EduSweep 2.6.2 Released

[bizzel]

The second maintenance release for EduSweep 2.6 is available. This update contains further improvements to the user interface, especially around list sorting and selection and it fixes a handful of bugs.

 

If you haven't used the application before then see this overview page that summarises the features: http://www.edugeek.net/forums/edusweep/201485-edusweep-overview-links-downloads.html

 

Full release notes at: https://github.com/paulbeesley3/EduSweep/releases/tag/2.6.2

 

EduSweep 2.6.2 Installer

 

Thanks for using EduSweep!

[johnabbots]

Thanks for all your hard work I've checked it out and have one suggestion for the next release if this is possible?

 

I searched our shared drive for a swf game and it returned a lot of results and I choose to quarantine all of the examples of the same file. The quarantine must put them all in one place but not with the folder names as it will only quaratine one example of the same file name if that makes sense before it comes up with an error that the file already exists.

 

 

Cheers

[bizzel]

That's interesting. I think that in older versions there was some renaming done when a file got moved into quarantine but perhaps that was lost when everything was overhauled for 2.6. I'll take a look into it.

[bizzel]

I took a look this evening. There was/is code to deal with this kind of thing, so files with the same name should have been renamed with a numbered suffix on import: file, file (1), file (2), and so on.

 

I was unable to recreate the issue on my dev machine - multiple files with the same names but from different directories were quarantined fine, with the names I expected. My best guess is that there is some kind of timing issue where the filename check ("what should I name this on import?") for the second file completes before the first file has been fully moved. That would cause the second file's name to be the same as that of the first. This is much more likely on a network home drive or a slow mechanical drive, I imagine, due to the latencies involved. Over here I'm testing on an SSD which isn't really the best comparison.

 

I have some ideas though. First, I've changed the behaviour of the import process a bit in this patch: https://github.com/paulbeesley3/EduSweep/commit/2d80aa96e454b65f3b1b22706725085594518157 . This should be a pretty safe change, and backwards compatible too.

 

If you'd like to give it a try then that would really help me out. The installer exe is here: https://ci.appveyor.com/project/paulbeesley3/edusweep/builds/23580015/artifacts . This is exactly the same as 2.6.2 otherwise, no other changes. You'll need to remove the existing EduSweep installation first but all your settings and files will be preserved.

Edited April 3, 2019 by bizzel

[johnabbots]

@bizzel excellent this has done the trick and now all is working amazing!

[johnabbots]

Just thinking now the only way this could be improved for me would the ability to schedule the running of tasks so when the lovely darling copy the files back the next night or whenever they get removed again.

 

Is this possible perhaps in a later release?

[bizzel]

Just thinking now the only way this could be improved for me would the ability to schedule the running of tasks so when the lovely darling copy the files back the next night or whenever they get removed again.

 

Is this possible perhaps in a later release?

 

That's a bit of a larger change ;-) It's on the roadmap but quite a way off. My plan, in a rough order, is:

 

 

1. Use a database (self-contained or an existing server on your network) in 2.7 for tasks, reports, quarantine, etc. This should improve performance quite a lot compared to storing and retrieving lots of objects from JSON files. It's also needed to let the task results tab handle thousands of files without locking up.
   
2. Enhance the scan engine with new features using the Analyzer interface (I shipped this silently in 2.6.2). Analyzers act on files that have already been detected, before they are displayed in the results summary. Examples would be: calculating file checksums, reading MIME types, extracting EXIF tags from images, etc. Think of the old File Inspector integrated into the scan results.
   
3. Overhaul the scan engine with rule based detections. This allows composite queries like: "detect all files that are over 30MB, that are images or videos, but don't have the word 'backup' in the name". Regular style queries aren't going anywhere though.
   
4. Background service mode? Regular filesystem indexing and report production on a schedule? This would need further database enhancements and a new UI at the least so it's several months of hobby work.
   

Hope that gives you some interesting insight into the pipeline, if nothing else. I should write this up in a post at some point.

[johnabbots]

Well I've had great fun today quarantining 25k flash games so its certainly not slow :-)

 

Thanks again!

[Fazza]

I'm getting Page not found on GitHub - it looks like all references to EduSweep have been removed/deleted

[lmgtfy]

I'm getting Page not found on GitHub - it looks like all references to EduSweep have been removed/deleted

I can probably hunt out the download if you still want it?