[13] RFR 8215937: Check usages of security-related Resources files

Weijun Wang weijun.wang at oracle.com
Fri Jan 18 02:23:03 UTC 2019



> On Jan 18, 2019, at 2:22 AM, Sean Mullan <sean.mullan at oracle.com> wrote:
> 
> This is a nice cleanup. Just a couple of comments:
> 
> - Update copyrights now that it is 2019

Will change.

> 
> - For the test, is the source code always guaranteed to be there? I was not sure if that was a requirement. Or does the test still pass if it can't find the source code?

 132     public static void main(String[] args) {
 133         if (Files.exists(SRC)) {
 134             MAP.forEach(Usages::check);
 135         } else {
 136             System.out.println("No src directory. Test skipped.");
 137         }
 138     }

But I remember asked about a similar case before and with Mach5 the src/ directory is always available.

Thanks,
Max

> 
> Thanks,
> Sean
> 
> On 12/27/18 3:11 AM, Weijun Wang wrote:
>> Please take a review at
>>    https://cr.openjdk.java.net/~weijun/8215937/webrev.00/
>> A new Usages.java test is added to make sure the strings in various Resources.java files are exactly what are used in security-related source files, no more no less.
>> Two old tests are removed. NewNamesFormat.java checks for format and Usages.java covers it. NewResourcesNames.java is a manual test and is too stale and not easy to run.
>> Several calls in keytool and jarsigner are modified to follow a more consistent calling convention (always rb.getString(string_literal)) so that they can be detected by Usages.java more easily. There are still several places in PolicyFile.java calling 'LocalizedMessage.getNonlocalized(POLICY + "...", source)' but I left them unchanged and dealt with it specially in Usages.java.
>> Many useless strings in Resources.java files are removed. I've double checked each and made sure the related calls were removed some time ago.
>> Thanks
>> Max




More information about the security-dev mailing list