October 2009 - Wine-devel - list.winehq.org

Business professors preaching Wine
by Zachary Goldberg 13 Oct '09

13 Oct '09

Thought you guys might enjoy this short story from my day: So I was in a class on product design (at the school of business @ the U of Penn) and my prof is going on about some software we'll need and how to use it. Out of the blue he says "And unfortunately it only works on Windows. But there is this great open source 'stuff' out there called Wine that acts like a middle layer between the now ubiquitous Intel hardware and the windows application to let it run on Linux". He went on like this about Wine, only partially accurate in technical fact, for about 5 minutes. Needless to say the entire classroom (of about ~50) now knows about Wine and that it can run Windows applications on Linux & Mac. I swear, I did not talk to him beforehand and encourage him to do this. Huzzah! -Z

1 0

Re: [1/3] mshtml: Add test suite for IHTMLLocation
by Nikolay Sivov 13 Oct '09

13 Oct '09

aeikum(a)codeweavers.com wrote: > --- > dlls/mshtml/tests/Makefile.in | 1 + > dlls/mshtml/tests/htmllocation.c | 383 > ++++++++++++++++++++++++++++++++++++++ > 2 files changed, 384 insertions(+), 0 deletions(-) > create mode 100644 dlls/mshtml/tests/htmllocation.c > > diff --git a/dlls/mshtml/tests/Makefile.in b/dlls/mshtml/tests/Makefile.in > index 06c6824..dc86c18 100644 > --- a/dlls/mshtml/tests/Makefile.in > +++ b/dlls/mshtml/tests/Makefile.in > @@ -9,6 +9,7 @@ CTESTS = \ > dom.c \ > events.c \ > htmldoc.c \ > + htmllocation.c \ > misc.c \ > protocol.c \ > script.c > diff --git a/dlls/mshtml/tests/htmllocation.c > b/dlls/mshtml/tests/htmllocation.c > new file mode 100644 > index 0000000..c3a5e82 > --- /dev/null > +++ b/dlls/mshtml/tests/htmllocation.c > @@ -0,0 +1,383 @@ > +/* > + * Copyright 2009 Andrew Eikum for CodeWeavers > + * > + * This library is free software; you can redistribute it and/or > + * modify it under the terms of the GNU Lesser General Public > + * License as published by the Free Software Foundation; either > + * version 2.1 of the License, or (at your option) any later version. + * > + * This library is distributed in the hope that it will be useful, + * > but WITHOUT ANY WARRANTY; without even the implied warranty of + * > MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * > Lesser General Public License for more details. > + * > + * You should have received a copy of the GNU Lesser General Public + * > License along with this library; if not, write to the Free Software + * > Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA > + */ > + > Hi, Andrew. This series is broken, patch body is formatted.

