Optimize report viewer dto size new

[nestabentum]

Ready for review version of PR #507.

The new file format . Consists of

• an overview.json file: it contains all data for the overview page (metadata about the JPlag run, distributions, top comparisons)
• a file submissionX-submissionY.json for each comparison. The json contains info about the matches in a comparison. This file does not contain any submission code anymore.
• a folder Submissions that has a subfolder submissionX for each submission. The subfolder submissionX contains all original files of the submission.

This file format significantly decreases the report's total size needed.
This PR should also fix crashes/lagging while reporting when handling large submissions.

[dfuchss]

Code review of the java part (not tested yet :))

[sebinside]

Here are a few comments on the code.

I also tested your version with a zip file with 50 submissions and around 1000 comparisons:

• No more file already exists exceptions, thank you
• Switching to max similarity (and back) freezes the site for about 30 seconds
• Is the distribution y-axis flipped?
• I still get an error on the comparison view, here is the log:

  Uncaught (in promise) TypeError: undefined is not iterable (cannot read property Symbol(Symbol.iterator))
  at Function.from (<anonymous>)
  at Object.eval (store.ts?502f:32:1)
  at ComparisonFactory.getComparison (ComparisonFactory.ts?d6d7:5:1)
  at setup (index.js?e1de:60:1)
  at callWithErrorHandling (runtime-core.esm-bundler.js?d2dd:155:1)
  at setupStatefulComponent (runtime-core.esm-bundler.js?d2dd:7165:1)
  at setupComponent (runtime-core.esm-bundler.js?d2dd:7119:1)
  at mountComponent (runtime-core.esm-bundler.js?d2dd:5473:1)
  at processComponent (runtime-core.esm-bundler.js?d2dd:5448:1)
  at patch (runtime-core.esm-bundler.js?d2dd:5038:1)
  

[nestabentum]

@sebinside thanks for the feedback!

• Switching to max similarity (and back) freezes the site for about 30 seconds

I will look into what causes this.

• Is the distribution y-axis flipped?

I swore I had flipped the labels it alongside the data generation.. Fixed the labels.

• I still get an error on the comparison view, here is the log:

  Uncaught (in promise) TypeError: undefined is not iterable (cannot read property Symbol(Symbol.iterator))
  at Function.from (<anonymous>)
  at Object.eval (store.ts?502f:32:1)
  at ComparisonFactory.getComparison (ComparisonFactory.ts?d6d7:5:1)
  at setup (index.js?e1de:60:1)
  at callWithErrorHandling (runtime-core.esm-bundler.js?d2dd:155:1)
  at setupStatefulComponent (runtime-core.esm-bundler.js?d2dd:7165:1)
  at setupComponent (runtime-core.esm-bundler.js?d2dd:7119:1)
  at mountComponent (runtime-core.esm-bundler.js?d2dd:5473:1)
  at processComponent (runtime-core.esm-bundler.js?d2dd:5448:1)
  at patch (runtime-core.esm-bundler.js?d2dd:5038:1)
  

That's weird I don't get an error...
What version/branch of the ReportViewer did you test? This branch?

[nestabentum]

@sebinside I don't think there is a quick fix for the slow performance when switching metrics: rendering 1000 DOM Elements (especially with Vue's v-for) is expensive. Possible solutions I found include:

• a Clusterize.js-like plugin
• Pagination

Of course, we should not build Pagination ourselves but rely on an existing implementation - we could and maybe even should adopt a UI Framework like BootStrapVue, Vuetify, etc...

[sebinside]

@nestabentum I totally understand that rendering so many elements takes time. However, the initial loading of the site (where the AVG metrics are rendered) takes significantly less time. That's why I wondered...

Regarding the error: I checked out the branch of this PR and built the web viewer locally... I might have another look at it next week

[dfuchss]

@nestabentum can you merge the current master branch into you branch. Afterwards, the SonarCloud scan will work again :)

[dfuchss]

Some minor remarks

[tsaglam]

As far as I can see, the addressed requested changes need approval by @dfuchss and @sebinside here.

[sebinside]

@nestabentum From looking at the changes, everything should be fine. However, the error in the comparison viewer still persists at least for me. I checked out a fresh version of this PR, tested multiple data sets, no success. I don't know whether this is only a problem in my installation.

So I would guess, if no one is able replicate this problem, and the merge conflicts are resolved, this PR can be merged.

[tsaglam]

So I would guess, if no one is able replicate this problem, and the merge conflicts are resolved, this PR can be merged.

There are some conflicts with the master branch. @nestabentum needs to solve them before we can merge.

[dfuchss]

As my comments are not that critical , I approve it :)

[nestabentum]

Resolved all conversations and all merge conflicts, so this is good to go.

@sebinside Very peculiar, but I'm sure eventually, we'll get to the bottom of this :)

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

13.1% Coverage
0.0% Duplication