- Pharo-dev - lists.pharo.org

watcher window ?
by Nicolai Hess May 4, 2015

May 4, 2015

What is a watcher window ? (UITheme>>#watcherWindowActiveFillStyleFor: aWindow) The only sender is GLMWatcherWindow, but the theme setting is in package Polymorph-Widgets, not in a glamour extension.

1 0

Re: [Pharo-dev] build pharo vm on windows (old mingw package)
by Nicolai Hess May 4, 2015

May 4, 2015

FINALLY! Got it working. I took another mingw distribution from http://tdm-gcc.tdragon.net/ (installer version tdm-gcc-4.7.1-2.exe) With mingwsys from my previos mingw installation. The gcc version this distribution uses: Using built-in specs. COLLECT_GCC=c:\mingw\bin\gcc.exe COLLECT_LTO_WRAPPER=c:/mingw/bin/../libexec/gcc/mingw32/4.7.1/lto-wrapper.exe Target: mingw32 Configured with: ../../src/gcc-4.7.1/configure --build=mingw32 --enable-languages=c,c++,ada,fortran,objc,obj-c++ --enable-threads=win32 --enable-libgomp --enable-lto --enable-fully-dynamic-string --enable-libstdcxx-debug --enable-version-specific-runtime-libs --with-gnu-ld --disable-nls --disable-win32-registry --disable-symvers --disable-build-poststage1-with-cxx --disable-werror --prefix=/mingw32tdm --with-local-prefix=/mingw32tdm --enable-cxx-flags='-fno-function-sections -fno-data-sections' --with-pkgversion=tdm-1 --enable-sjlj-exceptions --with-bugurl=http://tdm-gcc.tdragon.net/bugs Thread model: win32 gcc version 4.7.1 (tdm-1) All is working now, no crash on startup, no missing dependencies to the mingw runtime. 2015-05-04 0:06 GMT+02:00 Nicolai Hess <nicolaihess(a)web.de>: > Or what is the output of > depends libfreetype-6.dll > > 2015-05-03 23:47 GMT+02:00 Nicolai Hess <nicolaihess(a)web.de>: > >> No SDL2? >> >> This is want depends shows for me. >> (but it is the same output as for the vm from the build server - and that >> one >> works). There is a dependency on libgcc_s_dw2-1.dll, only for some >> modules that are >> loaded at runtime. Can you rename your mingw installation directory and >> start pharo (and look that >> freetype/cairo/SDL are loadable? >> >> thank you for your help :) >> >> >> 2015-05-03 22:56 GMT+02:00 phil(a)highoctane.be <phil(a)highoctane.be>: >> >>> [image: Inline image 1] >>> >>> >>> Pharo.exe - 5 363 606 bytes >>> >>> Phil >>> >>> >>> On Sun, May 3, 2015 at 1:17 PM, Nicolai Hess <nicolaihess(a)web.de> wrote: >>> >>>> >>>> >>>> 2015-05-01 23:05 GMT+02:00 phil(a)highoctane.be <phil(a)highoctane.be>: >>>> >>>>> Did you tried this ? >>>>> >>>>> https://github.com/pharo-project/pharo-vm/blob/master/README-Win32.md >>>>> >>>>> (I wrote that thing a while ago). >>>>> >>>>> >>>>> http://sourceforge.net/projects/mingw/files/Automated%20MinGW%20Installer/m… >>>>> was where I got the env from. >>>>> >>>>> I've got a PharoVM built with gcc 4.8.1 >>>>> >>>> >>>> Can you check that this vm does not depend on the mingw runtime. >>>> (I managed to create a vm that opens, but if I start this without the >>>> mingw installation directory in my path, >>>> I get the following error: >>>> libgcc_s_dw2-1.dll is missing ) >>>> >>>> The vm from the build server works even without a local mingw >>>> installation. >>>> >>>> >>>>> >>>>> Uploading the thing in dropbox. >>>>> >>>>> Will send a link when done. >>>>> >>>>> Phil >>>>> >>>>> >>>>> On Fri, May 1, 2015 at 9:04 PM, Nicolai Hess <nicolaihess(a)web.de> >>>>> wrote: >>>>> >>>>>> >>>>>> >>>>>> 2015-04-30 11:33 GMT+02:00 Esteban Lorenzano <estebanlm(a)gmail.com>: >>>>>> >>>>>>> arenât you able to install an older version