2 1

Re: Help me get a patch into Wine
by Charles Davis 13 Oct '09

13 Oct '09

Keith Muir wrote: > Charles Davis wrote: >> Damn it, I keep forgetting to reply all. >> >> Charles Davis wrote: >> >>> Vitaliy Margolen wrote: >>> >>>> Charles Davis wrote: >>>> >>>>> Hi, >>>>> >>>>> My name is Charles Davis, but you all can call me "Chip". >>>>> >>>>> I'm new to Wine development, and frankly, I'm kinda scared because I've >>>>> heard and read that getting patches into Wine is very difficult. In >>>>> fact, every time I got ready to send this first patch, I chickened out >>>>> and couldn't do it. So, I want someone to review this patch to see if >>>>> there's anything wrong with it. >>>>> >>>>> >>>> Thanks for writing the patch. >>>> >>>> Few things about your patch: >>>> >>>>> + * Copyright 2008, 2009 Charles Davis >>>>> >>>> You need to add substantial amount of code to or be working for a long time >>>> on this file to add yourself to the copyright note. As for your contribution >>>> to the project, all authors are tracked and AUTHORS file updated automatically. >>>> >>> Oh. I see. But I have many more patches for this file. >>> >>>>> + switch(fmt->Format) >>>>> + { >>>>> + case IOCTL_CDROM_CURRENT_POSITION: >>>>> + memset(&data->CurrentPosition, 0, sizeof(data->CurrentPosition)); >>>>> + break; >>>>> + case IOCTL_CDROM_MEDIA_CATALOG: >>>>> + memset(&data->MediaCatalog, 0, sizeof(data->MediaCatalog)); >>>>> + break; >>>>> + case IOCTL_CDROM_TRACK_ISRC: >>>>> + memset(&data->TrackIsrc, 0, sizeof(data->TrackIsrc)); >>>>> + } >>>>> + /* We need IOCDAudioControl for IOCTL_CDROM_CURRENT_POSITION */ >>>>> + if (fmt->Format == IOCTL_CDROM_CURRENT_POSITION) >>>>> + { >>>>> + ERR("This version of Mac OS X does not support IOCDAudioControl\n"); >>>>> + return STATUS_NOT_SUPPORTED; >>>>> + } >>>>> >>>> This doesn't seems right. If it's not supported (aka error) you shouldn't be >>>> touching returned data. Also why are you zeroing return buffers? I don't see >>>> other implementations do it. >>>> >>> Good catch. I should have done the check first. >>> >>>>> + case IOCTL_CDROM_MEDIA_CATALOG: >>>>> + memset(&ioc, 0, sizeof(ioc)); >>>>> >>>> and >>>> >>>>> + case IOCTL_CDROM_TRACK_ISRC: >>>>> + memset(&ioc, 0, sizeof(ioc)); >>>>> >>>> You shouldn't need to zero buffer you pass to system ioctl. >>>> >>> Good point. >>> >>>>> + if((io = ioctl(fd, DKIOCCDREADMCN, &ioc.mcn)) == -1) break; >>>>> + if((io = ioctl(fd, DKIOCCDREADISRC, &ioc.isrc)) == -1) break; >>>>> >>>> Please keep formatting consistent with the rest of the function/file. Here >>>> you need to add space after "if". >>>> >>> You're right. I read about that, but that must have slipped by me. >>> >>>> Vitaliy. >>>> >>> Thanks for your input. >>> >>> Chip >>> >> >> >> >> >> >> > > The Julliard rank is something worth explaining. Wine developers have > long wondered exactly *HOW* Alexandre determines whether or not a > specific patch gets committed. The process is something as follows: > > * Does the patch use C++ style comments? Yes: REJECT > * Does the patch change coding style from the surrounding code? Yes: > REJECT > * Is the code obviously wrong in any way? Yes: REJECT > * Does it take more than a few seconds to see that the code is > correct? If today is a busy day: IGNORE > * Is the author on the s**t list? Yes: REJECT > * Are the planets aligned in just the right way and none of the > above reject? Yes : ACCEPT > > And of course now we need some sort of measurement for inclusion in the > crap list. For every 'good' patch set an author moves further off the > list, for every bad patch the author's reputation is diminished and put > further onto the list. Most Wine developers have come to realize that > that the Julliard Rank is very similar to Google's PageRank algorithm: > everybody who cares understands how it is 'supposed' to work but could > never replicate it in a million years. > OK, I've sent a better version of my patch to wine-patches. Now we'll see what AJ thinks of it. Again, thanks for your input, guys. And the fact that you said something really helped me. Chip

2 1

Re: rsaenh: Fix the accepted RC2 key length, based on a test case by Dan Kegel.
by Hans Leidekker 13 Oct '09

13 Oct '09

On Tuesday 13 October 2009 08:46:59 am Dmitry Timoshkov wrote: > + ret = pCryptImportKey(provider, (BYTE*)&key_blob, > + sizeof(BLOBHEADER)+sizeof(DWORD)+key_length, > + 0, CRYPT_IPSEC_HMAC_KEY, &hkey); > + ok(ret, "CryptImportKey error %u\n", GetLastError()); > + > + pCryptReleaseContext(provider, 0); You should clean up the key as well. -Hans

