- 03 Jun, 2010 1 commit
-
-
Lasse Holmstedt authored
Reviewed-by: erikv
-
- 02 Jun, 2010 38 commits
-
-
Robert Loehning authored
-
Robert Loehning authored
Reviewed-by: Oswald Buddenhagen
-
Robert Loehning authored
-
Roberto Raggi authored
-
Roberto Raggi authored
-
hjk authored
-
Oswald Buddenhagen authored
that way the result vector can be pre-allocated - which gives zero resizes if no expansions are in the expression. for the joined argument lists of built-in functions, pre-allocate a fixed 5 elements, which covers all cases and more. the argument lists of user-defined functions are not pre-allocated yet.
-
Oswald Buddenhagen authored
this gives a rather significant boost, in particular for expressions which are evaluated repeatedly (includes).
-
Oswald Buddenhagen authored
some replace functions can return empty elements which should be zapped here (according to qmake, anyway).
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
this saves quite some mallocs in the parsing pass.
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
it's private to ProFileEvaluator::Private since the getting rid of the visitor pattern.
-
Oswald Buddenhagen authored
ProString is almost a QStringRef, except that it keeps a copy of instead of a pointer to the QString - to make it refcountable. additionally, it holds a hash so it can be efficiently used for repetetive hash lookups.
-
Oswald Buddenhagen authored
this function class just joins each string list argument into a single string. so instead of first building a list and joining it right afterwards, create a single string in the first place.
-
Oswald Buddenhagen authored
that's actually kinda braindead, but that's qmake as usual.
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
no point in it being virtual ...
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
turn else, for(), defineTest() and defineReplace() into own node types instead of treating them as magic conditionals/functions. on top of that, introduce a proper branching node type, so finding the alternative code paths can be moved into the parser instead of burdening the evaluator with it.
-
Oswald Buddenhagen authored
... which basically means "break it".
-
Oswald Buddenhagen authored
-
Oswald Buddenhagen authored
there is no point in the split, given that there is nothing virtual any more.
-
Oswald Buddenhagen authored
this is consistent with the top-level scope evaluation.
-
Oswald Buddenhagen authored
it's pretty trivial and not used a lot
-
Oswald Buddenhagen authored
it wouldn't work as expected anyway, as the splitting rule (using the semicolon in addition to whitespace) applied only to the string literal from the pro file, but not any expanded [environment] variables, etc. (i.e., where it might any make sense). so just drop it, as it would considerably complicate following optimizations.
-
kh1 authored
We missed some environment vars that had to be set to make the script work. NOTE: still does not work on windows, since madbin\pkg-config.cmd needs to be too. Reviewed-by: kh
-
hjk authored
Conflicts: doc/qtcreator.qdoc src/plugins/debugger/debuggermanager.cpp src/plugins/texteditor/basetexteditor.cpp
-
hjk authored
-
hjk authored
-
- 01 Jun, 2010 1 commit
-
-
kh1 authored
We missed some environment vars that had to be set to make the script work. NOTE: still does not work on windows, since madbin\pkg-config.cmd needs to be too. Reviewed-by: kh
-