Who use the XML provided by irrLicth ??
Who use the XML provided by irrLicth ??
I really don't understand ... I reported in this topic:
http://irrlicht.sourceforge.net/phpBB2/ ... hp?t=17313
that the CXMLWriter has many bugs ... but the irrlicht 1.2 is out and nothing is changed !!
Some of the bugs arises only on Linux machines... and maybe I'm the only person that use irrlicht's XML on Linux...
... but there are a serious bug that introduce spurious "\t" in the file that arises on all machines !!!
So, who use the XML provided by irrLicth ?!?! Nobody... because I don't understand why I'm the only person that notice that bug
http://irrlicht.sourceforge.net/phpBB2/ ... hp?t=17313
that the CXMLWriter has many bugs ... but the irrlicht 1.2 is out and nothing is changed !!
Some of the bugs arises only on Linux machines... and maybe I'm the only person that use irrlicht's XML on Linux...
... but there are a serious bug that introduce spurious "\t" in the file that arises on all machines !!!
So, who use the XML provided by irrLicth ?!?! Nobody... because I don't understand why I'm the only person that notice that bug
-
hybrid
- Admin
- Posts: 14143
- Joined: Wed Apr 19, 2006 9:20 pm
- Location: Oldenburg(Oldb), Germany
- Contact:
But you cannot force us to postpone any enhancements just because you found a conceptual bug
Any whitespaces introduced are also no real problem because XML should be able to discard them on the fly. and moreover is your bugfix just a hack replacing the other hack. As I said elsewhere I have also worked on this thing already, but the release had been scheduled before and last minute bugfixes are really bad. Thus, the new release was made with all these known bugs in mind - we will simply keep on working on them
any please keep things related to the same topic in one thread next time. I move this thread to general discussion because it does not really report a bug...
Any whitespaces introduced are also no real problem because XML should be able to discard them on the fly. and moreover is your bugfix just a hack replacing the other hack. As I said elsewhere I have also worked on this thing already, but the release had been scheduled before and last minute bugfixes are really bad. Thus, the new release was made with all these known bugs in mind - we will simply keep on working on them
any please keep things related to the same topic in one thread next time. I move this thread to general discussion because it does not really report a bug...
It's false !! Because withspaces indise tags are part of the contents of tag !!Any whitespaces introduced are also no real problem because XML should be able to discard them on the fly
I mean:
Code: Select all
<TAG>Content ciao </TAG>
So... if I write with the XMLWriter
Code: Select all
xml->writeContent( "Content ciao")
xml->closingTag( "TAG")
Code: Select all
<TAG>Content ciao </TAG>
Code: Select all
"Content ciao ";
And It's a bug remove the spaces and the end of that string, because if I want to write a content that end with a single space, I cant !!
In conclusion... It's a bug !!
Ciao,
Gianluca
P.S.: Please... move this topic to bug reports!!!
-
Eternl Knight
- Posts: 313
- Joined: Tue Nov 01, 2005 5:01 am
Well, while I can handle the spaces issue for my purposes - Gianluca is correct. According to the XML specs - this is a bug, not a "conceptual" thing. The example provided is proof why. A content string with whitespace at the end needs to be correctly represented.
Arbitrary whitespace can only exist between tags - not as content. Personally, I don't use irrXML for anything outside Irrlicht, so I haven't come across the problem. In fact, due to much more widely used & tested XML libraries, I cannot actually see why one would.
--EK
Arbitrary whitespace can only exist between tags - not as content. Personally, I don't use irrXML for anything outside Irrlicht, so I haven't come across the problem. In fact, due to much more widely used & tested XML libraries, I cannot actually see why one would.
--EK
-
hybrid
- Admin
- Posts: 14143
- Joined: Wed Apr 19, 2006 9:20 pm
- Location: Oldenburg(Oldb), Germany
- Contact:
I won't move this one back, because there is already a thread on this topic in the bug forum.
Please write your bug reports more precisely next time so one can understand what really the bug is. This is what you wrote in the other thread:
And the conceptual bug was not the problem with tabs, but the implicit assumption of wchar_t having 16bit sizes on all systems. But that's another problem - and it works on Linux since the output routines simply chop the correct two bytes unless you have a big endian system.
Please write your bug reports more precisely next time so one can understand what really the bug is. This is what you wrote in the other thread:
So how should one see where the tabs are actually added (in fact if they were added they were outside the tag).In details, the writeClosingTag always add "\t" characters when Tabs>0 condition is satisfied, but it's an error because the writeClosingTag should add "\t" chars only when a new line is just started.
And the conceptual bug was not the problem with tabs, but the implicit assumption of wchar_t having 16bit sizes on all systems. But that's another problem - and it works on Linux since the output routines simply chop the correct two bytes unless you have a big endian system.
First of all, english is not my first language... so, I always make mistake in writing ... but anyway... you can't take only a piece of my long topic as a proof that I not explained well where is the bug... because I reported three lines of my code where the problem appear ... and if the author of CXMLWriter see that three line of codes I think that he should be able to understande the problem in one seconds !!Please write your bug reports more precisely next time so one can understand what really the bug is.
