On 2020/03/08 09:37:23, hanwenn wrote: > On 2020/03/08 09:33:24, hahnjo wrote: > > On 2020/03/07 23:15:09, hanwenn wrote: > > > > https://codereview.appspot.com/555420043/diff/549710049/lily/text-interface.cc > > > File lily/text-interface.cc (right): > > > > > > > > > https://codereview.appspot.com/555420043/diff/549710049/lily/text-interface.cc#newcode70 > > > lily/text-interface.cc:70: SCM ligature = ly_assoc_get (substr, > > > replacement_alist, SCM_BOOL_F); > > > this whole thing is a bit clumsy. Couldnt we do > > > > > > for k, v in replacement-alist > > > str = re.sub(k, v, str) > > > > > > (python code, but you get the idea.) > > > > While simpler, your code does not do the same. The currently implemented > > algorithm is really text replacement per textbook: Try the longest replacement > > first to make sure we match the longest substring at each position. This > > warrants the "clumsy" structure as you call it. > > Well, what if you sort the replacements by decreasing size?
Yes, but you still have to check every position and not do a global search-and-replace. > If you want to keep this around, can you add a comment why it needs to be like > this, > with an example where my simplistic approach wouldn't work. https://codereview.appspot.com/555420043/
