google / google-toolbox-for-mac Goto Github PK

What steps will reproduce the problem?
1. Run RunIPhoneUnitTest.sh with a failed test.
2. Check return value $?

What is the expected output? What do you see instead?
Should be non-zero

What version of the product are you using? On what operating system?
trunk/r76; Mac OS X 10.5.5

Please provide any additional information below.
Attached a diff, which is how we patched this locally.

Neither the project home page nor the wiki actually give any clue as to what 
the purpose of this is, 
or what application(s) are included.

If a developer wants to make use of the unit testing additions in GTM, they 
must directly compile and include the 
GTM/UnitTesting modules in their project. It seems that, like the 
GoogleToolboxForMac.framework, there should 
be a parallel GTMTesting.framework so that developers can track GTM's SVN 
trunk/ in their unit testing code.

Use case:
Barry creates a unit testing bundle in a new project. He adds 
GTMTesting.framework as a target dependency of 
the unit testing bundle and adds GTMTesting.framework to the "Link Binary with 
Libraries" build phase of their 
unit testing bundle. GTMTesting.framewok is installed at @rpath (so that it can 
be installed either in an app test 
harness' frameworks directory (i.e. @executable_path/../Frameworks) or directly 
in the unit test bundle (ie. 
@loader_path/../Frameworks), so the Barry adds either 
@executable_path/../Frameworks or 
@loader_path/../Frameworks to the unit test bundle's runpath search path 
depending on whether his unit tests are 
run in an application harness or not.

Barry can now use GTM's unit testing additions in his own unit test code, while 
tracking improvements to GTM's 
unit testing additions via an svn:external.

Resolution of this issue would add a framework new target to GTM. This 
framework would include the 
UnitTesting/ modules and would link SenTestingKit.framework.

What steps will reproduce the problem?

1. Link a non-application bundle against GTM.framework, and copy GTM.framework 
to e.g. 
@loader_path/Frameworks for a framework target. Dynamic linking will fail 
because 
GTM.framework is built with installation path = @executable_path/../Frameworks.

Please provide any additional information below.

As far as I know, the only solution to this problem is to set the GTM.framework 
installation path 
to be @rpath. Applications would then set *their* "Runpath search paths" 
setting to include 
@executable_path/../Frameworks. Frameworks could set their "Runpath search 
paths" to include 
@loader_path/Frameworks. This way GTM.framework can be bundled in app and 
non-app 
bundles.

What steps will reproduce the problem?
Send GTMSystemVersion a +getMajor:minor:bugFix: message with a valid
|major| pointer but nil |bugFix| pointer:

SInt32 major = 0;
[GTMSystemVersion getMajor:&major minor:nil bugFix:nil];
NSLog(@"Major version: %ld", major);

What is the expected output?
$ ./a.out
2009-02-27 15:08:56.966 a.out[20508:10b] Major version: 10

What do you see instead?
$ ./a.out
Bus error

What version of the product are you using? On what operating system?
r86 (GTMSystemVersion.m last changed by r73) running under Mac OS X 10.5.6
(9G55)

Please provide any additional information below.
The attached file demonstrates the crashing behavior.

The error is caused by an erroneous test for |major| instead of |bugFix| at
GTMSystemVersion.m:105.

The current unit tests fail to catch this because they request all three
values at once, without trying any other combination of the requested outputs.

In GTMDefines.h, the _GTMDevAssert macro calls [NSString stringWithCString], 
which has been 
deprecated in Mac OS X v10.4. Documentation says that 
stringWithCString:encoding: should be 
used instead.

I have attached a patch that resolves the compilation warning.

The CorePlot project has moved to Mercurial.  One downside of this is we can no 
longer (trivially) 
include an external SVN repository.  Will GTfM be moving to Hg anytime soon, so 
we should pursue 
that route?  Or should be researching how to bridge SVN external repositories 
to Mercurial?

There seem to be workarounds for GIT, but I haven't seen anything for Mercurial.

http://panthersoftware.com/articles/view/4/git-svn-dcommit-workaround-for-git-su
bmodules

Lack of documentation.

Read in a thread somewhere that GTM has base64 encoder/decoder.  

Downloaded Google Toolbox for Mac...an installed the plug-in.  (Thanks for 
saying that the 
directory doesn't exist if you never installed a plug-in and you have to create 
it)

Now I'd like to try to use whatever it has for Base64 but of course...no 
documentation.  Any 
ideas?


What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?


Please provide any additional information below.

What steps will reproduce the problem?
1. Search for music file
2. play music file

What is the expected output? What do you see instead?
iTunes plays song from Library/DJ playlist like every other search program.
GQSB instead creates a 'Google' playlist with the one song and plays it,
causing music to stop after that one song and adding a useless playlist.

What version of the product are you using? On what operating system?
2.0.0.1447

Please provide any additional information below.

