So far the pre-compiled headers were processed (thus defines from those
headers were visible), but the actual includes for the documents were
not added, which is necessary for lookup/completion.
Note that this will be only done if pre-compiled headers are not ignored
(Options > C++ > Code Model > [] Ignore pre-compiled headers).
Change-Id: I54a8e6b00597af164d958e3e9f2a1075ea187788
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
In most cases we need to work with the utf16 indices. Only in
cppfindreferences the byte interface is still needed since there we read
in files and work on a QByteArray to save memory.
Change-Id: I6ef6a93fc1875a8c9a305c075d51a9ca034c41bb
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
* If the macro is defined before, track its reference
* Synchronize environment line before calling remove, which
currently sets incorrect line
* Set macro offset
Task-number: QTCREATORBUG-10454
Change-Id: I480d16423a976a025bb8c71046610a46f9d7b0fd
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@digia.com>
The following snippet demonstrates the problem:
--- snip ---
// comment \
#include <something.h>
...
class Foo
{
...
};
--- snap ---
If there are >=9 empty/preprocessor lines, the preprocessed source
becomes
// comment \
# 12 "file.cpp"
...
The lexer considers the line marker as a continued C++ comment, and
highlighting is broken
Change-Id: I30a2fc7d19b279316e9273697179c90d81099573
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
Easier to compare when indentation is aligned
Change-Id: I216073dc167c0ed785616ae308b566db7f910592
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
This takes too much memory. For qtcreator.pro the numbers are as
follows:
Patch applied: ~ 1600MB (RES)
Patch reverted: ~ 510MB (RES)
This reverts commit 4c2daa90ce.
Task-number: QTCREATORBUG-10973
Change-Id: I843bd7c1ea4a26a1ec55ddc14c2a34a98d040922
Reviewed-by: hjk <hjk121@nokiamail.com>
Reviewed-by: Robert Loehning <robert.loehning@digia.com>
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Reviewed-by: Orgad Shaneh <orgads@gmail.com>
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
This adds definitions for the macros __FILE__, __LINE__, __DATE__ and
__TIME__ on demand.
As a side effect, this also introduces highlighting for the uses of
these macros.
Task-number: QTCREATORBUG-8036
Change-Id: Ib7546c7d45d2eecbc50c7883fc684e3497154405
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Reviewed-by: hjk <hjk121@nokiamail.com>
Breaks highlighting for macros using the predefined macros.
This reverts commit 1d834c1126.
Change-Id: Ic13c407e293a806a63ff30153864530df6a32e47
Reviewed-by: hjk <hjk121@nokiamail.com>
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
Preprocessor variables __LINE__,__FILE__,__TIME__,__DATE__ where destroying
the following systems when affected variables were standing within the
same line with those variables:
* highlighting
* refactoring
* local renaming
Task-number: QTCREATORBUG-8036
Change-Id: I1a4b919d15812872ca5a8e63b1031ec1ab144c22
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@digia.com>
By lexing the first token after a macro call (meaning: the token after
the closing parenthesis (which was passed to handleFunctionLikeMacro
which in turn pushed it back into the token buffer)), a token buffer
might be popped, which unblocks the macro that generated the actual
param pack. The effect was that if this happens in the expansion of a
recursive macro (with parameters!), the preprocessor ended up in an
infinite loop.
Task-number: QTCREATORBUG-9015
Task-number: QTCREATORBUG-9447
Change-Id: I0d83c59188ec15c4a948970e9fa944a17d765475
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@digia.com>
Doing so resulted in an incorrect position for the EOF token when the
preprocessed output would be parsed. That in turn leads to incorrect
insertion positions for refactoring actions.
This is especially true when a file contains only preprocessor
directives: the EOF token would point to line 1 column 1, which is
usually not the place where code should be inserted.
Change-Id: I7d359aa7a6c04bc52c8b873fd49ad6afc3a77319
Reviewed-by: hjk <hjk121@nokiamail.com>
Preprocessor did not correctly handle when variadic macro arguments were not
provided at all, if there were other arguments: macro was not expanded
in case only the non variadic arguments were given.
#define MACRO(...) used to work fine for 0 or more arguments.
#define MACRO(ARG0, ...) used to work only for 2 or more arguments, now fixed.
Change-Id: I64e9199ceccae05618a49931c2adad8e4f9471ba
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@digia.com>
... and adjust INCLUDEPATH accordingly.
while i'm at messing with include statements, also re-order the include
blocks according to policy and sort them within bigger blocks.
Change-Id: I7762abfd7c4ecf59432b99db2f424e4fa25733a5
Reviewed-by: Tobias Hunger <tobias.hunger@digia.com>
Reviewed-by: Eike Ziller <eike.ziller@digia.com>
Track the typical #ifndef/#define/#endif usage in header files to see if
the macro is an include guard. If so, store it in the Document. No
behavioural change, just recording the name.
This can be used in the future to track if a file needs to be re-parsed
when a macro changes: if it was used in the file, and not defined in it
nor being the include-guard, a file should be re-preprocessed and
re-parsed.
It can also be used to check if two files have the same include guard.
