Re: [libvirt] [PATCH 3/7] Fix inconsistency in using curly braces around functions

On Mon, Mar 17, 2014 at 09:25:13AM -0600, Eric Blake wrote: I'll send a v2 split into smaller patches and I'll wrap long lines too, no problem with that, but... I was trying to tune the regexp to achieve 0 false positives and we would need to use PCRE regexp which I don't know whether it's supported by the sc_prohibit_ syntax checks. The resulting regexps (or rather the whole command) look like this (sorry for the long line): git grep -nHP -e '^\s*\*?[_a-zA-Z0-9]+(,? \*?[_a-zA-Z0-9]+)+\) ?\{' -e '^\s*(?!([a-zA-Z_]*for_?each[a-zA-Z_]*) ?\()[_a-zA-Z0-9]+( [_a-zA-Z0-9]+)* ?\(\*?[_a-zA-Z0-9]+(,? \*?[_a-zA-Z0-9\[\]]+)+\) ?\{' or (listing all allowed foreach methods: git grep -nHP -e '^\s*\*?[_a-zA-Z0-9]+(,? \*?[_a-zA-Z0-9]+)+\) ?\{' -e '^\s*(?!(libxl_for_each_set_bit|udev_list_entry_foreach|nla_for_each_nested) ?\()[_a-zA-Z0-9]+( [_a-zA-Z0-9]+)* ?\(\*?[_a-zA-Z0-9]+(,? \*?[_a-zA-Z0-9\[\]]+)+\) ?\{' Let me know if we can somehow incorporate it inside cfg.mk, I'll gladly do that. If not, bracket-spacing.pl will probably do the trick, but anyway, we have to filter to *.[hc] files only. Martin