So, why you've just ignore my bug reports without try the lines of code reported... or asking my some examples to understand more deeply where the problem is ?!?!!
Code: Select all
XMLWriter->writeElement( TAG_NAME );
XMLWriter->writeText( TEXT_CONTENT );
XMLWriter->writeClosingTag( TAG_NAME );
So, why did you not ask me the files to look at the problem ???
Ciao,
Gianluca
[/code]
Provide the diff, and stop bitching!
It's not like they are ignoring you. They are simply working on other stuff. And, besides, it's not like irrXML is the most fundamental part of Irrlicht. If CXMLWriter is buggy, then DON'T use it. There are many others MORE flexible librarys out there.
p.s.: as a general rule, NEVER use trailing spaces in a string... but maybe it's just me.
p.s.: as a general rule, NEVER use trailing spaces in a string... but maybe it's just me.
-
hybrid
- Admin
- Posts: 14143
- Joined: Wed Apr 19, 2006 9:20 pm
- Location: Oldenburg(Oldb), Germany
- Contact:
Neither did you provide it, you are only crying that your bug reports are not acknowledged and immediately fixed within minutes. You're completely free to do with the sources whatever you want. If you want to participate and contribute to the community you should do so, this is always appreciated. But you cannot expect that all developers will do just as you said and possibly immediately or even faster. If you pay money for it you might expect some more, but right now we're all doing this as a hobby in our spare time. Moreover, the release was due and we had to put some efforts into actually testing the release without blindly doing last minute patches (just as I said).
In order to give you some (hopefully helpful) hints on how the bug fixing usually works here's a list:
Provide a bug report in the bug forum. Give us all information you have, code, examples, console logs, everything which might help.
Add a bug tracker report on the SourceForge project pages of Irrlicht
Discuss the topic on the forum.
In case your bug is not fixed within 2 month or in the next release of Irrlicht (whichever comes last) you might provide even more information, thereby up'ing the thread again. In case all your reports are completely ignored you might create a patch page where you can host the code you believe would fix this bug
After waiting 6 month without the slightest improvements you might give it a last try and provide new information before choosing a different engine.
As you can see, it might take a while before we can do anything for you. But we all do our best, and if you provide enough information and a stable fix it usually takes not too long before we can do something. But never think that stupid complaining will help.
In order to give you some (hopefully helpful) hints on how the bug fixing usually works here's a list:
Provide a bug report in the bug forum. Give us all information you have, code, examples, console logs, everything which might help.
Add a bug tracker report on the SourceForge project pages of Irrlicht
Discuss the topic on the forum.
In case your bug is not fixed within 2 month or in the next release of Irrlicht (whichever comes last) you might provide even more information, thereby up'ing the thread again. In case all your reports are completely ignored you might create a patch page where you can host the code you believe would fix this bug
After waiting 6 month without the slightest improvements you might give it a last try and provide new information before choosing a different engine.
As you can see, it might take a while before we can do anything for you. But we all do our best, and if you provide enough information and a stable fix it usually takes not too long before we can do something. But never think that stupid complaining will help.
Of course, I know that it's impossible to solve all problems immediately as people reports it... it is not the time about I'm complain about... but the lacks of discussions and informations on how to contribute.
I'm not familiar with sourceforge tools provided and I don't know how, where and which format send my fixes or report bugs.
But anyway... my topic was moved from bug reports thread ... but it reports a bug also if I write bad and non-properly.
So, complain about this move for me is not a stupid complain... because I missing to help irrlicht project if a bug is not considered a bug.
And my first topic, I know that it was long, but it reports one subject... that there is the wchar_t size and wide string encoding issue... that ok, it's another problem... and you already working on it, so I wait for solution doesn't matter if it requires one year, but now I know that you have the problem in mind.
So, I'm happy... and I not complain about that 1.2 version is not fixed from this point of view... I still have my patch
Ciao,
Gianluca
I'm not familiar with sourceforge tools provided and I don't know how, where and which format send my fixes or report bugs.
But anyway... my topic was moved from bug reports thread ... but it reports a bug also if I write bad and non-properly.
So, complain about this move for me is not a stupid complain... because I missing to help irrlicht project if a bug is not considered a bug.
And my first topic, I know that it was long, but it reports one subject... that there is the wchar_t size and wide string encoding issue... that ok, it's another problem... and you already working on it, so I wait for solution doesn't matter if it requires one year, but now I know that you have the problem in mind.
So, I'm happy... and I not complain about that 1.2 version is not fixed from this point of view... I still have my patch
Ciao,
Gianluca
-
hybrid
- Admin
- Posts: 14143
- Joined: Wed Apr 19, 2006 9:20 pm
- Location: Oldenburg(Oldb), Germany
- Contact:
This thread is no bug report. It simply refrains the bug report made some days before. I would have removed it right away if you'd made fewer words, but I could not avoid to answer ir directly (also because I liked the original bug report). So we should possibly stop it right here and continue working on the bugs. and please continue reporting information regarding the wchar_t problems in the original thread 