this tremendously speeds up loading of qt 5.1 based projects (including
qt itself) under mac os, as these look up the sdk dynamically, and use
caching to avoid doing that in every subproject.
Change-Id: I833253f81c3159056fab2ff888f293b36cc2ef56
Reviewed-by: Daniel Teske <daniel.teske@digia.com>
it was merely an artifact of using QString::simplified() on the
unparsed (!) project code. there is no reason why anyone should actually
rely on it, so just remove it.
Change-Id: If9b957c4b1263f3990a2331f8851bb1c06154ea8
Reviewed-by: Qt Doc Bot <qt_docbot@qt-project.org>
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
(cherry picked from qtbase/c3c4f1eb5301a4fda17a9a48012df45f9a0ee7a3)
When writing a file with write_file() we have to inform the pro file parser
cache to discard the file if it's existent in the cache, to ensure that
calling include() after write_file() always works.
Change-Id: I7d09269a57de55ca30b0e11dd40770de9f919f64
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
(cherry picked from qtbase/f45d6414ad6ef1d75f049a61411cd954bcae29c3)
Reviewed-by: Daniel Teske <daniel.teske@digia.com>
this allows skipping an entire feature file if a condition is not
fulfilled, without putting the whole body inside a conditional.
Change-Id: I84fe9c94dda58c794fb52c3f163b40563b0db30a
Reviewed-by: Joerg Bornemann <joerg.bornemann@digia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@digia.com>
(cherry picked from qtbase/525eb337670f61078383c20ad9be274dcf2d0928)
#!/usr/bin/env ruby
Dir.glob('**/*.cpp') { |file|
# skip ast (excluding paste, astpath, and canv'ast'imer)
next if file =~ /ast[^eip]|keywords\.|qualifiers|preprocessor|names.cpp/i
s = File.read(file)
next if s.include?('qlalr')
orig = s.dup
s.gsub!(/\n *if [^\n]*{\n[^\n]*\n\s+}(\s+else if [^\n]* {\n[^\n]*\n\s+})*(\s+else {\n[^\n]*\n\s+})?\n/m) { |m|
res = $&
if res =~ /^\s*(\/\/|[A-Z_]{3,})/ # C++ comment or macro (Q_UNUSED, SDEBUG), do not touch braces
res
else
res.gsub!('} else', 'else')
res.gsub!(/\n +} *\n/m, "\n")
res.gsub(/ *{$/, '')
end
}
s.gsub!(/ *$/, '')
File.open(file, 'wb').write(s) if s != orig
}
Change-Id: I3b30ee60df0986f66c02132c65fc38a3fbb6bbdc
Reviewed-by: hjk <qthjk@ovi.com>
it's a pretty braindead thing to implement control flow statements as
(built-in) functions.
as a "side effect", this fixes return() value handling for lists.
Change-Id: I59c8efa0e4d65329327115f7f8ed20719e7f7546
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
this was forgotten when the value lists were extended by a size hint.
Change-Id: I6f9b55ed671224a9b8735c8d937f94aac4a73a42
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
this is a reduced RHS grammar to be used with
expandVariablesReferences() to implement the expansions in
QMAKE_SUBSTITUTES.
using the parser+evaluator for that is consistently up to 25% slower
than a purely interpreted approach - except for plain strings (with some
quoting), which happens to be exactly what is found in the template
files to substitute.
Change-Id: I219266460b970c6ddcb43cf85a914fbde6540271
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
"(eval)" does not exactly help to find the error source.
Change-Id: Iecd03e6a4909ca6d7eab846844ca4415ebfa3429
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
we don't really use the current character at this point, so don't fake
it and use the cur pointer as a flag.
Change-Id: I0dd31ff07fafe0748b88b3a573d25b689f0c3748
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
the performance is about the same (depending on the expression type it's
better or worse), but a lot of code just disappears.
Change-Id: I60eb9b87f23cc811d3f9577841c38966ecfd8e43
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
everything which is not stored in the parser cache is assumed to be
disposed of rather soon anyway.
this increases peak memory usage per evaluator by a few kilobytes
(something in the order of five times the file size). as only manual
spec parsing and eval() statemenents in projects use non-cached parsers,
the impact is pretty much insignificant.
Change-Id: I326f312f4dd98b30b692d219de7ae092b6ad3584
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
instead of having a bazillion different callbacks, have only one with a
type parameter. the drain typically multiplexes all into one stream anyway.
Change-Id: I963daefc5a266c91334a4cc599570ed26b603d5d
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
this is clearer and not necessarily more code.
Change-Id: Ic698a8076658ae986d0dbdeebb1f4760dd806e35
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
with the removal of the skip jump label some time ago, this condition
became insatisfiable.
Change-Id: I4fc52ca8a38e048fd37c2ae6bfaae69acf09ada0
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
in the context of possibly using the same parse result for multiple
build configurations, resolving env variables already during parsing
would be just wrong.
Change-Id: I49367b5eff5868a38c026b8bd74148e0b359fffb
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
they are "compiler defines", with no dependency on the evaluation context.
Change-Id: I25bf006347ecd2edb501a344820e2ac11ff389e9
Reviewed-by: Daniel Teske <daniel.teske@nokia.com>
Reviewed-by: Oswald Buddenhagen <oswald.buddenhagen@nokia.com>
