RFR 8193255: Root Certificates should be stored in text format and assembled at build time

[Weijun Wang]

Sure. How many info do you want to see?

I can prepend `keytool -printcert` but that's too much. At least I think the extensions part is not needed. Also, I don't wish people reading the fingerprint inside as genuine and does not calculate it from the cert itself.

So, I'm thinking of

Owner: CN=XRamp Global Certification Authority, O=XRamp Security Services Inc, OU=www.xrampsecurity.com, C=US
Issuer: CN=XRamp Global Certification Authority, O=XRamp Security Services Inc, OU=www.xrampsecurity.com, C=US
Serial number: 50946cec18ead59c4dd597ef758fa0ad
Valid from: 1 Nov 2004 17:14:04 GMT until: 1 Jan 2035 05:37:19 GMT
Signature algorithm name: SHA1withRSA
Subject Public Key Algorithm: 2048-bit RSA key
Version: 3

Is that OK?

Thanks,
Max

p.s. `keytool -printcert` shows validity in local timezone. Does not look good to me.

> On May 31, 2019, at 6:51 AM, Sean Mullan <sean.mullan at oracle.com> wrote:
> 
> One suggestion is to put a printable form of the contents of the certificate at the top of each of the PEM files. It would be nice as a quick-look to see what is in the certificate. Of course, you can also use keytool -printcert to do that, but if I am just perusing the source code via a browser or something like that, it would be nice to not have to do that.
> 
> --Sean
> 
> On 5/30/19 9:01 AM, Weijun Wang wrote:
>> Please take a review at
>>    http://cr.openjdk.java.net/~weijun/8193255/webrev.00/
>> Please pay attention to the 1st 3 and the last 2 files. Others are PEM files for all certs inside the original cacerts.
>> There is one thing I cannot get correct. If I update the GenerateCacerts.java file and rerun make, the cacerts file is unchanged. I thought the following line
>>    $(GENDATA_CACERTS): $(BUILD_TOOLS) $(GENDATA_CACERTS_SRC)
>> means when when the tool is changed, GENDATA_CACERTS will be called.
>> Thanks,
>> Max