Iris devs should declare all the information now, there is no option to "leave something out" anymore, it's for your own good. Version is not changed yet, giving you time to see that changelog and do the necessary changes to your codebase, it will written to HISTORY.md too of course.
Before:
app.Run(iris.AutoTLS(":443"))
Now:
app.Run(iris.AutoTLS(":443", "example.com", "mail@example.com")
Commit Change: f7b655f145 [formerly f5314a8c1f8303d7216481d05129eb8a62766e14]
Happy weekend!
Former-commit-id: bf974dc486d05ff008c9ed13aaf96e716b1e55c6
Iris devs should declare all the information now, there is no option to "leave something out" anymore, it's for your own good. Version is not changed yet, giving you time to see that changelog and do the necessary changes to your codebase, it will written to HISTORY.md too of course.
Before:
app.Run(iris.AutoTLS(":443"))
Now:
app.Run(iris.AutoTLS(":443", "example.com", "mail@example.com")
Happy weekend!
Former-commit-id: f5314a8c1f8303d7216481d05129eb8a62766e14
Iris devs should declare all the information now, there is no option to "leave something out" anymore, it's for your own good. Version is not changed yet, giving you time to see that changelog and do the necessary changes to your codebase. Happy weekend!
Former-commit-id: 490ce14a1022a2b81d347d7f59c2bb5412cfcdf2
Also disable version updater on Iris benchmark source code files. It may runs ever faster than before if you started the benchmarks immediately after the banner 👍
Former-commit-id: a55dc1e0b658d7386229c32ba6953b1ea60f2872
My second mistake on redis session db because I'm not personally use redis, so I'm waiting for redis users to find these type of errors inside it, such as @sy264115809
Former-commit-id: df84348df7509c0f863f41f01de6ed6db6e8133e
I don't know why I did add LifeTime.Second() back then, it was silly of me but didn't receive any reports except a commit comment 25 minutes ago, so we can assume that redis is almost "dead" nowdays. Thank you for the notice @sy264115809
Former-commit-id: 140caa2a42aa16b1e80905d24976c481eae24c69
No such a difference (before: 8.61MB/s now: 8.91MB/s).
I think we can't do much more, I already removed the heavy parts on the first run... but still I'm open for any suggestions. Thank you for the feedback!
Former-commit-id: 0b0c8ba40c8380b530b6c77a5ee75ecfb4357f16
No need to format using Writef on benchmarks for a simple string value without dynamic text, just use the WriteString which is the correct function for these apps.
Former-commit-id: 896ec514570d533d8da91af3ad199fc4f823fe74
Maybe we'll have to investigate it more some time, but for now I think we are ok, it's working as expected except the content-length canno be reported because we get the compressed size after the gzip writer.Write and we can't send headers after its underline writer which is the ResponseWriter, a solution for that is to use a temp bytes.Buffer but this will slow down the performance, users who need to know the size can use a temp writer instead for theirselves.
Former-commit-id: a4400ca5db526998a646fd5880ca67358fa72c0f
