You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As user I want the SLOC to be consistent across formats so that number are comparable.
Discussion
As pointed out in #122, the SLOC presented depends on the output --format. The initial reasoning was that including pure string lines make keeps the count similar to tools like SLOCount and cloc, allowing pygount to act as drop-in replacement. However, with the summary format, no such compatibility was necessary and a allowed my opinion of considering pure string lines to be trivial enough to consider them empty lines similar to no operations and lines containing only white characters. For more details, see the documentation chapter on "How pygount counts code".
With version 2.0, we could introduce a breaking change that would allow a more consistent behavior across format but breaking compatibility with other tools.
Goals
❓ This is a proposal open for comments.
When counting SLOC, all formats use the same number.
By default, this is the code count (excluding pure string lines).
If a user specifies the command line option --count-pure-strings, pure string lines are included in the code count.
The --format=json keep giving access to both counts.
The text was updated successfully, but these errors were encountered:
Story
As user I want the SLOC to be consistent across formats so that number are comparable.
Discussion
As pointed out in #122, the SLOC presented depends on the output
--format
. The initial reasoning was that including pure string lines make keeps the count similar to tools like SLOCount and cloc, allowing pygount to act as drop-in replacement. However, with thesummary
format, no such compatibility was necessary and a allowed my opinion of considering pure string lines to be trivial enough to consider them empty lines similar to no operations and lines containing only white characters. For more details, see the documentation chapter on "How pygount counts code".With version 2.0, we could introduce a breaking change that would allow a more consistent behavior across format but breaking compatibility with other tools.
Goals
❓ This is a proposal open for comments.
--count-pure-strings
, pure string lines are included in the code count.--format=json
keep giving access to both counts.The text was updated successfully, but these errors were encountered: