[cgi] split the query in words if needed and add them to the argv

[iri] accept also : and @ again, to be RFC3986 compliant.

don't %-decode the query

bring the CGI implementation in par with GLV-1.12556

ensure isn't NULL

accept a wider range of UNICODE codepoints while parsing hostnames

legibility: use p[n] instead of (*(p + n))

trim_req_iri: set error string

trim initial forward slashes this parse gemini:// into an IRI whose path is "foo". I'm not 100% this is standard-compliant but: 1. it seems a logical consequence of the URI/IRI cleaning algo (where we drop sequential slashes) 2. practically speaking serving file a sequence of forward slashes doesn't really make sense, even in the case of CGI scripts


check also that the port number matches


normalize host name when parsing the IRI RFC3986 3.2.2 "Host" says that > Although host is case-insensitive, producers and normalizers should > use lowercase for registered names and hexadecimal addresses for the > sake of uniformity, while only using uppercase letters for > percent-encodings. so we cope with that.

normalize schema when parsing the IRI RFC3986 in section 3.1 "Scheme" says that > Although schemes are case-insensitive, the canonical form is > lowercase and documents that specify schemes must do so with > lowercase letters. An implementation should accept uppercase > letters as equivalent to lowercase in scheme names (e.g., allow > "HTTP" as well as "http") for the sake of robustness but should only > produce lowercase scheme names for consistency. so we cope with that. The other possibility would have been to use strcasecmp instead of strcmp when checking on the protocol, but since the "case" version, although popular, is not part of any standard AFAIK I prefer downcasing while parsing and be done with it.

remove infinite loop