On 6/13/18 9:05 PM, Jim Young wrote: > On Wed, Jun 13, 2018 at 6:06 PM, Gerald Combs <[email protected] > <mailto:[email protected]>> wrote: > > > Does anyone know what we did between July 2012 and 2013 that would > affect SLOCCount? > > > Using git bisect and defining "bad" as lower SLOC and "good" as higher SLOC > this is what I came up with: > > $ git bisect start 6d7a3a8ceef0fbcf3fc0ced1b92834c505dc135a > 5ca89cfed3aa9f0498de2783673abc4c69b4271b > . > . > . >> a37c04420322102b2c677d5d7943c7575ebc898d is the first bad commit >> commit a37c04420322102b2c677d5d7943c7575ebc898d >> Author: Pascal Quantin <[email protected] > <mailto:[email protected]>> >> Date: Thu Apr 11 18:14:53 2013 +0000 >> >> Update the ASN.1 based dissectors following the changes done in r48812 >> >> svn path=/trunk/; revision=48820 > > ;-)
OK, that's due to a6131151f5, which changed the varying outputs of our code generation scripts to a consistent "Generated automatically...". It was intended to fix licensecheck but it also ended up fixing SLOCCount's behavior. As far as I can tell neither Tokei nor cloc check file contents for autogen patterns so their numbers include automatically generated code. ___________________________________________________________________________ Sent via: Wireshark-dev mailing list <[email protected]> Archives: https://www.wireshark.org/lists/wireshark-dev Unsubscribe: https://www.wireshark.org/mailman/options/wireshark-dev mailto:[email protected]?subject=unsubscribe
