How to measure code metrics ??
Danese had the question : "Do I read correctly that there are nearly 12 million lines of code on OpenOffice.org? That's a WHOLE LOT OF CODE. Is that really correct?"
The simple answer to this question is NO.
StatCVS is simply line counting on the base of the "cvs log" information. This surly gives you a rough estimate and some hints regarding the trend the code is growing, but IMHO it can't be considered as exact measurement.
So StatCVS gives you also a count of lines of comments, empty lines and even more worse, it gives you quite irregular numbers for binary files, e.g. Icon files, OpenOffice documents.
Common understanding is that the measurement of Lines of Code (LOC) is unreliable because it depends on coding style and measurement parameters so comparison of project is almost impossible.
The simple answer to this question is NO.
StatCVS is simply line counting on the base of the "cvs log" information. This surly gives you a rough estimate and some hints regarding the trend the code is growing, but IMHO it can't be considered as exact measurement.
So StatCVS gives you also a count of lines of comments, empty lines and even more worse, it gives you quite irregular numbers for binary files, e.g. Icon files, OpenOffice documents.
Common understanding is that the measurement of Lines of Code (LOC) is unreliable because it depends on coding style and measurement parameters so comparison of project is almost impossible.