Hi, thanks for the feedback. It is very welcome!
Gruss Richard Am Donnerstag, den 25.03.2021, 19:53 -0700 schrieb David Blevins: > > On Mar 25, 2021, at 7:53 AM, Zowalla, Richard < > > [email protected]> wrote: > > > > Hi David, Hi Jon, > > > > thanks for your replies. > > > > I just pushed some (short term) fixes to the tomee-site-generator: > > > > (a) Added an (experimental) generator to create the download page > > based > > on downloads.apache.org or archive.apache.org (rather than using > > repo.maven.apache.org), so we can be sure, that the links are > > working. > > > > (b) I updated the downloads-ng.adoc with (1) the feedback by Sebb > > and > > Romain from [1] - so it is now a bit "cleaner" (imho). > > > > (c) I updated and automatically converted the previous download- > > archive.adoc to ensure, that the links are working again. We had > > some > > mismatches related to "http" for the repo.maven.apache.org links > > for > > very old releases and some broken links due to the request by INFRA > > to > > remove very old relases from the mirroring system. Old releases, > > which > > previously linked to the mirrors are now linking to > > archive.apache.org. > > > > I will try to update https://github.com/apache/tomee-site-pub with > > the > > changes as well. > > > > As stated above: This is only a "short term" fix to comply with the > > infra requirements. I will respond on the other mail as well. > > Took a quick look and love the clean and direct approach you have in > the code! I'm a big fan of static inner classes to keep code tight > and easy to digest. > > Some style advice, get in the habit of using the `final` keyword on > all fields, variables and parameters. It'll look strange at first, > but you'll get used to it and it'll change the way you code in a good > way. > > For example these variables should be marked final and not > initialized immediately: > > - > https://github.com/apache/tomee-site-generator/commit/27a46ae7d958867431398f86abf1ed55df4a3f74#diff-30e729b63bba575db7b6e9681e02c8ff1faf81f33d27c610c2bbeb4fe333cc23R149-R151 > > Once you did that all your code would work ok, minus this chunk which > the compiler would force you to deal with differently. You'd be > forced to handle the situation better than simply a System.out no one > will read. > > - > https://github.com/apache/tomee-site-generator/commit/27a46ae7d958867431398f86abf1ed55df4a3f74#diff-30e729b63bba575db7b6e9681e02c8ff1faf81f33d27c610c2bbeb4fe333cc23R162-R164 > > If it's not going to work, better it really doesn't work and just > throw an exception with "this should never happen." That way the > future Jenkins job will appear red, not green, and we don't update > the website with any potentially invalid content. Instead of someone > coming to the website and seeing links with no names, they'd simply > see the old-but-valid links. The developer who attempted the website > update can just as easily determine the update didn't work by > noticing the links are old as they can by the links are > invalid. However if they don't notice by looking at the website, > they can notice by seeing failed Jenkins job. Until they notice, the > website stays valid (assuming the release wasn't removed from the > mirror of course). > > Moreover, if someone who is not you tried to use this code and that > "should never happen" situation occurred, they'd have no exception or > stack trace to immediately point them in the right direction. They'd > have to be keen to spot "This should not happen. New extensions?" > among the output, know it is significant, and search for the line in > the code where it occurs. Often an exception does occur, just later > in the code and that often means people more easily miss the "This > should not happen" output, because their attention is drawn to the > stacktrace which is actually just a symptom. > > Avoiding mutable fields and favoring final would have forced you down > the path of eagerly failing vs continuing with invalid defaults. > > Love the update, I see you did use final in many places! You're on > the right track. Lean into that more and try to force yourself to > use it even when it at first feels odd. I hope the feedback is > welcome and you see it as a compliment. You've been doing so much > great work and spending your time helping the project, it's my way of > saying thank you and you're worth the time to mentor :) For whatever > that's worth :) > > Keep up the truly awesome work. If we had 5 more Richards this > project would be signing. We'll get there :) > > > -David > -- Richard Zowalla, M.Sc. Research Associate, PhD Student | Medical Informatics Hochschule Heilbronn – University of Applied Sciences Max-Planck-Str. 39 D-74081 Heilbronn phone: +49 7131 504 6791 mail: [email protected] web: https://www.mi.hs-heilbronn.de/
smime.p7s
Description: S/MIME cryptographic signature