Change-Id: I2715f529997a7b24a11bdbc6150652e2669f1a46
Reviewed-by: Nikolai Kosjar <nikolai.kosjar@digia.com>
When in 'keep comments' mode, the preprocessor does not properly handle macro calls with
comments between the macro name and the opening parenthesis: "FOO /*something to say*/
(45)".
Change-Id: I6fe733242e4d2ccff2985d17399d0a084917415a
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
For shadow builds some tests couldn't find their test data. Solved by
using the same pattern as in Qt tests: Inject $$PWD as SRCDIR via a
DEFINES statement.
Note: It was considered to use QFINDTESTDATA for Qt5, but this was
rejected due to:
1) It's Qt5 only (would mean to introduce some wrapper...)
2) As the doc states, it will not work with QTEST_APPLESS_MAIN tests.
Change-Id: Ie6cf59570fe61725b7f6f83b7da650331ddb38fc
Reviewed-by: Erik Verbruggen <erik.verbruggen@digia.com>
Notice that a similar problem still exists for which we
need to fix the lexer when there's a C style commend which
ends with a backslash-newline.
Task-number: QTCREATORBUG-7713
Change-Id: I0f6d561703984f917fa5ed29de020ad0bdc5aaf0
Reviewed-by: David Schulz <david.schulz@nokia.com>
Reviewed-by: hjk <qthjk@ovi.com>
Summary of most relevant items:
- Preprocessor output format change. No more gen true/false. Instead
a more intuitive and natural expansion (like from a real compiler) is
performed directly corresponding to the macro invocation. Notice that
information about the generated tokens is not lost, because it's now
embedded in the expansion section header (in terms of lines and columns
as explained in the code). In addition the location on where the macro
expansion happens is also documented for future use.
- Fix line control directives and associated token line numbers.
This was not detected in tests cases because some of them were
actually wrong: Within expansions the line information was being
considered as originally computed in the macro definition, while
the desired and expected for Creator's reporting mechanism (just
like regular compilers) is the line from the expanded version
of the tokens.
- Do not allow for eager expansion. This was previously being done
inside define directives. However, it's not allowed and might
lead to incorrect results, since the argument substitution should
only happen upon the macro invocation (and following nested ones).
At least GCC and clang are consistent with that. See test case
tst_Preprocessor:dont_eagerly_expand for a detailed explanation.
- Revive the 'expanded' token flag. This is used to mark every token
that originates from a macro expansion. Notice, however, that
expanded tokens are not necessarily generated tokens (although
every generated token is a expanded token). Expanded tokens that
are not generated are those which are still considered by our
code model features, since they are visible on the editor. The
translation unit is smart enough to calculate line/column position
for such tokens based on the information from the expansion section
header.
- How expansions are tracked has also changed. Now, we simply add
two surrounding marker tokens to each "top-level" expansion
sequence. There is an enumeration that control expansion states.
Also, no "previous" token is kept around.
- Preprocessor client methods suffered a change in signature so
they now receive the line number of the action in question as
a paramater. Previously such line could be retrieved by the client
implementation by accessing the environment line. However, this
is not reliable because we try to avoid synchronization of the
output/environment lines in order to avoid unnecessary output,
while expanding macros or handling preprocessor directives.
- Although macros are not expanded during define directives (as
mentioned above) the preprocessor client is now "notified"
when it sees a macro. This is to allow usage tracking.
- Other small stuff.
This is all in one patch because the fixes are a consequence
of the change in preprocessing control.
Change-Id: I8f4c6e6366f37756ec65d0a93b79f72a3ac4ed50
Reviewed-by: Roberto Raggi <roberto.raggi@nokia.com>
This reverts commit 44d8e7bef7.
The original test case was actually right. I guess I was mislead
by the recently added preprocessor tests, which are the ones
that should be corrected (together with the preprocessor, which
seems to have issues with # lineno generation).
Conflicts:
tests/auto/cplusplus/preprocessor/tst_preprocessor.cpp
Change-Id: I47f82ed23a37086d0d81c4b3ea2cac48fb753451
Reviewed-by: hjk <qthjk@ovi.com>
Reviewed-by: Francois Ferrand <thetypz@gmail.com>
Do not expand function-like macros at all when there's a mismatch
in the parameter/argument count.
The report below raises the issue but its expected result is not
correct. This would be the more appropriate fix.
Task-number: QTCREATORBUG-7225
Change-Id: Ide8580faa7b724d3e8b396ec1f899cc5ca7f9e7e
Reviewed-by: hjk <qthjk@ovi.com>
Make sure the environment line is consistent during preprocessor
directives and identifier handling so clients can rely on consistent
information. Particularly important for macro usages.
New tests also added.
Change-Id: I962a39a86cd17b8d945d2959c2c95e2d258ea3e6
Reviewed-by: hjk <qthjk@ovi.com>
I believe those were actually incorrect, since the # mark
from the generated tokens `int f` should be relative to <stdin>
on the line they are defined, which is 1.
Change-Id: I663ef49ad75eb8bb0a4a4b18d4899a952011536c
Reviewed-by: Francois Ferrand <thetypz@gmail.com>
Reviewed-by: hjk <qthjk@ovi.com>
