Author: revane
Date: Sun Nov 17 10:08:04 2013
New Revision: 194968
URL: http://llvm.org/viewvc/llvm-project?rev=194968view=rev
Log:
Relax some preconditions for using FixedCompilationDatabase.
FixedCompilationDatabase (FCD) requires that the arguments it consumes after
'--' must not include
Committed in r194968.
http://llvm-reviews.chandlerc.com/D2039
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
- Applying reviewer suggestions
Hi klimek, chandlerc, ddunbar,
http://llvm-reviews.chandlerc.com/D2039
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D2039?vs=5318id=5430#toc
Files:
lib/Tooling/CompilationDatabase.cpp
unittests/Tooling/CompilationDatabaseTest.cpp
Index:
Replaced by D2039.
http://llvm-reviews.chandlerc.com/D1877
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
- Polishing up patch
Hi klimek, chandlerc, ddunbar,
http://llvm-reviews.chandlerc.com/D2039
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D2039?vs=5212id=5318#toc
Files:
lib/Tooling/CompilationDatabase.cpp
unittests/Tooling/CompilationDatabaseTest.cpp
Index:
What would the exposed interface in Driver look like? strip positional
arguments?
http://llvm-reviews.chandlerc.com/D2039
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
I was hoping to have process-based parallelism as a 3.4 feature of the C++
modernizer. This work on the FixedCompilationDatabase is blocking that.
http://llvm-reviews.chandlerc.com/D2039
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
Hi klimek, chandlerc, ddunbar,
FixedCompilationDatabase (FCD) requires that the arguments it consumes
after '--' must not include positional parameters or the argv[0] of the
tool. This patch relaxes those restrictions.
Note to reviewers: No tests yet as I wanted to get a preliminary this
is the
- Fixed broken unit tests
Hi klimek, chandlerc, ddunbar,
http://llvm-reviews.chandlerc.com/D2039
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D2039?vs=5198id=5202#toc
Files:
lib/Tooling/CompilationDatabase.cpp
unittests/Tooling/CompilationDatabaseTest.cpp
Index:
- New unit tests
Hi klimek, chandlerc, ddunbar,
http://llvm-reviews.chandlerc.com/D2039
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D2039?vs=5202id=5212#toc
Files:
lib/Tooling/CompilationDatabase.cpp
unittests/Tooling/CompilationDatabaseTest.cpp
Index:
Author: revane
Date: Thu Oct 17 12:57:36 2013
New Revision: 192905
URL: http://llvm.org/viewvc/llvm-project?rev=192905view=rev
Log:
Refactoring transform-specific options
Making the user null macros command-line option visible to the
UseNullptrTransform class instead of being visible only to
:34 AM, Edwin Vane edwin.v...@intel.com wrote:
The problem is the FixedCompilationDatabase mangles the arguments you
pass. FixedCompilationDatabase as a result has some preconditions: the name
of the compiler should not be present nor should be the filename to operate
on. Check
Renamed SubprocessCompilationDatabase to SingleCompilationDatabase
- Also improved tests to ensure -c as a compilation flag works with
clang-modernize's -c option.
Hi tareqsiraj, arielbernal, Sarcasm,
http://llvm-reviews.chandlerc.com/D1877
CHANGE SINCE LAST DIFF
The problem is the FixedCompilationDatabase mangles the arguments you pass.
FixedCompilationDatabase as a result has some preconditions: the name of the
compiler should not be present nor should be the filename to operate on. Check
CompilationDatabase.h for a better description. As stated in
Regarding paths with spaces, this is part of a more general problem of
escaping command lines which I thought would be properly handled auto-magically
but I was wrong. I need to do some work to clang first before this will work
properly.
http://llvm-reviews.chandlerc.com/D1860
BRANCH
Hi tareqsiraj, arielbernal, Sarcasm,
To facilitate passing compile command lines from driver process to
clang-modernizer subprocesses, a simple compilation database like
FixedCompilationDatabase that doesn't do any command-line mangling is
required.
http://llvm-reviews.chandlerc.com/D1877
- Reworked examples and addressed reviewer comments
- @Sarcasm: excellent suggestions. Thanks!
Hi arielbernal, tareqsiraj, Sarcasm,
http://llvm-reviews.chandlerc.com/D1849
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1849?vs=4706id=4732#toc
Files:
@@ -9,5 +9,5 @@ beautification by scripts. The fields a
(S).