of gcc in your newer >>>>>>> version of mingw? >>>>>>> >>>>>> >>>>>> hm, downgrading the compiler was suprisingly easy. But I have still >>>>>> the same error. >>>>>> This is the gcc version information: >>>>>> >>>>>> Using built-in specs. >>>>>> COLLECT_GCC=c:\MinGW\bin\gcc.exe >>>>>> >>>>>> COLLECT_LTO_WRAPPER=c:/mingw/bin/../libexec/gcc/mingw32/4.6.2/lto-wrapper.exe >>>>>> Target: mingw32 >>>>>> Configured with: ../gcc-4.6.2/configure >>>>>> --enable-languages=c,c++,ada,fortran,objc,obj-c++ --disable-sjlj-exceptions >>>>>> --with-dwarf2 --enable-shared --enable-libgomp --disable-win32-registry >>>>>> --enable-libstdcxx-debug --enable-version-specific-runtime-libs >>>>>> --build=mingw32 --prefix=/mingw >>>>>> Thread model: win32 >>>>>> gcc version 4.6.2 (GCC) >>>>>> >>>>>> >>>>>> Anyone knows what could be the problem ? Stacktrace from the crash, >>>>>> this happens >>>>>> directly after the start: >>>>>> >>>>>> #0 0x772359dd in ntdll!RtlpNtMakeTemporaryKey () >>>>>> from C:\Windows\system32\ntdll.dll >>>>>> #1 0x77201d8f in ntdll!EtwSetMark () from >>>>>> C:\Windows\system32\ntdll.dll >>>>>> #2 0x77236a37 in ntdll!RtlpNtMakeTemporaryKey () >>>>>> from C:\Windows\system32\ntdll.dll >>>>>> #3 0x771fa3ef in ntdll!EtwSetMark () from >>>>>> C:\Windows\system32\ntdll.dll >>>>>> #4 0x771c6766 in wcsnicmp () from C:\Windows\system32\ntdll.dll >>>>>> #5 0x76cdc484 in KERNEL32!HeapFree () from >>>>>> C:\Windows\system32\kernel32.dll >>>>>> #6 0x6c24c4d7 in free () from C:\Windows\system32\crtdll.dll >>>>>> #7 0x00850000 in ?? () >>>>>> #8 0x00542065 in glob_match () >>>>>> #9 0x00542604 in __mingw_glob () >>>>>> #10 0x00540f17 in _setargv () >>>>>> #11 0x004011ca in __mingw_CRTStartup () >>>>>> #12 0x004012d5 in WinMainCRTStartup () >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> nicolai >>>>>> >>>>>> >>>>>> >>>>>> >>>>>>> I do not have time right now to update the build into a new >>>>>>> compiler, sorry (also, that is not always trivial/desired, he) >>>>>>> >>>>>>> Esteban >>>>>>> >>>>>>> > On 30 Apr 2015, at 08:38, Nicolai Hess <nicolaihess(a)web.de> wrote: >>>>>>> > >>>>>>> > I managed to trash my mingw build environment >>>>>>> > -> uninstalled /removed mingw/mingsys >>>>>>> > -> reinstall from current mingw-installer >>>>>>> > -> bad idea >>>>>>> > >>>>>>> > I can build a pharo windows vm but it immediately crashes on >>>>>>> startup. I suspect >>>>>>> > some changes between gcc 4.6 (old environment) gcc 4.8 (new >>>>>>> environment). >>>>>>> > (I think there are some changes related to file globbing (I >>>>>>> managed to get >>>>>>> > the vm somehow working to not crash on startup but on shutdown >>>>>>> instead - with >>>>>>> > _CRT_glob = 0. And gcc-4.8 uses dfferent (dwarf vs. sjlj ) >>>>>>> exception handling, don't know if this is >>>>>>> > related). >>>>>>> > >>>>>>> > I tried to find a ming-installer for the old version but could not >>>>>>> find anything and >>>>>>> > I tried to manually setup a ming installation from the old >>>>>>> packages (dozen of packages >>>>>>> > for mingw / core-utils / bintutils /msys from different places (no >>>>>>> fun)). >>>>>>> > >>>>>>> > 1) did anyone get a windows vm from mingws current version (with >>>>>>> gcc 4.8) working? >>>>>>> > >>>>>>> > 2) can someone package/zip and upload an old mingw installation >>>>>>> (maybe that one that >>>>>>> > is used on the build server? >>>>>>> > >>>>>>> > >>>>>>> > regards >>>>>>> > nicolai >>>>>>> >>>>>>> >>>>>> >>>>> >>>> >>> >>> >>> >> >

