Erik changelog 0.1.17-49

[SchSascha]

• [ ] #40 (automated index updates)
• [x] #70 (resource versioning)
• [x] #73 (extract Neo4j concept project)
• [x] #83 (outline a solution for #73)

[khituras]

Overall Report

Failed to meet time constraints.

In week 17-49 I worked on #73 (the extraction of the Neo4j concept project) as this is a prerequisite for #70. The process took much longer than original estimated because:

• We (currently) need the semedico-resource-management project which depended on semedico-core which has compiler issues right now and thus Maven refuses to build any of the projects. Circumvention via manual classpaths only helped so far (not far enough).
• I didn't want to use an older (but functional) version of semedico since the changes I make should be available in the newest version and I was afraid of merge conflicts and general confusion.
• Resolving the above issues was annoying and time consuming.
• I did create a new project with a whole new architecture apart from the semedico-resource-management tools to be freed of such issues in the future. This was a major refactoring that also took longer than expected.
• Other dependency issues came up in semedico and the julielab neo4j server plugins that are required for the gene Neo4j database.

Nontheless, I did finish the code relevant for GePi (not the whole project as used by Semedico) on friday of the 49th week. However, since the gene resources need GeNo output that is currently not stored on our HDD (I deleted them at one point in GeNo development) I currently need to build them anew. Probably new issues will come up once I actually try to build the database (maybe not, but...) so I think I will finish this at the end of Thursday, 12-12, week 50.

Plans for week 50

Until Tuesday evening: Finish #73, #70 Until Friday evening: Bring #40 into a state that Benjamin needs to do the administrative stuff (chron jobs, downloads, SLURM, new Linux on s-machines, automatic on-off, whatever).

[khituras]

Until now, I wanted to have finished #73 and #70. I did not succeed because the GeNo resource building had an error I didn't immediately recognize. Since it takes quite a while before the error comes up after starting to build the resources, I did not finish yet. New deadline for #73 and #70: Wednesday evening.

[SchSascha]

Please review changes for pull request https://github.com/khituras/gepi/pull/81.

[khituras]

Made progress on #73 and #70, not finished yet. The gene database now can be built and used employing the new stand alone database tool. Not yet a solution is there to get the resource files out of the database. There also won't be this week because I now really have to focus on my talk on friday since yesterday evening there was no time (long talks with the boss). Created new issue #83 because I actually don't think I can just come up with a realistic timeline for #73 before I have sorted out the issues mentioned in the new issue. Deadline for #83 is monday evening. Other issues are postponed until then.

[SchSascha]

Since #83 is closed (nice!) and I suppose resolved, is it possible for you to make a better estimate for #73?

[khituras]

Deadline for #73 is Friday the 23rd. Last (official) working day this year.

70 is then due to the 3rd of January.

40 is due to the 5th of January.

[khituras]

73 is done for all that GePi needs.

[SchSascha]

If so it is your honour to close the issue itself as well :-)