N: Edwin Vane
-E: edwin.v...@intel.com
+E: rev...@gmail.com
D: clang-modernize
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
Hi tareqsiraj, arielbernal, Sarcasm,
To facilitate passing compile command lines from driver process to
clang-modernizer subprocesses, a simple compilation database like
FixedCompilationDatabase that doesn't do any command-line mangling is
required.
http://llvm-reviews.chandlerc.com/D1860
Closed by commit rL192253 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1849?vs=4732id=4747#toc
http://llvm-reviews.chandlerc.com/D1849
COMMIT
http://llvm-reviews.chandlerc.com/rL192253
___
cfe-commits
Author: revane
Date: Tue Oct 8 18:31:05 2013
New Revision: 192253
URL: http://llvm.org/viewvc/llvm-project?rev=192253view=rev
Log:
clang-modernize: Clean up help output
Now hiding options clang-modernize doesn't use and didn't create. Version
printer specialized for clang-modernize. EXAMPLES
Author: revane
Date: Mon Oct 7 08:40:19 2013
New Revision: 192100
URL: http://llvm.org/viewvc/llvm-project?rev=192100view=rev
Log:
clang-modernize: Fixing doxygen warnings
Modified:
clang-tools-extra/trunk/clang-modernize/Core/ReplacementHandling.h
Modified:
Hi arielbernal, tareqsiraj, Sarcasm,
* Now hiding options clang-modernize doesn't use and didn't create.
* Version printer specialized for clang-modernize.
* EXAMPLES text fixed and brought up-to-date.
http://llvm-reviews.chandlerc.com/D1849
Files:
clang-modernize/Core/Transforms.cpp
Author: revane
Date: Mon Oct 7 10:56:25 2013
New Revision: 192105
URL: http://llvm.org/viewvc/llvm-project?rev=192105view=rev
Log:
Update clang-modernizer docs
Summary: Docs updated to reflect new behaviour and new options.
Differential Revision: http://llvm-reviews.chandlerc.com/D1841
Closed by commit rL192105 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1841?vs=4683id=4707#toc
http://llvm-reviews.chandlerc.com/D1841
COMMIT
http://llvm-reviews.chandlerc.com/rL192105
___
cfe-commits
Closed by commit rL192032 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1836?vs=4684id=4688#toc
http://llvm-reviews.chandlerc.com/D1836
COMMIT
http://llvm-reviews.chandlerc.com/rL192032
___
cfe-commits
Author: revane
Date: Sat Oct 5 08:31:31 2013
New Revision: 192033
URL: http://llvm.org/viewvc/llvm-project?rev=192033view=rev
Log:
clang-modernize: Fix clang-apply-replacements invocation bug
The command-line for clang-apply-replacements was being mangled due to an
uninitialized variable. gasp!
Comment at: clang-modernize/Core/ReplacementHandling.cpp:108
@@ +107,3 @@
+ bool ExecutionFailed = false;
+ int ReturnCode = ExecuteAndWait(CARPath.c_str(), Argv.data(), /* env */ 0,
+ /* redirects */ 0,
Tareq A. Siraj wrote:
Can somebody mark the patch as accepted if it LG?
Btw, doc updates are coming in another patch.
http://llvm-reviews.chandlerc.com/D1836
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
Hi tareqsiraj, arielbernal, Sarcasm,
Docs updated to reflect new behaviour and new options.
http://llvm-reviews.chandlerc.com/D1841
Files:
docs/ModernizerUsage.rst
docs/clang-modernize.rst
Index: docs/ModernizerUsage.rst
===
Could you expand rel to relative wherever it appears? LLVM style is
shorthand-averse.
Comment at: clang-modernize/Core/Transform.cpp:102-105
@@ +101,6 @@
+
+ llvm::SmallString256 FilePath(Filename);
+ llvm::error_code EC = llvm::sys::fs::make_absolute(FilePath);
+
Author: revane
Date: Tue Oct 1 08:21:39 2013
New Revision: 191754
URL: http://llvm.org/viewvc/llvm-project?rev=191754view=rev
Log:
clang-apply-replacements: Fixing doxygen warning
Modified:
clang-tools-extra/trunk/clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp
Modified:
Author: revane
Date: Mon Sep 30 08:31:48 2013
New Revision: 191666
URL: http://llvm.org/viewvc/llvm-project?rev=191666view=rev
Log:
Moving style option formatting to libFormat
The help text for clang-format's -style option and the function that processes
its value is moved to libFormat in this
Committed in r191666.
http://llvm-reviews.chandlerc.com/D1773
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
Comment at: clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp:79
@@ +78,3 @@
+///
+/// \param[in] Replacements Replacements to apply
+/// \param[in] Rewrites Rewriter to use to apply replacements.
