MemeCode - Blog

Scribe v2.0 Final

Date: 14/4/2010

Well according to the Bug DB all the P1 and P2 bugs are fixed (although not all fixes released yet) and there are just 5 P3's standing between me and a v2 final release. If I go by my self imposed metric of all P1, 2 and 3 bugs fixed for the final release.

So far no one is complaining about that so yeah I forge on ahead. This could mean that if I knock those over I need to setup the upgrade process for people. I recently fixed the InScribe purchase page to correctly show the key and download on checkout. But I'll have to revisit all that for people upgrading to v2 from v1.

I'm hoping that Beta23 will be the last beta, and basically the next build after that will enable the new key system for v2 and that'll be the finished retail build. Some 'n' years late... I don't know what 5 years ago I started this re-factoring thing? A few failed attempts and then I finally settled on an API and a new backend DB format.

Update: 2 left... and I'm wondering how important this one really is. The mail3 format doesn't allow folder loads to be done incrementally, in that I run the query, and control doesn't return to my code until the query has finished executing... meanwhile the app hangs. So there is no opportunity for me to update a progress bar.

Finished? Maybe... the bugs are now all dealt with, but there are some non-bug related things to process before the last (??) beta becomes the "Final" build:

Easy migration path for mail2->mail3. Maybe a nice little "Do you want to upgrade your folders? Yes/No" on starting v2 for the first time.
Collecting final submissions from all the translators.
Preparing the codebase and website for v2 upgrade sales.
Maybe some new icons / colours in the UI and documentation. Required: 1 reasonably talented graphic artist.
Revise all the help files for accuracy and readability. Required: 1 reasonably competent technical writer.

So you see the next stage could still take a while, but it's the home stretch... and in the mean time the users have a stable beta... no "release candidate" to use :)

Comments:

jbw
05/05/2010 10:00am

I'm definitely not an artist, but I'm often processing and displaying graphics, then my question is: why not to replace fusty gif format with modern png? If you think seriously about skins, let people to have full color and true transparency. And even when retaining indexed mode of gif, pngs are always smaller.

jbw
05/05/2010 11:21am

BTW, progress bar can be updated with callback function, see http://www.sqlite.org/c3ref/progress_handler.html

fret
05/05/2010 11:45pm

My main objection with PNG is having to bundle the libpng and zlib DLL's to be able to load them. I'm about to look at whether I can get some installed component of the system to load them instead so that I don't have to bundle support DLL's (and increase download size). There is a COM object that can load PNG's but it doesn't support transparency at all. So I'm told I should use GDI+, which might solve the issue easily. However I'm undecided on whether I make GDI+ a hard dependency on windows or not. It's probably safe, but I can't be sure. I've already dropped official support for Win9x, maybe it's safe to assume GDI+ is present too.

As for Sqlite progress callbacks, yes that would be nice to have... but I'm also interesting in pursuing not requiring them due to cursors/load on demand etc first.

From:

Email (optional):

(Will be HTML encoded to evade harvesting)

Message:

Remember username and/or email in a cookie. Notify me of new posts in this thread via email.	BBcode: [q]text[/q] [url=link]description[/url] [img]url_to_image[/img] [pre]some_code[/pre] [b]bold_text[/b]

Sitemap