Just add this line to the beginning RuniPhoneUnitTest.sh

ps -x | egrep "iPhone Simulator" | awk '{ print $1 }' | xargs kill -9

In TestMerge (part of the Core Plot project), we've written a tool to manage 
the section of new 
output/reference/failed output files from GTM's image and state assertion 
macros. Because the 
algorithm for finding the most specific name(.arch)*(.os)*.extension path for a 
named output is 
encapsulated in GTMNSObject+UnitTesting which imports SenTesting/SenTesting.h, 
we can't use 
the existing algorithm in a non-unit testing context (i.e. within 
TestMerge.app). We've duplicated 
the code within TestMerge, but this is obviously non-idea.

It would be beneficial for us if that algorithm was exposed via a non-unit 
testing-specific API. For 
reference the use of the copied code in TestMerge is in -[OutputGroup  
mostSpecificGTMUnitTestOutputPathInSet:(NSSet*)paths name:(NSString*)name 
extension:(NSString*)ext] (http://code.google.com/p/core-
plot/source/browse/TestMerge/MOClasses/OutputGroup.m).

What steps will reproduce the problem?
1. [[[GTMHTTPFetcher alloc] init] beginFetchWithDelegate:self ...];

What is the expected output? What do you see instead?

Expect a callback to the failed selector. Instead, I get a crash as the
null selector is sent to the delegate.

What version of the product are you using? On what operating system?

R70

Please provide any additional information below.

GTMHTTPFetcher.m:288 guards on failedSEL but actually uses failedSEL_,
resulting in the null selector. 

What steps will reproduce the problem?
* Following directions to add new run script to unit testing target.

What is the expected output? What do you see instead?
* I expected for test case to fail assertion but instead got the following
error:
Running custom shell script (1 error)
line 23: 82687 Abort trap "$TARGET_BUILD_DIR/$EXECUTABLE_PAT"
-RegisterForSystemEvents
* This also seems to cause the iPhone simulator to crash before it comes up
since I get a dialog window saying so.


What version of the product are you using? On what operating system?
Mac OS X 10.5.6 with XCode 3.1.2

Please provide any additional information below.
I saw that the comments for writing iPhone unit tests includes this same
problem from another person.

GTMTheme in needs a general redefinition and purpose, or we should toss it 
entirely. Right now it 
is being used by Chrome, but it really needs to be cleaned up.

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?


Please provide any additional information below.

With OSX 10.6, the callStackSymbols call gives a nicely formatted stack
trace, whereas the GTM one is a bit hard to read because of not handling
the length of 64 bit frame addresses, and using the full paths to
executables, and some formatting issues. I therefore altered the formatting
to match what callStackSymbols gives as follows:

static NSString *GTMStackTraceFromAddressDescriptors(struct
GTMAddressDescriptor descs[],
                                                     NSUInteger count) {
    NSMutableString *trace = [NSMutableString string];

    for (NSUInteger i = 0; i < count; i++) {
        // Newline between all the lines
        if (i) {
            [trace appendString:@"\n"];
        }
        if (descs[i].class_name) {
            [trace appendFormat:@"%-2u %-35s %#016lx %s[%s %s]",
             i, 
             (descs[i].filename ? [[[NSString stringWithCString:descs[i].filename
encoding:NSUTF8StringEncoding] lastPathComponent]
cStringUsingEncoding:NSUTF8StringEncoding] : "??"),
             descs[i].address, 
             (descs[i].is_class_method ? "+" : "-"),
             descs[i].class_name,
             (descs[i].symbol ? descs[i].symbol : "??")];
        } else {
            [trace appendFormat:@"%-2u %-35s %#016lx %s()",
             i, 
             (descs[i].filename ? [[[NSString stringWithCString:descs[i].filename
encoding:NSUTF8StringEncoding] lastPathComponent]
cStringUsingEncoding:NSUTF8StringEncoding] : "??"),
             descs[i].address,
             (descs[i].symbol ? descs[i].symbol : "??")];
        }
    }
    return trace;
}


I hope this is of some use to improve the source.

There is no analogue for the AddressBook API's
ABAddressBookCopyPeopleWithName() in GTMABAddressBook. This would be
trivial to add.

I'm working on a patch that I'll submit shortly - just need to add unit
testing and figure out how to jump through all the submission hoops.

We are building the Unit-Tests as a static library and link this into test 
project.
The app will crash because GTMIPhoneUnitTestDelegate is not present.

----
2009-05-25 18:59:22.633 RABSQLite[23586:10b] *** Assertion failure in 
UIApplicationMain(), 
/SourceCache/UIKit/UIKit-738/UIApplication.m:1084
2009-05-25 18:59:22.634 RABSQLite[23586:10b] *** Terminating app due to 
uncaught 
exception 'NSInternalInconsistencyException', reason: 'Unable to instantiate 
the UIApplication 
delegate instance. No class named GTMIPhoneUnitTestDelegate is loaded.'
2009-05-25 18:59:22.636 RABSQLite[23586:10b] Stack: (
    2484961451,
    2486152763,
    2484960907,
    2493939445,
    816221235,
    61549,
    8210
)
----

The file is only referenced by name and never used, the linker strips the 
delegate.
When using the delegate inside the main function this problem is solved,

----
int main(int argc, char *argv[]) {
  NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];

  // Ensures that the app delegate will be linked correctly.
  [GTMIPhoneUnitTestDelegate description];

  int retVal = UIApplicationMain(argc, argv, nil, @"GTMIPhoneUnitTestDelegate");
  [pool release];
  return retVal;
}

---

This little change will have no negative effects and helps to use the 
Units-Tests as a static 
library.