Daniel Jasper wrote:
Is there a reason for this
Committed in r191667.
http://llvm-reviews.chandlerc.com/D1730
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
Author: revane
Date: Mon Sep 30 08:59:21 2013
New Revision: 191667
URL: http://llvm.org/viewvc/llvm-project?rev=191667view=rev
Log:
clang-apply-replacements: Add code formatting functionality
The tool now supports a collection of arguments to turn on and provide settings
for the formatting of
Author: revane
Date: Mon Sep 30 09:29:28 2013
New Revision: 191669
URL: http://llvm.org/viewvc/llvm-project?rev=191669view=rev
Log:
Fix build by adding dep on TransformUtils
clang-apply-replacements unittest Makefile wasn't linking in TransformUtils.
Modified:
Author: revane
Date: Fri Sep 27 12:52:45 2013
New Revision: 191536
URL: http://llvm.org/viewvc/llvm-project?rev=191536view=rev
Log:
Adding pre/post conditions for some Replacement handling functions
The vector version of shiftedCodePosition() requires Replacements to be
in sorted order.
Turned
Hi djasper, alexfh,
The help text for clang-format's -style option and the function that
processes its value is moved to libFormat in this patch. The goal is to
enable other tools that use libFormat and also have a -style option to
behave consistently with clang-format.
Patch forthcoming that addresses the other concerns.
Comment at: clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp:255
@@ +254,3 @@
+if (!applyReplacements(I-getValue(), NewFileData, Diagnostics)) {
+ errs() Failed to apply replacements to I-getKey()
- Re-use much stuff from libFormat and clang-format:
- Use format::reformat() instead of reformatRanges().
- Handling of format command-line options has has a major overhaul. Now
using some of clang-format's functionality now in libFormat.
- Range coalescing doesn't
Comment at: clang-apply-replacements/lib/Tooling/ApplyReplacements.cpp:289
@@ +288,3 @@
+
+void calculateChangedRanges(
+const std::vectorclang::tooling::Replacement Replaces,
Daniel Jasper wrote:
I think a lot of the complexity stems from this function
- Simplified calculateChangedRanges() and tool's main()
Hi tareqsiraj, arielbernal, Sarcasm, klimek, djasper,
http://llvm-reviews.chandlerc.com/D1730
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1730?vs=4457id=4491#toc
Files:
clang-apply-replacements/CMakeLists.txt
- Addressed reviewer comments. As a note to @djasper, we can't just reuse
applyReplacements() within applyFormatting() since we need access to the
SourceManager
to override the file contents.
Hi tareqsiraj, arielbernal, Sarcasm, klimek, djasper,
http://llvm-reviews.chandlerc.com/D1730
Author: revane
Date: Thu Sep 26 14:10:04 2013
New Revision: 191448
URL: http://llvm.org/viewvc/llvm-project?rev=191448view=rev
Log:
clang-modernize: Reset LoopConvert's TU tracking info per TU
The LoopConvert transform makes use of data structures it builds up over
the course of transforming a
Author: revane
Date: Thu Sep 26 14:26:58 2013
New Revision: 191451
URL: http://llvm.org/viewvc/llvm-project?rev=191451view=rev
Log:
clang-modernize: Fixing a few left over cpp11-migrate references
Build files for unit tests hadn't been updated yet.
Modified:
Committed as r191448.
http://llvm-reviews.chandlerc.com/D1758
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
@klimek As you wish. I personally see no problem in readability.
http://llvm-reviews.chandlerc.com/D1730
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
- Returning containers by value as requested.
Hi tareqsiraj, arielbernal, Sarcasm, klimek, djasper,
http://llvm-reviews.chandlerc.com/D1730
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1730?vs=4493id=4496#toc
Files:
clang-apply-replacements/CMakeLists.txt
Hi tareqsiraj, arielbernal,
The LoopConvert transform makes use of data structures it builds up over
the course of transforming a TU. Until now, these data structures
weren't being cleared out before the next TU was being processed.
