Bug 8529 – ctRegex - named submatch - hash-table exception

Reported on v2.060 (but no version seems to work). The code is based on the opIndex example from: http://dlang.org/phobos/std_regex.html#regex The normal regex version is functional, but when converting to ctRegex, the hashtable lookup throws.( auto m = match("a = 42;", ctRegex!(`(?P<var>\S+)\s*=\s*(?P<value>\d+);`)); auto c = m.captures; // c["var"]; => [email protected](2145): Range violation // c["value"]; => Object.Exception@C:\D\dmd2\windows\bin\..\..\src\phobos\std\regex.d(2145): no submatch named value

(In reply to comment #0) > Reported on v2.060 (but no version seems to work). > > The code is based on the opIndex example from: > http://dlang.org/phobos/std_regex.html#regex > > The normal regex version is functional, but when converting to ctRegex, the > hashtable lookup throws.( > > auto m = match("a = 42;", > ctRegex!(`(?P<var>\S+)\s*=\s*(?P<value>\d+);`)); > auto c = m.captures; > > // c["var"]; => > [email protected](2145): Range violation > > // c["value"]; => > Object.Exception@C:\D\dmd2\windows\bin\..\..\src\phobos\std\regex.d(2145): > no submatch named value Just spent an hour duplicating this one. Bump. Since ctRegex is an "Experimental feature.", it may be worth documenting the existence of this bug in http://dlang.org/phobos/std_regex.html#.StaticRegex. If we don't know how to fix this, it may also be worth temporarily and statically turning down named captures in static regexes, with a "sorry" message.

Bump for Dmitry.

Issue 9280 shows it's a general problem in CTFE version of regex parser *** This issue has been marked as a duplicate of issue 9280 ***

(In reply to comment #1) > (In reply to comment #0) > > Reported on v2.060 (but no version seems to work). > > > > The code is based on the opIndex example from: > > http://dlang.org/phobos/std_regex.html#regex > > > > The normal regex version is functional, but when converting to ctRegex, the > > hashtable lookup throws.( > > Turns out it's not hashtable but rather binary search on a broken array :( The good news is that the fix turns out to be trivial.