Max

What steps will reproduce the problem?
1. In Xcode, open Preferences -> Building and set "Place Build Products
In:" to "Customized location:", then give an absolute path (e.g.
/Users/cdonnelly/Temp/Build)
2. Create a new project (or open an existing one -- shouldn't matter)
3. Build -> Show Code Coverage For -> Current project

What is the expected output? What do you see instead?

Instead of running, it errors with
"The path we needed didn't exist.
'/Users/cdonnelly/path/to/MyProject//Users/cdonnelly/Temp/Build'

It should detect that the path is absolute, but it is not.


What version of the product are you using? On what operating system?
GTM Xcode Plugin, Version 10.0.1
Xcode, Version 3.2.1, 64-bit
Component versions
* Xcode IDE: 1613.0
* Xcode Core: 1614.0
* ToolSupport: 1591.0

OS:
  System Version:   Mac OS X 10.6.2 (10C2234)
  Kernel Version:   Darwin 10.2.0
  Boot Volume:  Macintosh HD
  Boot Mode:    Normal
  Secure Virtual Memory:    Not Enabled
  64-bit Kernel and Extensions: No

Please provide any additional information below.

Hardware Overview:

  Model Name:   Mac Pro
  Model Identifier: MacPro4,1
  Processor Name:   Quad-Core Intel Xeon
  Processor Speed:  2.26 GHz
  Number Of Processors: 2
  Total Number Of Cores:    8
  L2 Cache (per core):  256 KB
  L3 Cache (per processor): 8 MB
  Memory:   12 GB
  Processor Interconnect Speed: 5.86 GT/s
  Boot ROM Version: MP41.0081.B07
  SMC Version (system): 1.39f5
  SMC Version (processor tray): 1.39f5

What steps will reproduce the problem?


1. Write code like this in an iPhone unit test:

- (void)testStrings
{
    NSString* s = @"mailto:[email protected]?subject=Hello%20World";
    STAssertEqualStrings( @"Something else", s, nil );
}


2. The error log from the test run contains gibberish in place of the URL
escape sequence. Presumably this is because %20W has been interpreted as a
printf-style substitution parameter; the escape sequence %2f yields a
floating point gibberish number, for example. 

/Users/ryan/projects/lib-iphone/UnitTests/IFMailMessageTest.m:15: error:
failure in raise (@"Something else") == (s)
(/Users/ryan/projects/lib-iphone/UnitTests/IFMailMessageTest.m:15: error:
Something else != mailto:[email protected]?subject=Hello2orld : ) : 


What is the expected output? What do you see instead?

I see "Hello2orld" instead of "Hello%20World".


What version of the product are you using? On what operating system?

1.5.1 on Mac OS X 10.5.5


Please provide any additional information below.

svn version 254 demonstrates a some issues when compiling for iPhone
simulator 3.0 and 3.1.2: 

the compilation output is listed below. 

What steps will reproduce the problem?
1. xcode 3.1.2
2. compile for 10.1.6


suggested patch:

Index: Foundation/GTMNSObject+KeyValueObserving.m
===================================================================
--- Foundation/GTMNSObject+KeyValueObserving.m  (revision 254)
+++ Foundation/GTMNSObject+KeyValueObserving.m  (working copy)
@@ -156,6 +156,9 @@

 @end

+extern int objc_atomicCompareAndSwapGlobalBarrier(void*, void*, void*); 
+
+
 @implementation GTMKeyValueObservingCenter

 + (id)defaultCenter {
Index: Foundation/GTMLoggerTest.m
===================================================================
--- Foundation/GTMLoggerTest.m  (revision 254)
+++ Foundation/GTMLoggerTest.m  (working copy)
@@ -295,7 +295,7 @@
   [fh logMessage:@"test 4" level:kGTMLoggerLevelAssert];
   [fh closeFile];

-  NSString *contents = [NSString stringWithContentsOfFile:path_];
+  NSString *contents = [NSString stringWithContentsOfFile:path_
encoding:NSUTF8StringEncoding error:nil];

   STAssertEqualObjects(@"test 0\ntest 1\ntest 2\ntest 3\ntest 4\n",
contents, nil);
 }
@@ -458,7 +458,7 @@
   [fh logMessage:@"test 6" level:kGTMLoggerLevelInfo];
   [fh closeFile];