This resolves PR17253.
http://llvm-reviews.chandlerc.com/D1758
I'm requesting some big organizational changes to make the document a little
clearer. At some point I stopped making grammar fixes once I saw bigger changes
were necessary in those sections.
The over-arching thing to keep in mind is the audience for this document:
someone changing their
Hi tareqsiraj, arielbernal,
Options that leak from other parts of LLVM are now pruned out of -help.
-version output is specific to clang-apply-replacements now.
http://llvm-reviews.chandlerc.com/D1747
Files:
clang-apply-replacements/tool/ClangApplyReplacementsMain.cpp
Index:
Author: revane
Date: Tue Sep 24 13:14:54 2013
New Revision: 191322
URL: http://llvm.org/viewvc/llvm-project?rev=191322view=rev
Log:
clang-apply-replacements: Clean up -help and -version
Options that leak from other parts of LLVM are now pruned out of -help.
-version output is specific to
- Addressed user comments. Ported unit tests.
Hi tareqsiraj, arielbernal, Sarcasm, klimek,
http://llvm-reviews.chandlerc.com/D1730
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1730?vs=4423id=4446#toc
Files:
clang-apply-replacements/CMakeLists.txt
Oops. I see a swap file got added somehow. Will remove that.
http://llvm-reviews.chandlerc.com/D1730
BRANCH
mv-reformat
ARCANIST PROJECT
clang-tools-extra
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
As you guessed, this functionality is being moved from clang-modernize. First
step is to add it here, next step is to remove it from clang-modernize.
Comment at: clang-apply-replacements/lib/Tooling/ApplyReplacements.cpp:324
@@ +323,3 @@
+
+ // cleanups unecessary ranges to
- Addressed reviewer comments
- Also changed -format-style into -format which now has an optional value
that
defaults to the llvm style.
Hi tareqsiraj, arielbernal, Sarcasm, klimek,
http://llvm-reviews.chandlerc.com/D1730
CHANGE SINCE LAST DIFF
Have you tried this on Windows yet?
http://llvm-reviews.chandlerc.com/D1710
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
LGTM.
http://llvm-reviews.chandlerc.com/D1710
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
If you see value in it, I'd suggest splitting this main.cpp into a bunch of
smaller files that do one test each. Easier to identify errors.
Comment at: test/clang-modernize/Compilations/main.cpp:17
@@ +16,3 @@
+
+// RUN: cp %S/Inputs/compilations_expected.cpp %T
+// RUN: cp
Hi tareqsiraj, arielbernal, Sarcasm, klimek,
Updated libclangApplyReplacements interface with two new functions:
* calculateChangedRanges()
* reformatRanges()
which are used to do code formatting on a single file.
clang-apply-replacements uses these functions to apply code formatting
after
Have you tried this test on Windows?
These tests should be added:
# `-p` is specified with no files and no `-include`.
# `-p` is specified with *both* files and `-include`.
# Until you fix the compilation database auto-detect bug, You should have
another XFAIL test for when `-p` is
Closed by commit rL190158 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1610?vs=4072id=4103#toc
http://llvm-reviews.chandlerc.com/D1610
COMMIT
http://llvm-reviews.chandlerc.com/rL190158
___
cfe-commits
Author: revane
Date: Fri Sep 6 09:23:56 2013
New Revision: 190158
URL: http://llvm.org/viewvc/llvm-project?rev=190158view=rev
Log:
clang-modernize: Remove -headers option
clang-modernize can now transform headers properly and the experimental
-headers option is no longer necessary.
Remember,
Hi tareqsiraj, arielbernal, Sarcasm,
-include/-exclude and friends have been marked as hidden options until
this point. This is no longer necessary. Update the docs to describe
their effect.
http://llvm-reviews.chandlerc.com/D1621
Files:
clang-modernize/tool/ClangModernize.cpp
Closed by commit rL190194 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1621?vs=4115id=4117#toc
http://llvm-reviews.chandlerc.com/D1621
COMMIT
http://llvm-reviews.chandlerc.com/rL190194
___
cfe-commits
Author: revane
Date: Fri Sep 6 14:26:25 2013
New Revision: 190194
URL: http://llvm.org/viewvc/llvm-project?rev=190194view=rev
Log:
clang-modernize: Update docs with -include/-exclude
-include/-exclude and friends have been marked as hidden options until this
point. This is no longer necessary.
