Sam Varshavchik | 1 Feb 02:49

Re: NOTE to C++ programming

Ed Sweetman writes:

> Ed Sweetman wrote:
>> namespace declarations do not go in header files. Period. I thought they 
>> taught that in intro C++ courses. It causes namespace pollution. I'm 
>> reverting all namespace additions to headerfiles, compiling ...seeing if 
>> that happens to work now since it wont compile as it is, then adding the 
>> namespace std lines in the appropriate places in the source files and 
>> testing that.
>> 
> 
> 
> 
> Give me a few more hours..   I should have the entire unix/shared source 
> tree converted to the way stl is supposed to be used. I'll be touching 
> just about every file so i'll probably have to commit the changes in 
> chunks.

This is very good news.  Although I haven't had a chance to look at Zinf 
sources lately, I recall that I was able to get the last public release of 
Zinf compiled with gcc 3.2, although it spit out a bucketful of warnings, 
due to namespace issues.

Converting pre-ISO/C++ code to the new STL and ISO/C++ namespacing is 
certainly a painful experience.  I've gone through it myself with several 
large chunks of code.  My general approach usually goes something like this:

* Fix all the includes: iostream.h >iostream, etc...  Nothing will compile 
now.

(Continue reading)

Ed Sweetman | 1 Feb 03:22
Favicon

Re: Re: NOTE to C++ programming

Ed Sweetman wrote:
> Ed Sweetman wrote:
> 
>> namespace declarations do not go in header files. Period. I thought 
>> they taught that in intro C++ courses. It causes namespace pollution. 
>> I'm reverting all namespace additions to headerfiles, compiling 
>> ...seeing if that happens to work now since it wont compile as it is, 
>> then adding the namespace std lines in the appropriate places in the 
>> source files and testing that.
>>
> 
> 
> 
> Give me a few more hours..   I should have the entire unix/shared source 
> tree converted to the way stl is supposed to be used. I'll be touching 
> just about every file so i'll probably have to commit the changes in 
> chunks.
> 

i've finished converting to using STL the correct way.

But as before altering the source i get the same gtk errors. Does anyone 
else compile zinf with gcc 3.2 with gtk libs compiled by 2.95 and is 
there anything that the ./configure script doesn't do that i need to do 
by hand with the current cvs pull?

-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
(Continue reading)

Ed Sweetman | 1 Feb 04:28
Favicon

Re: Re: Re: NOTE to C++ programming

Ed Sweetman wrote:
> Ed Sweetman wrote:
> 
>> Ed Sweetman wrote:
>>
>>> namespace declarations do not go in header files. Period. I thought 
>>> they taught that in intro C++ courses. It causes namespace pollution. 
>>> I'm reverting all namespace additions to headerfiles, compiling 
>>> ...seeing if that happens to work now since it wont compile as it is, 
>>> then adding the namespace std lines in the appropriate places in the 
>>> source files and testing that.
>>>
>>
>>
>>
>> Give me a few more hours..   I should have the entire unix/shared 
>> source tree converted to the way stl is supposed to be used. I'll be 
>> touching just about every file so i'll probably have to commit the 
>> changes in chunks.
>>
> 
> i've finished converting to using STL the correct way.
> 
> But as before altering the source i get the same gtk errors. Does anyone 
> else compile zinf with gcc 3.2 with gtk libs compiled by 2.95 and is 
> there anything that the ./configure script doesn't do that i need to do 
> by hand with the current cvs pull?

It appears that old preferences when used with new zinf can lead to 
strange errors.  Best to start clean with a new preferences until the 
(Continue reading)

Ed Sweetman | 1 Feb 06:21
Favicon

part1 of code cleanup phase

Part1 is underway for code cleanup in 2.x

some of the changes that didn't have to do with putting namespaces where 
they belong are:

add id3lib to the plugin makefile so it is cleaned up with make clean.

cleanup alsa's device selector so that it works. (needs some ui cleanup 
to reflect changes but should still work intuitively)

try and fix strange theme path issues.

Numerous and various missing header includes in files as well as code 
style cleanups and the beginnings of forcing lines to 80 char lengths.

-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
Ed Sweetman | 1 Feb 17:58
Favicon

Re: part1 of code cleanup phase

i've also noticed some strange namings of variables.  Some variables are 
named the same as variable types which is not only confusing but 
annoying.  I'm talking about variables named string that are char arrays 
and such.  This should at least be changed to capital String or string1 
  something to change it from string.   there are other poor variable 
names in there as well.