-  NSString *contents = [NSString stringWithContentsOfFile:path_];
+  NSString *contents = [NSString stringWithContentsOfFile:path_
encoding:NSUTF8StringEncoding error:nil];
   STAssertNotNil(contents, nil);
   STAssertEqualObjects(@"test 1\ntest 2\ntest 3\ntest 4\ntest 5\ntest
6\n", contents, nil);  
 }
Index: Foundation/GTMNSFileManager+PathTest.m
===================================================================
--- Foundation/GTMNSFileManager+PathTest.m  (revision 254)
+++ Foundation/GTMNSFileManager+PathTest.m  (working copy)
@@ -141,7 +141,7 @@
     }
     for (size_t j = 0; j < sizeof(testFiles) / sizeof(NSString*); j++) {
       NSString *testFile = [testDir
stringByAppendingPathComponent:testFiles[j]];
-      STAssertTrue([@"test" writeToFile:testFile atomically:YES], nil);
+      STAssertTrue([@"test" writeToFile:testFile atomically:YES
encoding:NSUTF8StringEncoding error:nil], nil);
     }
   }




CompileC
build/GTMiPhone.build/Debug-iphonesimulator/GTMiPhoneUnitTesting.build/Objects-n
ormal/i386/GTMNSFileManager+PathTest.o
Foundation/GTMNSFileManager+PathTest.m normal i386 objective-c
com.apple.compilers.gcc.4_2
cd /Users/mwinter/tmp/google-toolbox-for-mac
setenv LANG en_US.US-ASCII
setenv PATH
"/Developer/Platforms/iPhoneSimulator.platform/Developer/usr/bin:/Developer/usr/
bin:/opt/local/bin:/opt/local/sbin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:
/Developer/usr/bin:/usr/X11/bin"
/Developer/Platforms/iPhoneSimulator.platform/Developer/usr/bin/gcc-4.2 -x
objective-c -arch i386 -fmessage-length=0 -pipe -std=c99 -Wno-trigraphs
-fpascal-strings -fasm-blocks -O0 -mdynamic-no-pic -Werror -Wreturn-type
-Wmissing-braces -Wparentheses -Wswitch -Wunused-function -Wunused-label
-Wunused-variable -Wunused-value -Wunknown-pragmas -Wshadow -Wnewline-eof
-D_GLIBCXX_DEBUG_PEDANTIC -D_GLIBCXX_DEBUG -D_GLIBCPP_CONCEPT_CHECKS
-DDEBUG=1 -D__IPHONE_OS_VERSION_MIN_REQUIRED=30000 -isysroot
/Developer/Platforms/iPhoneSimulator.platform/Developer/SDKs/iPhoneSimulator3.0.
sdk
-fvisibility=hidden -mmacosx-version-min=10.5 -gdwarf-2 -iquote
/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesimu
lator/GTMiPhoneUnitTesting.build/GTMiPhoneTest-generated-files.hmap
-I/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesi
mulator/GTMiPhoneUnitTesting.build/GTMiPhoneTest-own-target-headers.hmap
-I/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesi
mulator/GTMiPhoneUnitTesting.build/GTMiPhoneTest-all-target-headers.hmap
-iquote
/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesimu
lator/GTMiPhoneUnitTesting.build/GTMiPhoneTest-project-headers.hmap
-Wall -Wendif-labels -Winvalid-pch -Wformat=2 -Wmissing-format-attribute
-Wwrite-strings -Wstack-protector -Wstrict-aliasing=2 -Wpacked
-Wmissing-field-initializers -Wextra -Wno-unused-parameter -Wpointer-arith
-Wdisabled-optimization -Wfloat-equal
-F/Users/mwinter/tmp/google-toolbox-for-mac/build/Debug-iphonesimulator
-I/Users/mwinter/tmp/google-toolbox-for-mac/build/Debug-iphonesimulator/include
-I/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesi
mulator/GTMiPhoneUnitTesting.build/DerivedSources/i386
-I/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesi
mulator/GTMiPhoneUnitTesting.build/DerivedSources
-Wdiv-by-zero -Wbad-function-cast -Wnested-externs -Wold-style-definition
-fstack-protector -fstack-protector-all -include
/var/folders/h+/h+AKwl6EGhuQQkkrG6PCMk+++TI/-Caches-/com.apple.Xcode.501/SharedP
recompiledHeaders/GTM_Prefix-aujutpcpiljyurcgaitldeskopbz/GTM_Prefix.pch
-c
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSFileManager+PathTest.m
-o
/Users/mwinter/tmp/google-toolbox-for-mac/build/GTMiPhone.build/Debug-iphonesimu
lator/GTMiPhoneUnitTesting.build/Objects-normal/i386/GTMNSFileManager+PathTest.o

cc1obj: warnings being treated as errors
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSFileManager+PathTest.m
:
In function '-[GTMNSFileManager_PathTest
testfilePathsWithExtensionsInDirectory]':
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSFileManager+PathTest.m
:144:
warning: 'writeToFile:atomically:' is deprecated (declared at
/Developer/Platforms/iPhoneSimulator.platform

cc1obj: warnings being treated as errors
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMLoggerTest.m: In
function '-[GTMLoggerTest testFileHandleWriter]':
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMLoggerTest.m:298:
warning: 'stringWithContentsOfFile:' is deprecated (declared at
/Developer/Platforms/iPhoneSimulator.platform/Developer/SDKs/iPhoneSimulator3.0.
sdk/System/Library/Frameworks/Foundation.framework/Headers/NSString.h:357)
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMLoggerTest.m: In
function '-[GTMLoggerTest testFileHandleCreation]':
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMLoggerTest.m:461:
warning: 'stringWithContentsOfFile:' is deprecated
cc1obj: warnings being treated as errors
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSObject+KeyValueObservi
ng.m:
In function '+[GTMKeyValueObservingCenter defaultCenter]':
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSObject+KeyValueObservi
ng.m:169:
warning: implicit declaration of function
'objc_atomicCompareAndSwapGlobalBarrier'
/Users/mwinter/tmp/google-toolbox-for-mac/Foundation/GTMNSObject+KeyValueObservi
ng.m:169:
warning: nested extern declaration of 'objc_atomicCompareAndSwapGlobalBarrier'

Will I see the ability to access Google notes from my application any time 
soon?  I want to sync 
to/from Google notes on the web and via iPhone.

