Bug: Exported cc.json include data that is not supposed to be there

Bug

Expected Behavior

• The node of "type": "Folder" has empty "attributes": { }
• The node of "type": "File" does not have the attribute unary
• The node of "type": "File" does not have the "origin": "sample1.cc.json"

The visualization is not supposed to modify the "nodes": [] or "edges": []. These are generated by the analysis. More specifically the problems are:

• A folder is not supposed to have any attributes. The Visualization is supposed to either sum or average the "file"s inside a “folder” but should not store these values in the cc.json.
• The unary attribute is a virtual attribute that is 1 for each “file”. It’s not supposed to be in the cc.json.
• The origin is very useful to differentiate in Multiple Mode where the file came from. It’s weird to have it when I opened the file in Single Mode.

Now for the bigger question: If I had openend multiple files in Multiple Mode should the “origin” be there for each “file”? I don’t think it should. Currently the download button in Multiple Mode can be used to merge files but that was never the intention of *Multiple Mode. Merging files is supposed to be done by analysis and the merge filter. I think we should disable the download button for Multiple and Delta Mode because it’s unclear what the resulting file will contain.

Actual Behavior

When I save a cc.json from the visualization it includes data it’s not supposed to include:

• The node of "type": "Folder" has "attributes": { "unary": 4, "avgCommits": 146, "rloc": 600, ... }
• The node of "type": "File" has this "attributes": { "unary": 4 }
• The node of "type": "File" has this "origin": "sample1.cc.json"

Steps to Reproduce the Problem

1. Open CodeCharta with mini-map
2. Press the download button
3. Press save

Specifications

• Version: 1.32.0
• Platform: Web
• Subsystem: Visualization
• Browser: Firefox

[fatihi]

I’m with @alschmut here, the file is not intended to be read manually. If someone still wants to do that, there are programs that format json automatically.

[alschmut]

Short side note: I imported a cc.json file with 1.5MB after using the analysis and when I exported that map, it had 8.6MB. So while removing duplicated data with this issue, we should also consider minimizing the file size by removing whitespace/line breaks etc.

With this issue implemented, I just shortly wanna mention, how much smaller in size our exported files just got. When I imported the same 1.5MB cc.json file and exported it afterwards, it had surprisingly just 1.3MB. I don’t have an answer to why it is even smaller than 1.5MB, but I guess its a pretty good improvement from exporting a file which had 8.6MB before.