mauke

localtime(HUGEVAL) reports wrong number in error message

3

**Description** If the number you pass to `localtime` is too big, you get a warning (and an error) message that contains the invalid argument. But it doesn't always print the...

make unsnoc handle infinite lists

1

Before: ghci> take 10 . fst unsnoc [1..] Just *** Exception: stack overflow After: ghci> take 10 . fst unsnoc [1..] Just [1,2,3,4,5,6,7,8,9,10]

I want to safely run certain queries from multiple concurrent clients. For example, consider the following pseudo-code: ```sql x

Coveralls badge doesn't update on github

1

I just set up coveralls integration with a repository of mine. The coveralls page https://coveralls.io/github/mauke/HTML-Blitz shows that my test coverage after the latest commit is 92%, but my github page...

`exit` has higher precedence than `?:`, so `exit $connected ? 1 : 0` parses as `(exit $connected) ? 1 : 0`, which is not what was intended.

support netbsd-curses terminfo format

2

Look into supporting other terminfo formats than ncurses, such as netbsd-curses. Related: https://github.com/sabotage-linux/netbsd-curses/

This patch adds a failing test for line numbers. Unfortunately I have no idea what's happening or how to fix it.

`exit` has higher precedence than `?:`, so `exit $result->was_successful ? 0 : 1` parses as `(exit $result->was_successful) ? 0 : 1`, which is not what was intended.

S_ENFMT properly belongs to the group of permission bits (like setuid/setgid), not file types. On systems that have it (like AIX), it can be set/cleared with chmod(). (In fact, it...

S_ISENFMT($mode) has no equivalent C macro. It was implemented (in XS) as the equivalent of `($mode & S_IFMT) == S_ENFMT`, which is always false because S_ENFMT is not a file...