thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-23 16:12:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

Return scheme logic

Browse Source
This commit is contained in:
alesapin 2019-06-24 14:16:08 +03:00
parent 56a7595254
commit 331e17d56a
1 changed files with 16 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
dbms/src/Functions/URL/domain.h
View File

@ -19,7 +19,6 @@ inline StringRef buildFound(const Pos & pos, const Pos & dot_pos, const Pos & st
if (after_dot == ':' || after_dot == '/' || after_dot == '?' || after_dot == '#') if (after_dot == ':' || after_dot == '/' || after_dot == '?' || after_dot == '#')
return StringRef{}; return StringRef{};
return StringRef(start_of_host, pos - start_of_host); return StringRef(start_of_host, pos - start_of_host);
} }
@ -34,16 +33,26 @@ inline StringRef getURLHost(const char * data, size_t size)
if (*(end - 1) == '.') if (*(end - 1) == '.')
return StringRef{}; return StringRef{};
StringRef scheme = getURLScheme(data, size);
if (scheme.size != 0) Pos slash_pos = find_first_symbols<'/'>(pos, end);
if (slash_pos != end)
pos = slash_pos;
else
pos = data;
if (pos != data)
{ {
StringRef scheme = getURLScheme(data, size);
Pos scheme_end = data + scheme.size; Pos scheme_end = data + scheme.size;
pos = scheme_end + 1; if (pos - scheme_end != 1 || *scheme_end != ':')
if (*scheme_end != ':' || *pos != '/') {
std::cerr << "RETURNING HERE\n";
return StringRef{}; return StringRef{};
} }
}
if (end - pos > 2 && *pos == '/' && *(pos + 1) == '/') // Check with we still have // character from the scheme
if (end - pos > 2 && *(pos) == '/' && *(pos + 1) == '/')
pos += 2; pos += 2;
auto start_of_host = pos; auto start_of_host = pos;
@ -61,7 +70,7 @@ inline StringRef getURLHost(const char * data, size_t size)
case '#': case '#':
return buildFound(pos, dot_pos, start_of_host); return buildFound(pos, dot_pos, start_of_host);
case '@': /// myemail@gmail.com case '@': /// myemail@gmail.com
start_of_host = pos; start_of_host = pos + 1;
break; break;
case ' ': /// restricted symbols case ' ': /// restricted symbols
case '\t': case '\t':