Hello everyone!
It has been a bit since last time, I post in hopes that everyone is blessed and can only remember the good times!
I installed Squeak5.3rc2-19422-64bit and ran the shared image, selected Solarized (Dark) then selected save. I added a local repo and Cryptography repo and loaded Cryptography-rww.115.mcz and ran the tests. Some failures with DateAndTime I fixed (nanos disappeared) so some method call in Cryptography, ASN1UTCType was changed. Then I ran into a problem with MultiByteFileStream>>#primSize: which failed some more Cryptography tests but also prevents me from seeing the Cryptography changes I have just mentioned. I can't determine the cause of primSize: failure, so I thought to report to you the issue. I appreciate any advice in helping validate & pass Cryptography tests. Could these tests be run by the CI server? That would be most helpful, to catch early issues.
Kindly, Rabbit
Hello all,
I was able to switch uses of #clone to #shallowCopy. I updated my DateAndTime>>#milliseconds method away from using the nanos ivar (which went away) to be
#milliseconds ^ self utcMicroseconds // 1000
I have a remaining issue that the size of the DateAndTime encoded to ASN1 has changed. I am looking into this issue with a modification back to the expected output. I believe this to be the last remaining issue with Cryptography.. A new version of the Cryptography package will be released for Squeak 5.3.
K, r
On 2/28/20 1:58 PM, Robert via Squeak-dev wrote:
Hello everyone!
It has been a bit since last time, I post in hopes that everyone is blessed and can only remember the good times!
I installed Squeak5.3rc2-19422-64bit and ran the shared image, selected Solarized (Dark) then selected save. I added a local repo and Cryptography repo and loaded Cryptography-rww.115.mcz and ran the tests. Some failures with DateAndTime I fixed (nanos disappeared) so some method call in Cryptography, ASN1UTCType was changed. Then I ran into a problem with MultiByteFileStream>>#primSize: which failed some more Cryptography tests but also prevents me from seeing the Cryptography changes I have just mentioned. I can't determine the cause of primSize: failure, so I thought to report to you the issue. I appreciate any advice in helping validate & pass Cryptography tests. Could these tests be run by the CI server? That would be most helpful, to catch early issues.
Kindly, Rabbit
Heyho! :-)
A new version of the Cryptography package will be released for Squeak 5.3.
This is great news! Thank you for looking into it.
Best, Marcel Am 02.03.2020 17:13:25 schrieb Robert via Squeak-dev squeak-dev@lists.squeakfoundation.org: Hello all,
I was able to switch uses of #clone to #shallowCopy. I updated my DateAndTime>>#milliseconds method away from using the nanos ivar (which went away) to be
#milliseconds ^ self utcMicroseconds // 1000
I have a remaining issue that the size of the DateAndTime encoded to ASN1 has changed. I am looking into this issue with a modification back to the expected output. I believe this to be the last remaining issue with Cryptography.. A new version of the Cryptography package will be released for Squeak 5.3.
K, r
On 2/28/20 1:58 PM, Robert via Squeak-dev wrote:
Hello everyone!
It has been a bit since last time, I post in hopes that everyone is blessed and can only remember the good times!
I installed Squeak5.3rc2-19422-64bit and ran the shared image, selected Solarized (Dark) then selected save. I added a local repo and Cryptography repo and loaded Cryptography-rww.115.mcz and ran the tests. Some failures with DateAndTime I fixed (nanos disappeared) so some method call in Cryptography, ASN1UTCType was changed. Then I ran into a problem with MultiByteFileStream>>#primSize: which failed some more Cryptography tests but also prevents me from seeing the Cryptography changes I have just mentioned. I can't determine the cause of primSize: failure, so I thought to report to you the issue. I appreciate any advice in helping validate & pass Cryptography tests. Could these tests be run by the CI server? That would be most helpful, to catch early issues.
Kindly, Rabbit
squeak-dev@lists.squeakfoundation.org