1 0

native retest needed: GetDeviceIdentifier()
by Markus Stockhausen 13 Oct '09

13 Oct '09

According to Stefans suggestion I have implemented a test for the string copies into the identifier structure. Test is designed to fail to give me a hint what is going on. Hopefully someone can post the results of a test run in native environment. Thank you

4 3

IsWindowUnicode() is always true after RegisterClassW
by Nikolay Sivov 13 Oct '09

13 Oct '09

Is it ok that IsWindowUnicode() returns 1 for windows created with -A calls? I've just another unrelated issue for Comboex and spotted then I'm apparently using unicode window (as reported by this call). But it was created with --- CreateWindowExA and class WC_COMBOBOXEXA --- Currently comboex code depends on being unicode window or not, so it could cause problems. Looking at user32 code: --- BOOL WINPROC_IsUnicode( WNDPROC proc, BOOL def_val ) { WINDOWPROC *ptr = handle_to_proc( proc ); if (!ptr) return def_val; if (ptr->procA && ptr->procW) return def_val; /* can be both */ return (ptr->procW != NULL); } --- If I'm creating with -A call def_val = = FALSE and the only reason why I got 1 from IsWindowUnicode is that ptr->procA is null. Further WINPROC_AllocProc() with both not null parameters is used for "builtin" user32 classes and default procedure. Since we're registering comctl32 classes as unicode with RegisterClassW, IsWindowUnicode() becomes noop for that, cause it's always unicode. This matches windows behavior actually (at least for ComboEx on XP SP2) and that's how it works. The question is how could I handle CB_GETLBTEXT properly for ComboEx? For Combobox it's obvious cause we know is it unicode or not, for Comboex caller should rely on IsWindowUnicode() for this control for buffer format or what? Any thoughts how to deal with this or am I missing something?

3 4

(no subject)
by KINEPOLIS MYSTERY SHOPPING 13 Oct '09

13 Oct '09

KINEPOLIS MYSTERY SHOPPING. RC: 2004/628/CM. Ref: MYT-200657778 www.kinepolis.com You have been chosen by 0ur HR Manager as one of our possible candidate suitable for employment and being one of our chosen candidates your first assignment as a company representative. Duties:- We want you to run a survey on two prominent companies in your area. 1st: The Money Gram Location: 2nd: The Western union Location *There has been reports about laps in the services of their Management and some of their staffs, *Their complains were based on reports which their customers forwarded anonymously and phone calls which were also made to the head Office. About Us: Kinepolis was founded in 1987 to help leading companies close the gap between the promise of service excellence and its actual delivery. Today, Kinepolis is one of the nation?s largest customer experience measurement and mystery shopping companies, and our Fortune 500 clients represent some of the most innovative and successful customer-focused organizations in the world. Simply put, our approach is designed to isolate the key behaviors that drive customer delight and show our clients how to leverage this information to increase customer loyalty and profitability. Since our inception, we?ve been helping good companies become great companies with our comprehensive arsenal of proven surveying, auditing, and management systems. Our programs allow our clients not only to gather incident-specific information about the moment of truth when a customer comes in contact with an organization, but also to track and trend attributes and behaviors that impact long-term guest value, satisfaction and advocacy. The Money Gram location was reported for evaluation for the following reasons: * Customers have reported their money missing * Slow services * Unbalanced transfer charges Second company was reported to be rendering *Poor services *Rudeness to customers *Excess charges *Late opening time and Closing before time. Your Secret Evaluation would be: * To make a transfer of fund from this Money Gram location or western union location to our Mystery shopper, and the funds would be picked up by another mystery shopper at the exact location Which a customer reported her funds missing. * You would have to record the time at which you go to the location and how many minutes it took you to get service. * The check for the sum of $5000.00 will be send to you, As soon as you receive the check , You will have to deposit the check at your bank and deduct your $500 and use the rest of the money for the services *** First Offer of Mystery Shopper *** Pay-check of $5000 on your name & you will be the one that will be cashing any pay-check issued on your name & you will deduct 10% of the first offer & the rest will be transfer to our Accounting Department by MONEY GRAM OR WESTERN UNION TRANSFER . *Upon receiving the funds, the locations address would be forwarded to you, and also the Name and address of whom the Mystery shoppers transfer would be made to. You would have to keep a comprehensive report on every activity you carry out. * You would also provide me with the name of the cashier that attended to you. Kindly send us your details below to world_service.web(a)w.cn via email in order to have payment issued to you as soon as possible. The payment will be in form of a CERTIFIED CHECK or MONEY ORDER and it will be delivered to you before the week runs out. Thank You. Mr. Frank Anderson. KINEPOLIS MYSTERY SHOPPING /Human resource Manager ================================ Your are expected to send the information below to us. FULL NAME -- FULL RESIDENTIAL ADDRESS -- CITY -- STATE -- ZIP CODE -- COUNTRY -- PHONE NUMBER -- GENDER -- MARITAL STATUS -- AGE -- NATIONALITY -- OCCUPATION -- Kindly send us your details above to world_service.web(a)w.cn Apply Now ...Part-Time Employment Offer...!

