aprantl added a comment.
I'll take that back: make_absolute only copies when the path isn't already
absolute.
https://reviews.llvm.org/D45977
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/l
aprantl added a comment.
> Also not sure if the LLVM stuff will try to substitute in the current working
> directory for "."?
No it won't, remove_dots() has no side effects. There's a separate
make_absolute() function. That one will cause another copy. Personally, if I
have a choice between a
aprantl added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
clayborg wrote:
> clayborg wrote:
>
clayborg added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
clayborg wrote:
> aprantl wrote:
>
clayborg added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
aprantl wrote:
> clayborg wrote:
>
aprantl added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
clayborg wrote:
> I am fine switchi
clayborg added a comment.
In https://reviews.llvm.org/D45977#1076048, @aprantl wrote:
> One general question: why is this form of normalization preferred over
> calling realpath?
Normalization is everything we can do to fix up a path without knowing anything
about the current working director
clayborg added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
I am fine switching to using the l
aprantl added a comment.
One general question: why is this form of normalization preferred over calling
realpath?
https://reviews.llvm.org/D45977
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listin
aprantl added inline comments.
Comment at: source/Utility/FileSpec.cpp:107
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
+ i = path.find('/', i + 1)) {
+const auto nextChar = safeCharAtIndex(path, i+1);
clayborg wrote:
> no as the only ca
clayborg updated this revision to Diff 143643.
clayborg added a comment.
Fix comment typo
https://reviews.llvm.org/D45977
Files:
include/lldb/Core/FileSpecList.h
include/lldb/Utility/FileSpec.h
source/Breakpoint/BreakpointResolverFileLine.cpp
source/Core/FileSpecList.cpp
source/Plugin
clayborg added inline comments.
Comment at: source/Utility/FileSpec.cpp:105
+//--
+bool needsNormalization(const llvm::StringRef &path) {
+ for (auto i = path.find('/'); i != llvm::StringRef::npos;
aprantl added inline comments.
Comment at: source/Plugins/SymbolFile/DWARF/SymbolFileDWARF.cpp:189
+ // Always normalize our compile unit directory to get rid of redundant
+ // slashes and other path anonalies before we use it for path prepending
+ FileSpec local_spec(local_pa
clayborg updated this revision to Diff 143609.
clayborg added a comment.
Remove commented out code.
https://reviews.llvm.org/D45977
Files:
include/lldb/Core/FileSpecList.h
include/lldb/Utility/FileSpec.h
source/Breakpoint/BreakpointResolverFileLine.cpp
source/Core/FileSpecList.cpp
sou
aprantl added inline comments.
Comment at: lldb.xcodeproj/xcshareddata/xcschemes/darwin-debug.xcscheme:29
selectedLauncherIdentifier = "Xcode.DebuggerFoundation.Launcher.GDB"
+ language = ""
shouldUseLaunchSchemeArgsEnv = "YES">
Is this an acci
clayborg updated this revision to Diff 143608.
clayborg added a comment.
Remove unwanted file changes.
https://reviews.llvm.org/D45977
Files:
include/lldb/Core/FileSpecList.h
include/lldb/Utility/FileSpec.h
source/Breakpoint/BreakpointResolverFileLine.cpp
source/Core/FileSpecList.cpp
clayborg created this revision.
clayborg added reviewers: labath, jingham, davide, pranavb.
Herald added subscribers: JDevlieghere, aprantl.
Always normalizing lldb_private::FileSpec paths will help us get a consistent
results from comparisons when setting breakpoints and when looking for source
Author: davide
Date: Mon Apr 23 10:06:55 2018
New Revision: 330617
URL: http://llvm.org/viewvc/llvm-project?rev=330617&view=rev
Log:
[CMake] Add the missing `dsymutil` dependency when running tests.
Modified:
lldb/trunk/lit/CMakeLists.txt
Modified: lldb/trunk/lit/CMakeLists.txt
URL:
http://
labath created this revision.
labath added reviewers: JDevlieghere, aprantl.
Herald added a subscriber: eraman.
In the magic test duplicator, we were making the decision whether to
create a test variant based on the compiler and the target platform.
This meant that the set of known tests was diffe
Author: labath
Date: Mon Apr 23 04:22:44 2018
New Revision: 330578
URL: http://llvm.org/viewvc/llvm-project?rev=330578&view=rev
Log:
llgs-tests: Increase packet time out
The default packet timeout of 1 second is a bit too small for these
tests, particularly as they are working in ack-mode, which
Author: jdevlieghere
Date: Mon Apr 23 01:44:06 2018
New Revision: 330568
URL: http://llvm.org/viewvc/llvm-project?rev=330568&view=rev
Log:
[CMake] Gate 'dsymutil' dependency
Only add `dsymutil` as a test dependency when this is not a stand-alone
build (LLDB_BUILT_STANDALONE).
Modified:
lldb/
21 matches
Mail list logo
