[Core] archive extraction under windows/linux does not work

[peshkira]

Extracting archives under windows and linux seems not to work, as the temporary directory is not being properly created. This might be a permission problem and has to be tested

[artourkin]

Extraction works fine under Ubuntu.
The performances of processing zipped and raw files remain the same. No x10 speedup detected.

[peshkira]

this means, it was tar that did the magic? Any other ideas? How about, if you rerun twice on the same data? Is there any speedup?

[artourkin]

1. My version of c3po processes zipped files in iteration. Each iteration it extracts next 50 zips and puts data (~ 0.5GB) into mongo. Obviously, the linux cache has limitations, but I have no clues about it's size.
   I tried out the following:
   $ free -h
   total used free shared buffers cached
   Mem: 7.7G 6.2G 1.5G 0B 796M 2.4G
   -/+ buffers/cache: 3.1G 4.7G
   Swap: 18G 288K 18G

So,possibly extracting all the files at once will not work.

2. Rerunning c3po 0.5.0 on the same raw data with cleaning mongodb gives different execution time of the same order (again no x10 speedup). In some cases it processes different amount of files. Needs further inspection.

[proteann]

Hi @peshkira,
First of all, great job with c3po. I tried using the c3po on windows and I am not sure if it works - I do not find the c3po-cmd.jar, do I have to build and compile myself or is there way out? . Have the aforementioned bug been fixed? What is the remedy that you guys suggest?

Update: I used Maven to build the project as detailed in the Developers guide and a have c3po-cmd-0.5.0-SNAPSHOT-jar-with-dependencies.jar file, that I am using for now for running!

Thank you
Harish

[peshkira]

Thank you for your kind words, @proteann. I am glad that you have been able to resolve the problem. Unfortunately, I am no longer supporting c3po.

Maybe @artourkin can give you some more insight about the project and whether it is still developed/supported by the UT Vienna or OPF? Thanks @artourkin

best