incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Rist <andrew.r...@oracle.com>
Subject Re: Need to Revisit RAT Excludes and Wildcards
Date Thu, 23 Aug 2012 16:56:52 GMT

On 8/22/2012 5:43 PM, Rob Weir wrote:
> On Mon, Aug 20, 2012 at 6:05 PM, Dave Fisher <dave2wave@comcast.net> wrote:
>> Hi Folks,
>>
>> We are really getting ahead of ourselves.
>>
>> We have a legitimate -1 IPMC vote on our release, it might get changed - Marvin asked
for confirmation that the IPMC votes are based only on the official SOURCE release.
>>
>> I looked again at the rat-excludes and we should look into a small handful of binary
files that slip through. Most are "test" files and should be OK, but there are a small number
that are not. If we have a plan to remove them and/or show that they are unused and harmless
it may be that Marvin will be convinced to change his vote.
>>
>> Someone should inspect these wildcards in the source tree
>> **/*.dbf
>> **/*.dbt
>> **/*.jar
>> **/*.zip
>> **/*.class
>> **/*.dll
>> **/*.exe
>> **/*.mdb
>>
> A quick general observation:
>
> Wildcard exclusions are dangerous.  They may be useful but they are
> dangerous.  Does RAT have a mode where we can dump a list of all
> excluded files?  Not just the patterns, but the files?
Well, you can remove the wild cards.  Then all the files are listed in 
the report.
Not making any judgements or decisions, I'll describe how the RAT 
exclusions were generated.
I started out with excludes for almost every type of file, except for 
*.c *.h *.cpp *.hpp *.java
Then as the relicensing of the files went ahead, this list of excludes 
was slowly whittled down.
RAT doesn't handle binary files, so those excludes were kept to quiet 
down the scan.

This clean-up is looking really good, and arriving at a point where 
there are no wild cards and
explanations for each excluded file would be the best outcome.  One way 
to do that would be
to remove the remaining wildcards, and then work through cleaning up the 
report.

A.
>
> If so, that might be a good thing to enable, just so we're aware when
> reviewing the report, to make sure that more is not excluded than we
> intended.
>
> If this isn't an option in RAT, maybe this is a patch we can send them?
>
> -Rob
<snip>

Mime
View raw message