1 0

[SURVEY] Code Critics Usage
by Yuriy Tymchuk May 4, 2015

May 4, 2015

Dear Pharo developers, I kindly ask you to participate in the Code Critics mini-survey: https://www.esurveycreator.com/s/8af18d8. It will take only a coupe minutes of your time (or less), and will give us the better understanding of how Code Critics are currently used and how to improve them. Cheers! Yuriy

3 2

MC: Strange Problem
by Sean P. DeNigris May 4, 2015

May 4, 2015

I don't know how to describe it. Here is a screenshot from the MC browser showing part of it. I adopted a version from Pharo/TxText and this is what appeared instead of the package name: <http://forum.world.st/file/n4824138/Screenshot_2015-05-04_01.png> I thought maybe it didn't matter, but it really screws up the history, which now shows everything has changed since the previous version (but I only added a version and bumped #stable) ----- Cheers, Sean -- View this message in context: http://forum.world.st/MC-Strange-Problem-tp4824138.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

4 13

Playground request - ivar references to
by Ben Coman May 4, 2015

May 4, 2015

This doesn't hurt me enough to try to do it myself yet, but one thing I feel it would be natural to do in Raw view would be to have a <references to> context menu item for the instance variables, or even better like the debugger have <List methods using...> and <List methods storing into...>. btw I notice those last two actually are missing from GTDebugger and I miss them :). (or am I missing something functionality?) cheers -ben

2 1

Validation Artifacts
by Sean P. DeNigris May 4, 2015

May 4, 2015

It seems that only the report is saved. On failures, shouldn't we also save the Fuel-outs e.g. of the failed tests? ----- Cheers, Sean -- View this message in context: http://forum.world.st/Validation-Artifacts-tp4824219.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

2 1

New Versionner Tool and Authentication
by Sean P. DeNigris May 4, 2015

May 4, 2015

It seems maybe the "commit Xyz patch" tool is creating new instances of repositories instead of using the existing ones? Even though I am logged into sthub/Pharo/TxText, when I try to commit a config and new package versions for packages, dialogs pop up to enter my username and password, and even if I fill them out the credentials don't seem to be accepted. After cancelling, I can copy them over from my cache in the MC browser, so I'm definitely logged in. Any ideas? ----- Cheers, Sean -- View this message in context: http://forum.world.st/New-Versionner-Tool-and-Authentication-tp4824032.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

2 9

[sligt improvement] New Binary File Stream
by Guillermo Polito May 4, 2015

May 4, 2015

Hi guys, due to last week's discussions and some comments in the issue tracker, I wrote this morning the little improvement of the day: A simple BinaryFileStream. That is: A file stream that is - simple (no character encoding, no buffering) - binary (no automatic transformation to ascii characters) The good thing is that these features are not lost: this file stream is composable with Zn stream decorators: "encoding" fileStream := BinaryFileStream openFileNamed: 'test.txt'. writeStream := ZnCharacterWriteStream on: fileStream. writeStream nextPutAll: 'aÃ¡Ã¨OÃ»Ã¯'. writeStream nextPut: $a. fileStream := BinaryFileStream openFileNamed: 'test.txt'. readStream := ZnCharacterReadStream on: fileStream. readStream upToEnd. "Buffering" fileStream := BinaryFileStream openFileNamed: 'test.txt'. readStream := ZnBufferedReadStream on:(ZnCharacterReadStream on: fileStream). readStream upToEnd. On the other side, this stream does not handle UI (like the old FileStream that e.g., asks the user if a file already exists), since I believe this is a concern that should be of the tools and not the file stream ^^. This is related to issue 15486. A slice is in the inbox for it if you would like to try. Guille

1 0

Beyond resolving a case
by Sergio Fedi May 4, 2015

May 4, 2015

How do we state and/or document that further work is needed after a Case? For example, some things I had in mind after finishing this case: *While adding the About comment for the Versionner tool I've found that the About should be modeled (at least to encapsulate the about text and the about window title, but maybe also for other information such as the purpose of the tool, the author, external link, further reading, etc).* *Should we generate a test for this About behaviour ?* *Should we move the About functionality up in the hierarchy and integrate it more with the classes used to make Tools?* *I added a method that answers the tool name (as opposed to the tool class name, usually used in the About title)* *and I found out there was another implementor on another Tool, but which I didn't understand the relationship with the Versionner.* *Is there something that could be done? Maybe make an abstract method?* *Is there a place where to put all these comments and questions/comments/ramblings?* *If there is, where?*

4 4

TxSpan>>#isAtEnd
by Sean P. DeNigris May 4, 2015

May 4, 2015

Right now, there is: TxBasicSpan>>#isAtEnd ^ false and TxEndSpan>>#isAtEnd ^ true However, there are cases were a list of spans has no end marker e.g. the argument to #insert: I'm thinking of changing to: TxBasicSpan>>#isAtEnd ^ self next == self. This will allow changing e.g. the following slightly opaque loop condition: [ span next == span ] whileFalse: [ span := span next ]. to: [ span isAtEnd ] whileFalse: [ span := span next ]. All tests pass with this change, but I'm still very much beginning learning Tx, so I wanted to run it by everyone to see if I'm missing something... ----- Cheers, Sean -- View this message in context: http://forum.world.st/TxSpan-isAtEnd-tp4824134.html Sent from the Pharo Smalltalk Developers mailing list archive at Nabble.com.

2 2