Jim Lee
[email protected]

What steps will reproduce the problem?
1. Create a test case that uses STAssertTrue or STAssertFalse
2. Have the assertion fail

What is the expected output? What do you see instead?
For STAssertTrue:
The message should be something like expression should be TRUE.
Instead, the message says something like expression should be FALSE

What version of the product are you using? On what operating system?
SVN trunk as of 7/24/09. Mac OS 10.5.6 with iPhone OS 3.0 simulator

Please provide any additional information below.
Maybe I'm missing something, but it seems like the error messages for
STAssertTrue and STAssertFalse are reversed? I created the same tests with
OCUnit and the correct messages are displayed.

What steps will reproduce the problem?
1. Create a new file in XCode.
2. Add enough code so that there is a scrollbar.
3. Add a line that has trailing whitespace.
4. Save the file, noting the position of the text.

What is the expected output? What do you see instead?

It clears the whitespace just fine, but after doing so, it scrolls the document 
up, causing me to 
loose my position.  The cursor does not change position.

What version of the product are you using? On what operating system?

GTM XCode Plugin 10.0.3 in XCode 3.2.2 on Mac OS X 10.6.3.

What steps will reproduce the problem?
1. In Xcode preferences, select Building from the toolbar, and change 
"Place Build Products in:" value to /Users/[username]/build
2. Install GTM per wiki instructions for Snow Leopard.
3. (problem occurs with or without this step) Select Build -> Turn Code 
Coverage On
4. Select Build -> Show Code Coverage For -> Current project

What is the expected output? What do you see instead?
I don't know what it is supposed to do, but the output is an alert window 
with the following text:
"The path we needed didn't exist. '/Users/[username]/[project 
directory]//Users/[username]/build' "


What version of the product are you using? On what operating system?
GTM plugin 10.0.1 on OS X 10.6.2

Please provide any additional information below.

The change to include <objc/objc-auto.h> for 
objc_atomicCompareAndSwapGlobalBarrier
breaks compilation on the iPhone:
http://code.google.com/p/google-toolbox-for-mac/source/detail?r=326

The compiler complains about the redefinition of 
objc_atomicCompareAndSwapGlobalBarrier and 
objc_atomicCompareAndSwapInstanceVariableBarrier which are originally defined in
/Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS3.0.sdk/usr/includ
e/objc/objc
-auto.h
which is now included in all cases.

I guess these 2 definitions should be removed from GTMObjC2Runtime.h when 
targeting iPhone OS 
3.x (unless there's a good reason to have the custom implementations of course).

Simply an option to automatically run the full test suite after loading the 
iPhone application. This 
means we would not have to click the run button.

What steps will reproduce the problem?
1. Create a simple server app using GTMHTTPServer
2. Sends some 1 MB of data
3. The app will have a peak memory usage of 1 GB.

What is the expected output? What do you see instead?
The memory usage should be low.

What version of the product are you using? On what operating system?
Mac OS X 10.5.

Please provide any additional information below.
The problem is that -[GTMHTTPServer dataAvailableNotification:] creates new 
instances of 
NSData/CFData in various places (incl. when appending data to HTTP message by 
calling CFHTTPMessageAppendBytes). When there are a succession of received 
notifications, autorelease 
pool may not be drained timely, and the accumulated NSData/CFData objects start 
to choke.

A proposed fix is attached as follows. By adding an autorelease pool and by 
draining it at various 
exit points in -[GTMHTTPServer dataAvailableNotification:], the memory usage is 
now brought 
under control.

(Side note: the server can still be extremely slow when receiving body whose 
size > 2 MB)

What steps will reproduce the problem?
1. Use STAssertEuqals in Test (like so: STAssertEquals([field doubleValue], 
0.0, nil); )
2. Set Compiler Version to Clang LLVM 1.0 in Build Settings
3. Build

What is the expected output? What do you see instead?
A warning is shown for each use:
"result of comparison against @encode is unspecified (use strcmp instead)"

What version of the product are you using? On what operating system?
XCode 3.2 on Snow Lepoard, SDK 3.1

Please provide any additional information below.
I have attached a patch for GTMSenTestCase.h that uses strcmp instead of != for 
the comparison 
of types in STAssertEquals

What steps will reproduce the problem?
Run a test

What is the expected output? What do you see instead?

