Hi,
before submitting patches on gerrit, I would like to have your feed back
to make sure I'm right.
'OUString' have 'indexOf' that can find a substring (literal or
OUString) or a char.
IMO, looking for a 1 char long string is the same as looking for this
char itself. However, the later is faster. In some cases, we even
allocate memory before doing the search !
Example #1: we allocate memory to build a 1 char long string and search
for it
>>>>>>>>>>>>>>>>>>
diff --git a/sc/source/ui/vba/vbafilesearch.cxx
b/sc/source/ui/vba/vbafilesearch.cxx
index ddb5c04..220b2e2 100644
--- a/sc/source/ui/vba/vbafilesearch.cxx
+++ b/sc/source/ui/vba/vbafilesearch.cxx
@@ -180,7 +180,7 @@ sal_Int32 SAL_CALL ScVbaFileSearch::Execute( ) throw
(css::uno::RuntimeExceptio
if ( IsWildCard( aTempFileName ) )
{
bool bEndWithAsterisk = aTempFileName.endsWithAsciiL("*", 1);
- bool bStartWithAsterisk =
(aTempFileName.indexOf(::rtl::OUString("*")) == 0);
+ bool bStartWithAsterisk = (aTempFileName.indexOf('*')) == 0);
if ( !bEndWithAsterisk && !bStartWithAsterisk )
{
aTempFileName = ::rtl::OUString("*") + aTempFileName +
::rtl::OUString("*");
>>>>>>>>>>>>>>>>>>
Example #2: we search for a 1 char string
>>>>>>>>>>>>>>>>>>
diff --git a/shell/source/tools/lngconvex/lngconvex.cxx
b/shell/source/tools/lngconvex/lngconvex.cxx
index 5b2ecb2..dce42ab 100644
--- a/shell/source/tools/lngconvex/lngconvex.cxx
+++ b/shell/source/tools/lngconvex/lngconvex.cxx
@@ -200,7 +200,7 @@ public:
private:
void init()
{
- sal_Int32 idx = lang_.indexOf("-");
+ sal_Int32 idx = lang_.indexOf('-');
if (idx > -1)
{
>>>>>>>>>>>>>>>>>>
In both cases, I think it is safe to search only for the char irself.
There are a few of these that can be cleanup.
If you agree with this, I will provide a patch via gerrit.
Best regards
_______________________________________________
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
