Norman Gray

I notice that the associated patch respects environment variable `$HTTP_PROXY`. It would be be useful for it to instead (or, if necessary, additionally) respect `$http_proxy` (ie, lowercase). There is no...

Indeed, can I suggest the attached patch: [ioc_exec.txt](https://github.com/iocage/iocage/files/7936173/ioc_exec.txt)? This is as applied to `iocage_lib/ioc_exec.py` in released version 1.2 (FreeBSD 13.0). This defines `{http,https,no}_proxy` in the IOCExec environment, taking the value...

> This issue is just addressing the fetch/update process of iocage. > Both use methods which only take the uppercased environment variables into account. So technically no need to export...

I've now had the chance to check this again. See below. The correct proxy variables are defined: .../site-packages/iocage_lib# export|grep proxy http_proxy=http://blah.... https_proxy=http://blah... With the distributed ioc_exec.py, the `iocage fetch` appears...

The problem is that at least one useful error message is being discarded. I've solved my jail-creation problem by hacking `iocage_lib/ioc_create.py` and `iocage_cli/create.py` to add error reporting. With this, I...

Looking at `.../sqlite3/ffi.rkt`, I see that `sqlite3_prepare_v2` deliberately returns `tail?` rather than `tail`, so I'm not suggesting you're not aware of the issue. Also, it would be relatively straightforward to...

Righto – it seems clear that there's no way of getting the underlying SQL system to do the multi-statement parsing in a portable way. I take your point about bulk...

(...and _of course_ I forgot a test case: [extract-sql.rkt](https://github.com/racket/db/files/3573942/extract-sql.txt))

See eg https://wiki.apache.org/httpd/DirectoryListings

...and just to be clear, this isn't something that can be fixed by a patch to the code in this repository, but (if the above analysis is correct) only by...