|
|
Home » Community » Coffee corner » U++ public image - please read and discuss...
|
Re: U++ public image - please read and discuss... [message #1158 is a reply to message #1149] |
Tue, 21 February 2006 17:17 |
hojtsy
Messages: 241 Registered: January 2006 Location: Budapest, Hungary
|
Experienced Member |
|
|
I see these issues as the main obstacles for U++:
1) documentation coverage is the worst I have ever seen for any library
2) there is no well known big application built on it which would prove the concept (theIde is not a well known app!)
3) navigation in the topic browser is so inconvenient that even a single text file with all the docs would be better. At least a searchable panel with the class names should be provided ASAP as the first step.
4) it is perceived as a one-man project, meaning that:
4a) if something would happen to the one or two developpers the development would be discontinued
4b) some uncommon personal preferences appear which might not seem justified for the majority of developers. For example several other GUI toolkits get along quite well withouth inventing an own rich text format. Or that a hundred other text editors save only when the user requests it, but theIde knows better and saves anyway. Or the ColumnList scrolling vertically, while multi-column lists scrolls horizontally everywhere outside U++. Or that the root namespace is contaminated with 2000 global functions! This list could go on and on. I don't want to be offensive: I also have strange uncommon preferences but these should be suppressed when a library is developed for the public.
5) I feel that organization of website is unprofessional and confusing, but I can not pinpoint what exactly makes me feel this way. Altough I recently accepted to administer the updating of the website from the uppweb source, I am a developer, and not a web designer. So most probably my attempt to reorganize the site would result in something which is not any better.
I think that the strenghts of U++ are:
- very fast runtime & compile time
- feature rich
- some library services are very polished and forces a "wow": for example serialization, Nformat, xml parsing, logging, callbacks
- RAD if you already know how to use U++ (but learning is not easy enough)
I tried looking around the web for U++ related discussions. There are very few. I added a few lines to the wxWidgets wiki, mentioning some factual errors in their comparison to U++. http://www.wxwidgets.org/wiki/index.php/WxWidgets_Compared_T o_Other_Toolkits#wxWidgets_compared_to_Ultimate.2B.2B
Quote: | wxWidgets compared to Ultimate++
* Ultimate++ only supports Windows and Linux, not MacOS
* The comparision on http://upp.sourceforge.net/www$vswx$en-us.html isn't really fair. The (very old) wxWidgets sample they took is well-commented and well-structured to show the reader how to design a wxWidgets application. Their implementation is uncommented and doesn't even behave the same. Also, a small example like that doesn't show how the toolkit scales to bigger applications.
(The reference to code comments is incorrect. The 600 line wxWidget sample on U++ page contains a sum total of 4 comment lines, namely: 1 occurence of "// Constructor and destructor", and 3 of "//precompiled headers". This does not qualify as well-commented. Also the hidden hint to the U++ example being not well-structured is unfair: it just employs a more simple and straightforward structure. - Sandor Hojtsy)
| Could you comment on the "behaves differently" part? It would be more fair to reproduce the functionality of the example exactly.
|
|
|
|
|
Re: U++ public image - please read and discuss... [message #1182 is a reply to message #1158] |
Wed, 22 February 2006 09:00 |
|
hojtsy wrote on Tue, 21 February 2006 11:17 | I see these issues as the main obstacles for U++:
1) documentation coverage is the worst I have ever seen for any library
|
It is not so bad.. believe me I can even say that Bulider C++ which I am force to use at work is much worse in this case. Besides upp has only 2 main developers, who most of their time spend on developing commercial apps..
I think we need to write more tutorials. This what is the most needed at the very begining. I wanted to write one about creating own widgets but there is always lack of time..
Quote: | 2) there is no well known big application built on it which would prove the concept (theIde is not a well known app!)
|
what about 'application wrtitten in upp'. There are quite a few
Quote: | 3) navigation in the topic browser is so inconvenient that even a single text file with all the docs would be better. At least a searchable panel with the class names should be provided ASAP as the first step.
|
agreed. And if the text could be selectable.. And some docs in topic browser should be always visible (not only when that and no other package is open).. like docs for ide
One more thought: We could deliver the chm doc-file with an installation.
Quote: | 4) it is perceived as a one-man project, meaning that:
4a) if something would happen to the one or two developpers the development would be discontinued
|
Iteresting I think about it from time to time But the community is growing, the first patches appeared from outside. I hope it will be continual tendency.
Quote: | 4b) some uncommon personal preferences appear which might not seem justified for the majority of developers. For example several other GUI toolkits get along quite well withouth inventing an own rich text format. Or that a hundred other text editors save only when the user requests it, but theIde knows better and saves anyway. Or the ColumnList scrolling vertically, while multi-column lists scrolls horizontally everywhere outside U++. Or that the root namespace is contaminated with 2000 global functions! This list could go on and on. I don't want to be offensive: I also have strange uncommon preferences but these should be suppressed when a library is developed for the public.
|
Having your own rich text format is an adavantage in my meaning.. You have full control on your documents, raports, etc.. Besides qtf can be easily converted to any other format.
Saving the files is in fact different to any other editor but I don't think it is an very important thing. I didn't have any problems to get used to it (and now I even miss it in others editors )
|
|
|
Re: U++ public image - please read and discuss... [message #1183 is a reply to message #1107] |
Wed, 22 February 2006 09:14 |
|
fudadmin wrote on Mon, 20 February 2006 13:08 | 7. there is no Grid widget
8 there are no regular expressions
|
7 There is - ArrayCtrl. And I want to add my GridCtrl soon
8 I will be working on it (because I need it) and I will public my wrapper for pcre, but I think that Mirek probably will be interested in his own solution and better integration with the whole upp lib.
|
|
|
Re: U++ public image - please read and discuss... [message #1184 is a reply to message #1182] |
Wed, 22 February 2006 09:21 |
|
mirek
Messages: 14105 Registered: November 2005
|
Ultimate Member |
|
|
Quote: |
Quote: | 3) navigation in the topic browser is so inconvenient that even a single text file with all the docs would be better. At least a searchable panel with the class names should be provided ASAP as the first step.
|
agreed. And if the text could be selectable.. And some docs in topic browser should be always visible (not only when that and no other package is open).. like docs for ide
|
We had a chat about this with Tom and it seems like things can start improving there as soon as the next week. First two issues to solve: browser will show everything from current nests instead of active packages only (but active will be bold) and there will be the search tool.
Mirek
[Updated on: Wed, 22 February 2006 09:23] Report message to a moderator
|
|
|
|
|
|
Goto Forum:
Current Time: Fri Nov 01 00:31:01 CET 2024
Total time taken to generate the page: 0.02649 seconds
|
|
|