Author: revane
Date: Fri Sep 6 14:27:19 2013
New Revision: 190195
URL: http://llvm.org/viewvc/llvm-project?rev=190195view=rev
Log:
clang-modernize: Tweak docs after rename
* Removing sphinx warnings about docs not in a toctree.
* Adding more links to cpp11-migrate redirect docs.
Modified:
Comment at: clang-modernize/tool/ClangModernize.cpp:362
@@ +361,3 @@
+llvm::report_fatal_error(
+No sources or compilation database were provided.);
+
I'd just say Could not determine sources to transform.
Comment at:
Hi tareqsiraj, arielbernal, Sarcasm,
clang-modernize can now transform headers properly and the experimental
-headers option is no longer necessary.
Remember, at least -include is necessary for indicating which headers
are allowed to be changed.
http://llvm-reviews.chandlerc.com/D1610
Files:
To address the error comment I'd just tack on a test after initializing the
list of sources to transform: is the list empty then error out.
http://llvm-reviews.chandlerc.com/D1517
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
wrote:
Edwin Vane wrote:
Is all this extra work really necessary? Can't we rely on formatting to fix
whitespace problems?
I agree with you but the bug report complained about an extra space when it
was not necessary, so I figured out this solution wasn't too complicated and
may produce
Comment at: clang-modernize/PassByValue/PassByValueActions.cpp:143-146
@@ -137,2 +142,6 @@
// ~~~ ~~~^
-ValueStr += ' ';
+SourceLocation NextChar =
+Lexer::getLocForEndOfToken(ParamTL.getEndLoc(), 0, SM, LangOptions());
+if
The logic is too complex. I think the best way to clean this up is to
refactor it. What is is this code really trying to do? Two things:
# Initialize `Compilations`, the compilation database.
# Figure out the list of sources to transform.
I think you should split up the code to follow
Author: revane
Date: Tue Sep 3 12:58:19 2013
New Revision: 189832
URL: http://llvm.org/viewvc/llvm-project?rev=189832view=rev
Log:
Rename clang-replace - clang-apply-replacements
Made changes throughout clang-tools-extra for the renaming of
clang-replace to clang-apply-replacements as per
- Addressed reviewer comments
Hi klimek, chandlerc,
http://llvm-reviews.chandlerc.com/D1548
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1548?vs=3901id=3924#toc
Files:
clang-replace/CMakeLists.txt
clang-replace/Makefile
- Fixed header guards
Hi klimek, chandlerc,
http://llvm-reviews.chandlerc.com/D1548
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1548?vs=3924id=3925#toc
BRANCH
reorg
ARCANIST PROJECT
clang-tools-extra
Files:
clang-replace/CMakeLists.txt
clang-replace/Makefile
Closed by commit rL189671 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1548?vs=3925id=3926#toc
http://llvm-reviews.chandlerc.com/D1548
COMMIT
http://llvm-reviews.chandlerc.com/rL189671
___
cfe-commits
* The `data()` comment has already been addressed.
* The empty line comments are for files not in this patch. They're from a
rebase.
* Finally realized what you meant by doxygen-style comments and fixed them.
http://llvm-reviews.chandlerc.com/D1545
BRANCH
driver-model
ARCANIST PROJECT
Author: revane
Date: Fri Aug 30 10:40:26 2013
New Revision: 189681
URL: http://llvm.org/viewvc/llvm-project?rev=189681view=rev
Log:
clang-replace: Fix autoconf build failures
Updating autoconf files to reflect changes to directory structure
Modified:
Author: revane
Date: Fri Aug 30 09:33:56 2013
New Revision: 189671
URL: http://llvm.org/viewvc/llvm-project?rev=189671view=rev
Log:
clang-replace: Re-org of file structure
clang-replace is likely to move to clang proper one day soon. To facilitate
that move, renaming files and directory
Closed by commit rL189689 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1545?vs=3934id=3940#toc
http://llvm-reviews.chandlerc.com/D1545
COMMIT
http://llvm-reviews.chandlerc.com/rL189689
___
cfe-commits
Author: revane
Date: Fri Aug 30 14:45:49 2013
New Revision: 189691
URL: http://llvm.org/viewvc/llvm-project?rev=189691view=rev
Log:
cpp11-migrate: Fixing autoconf build after adding libclangReplace dependency
Modified:
clang-tools-extra/trunk/cpp11-migrate/Core/Makefile
Sorry for taking so long to respond to this review. I've been heads down
trying to prepare D1545 for review.