Ed Sweetman wrote:
> Part1 is underway for code cleanup in 2.x
> 
> some of the changes that didn't have to do with putting namespaces where 
> they belong are:
> 
> add id3lib to the plugin makefile so it is cleaned up with make clean.
> 
> cleanup alsa's device selector so that it works. (needs some ui cleanup 
> to reflect changes but should still work intuitively)
> 
> try and fix strange theme path issues.
> 
> Numerous and various missing header includes in files as well as code 
> style cleanups and the beginnings of forcing lines to 80 char lengths.
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
(Continue reading)

Ed Sweetman | 2 Feb 08:45
Favicon

developer talks

those who know how to use irc should get on irc.freenode.net and goto 
#zinf and try staying there as often as possible to talk about where 
attention should be focused and what needs to start getting done and all 
that good stuff.

-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
Peter Korf | 2 Feb 11:53
Picon

zinf fails starting if SaveCurrentPlaylistOnExit=true

Hi, Zinf Team

I have download the latest cvs version from zinf (but the error occours in 
earlier version (2.2.2), too.
When SaveCurrentPlaylistOnExit=true then the following errors are printed.
and zinf will not start.

I'm using SuSE 8.1
gcc3.2
gtk 1.2.10
gtk2 is also installed, but not found by configure.

Peter

GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
`hash_table != NULL' failed.

Gtk-WARNING **: gtk_type_create(): unknown parent type `21'.

GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
`hash_table != NULL' failed.

GLib-CRITICAL **: file ghash.c: line 152 (g_hash_table_insert): assertion 
`hash_table != NULL' failed.

GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
`hash_table != NULL' failed.

GLib-CRITICAL **: file ghash.c: line 152 (g_hash_table_insert): assertion 
`hash_table != NULL' failed.
(Continue reading)

Ed Sweetman | 2 Feb 20:12
Favicon

Re: zinf fails starting if SaveCurrentPlaylistOnExit=true

delete your current preference file. restart zinf.  I mentioned this in 
previous emails.  if you were following you could have saved yourself 
some time :)

Peter Korf wrote:
> Hi, Zinf Team
> 
> I have download the latest cvs version from zinf (but the error occours in 
> earlier version (2.2.2), too.
> When SaveCurrentPlaylistOnExit=true then the following errors are printed.
> and zinf will not start.
> 
> I'm using SuSE 8.1
> gcc3.2
> gtk 1.2.10
> gtk2 is also installed, but not found by configure.
> 
> Peter
> 
> 
> GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
> `hash_table != NULL' failed.
> 
> Gtk-WARNING **: gtk_type_create(): unknown parent type `21'.
> 
> GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
> `hash_table != NULL' failed.
> 
> GLib-CRITICAL **: file ghash.c: line 152 (g_hash_table_insert): assertion 
> `hash_table != NULL' failed.
(Continue reading)

Ed Sweetman | 2 Feb 20:13
Favicon

Re: zinf fails starting if SaveCurrentPlaylistOnExit=true


On second thought, having it enabled is the reason why this happens. 
deleting the preference file only disables the option.  I will fix this.

Peter Korf wrote:
> Hi, Zinf Team
> 
> I have download the latest cvs version from zinf (but the error occours in 
> earlier version (2.2.2), too.
> When SaveCurrentPlaylistOnExit=true then the following errors are printed.
> and zinf will not start.
> 
> I'm using SuSE 8.1
> gcc3.2
> gtk 1.2.10
> gtk2 is also installed, but not found by configure.
> 
> Peter
> 
> 
> GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
> `hash_table != NULL' failed.
> 
> Gtk-WARNING **: gtk_type_create(): unknown parent type `21'.
> 
> GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): assertion 
> `hash_table != NULL' failed.
> 
> GLib-CRITICAL **: file ghash.c: line 152 (g_hash_table_insert): assertion 
> `hash_table != NULL' failed.
(Continue reading)

Ed Sweetman | 2 Feb 20:56
Favicon

Re: Re: zinf fails starting if SaveCurrentPlaylistOnExit=true

problem solved.

Ed Sweetman wrote:
> 
> On second thought, having it enabled is the reason why this happens. 
> deleting the preference file only disables the option.  I will fix this.
> 
> 
> Peter Korf wrote:
> 
>> Hi, Zinf Team
>>
>> I have download the latest cvs version from zinf (but the error 
>> occours in earlier version (2.2.2), too.
>> When SaveCurrentPlaylistOnExit=true then the following errors are 
>> printed.
>> and zinf will not start.
>>
>> I'm using SuSE 8.1
>> gcc3.2
>> gtk 1.2.10
>> gtk2 is also installed, but not found by configure.
>>
>> Peter
>>
>>
>> GLib-CRITICAL **: file ghash.c: line 138 (g_hash_table_lookup): 
>> assertion `hash_table != NULL' failed.
>>
>> Gtk-WARNING **: gtk_type_create(): unknown parent type `21'.
(Continue reading)


Gmane