search

vazexqi / CodingSpectator

Watches and analyzes code edits in the Eclipse IDE non-invasively
http://codingspectator.cs.illinois.edu
Other
20 stars 14 forks source link

Estimate the number of hours of using CodingSpectator #321

Closed reprogrammer closed 13 years ago

reprogrammer commented 13 years ago

It would be nice to report the number of hours that our data covers. Is it possible to analyze CodinTracker logs to estimate the number of hours that every participant has programmed under the presence of CodingSpectator?

reprogrammer commented 13 years ago

@Wanderer777, @vazexqi: We can also consider the UDC data to estimate the number of hours of development.

reprogrammer commented 13 years ago

@Wanderer777: Thanks for providing UsageTimeAnalyzer. How can we use this analyzer?

Wanderer777 commented 13 years ago

@reprogrammer: You can use UsageTimeAnalyzer exactly the same way as you use RefactoringIntensityAnalyzer, i.e. specify the root folder and then run it as a JUnit Plugin Test. The combined result will appear in the specified root folder, the file's default name is combined.usage_time.

reprogrammer commented 13 years ago

@vazexqi, @Wanderer777: I suggest that we change the name of the column USAGE_TIME to USAGE_TIME_IN_MILLI_SECS to make the unit of measurement clear.

reprogrammer commented 13 years ago

@Wanderer777: I renamed USAGE_TIME to USAGE_TIME_IN_MILLI_SECS in the SQL script (See commit f2e39ec9d66dd8920d8869b98776d8924b89256e). I think you also need to update the column name of the CSV file that your analyzer generates accordingly.

Wanderer777 commented 13 years ago

@reprogrammer: The column name in the CSV file generated by the analyzer is 'usage time (ms)'. And as far as I can remember, this column always had this name. So, it never matched the SQL script name. Therefore, I am not sure why we need to match it now (its current name looks sufficiently descriptive to me).