Bug 1937 – std.uri.decode throws wrong exception

If the parameter of std.uri.decode contains %E4, which is a German umlaut ä, the exception "URI error" is thrown. This is wrong behaviour, as an URI may contains umlauts. Here's my C implementation which does the job: void decode(char *src, char *last, char *dest) { for (; src != last; src++, dest++) if (*src == '+') *dest = ' '; else if (*src == '%') { int code; if (sscanf(src+1, "%2x", &code) != 1) code = '?'; *dest = code; src +=2; } else *dest = *src; *dest = 0; } To my understanding, it's nothing else than a hex code to byte conversion, so there should be no reason to forbid certain codes and throw exceptions. Also reading the documentation I expected that at least std.uri.decodeComponent is a straightforward implementation, but it also throws exceptions.

My own web pages currently work with %E4 for umlaut 'ä', which is ASCII code, but standard is UTF-8 which makes it %C3%A4. So this issue seems to be invalid and I have to change things for the D port of my web application.

RFC 3986 states in section 2.1. Percent-Encoding For example, "%20" is the percent-encoding for the binary octet "00100000" (ABNF: %x20), which in US-ASCII corresponds to the space character (SP). So UTF-8 is not the standard used for URI percent encoding. Therefore, std.uri.decode should not throw an exception if its parameter contains %E4. It is allowed for browsers to encode German umlaut ä as %E4, which was the reason I ran into this problem (because my server side application crashed as soon as std.uri.decode was called).

It's unlikely this D1 issue will get worked on. If this bug applies to D2 and/or if anyone plans to work on it, please reopen.