summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLars Wirzenius <liw@liw.fi>2021-08-27 09:54:07 +0300
committerLars Wirzenius <liw@liw.fi>2021-08-27 09:54:07 +0300
commitd1812dbcc483e40843a0d8ed1fd915654cba0f89 (patch)
treed7d664301eef90bdc128c55a36e8d500c13f682a
parent0a4db253323b6d9c2199f9af6712c31736dbb6e7 (diff)
downloadsshca-web-d1812dbcc483e40843a0d8ed1fd915654cba0f89.tar.gz
blog: decision on scenario step implementation language
Sponsored-by: author
-rw-r--r--blog/2021/08/27/step-language.mdwn23
1 files changed, 23 insertions, 0 deletions
diff --git a/blog/2021/08/27/step-language.mdwn b/blog/2021/08/27/step-language.mdwn
new file mode 100644
index 0000000..9117d95
--- /dev/null
+++ b/blog/2021/08/27/step-language.mdwn
@@ -0,0 +1,23 @@
+[[!meta date="Fri, 27 Aug 2021 09:34:14 +0300"]]
+[[!tag decision]]
+[[!meta title="Decision: Use Rust for Subplot scenario step implementation"]]
+
+I will be using [Subplot][] to document the requirements and
+acceptance criteria for `sshca`, and for automating the verification
+that they are met. The approach in Subplot is to write semi-formal
+_scenarios_ that get translated into code, using functions that
+implement each scenario step. I can choose between the Python and Rust
+languages for writing those scenarios. I choose Rust, because it
+allows more concurrency when the verification code is running, and
+also I find I can write correct code more easily in Rust than in
+Python.
+
+The only reason to choose Python would be to make the step functions
+be more easily audited by people other than me, but I think Rust has
+already become sufficiently popular that Rust will not be a problem.
+It's more important that the actual requirements, acceptance criteria,
+and the scenarios are understood by anyone interested.
+
+The actual code for `sshca` will be written in Rust in any case.
+
+[Subplot]: https://subplot.liw.fi/