On Wed, Oct 22, 2008 at 03:33:20PM -0400, James Carlson wrote:
> Stephen Ostrowski writes:
> > I was told to do hg nits, how is that different then hg pbchk ?
> 
> hg pbchk checks more things -- try "hg help pbchk."
> 
> In particular, it checks these things that "hg nits" does not check:
> 
>     comchk, tagchk, branchchk, rtichk
> 
> It also warns about uncommitted changes.

The two hg subcommands, nits and pbchk, are derived from an earlier tool
wx which was used prior to the introduction of Mercurial/hg.  When I
wrote the wx pbchk subcommand I was thinking that wx nits, which existed
before I added pbchk, was useful for checking for certain issues in the
files that had been modified by the developer and should be run while
the files were in the edit state (note, SCCS was the version control
system at the time) since the developer would likely have to modify the
files that wx nits complained about.  And any file modification should
result in a build and test so it's a good idea to run hg nits regularly
to avoid having to build/test just because hg nits complained.

wx pbchk on the other hand was supposed to be run just prior to the
putback to the gate (a Teamware command similar to hg push).  In
addition to nits checks and pbchk ran additional checks to make sure
there were no SCCS or other workspace issues that violated ON (think
Solaris) gate putback rules.  The non-nits check warnings usually did
not require file content modification and thus did not require a
build/test after fixing the issue.

-- 
Will Fiveash
Sun Microsystems               Office x64079/512-401-1079
Austin, TX, 78727              (TZ=CST6CDT), USA
Internal Solaris Kerberos/GSS/SASL website: http://kerberos.sfbay.sun.com
http://opensolaris.org/os/project/kerberos/

Reply via email to