4 ms·
If we're iterating characters, we might as well use zlib-compressed 21-bit unicode characters. They save even more room than UTF8, and it's difficult to pretend
by geocar 11y ago
If we're iterating characters, we might as well use zlib-compressed 21-bit unicode characters. They save even more room than UTF8, and it's difficult to pretend that characters can be accessed by index.
- barrkel 11y agoThere's an order of magnitude difference (at least) between accessing memory vs unzipping. The constant space overhead isn't great either for small strings. For big strings like web pages, we already do use zlib (via gzip).