PureBasic 5.11 is out !

[Azul]

thanks

[SFSxOI]

Thanks for the release.

What got fixed or changed? The history in the help has this > "Fixed: Some compiler and libraries related bugs "

But what related bugs? Need to know, so I can evaluate if code in the field needs to be updated, rescinded, or if vulnerabilities exist that need to be immediately addressed. So what are the fixed bugs?

[luis]

But what related bugs? Need to know, so I can evaluate if code in the field needs to be updated, rescinded, or if vulnerabilities exist that need to be immediately addressed. So what are the fixed bugs?

PB docs never detailed the fixed bugs in its history, and it's become even more generic about that from 2007.
Painstakingly follow the forum it's the only way I know.

Anyway how did you do in the past ? Why it's different this time ?

[Pforzheimer]

Thanks team

[said]

Thanks PB team
Just had the chance to test a big project and it works like a charm

[oreopa]

Thx per usual.

~10,000 lines compiled without problems... (yes, 10,000 lines is not a huge project, but believe me, it gives me the same headaches of one that is several million lines long... hehe )

[SFSxOI]

Anyway how did you do in the past ? Why it's different this time ?

Its different this time.

We have reached the point where we have a large amount of PureBasic code deployed in everything from small utility apps to actual prototype/production code for anti-malware and network intrusion detection and fixes being used by major corporations and governments and even (in some cases) deployed forward by those corporations in various forms based upon our code in publically available fixes or detection for malware and network intrusion, then this PureBasic update comes along with simply a cryptic message outlining that bugs were fixed. That cryptic message blows any previous certification we used out of the water so now we are left with facing a recall of all the code we have deployed and retesting it to ensure the code is sound for continued certification which will cost us thousands of dollars, whereas if we knew what bugs were addressed and fixed in this version of PureBasic we would know where to focus to continue the previous certifications and simply issue updates if needed.

In this new release the help history is cryptic and only says:

19th March 2013 : Version 5.11

- Fixed: Some compiler and libraries related bugs

If bugs existed in previous versions of the product then what were those bugs and how do these bugs affect what we have already deployed? So the question remains, for very valid reasons, what were the fixes implemented and for what bugs?

That's what has changed.

Its a completely different world when the code comes from an INFOSEC and ISO certified and CERT supported/endorsed facility that provides services to governments and corporations, and in some cases ultimately to the public at large from those corporations, as every little word counts in documentation and everything that goes to the field needs certification. Its not like we can just slap code together to do a certain thing and say it works and send it out to the field to these entities.

[freak]

You knew all that beforehand though. You have been around for a while so this is not your first PB update, is it?

[SFSxOI]

@freak

That's suppose to mean, what?

Every time in the past when there was an update, fix, addition, change, it was noted what that was for or it was simply generalized as 'Fixed: Numerous bugs' or 'Fixed: many bugs'. However, we come to this cryptic update in the help 'history' of "Fixed: Some compiler and libraries related bugs" with no indications what so ever as to what they are for when it specifically now says '...and libraries related bugs' and that's different then just indicating general bug fixes as indicated in the past history. If it had said simply again 'Fixed: Numerous bugs' or 'Fixed: many bugs' then I would not be asking the question, but when a commercial programming language ('commercial' because its paid for) specifically points out a specific area for bug fixes (libraries in this case) then it gets specific and when specifics are identified in the official documentation (the help) for a commercial programming language/product it needs to be outlined what those specifics are because they are defects and there are many legal, certification, and other questions it raises and a product vendor is legally obligated to inform customers of defects (bugs in this cases) and specifically what those defects are (yes, even in Belgium, Germany, or the EU in general). Why do you think companies like Microsoft outline what the bug or problem was and what an update or fix is for? Its not because they need to do so, even if it is good customer relations, its because of the legal liability if they don't. It would be different if the language was free, but it isn't. So it does not matter what I knew or did not know in the past, it only matters what it says now and its not too much to ask what the bugs were because this is paid for and expected to be used to produce code for many purposes and its an expected and legal right to know what defects exist or were fixed (bugs, and obviously you know there were bugs fixed) that may or may not affect already deployed product/code liability or product usability.

So, i'm not going to argue this and you or anyone else is free to have all the different opinions they wish. I don't need to, others may think they are getting a good deal or something and leave it at that and don't mind getting treated like cannon fodder testers like you want to speak to them, they are customers who bought and paid for this (now defective) product and you basically work for them in certain terms. This is a serious matter and addressing customers with insulting posts like they work for you or are expected to grovel or not expect an answer is not the way to go here.

[dman]

Does it really matter if the PB team don't give details of what has been fixed? If it doesn't
affect your projects then it's all good.

Your getting a product that lets you pay ONCE and receive updates for no charge, pretty good deal me thinks, if they
choose not to give you exact details of what bugs have been fixed, give them a break dude!

Just code!

[SFSxOI]

Does it really matter if the PB team don't give details of what has been fixed? If it doesn't
affect your projects then it's all good.

Your getting a product that lets you pay ONCE and receive updates for no charge, pretty good deal me thinks, if they
choose not to give you exact details of what bugs have been fixed, give them a break dude!

Just code!

No saying its not a good deal. Yes, it does matter in this case because they got specific and specifically pointed to libraries for bug fixes without defining what libraries were affected or the bugs fixed. I don't know if it affects our projects or not, until we do one of two things, 1) recall all the code we have deployed and send it back through testing with different protocols to help find any bugs that may have been previously missed and then recertify or 2) we know what the bugs are so we can determine if they will affect the code we have deployed. Either way, we need to do one or the other to maintain the certification for our code, all because of this one 'specifically' worded (and probably misused) phrase in the help that means so much more than what the words themselves say. Both ways cost us something, the first way costs us thousands of $$$, the second costs us time mostly and maybe some $$$. I'm not trying to task them in any way, they should already know what bugs were fixed so its a simple matter of just saying what bugs were fixed or ,heck if that's too difficult, even simply what libraries were affected. Its the first time I've ever asked what the bugs were and its important to know in this case.

[skywalk]

...~...It would be different if the language was free, but it isn't.

No, PB is not free, but you cannot seriously compare its support to M$ and MSDN support fees and licenses? This has been discussed quite often and freak implied a while back that a support arrangement could be negotiated for a user's extra effort.
I am not enamored with PB's Bug tracking either and cannot imagine how the current method is easier than automation? But, for any user to make demands of such a small team without compensation is bordering on whining.
I applaud your accomplishments so far with PB and hope you can continue to use it in your business.

[SFSxOI]

Its not demands, small team or not, its a simple copy and paste (or should be if they maintained a document track of bug fixes). They already got compensated, the product was paid for. Its not any whining, its a simple question that is able to be answered easily. I'm not comparing anything with their support.

[rsts]

I don't see how telling you what bugs were "fixed" helps you at all, since there is no mention of what bugs were added. You have to test every release either way.

cheers

[SFSxOI]

Knowing what bugs were fixed enables us to focus on only those things which used the libraries associated with the fixed bugs in already deployed code that used previous versions of PureBasic. For those we can issue fixes/updates/replacements and still maintain certification if we know what bugs were fixed so we can develop the fixes/updates/replacements. Not knowing what bugs were fixed (or at least the libraries affected) means we have to recall every single piece of code in the field and decertify everything deployed we have done with PureBasic until its tested again which means thousands of dollars in cost not to mention the effect it would have on the end users using the code, there are also several other effects which I will not go into here. So yes, knowing what bugs were fixed makes a big difference not only in cost and to us, but to others as well.