With the following log, the error shows up twice in the xcode inline error 
display:
<code>
    /bin/sh -c /Users/mml/xcode_projects/.builds/LSTime.build/Debug-
iphonesimulator/UnitTests.build/Script-A1407FE90F0035FA00B120AF.sh
2008-12-23 12:08:23.414 UnitTests[13155:10b] Failed to load NSMainNibFile 
MainWindow.
Test Suite '/Users/mml/xcode_projects/Debug-iphonesimulator/UnitTests.app' 
started at 2008-
12-23 12:08:23 -0600
Test Suite 'LSTimeViewControllerTest' started at 2008-12-23 12:08:23 -0600
2008-12-23 12:08:23.420 UnitTests[13155:10b] *** -[UIRoundedRectButton text]: 
unrecognized selector sent to instance 0x4533c0
error: *** -[UIRoundedRectButton text]: unrecognized selector sent to instance 
0x4533c0
error: *** -[UIRoundedRectButton text]: unrecognized selector sent to instance 
0x4533c0
Test Case '-[LSTimeViewControllerTest testButtonIsWired]' failed (0.004 
seconds).
Test Case '-[LSTimeViewControllerTest testHasButton]' passed (0.001 seconds).
Test Case '-[LSTimeViewControllerTest testHasLabel]' passed (0.001 seconds).
Test Suite 'LSTimeViewControllerTest' finished at 2008-12-23 12:08:23 -0600.
Executed 3 tests, with 1 failures (1 unexpected) in 0.007 (0.007) seconds
Test Suite 'SenTestCase' started at 2008-12-23 12:08:23 -0600
Test Suite 'SenTestCase' finished at 2008-12-23 12:08:23 -0600.
Executed 0 tests, with 0 failures (0 unexpected) in 0.000 (0.000) seconds
Test Suite 'GTMTestCase' started at 2008-12-23 12:08:23 -0600
Test Suite 'GTMTestCase' finished at 2008-12-23 12:08:23 -0600.
Executed 0 tests, with 0 failures (0 unexpected) in 0.000 (0.000) seconds
Test Suite '/Users/mml/xcode_projects/Debug-iphonesimulator/UnitTests.app' 
finished at 
2008-12-23 12:08:23 -0600.
Executed 3 tests, with 1 failures (1 unexpected) in 0.008 (0.008) seconds
</code>

What version of the product are you using? On what operating system?
5.1, osx

Please provide any additional information below.
Running w/ "Build", not "build & go"

What steps will reproduce the problem?
1. Create a directory with a space in the name, e.g. "SVN Repositories"
2. Checkout GTM into that directory (or a subdirectory)
3. Open GTMiPhone.xcodeproj
4. Build

What is the expected output? 
A clean build.

What do you see instead?
- Build fails with "Command /bin/sh failed with exit code 1".
- last line of script output is "usage: dirname path"


What version of the product are you using? On what operating system?
svn trunk rev131 (also exhibits in 1.5.1, which is why I checked out the trunk)
MacOS X 10.5.7, iPhone SDK 2.2.1

Please provide any additional information below.

This is fixed by changing line 53 of RunIPhoneUnitTest.sh from:

  ScriptDir=$(dirname $(echo $0 | sed -e "s,^\([^/]\),$(pwd)/\1,"))

to:

  ScriptDir=$(dirname "$(echo $0 | sed -e "s,^\([^/]\),$(pwd)/\1,")")

What steps will reproduce the problem?
1. Installed product 
2. Spotlight search in folder with scpt files produces no results, with a 
common search term I 
know exists
3.

What is the expected output? What do you see instead?


What version of the product are you using? On what operating system?
Latest version fro this website

Please provide any additional information below.

When GTMScriptRunner is used to call a script which generates a lot of output 
in stderr, the 
run:standardError: method deadlocks because the [fromTask readDataToEndOfFile] 
waits for the 
script to terminate and the script waits for the runner to collect from the 
full stderr pipe.

What steps will reproduce the problem?
1. open a terminal window or a svn client
2. type on it 'svn checkout
http://google-toolbox-for-mac.googlecode.com/svn/trunk/'
3. svn server is unreachable

What is the expected output? What do you see instead?
Expected to download a read-only copy of trunk source code instead I can't
download anything, although I'm able to browse svn tree through http.

What version of the product are you using? On what operating system?


Please provide any additional information below.
I don't know if it's the correct place to inform, but I just wanted you to
know. Sorry for the inconvienence.

The AppleScript Spotlight importer installation fails since the 
app/binary/package of the Script Editor.app (whatever you'd prefer) has been 
moved and renamed:

In 10.5.x it is:
/Applications/AppleScript/Script Editor.app

In 10.6.x it changed to:
/Applications/Utilities/AppleScript Editor.app

So the installation script would need to accommodate both locations.

Thus said, even copying it manually does not make it work. Something else 
necessary for 10.6?