I'm going to break this down into two cases:
=== 1. Explicitly provided files ===
If a user explicitly lists a source on the command-line it should be
transformed. The only
Hi klimek, chandlerc,
clang-replace is likely to move to clang proper one day soon. To
facilitate that move, renaming files and directory structure layout to
ease transition for users of clang-replace and libclangReplace.
For now, functionality still exists in clang::replace namespace. Header
Replaced by D1548 which cleans up the patch and introduces a few more 'get
ready for clang' changes.
http://llvm-reviews.chandlerc.com/D1522
___
cfe-commits mailing list
cfe-commits@cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
Comment at: cpp11-migrate/Core/FileOverrides.cpp:192
@@ +191,3 @@
+std::string S = FileStates[FileName];
+S.clear();
+llvm::raw_string_ostream StringStream(S);
Tareq A. Siraj wrote:
Is it necessary to take a reference and clearing the string vs.
Comment at: docs/PassByValueTransform.rst:136
@@ +135,3 @@
+When delayed template parsing is enabled, constructors part of templated
+contexts aren't transformed. Delayed template parsing is enabled by default on
+Windows as a Microsoft extension: `Clang Compiler User’s Manual -
Closed by commit rL189493 (authored by @revane).
CHANGED PRIOR TO COMMIT
http://llvm-reviews.chandlerc.com/D1519?vs=3808id=3869#toc
http://llvm-reviews.chandlerc.com/D1519
COMMIT
http://llvm-reviews.chandlerc.com/rL189493
___
cfe-commits
Author: revane
Date: Wed Aug 28 12:19:10 2013
New Revision: 189493
URL: http://llvm.org/viewvc/llvm-project?rev=189493view=rev
Log:
clang-replace: Afford applying replacements in memory
For users of libclangReplace, this patch affords the ability to apply
replacements in memory instead of
Eventual inclusion into clang itself was a reason why I wanted to re-org the
directories to mirror clang. Thinking about it overnight I think there needs to
be an extra directory here: include/clang-replace/Replace/ instead of
include/Replace. clang-tools-extra seemed like a good place to
Author: revane
Date: Tue Aug 27 08:00:34 2013
New Revision: 189344
URL: http://llvm.org/viewvc/llvm-project?rev=189344view=rev
Log:
Adding const buffer iterator generators to Rewriter
Modified:
cfe/trunk/include/clang/Rewrite/Core/Rewriter.h
Modified:
- comment fix
- Using Rewriter to transfer data between applyReplacements and writeFiles
Hi klimek,
http://llvm-reviews.chandlerc.com/D1519
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1519?vs=3765id=3808#toc
Files:
clang-replace/ApplyReplacements.cpp
- Adding clang-replace subdirectory to 'include'
Hi klimek, chandlerc,
http://llvm-reviews.chandlerc.com/D1522
CHANGE SINCE LAST DIFF
http://llvm-reviews.chandlerc.com/D1522?vs=3768id=3809#toc
Files:
clang-replace/CMakeLists.txt
clang-replace/Makefile
Some changes are starting to leak from D1519 due to a rebase. All that
changed here was the introduction of the `clang-replace` subdirectory of
`include` and changes to `#include` directives where the header was included.
http://llvm-reviews.chandlerc.com/D1522
Main problem I see here is that if you don't explicitly include a directory,
no files get transformed. We should by default transform every file in the
compilation database. Since you're using the `IncludingExcludeInfo` structure
to identify modifiable sources and headers (which is fine I
Don't worry about the blurb on templates now. Now that the behaviour with
args that are not template args has been found to work fine no need to talk
about the case that obviously shouldn't work.
One, hopefully last, request: can you add a test where the constructor has
multiple args, 1
Author: revane
Date: Tue Aug 27 10:44:26 2013
New Revision: 189358
URL: http://llvm.org/viewvc/llvm-project?rev=189358view=rev
Log:
Adding a vector version of clang::tooling::shiftedCodePosition().
During the transition of clang::tooling::Replacements from std::set to
std::vector, functions such
Comment at: docs/PassByValueTransform.rst:135
@@ +134,3 @@
+
+When delayed template parsing is enabled, constructors with dependent types
+aren't transformed. Delayed template parsing is enabled by default on Windows
as
Need a reword on constructors with
1 - 100 of 654 matches
Mail list logo
