Bastien <b...@gnu.org> writes: > Hi Trevor, > > Trevor Murphy <trevor.m.mur...@gmail.com> writes: > >> + (not (eq (org-element-type (org-element-at-point)) 'src-block))))) > > I think `org-in-src-block-p', while a bit less reliable, will be > faster, and reliable/fast enough for this use-case. > > Let's see what others think/test.
For the record, here are the results of a very unscientific profiling of the patch above. On my woefully under-powered and aging Atom processor, For each scenario I typed without errors "The quick brown fox jumps over the lazy dog." With the old behavior (i.e., no test for source blocks): org-mode-flyspell-verify 27 0.005580378 0.0002066806 With the patch above: org-mode-flyspell-verify 27 0.35359755 0.0130962055 Using (not (org-in-src-block-p)), as Bastien suggests: org-mode-flyspell-verify 27 0.0112581490 0.0004169684 With org-in-src-block-p, a half of a hundredth of a second spread over 27 characters causes no noticeable slowdowns. But using org-element-at-point causes the cursor to lag a bit. So +1 is for (not (org-in-src-block-p)). Best, Matt