Steps I've tried:
1. Created directory with:
  sudo mkdir -m 755 -p /Applications/Utilities/AppleScript\ Editor.app/Contents/Library/Spotlight

2. Copied AppleScript.mdimporter to:
  /Applications/Utilities/AppleScript Editor.app/Contents/Library/Spotlight/AppleScript.mdimporter

3. Tried to register with: 
  su ${USER} -c "/usr/bin/mdimport -r '/Applications/Utilities/AppleScript\ Editor.app/Contents/Library/Spotlight/AppleScript.mdimporter'"

Listing the currently active mdimporters with "mdimport -L" shows that the 
bundle did *not* register.
system.log says nothing.
Searching for a known-string-inside-an-AppleScript.scpt yields nothing, so most 
likely the importer isn't working at all.
If because of not registering or because of another general problem, I'm not 
able to say.

rdar://6857649 - NSGradient fails when using LAB Colorspace

causes -[GTMTheme patternColorForStyle:active:] to fail on 10.6 , sample code:

NSColor *backgroundColor = [NSColor blueColor];
NSColor *startColor = [backgroundColor 
gtm_colorAdjustedFor:GTMColorationLightShadow faded:NO];
NSColor *endColor = [backgroundColor 
gtm_colorAdjustedFor:GTMColorationLightShadow faded:NO]; 
NSGradient *gradient = [[NSGradient alloc] initWithStartingColor:startColor 
endingColor:endColor];

The workaround for this issue is to convert colors to RGB colorspace before 
passing them to NSGradient.

The workaround really messes up the surrounding code with conversions (and also 
makes a bunch of unittests fail since they assume they're getting LAB colors), 
so for now we're holding off.  If the radar bug isn't fixed soon we'll need to 
fix this in GTM.

What steps will reproduce the problem?
1. [GTMABMutableMultiValue valueWithPropertyType: kABMultiStringPropertyType]

What is the expected output? What do you see instead?
Expected:  an instance of GTMABMutableMultiValue
Saw: nil


What version of the product are you using? On what operating system?
Where do I find the version?  Downloaded within the last 2 weeks, current 
version still has 
problem (see below)


Please provide any additional information below.

- (id) initWithPropertyType: (ABPropertyType) type

has the following test:
if (!(type & kABMultiValueMask) && (type != kABInvalidPropertyType))

if you remove the first !, all is well, i.e.
if ((type & kABMultiValueMask) && (type != kABInvalidPropertyType))

GTM compile phase using base SDK 10.6 shows a number of errors saying that
specific API used in GTM is deprecated starting from 10.6.

Steps to reproduce:
1. open GTM.xcodeproj project
2. select SnowLeopardOrLater-Debug configuration
3. select UnitTest-Foundation target
4. cmd+B to run compilation and tests

Expected result:
compilation succeed without errors (and warnings), tests passed

Actual result:
compilation failed with a number of errors reported

OS: Mac OS X 10.6.3 (10D573)
Xcode - v. 3.2.2 with iPhone SDK
(Component versions: Xcode IDE: 1650.0 Xcode Core: 1648.0 ToolSupport: 1631.0)
GTM - checkout of trunk r:335.

Please see errors reported in attached file.

What steps will reproduce the problem?
1. Use the UIImage category on a second thread

What is the expected output? What do you see instead?
Expected: EXC_BAD_ACCESS if another thread affects the context whilst drawing.
Actual: As above. Inconsistent reproduction, due to it requiring specific
thread interleaving.

What version of the product are you using? On what operating system?
Version: Latest source code.
OS: iPhone/Any.

Please provide any additional information below.
I suggest reading
http://vocaro.com/trevor/blog/2009/10/12/resize-a-uiimage-the-right-way/
for details on how to perform the same operations using pure CoreGraphics
functions, which are thread-safe.

What steps will reproduce the problem?

1. Open GTMiPhone.xcodeproj.
2. Attach an iPad configured for development.
3. Change the SDK to 3.2 and the target to device.
4. Build and run.

What is the expected output?

Expect the project to build, deploy to the iPad, and run.

What do you see instead?

Errors about the project targeting iPhone, instead of iPad; and armv6 instead 
of armv7.

What version of the product are you using? On what operating system?

XCode 3.2.2 64-bit with iPhone SDK 3.2 on Mac OS X 10.6.3.

Please provide any additional information below.

Patch to follow.

What steps will reproduce the problem?
1. Add framework to project as instructed
2. Execute target (runs script)

What is the expected output? What do you see instead?
Unexpected error appears:  Line 113: 1253 Bus error 
"$TARGET_BUILD_DRI/$EXECUTABLE_PATH" 
-RegisterForSystemEvents.  OS reports a crash.

What version of the product are you using? On what operating system?
SVN trunk, on OS X 10.5.6 with XCode 3.1.3

Please provide any additional information below.
Tests appear to run...  Crash dump attached

What steps will reproduce the problem?
1. Create a failing unit test using the iPhone unit test support.
2. Run the build in XCode. [Cmd+B]
3. Examine the build result.

What is the expected output? The build should report failure. What do you see 
instead? Instead 
reports success


What version of the product are you using? On what operating system?
Downloaded early Dec '08 or Late Nov '08, not sure.

Please provide any additional information below.
I have an update to the RunIPhoneUnitTest.sh that I'm inlining below. Feel free 
to add this to the 
project, mention me on the site, add me as a contributor, or donate $1 million 
in large bills. 
Whatever suits you.

