On Tue, 20 Sep 2016 08:28:24 -0500, Kevin Minerley wrote: >The packager admits to using 7-zip (yes, we all figured that out). > >Someone had suggested a "finger check" -- bingo. > >I've asked the to repackage and test it in Windows before uploading the >replacement. > Thanks for your quick action. A few thoughts:
o "Admits" is a slightly biased word. 7-Zip is not the problem. By experiment (a scientific sample of one test case) it works fine if one chooses the right options. o A visual inspection of the 7-Zip display of the generated archive should show no compression methods other than "stored" and "deflated". o For my experiment I skimmed the 7-Zip doc. It mentions a command line interface. That should imply that it's scriptable. A script makes the process more repeatable, even if it's only a copy-and-paste from instructions on the screen. o Test (also) on a platform other than Windows. It eliminates one more bias. (Perhaps "jar" on z/OS 2.2 was grandstanding, but I chose tne most alien test environment available.) Likewise, if the human resource is available, the tester should be a person other than the packager. Thanks again, gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
