to a workflow platform similar to Flowable.
The questions I have are the following:
- If I push a zip file, then make a small change, and push it again, what will happen? Will it push the difference, or the entire copy of the update ZIP file?
- If I merge branch B1 into B2 with the same ZIP file I mentioned above, what will happen? Will it increase the size in Bitbucket by double the size of the ZIP file or it will reuse the ZIP file in B1 to be present in B2?
- The process will repeat multiple times, that is push an updated version of the ZIP file to B1 multiple times, and then merge the changes from B1 to B2 and sometimes to B3. Do I have to be worried about bloating Bitbucket?
If the process flow above is problematic for Bitbucket, I am thinking to write a script that will use REST API to export the application component model from the workflow system, and unzip the file (it's all text in JSON and XML), and push the model to Bitbucket. Each ZIP file may contain literally hundreds of files (JSON and XML text files) that describe the application parts. The only problem with this approach is that it will take a long time, and there is a large number of changes that are not related to the programmatic features of the model. Such changes are to flag the metadata of each part within the component which is really not relevant at all to the programmatic change (application implementation details). I once did an experiment to make the application change to the unzipped files, zipped them back, and deployed the model to the workflow, and all was fine.
I appreciate your feedback.
Tarek