#Begin RunIPhoneUnitTest.sh
#!/bin/sh
#  RunIPhoneUnitTest.sh
#  Copyright 2008 Google Inc.
#  
#  Licensed under the Apache License, Version 2.0 (the "License"); you may not
#  use this file except in compliance with the License.  You may obtain a copy
#  of the License at
# 
#  http://www.apache.org/licenses/LICENSE-2.0
# 
#  Unless required by applicable law or agreed to in writing, software
#  distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
#  WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.  See the
#  License for the specific language governing permissions and limitations under
#  the License.
#
#  Runs all unittests through the iPhone simulator

export DYLD_ROOT_PATH="$SDKROOT"
export DYLD_FRAMEWORK_PATH="$CONFIGURATION_BUILD_DIR"
export IPHONE_SIMULATOR_ROOT="$SDKROOT"
export CFFIXED_USER_HOME="$USER_LIBRARY_DIR/Application Support/iPhone 
Simulator/User"

echo "Props"
echo "export DYLD_ROOT_PATH=$DYLD_ROOT_PATH"
echo "export DYLD_FRAMEWORK_PATH=$DYLD_FRAMEWORK_PATH"
echo "export IPHONE_SIMULATOR_ROOT=$IPHONE_SIMULATOR_ROOT"
echo "export CFFIXED_USER_HOME=$CFFIXED_USER_HOME"

echo "Starting the build"
"$TARGET_BUILD_DIR/$EXECUTABLE_PATH" -RegisterForSystemEvents 2>&1 | tee 
.theresults
PASSFAIL=`cat .theresults | tail -1 | awk '{if($5 != 0) print "fail"; else 
print "pass"}'`
rm .theresults
echo "Build passed? $PASSFAIL"
echo "$PASSFAIL=pass"
if [ "$PASSFAIL" = pass ];
    then
    echo "Exit w/ success"
    exit 0
else
    echo "Exit w/ failure"
    exit -1
fi
#End RunIPhoneUnitTest.sh

What steps will reproduce the problem?
1. Download the release file
2. Unzip
3. cat GTM Xcode Plugin.xcplugin/Contents/Info.plist
4. See 10.0.3 instead of 10.0.4

This appears to be fixed on trunk as the exit strategy is different.
However, for the release tag it is broken.

What steps will reproduce the problem?
1. Running tests against iPhone 3.0 Simulator using google test kit 1.5.1

What is the expected output? What do you see instead?

Expected:

Executed 193 tests, with 0 failures (0 unexpected) in 0.045 (0.045) seconds

Got:

Executed 193 tests, with 0 failures (0 unexpected) in 0.045 (0.045) seconds
2009-03-18 09:56:15.558 Tests[64567:10b] *** -[UIApplication terminate]:
unrecognized selector sent to instance 0xd14410
2009-03-18 09:56:15.558 Tests[64567:10b] *** Terminating app due to
uncaught exception 'NSInvalidArgumentException', reason: '***
-[UIApplication terminate]: unrecognized selector sent to instance 0xd14410'
2009-03-18 09:56:15.558 Tests[64567:10b] Stack: (
    807800427,
    2419809851,
    808176203,
    807751910,
    807604514,
    11166,
    814624733,
    814659967,
    814621115,
    810661610,
    807585791,
    807581512,
    838921506,
    838921703,
    814661496,
    13982,
    8698
)

What version of the product are you using? On what operating system?
iPhone 3.0 Simulator, MacOS X 10.5.6

Please provide any additional information below.

Apple changed their 'UIApplication' private api and renamed the 'terminate'
method to 'terminateWithSuccess'. This can be verified by using the
class-dump utility against the UIKit framework.

Attached is a patched version of GTMIPhoneUnitTestDelegate.m that makes the
necessary changes.

GTMProgressMonitorInputStream has no unittests. We need some.

What steps will reproduce the problem?
1. Starting with working unit tests using 2.2.1 SDK, use 3.0 GM SDK
2. Variety of issues: bad CPU, BPT trap, etc.
3.

What is the expected output? What do you see instead?
Expect build success, rather than compilation errors and test framework 
problems.

What version of the product are you using? On what operating system?
1.5.1 of GTM on OS X 10.5.7 with iPhone OS SDK 3.0 GM

Please provide any additional information below.
If someone can comment that they have this working with 3.0 SDK, I'll try a bit 
more, but just 
seems like the switch to 3.0 broke GTM iPhone unit tests.

What steps will reproduce the problem? - not a problem. it's a request
1. write a unit test which fails 
2. build
3. run and then step through code to see the failure.

What is the expected output? What do you see instead?

There are cases where I would like to be able to step through failed code
at runtime opposed to having to look into the error at compile time.

What version of the product are you using? On what operating system?
xcode 3.2 and did an svn checkout on 9-2

Please provide any additional information below.

We need someone to figure out how the ARM stack traces work, and add a little 
code.

It is common to enter placeholder values when adding bindings from IB.
GTMUILocalizer should support localizing these placeholder values.

We should be consistent in our case handling for iPhone.