1 0

Re: Bug 6971 in RedOrchestra Game.
by James McKenzie 13 Oct '09

13 Oct '09

Loren Kraushar wrote: > Ok Just trying to put it out there I have seen endless sights for this > one problem and still no consistent fix. I was trying to see if you > geniuses could throw it in on the next fixed issues list. > > Thank you for the response master. Please keep the sarcasm to a mimimum. My responses are based upon my own experiences and those of others. First, if you have a patch with testcase, please put it on the wine-patches mailing list. Second, please monitor the wine-devel mailing list for any feedback as well as IRC channel #winehackers on irc.freenode.net. Third, I do not make the final decision as to code acceptance/rejection. This is reserved ENTIRELY to AJ. We have NO say so in this process. This is not liked but is, for the time being, accepted. It is good for one person to be able to make these decisions. He has HIGH standards for acceptance and there are legal and moral reasons for this. The project is always looking for new code and developers. James McKenzie P.S. The mailing lists are bottom post only, and that is for a good reason, it is the way we do things here and is per RFC 1855.

1 0

Ballooning "dlls" folder
by Erich Hoover 12 Oct '09

12 Oct '09

With the separation of the 16-bit APIs recently, the "dlls" folder has ballooned quite a bit. Is there any particular reason that the 16-bit and 32-bit DLLs are not in separate folders? Erich Hoover ehoover(a)mines.edu

2 1

Re: CPP Check 1.1.31
by Nikolay Sivov 12 Oct '09

12 Oct '09

James McKenzie wrote: > Nikolay Sivov wrote: > >> James McKenzie wrote: >> >>> Nikolay Sivov wrote: >>> >>> >>>> chris ahrendt wrote: >>>> >>>> >>>>> [/home/cahrendt/wine-git/dlls/kernel32/resource.c:1279]: (error) >>>>> Division by zero >>>>> >>>>> >>>> False positive I suppose: >>>> >>>> --- >>>> static const BYTE pad[] = { >>>> >>>> 'P','A','D','D','I','N','G','X','X','P','A','D','D','I','N','G' }; >>>> DWORD i; >>>> >>>> -> for ( i = 0; i < size / sizeof pad; i++ ) >>>> --- >>>> >>>> >>> Should pad be enclosed in parenthesis? >>> >>> James McKenzie >>> >>> >> No, why? >> >> >> > I was thinking that c specifies sizeof needs the parameter to be > enclosed in parathesis like this sizeof (pad) but I may be incorrect. > > James McKenzie > This is for type names, variables could be used without enclosing.

3 2