YesNoOk
avatar

Test release (Read 10709 times)

Started by Nightwish, January 07, 2005, 17:26:05

0 Members and 1 Guest are viewing this.
Share this topic:
#1
Test release |
January 07, 2005, 17:26:05
I have uploaded a new test release. There are no new features (except some internal things in the event api which might be useful for developers of other plugins). Please test it and report bugs only. No feature request are accepted for this version, because, based on this test release, I plan to upload a "official" beta to the plugin page on miranda-im.org.

New features and bigger changes will be added later, but first I want to have a version usable by all those people who only check the plugin site using wassup (most of them are still using 0.0.8 ).
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#2
Test release |
January 07, 2005, 21:01:47
I have a bug to report...
I am using the tabs at bottom and hiding them if there's only one tab... the problem I have it's that when the message log is full (fullfilling the window), then I open a new tab and then turn back to the previous tab, the log isn't scrolled down, even if you receive a message in the mean time...

that's it, hope it's fixable
#3
Test release |
January 08, 2005, 06:55:22
I tested the new version and now smileys are well showed in chat and historyviewer windows, that is not surronded by black space.

Great! Thank you and keep up the good work!
#4
Test release |
January 08, 2005, 20:10:50
I updated it with a new build.

This includes basic support for the metacontacts plugin. A new status bar indicator will show you the active protocol. If the current contact is a metacontact, then you can right click the protocol icon in the status bar and get a little context menu for switching protocols.

NOTE: The "force" option doesn't seem to work yet - I'll have to talk to the author of metacontacts again, but I uploaded this version so he can download it for testing purposes. Setting the default protocol from the message window does seem to work so far.

Nothing else has changed, except for some minor things and code cleanups.

There is also an updated version of NewEventNotify included in the archive - it should help against the double popups when using metacontacts.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#5
Test release |
January 09, 2005, 05:00:47
what about my bug report?
#6
Test release |
January 09, 2005, 12:20:36
"neooffs" said:
what about my bug report?
Not reproduceable here. It always scrolls down nicely...
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#7
Test release |
January 09, 2005, 13:56:01
I'll try to explain it again... hope I don't repeat myself...

ok
I have tabs on the bottom of the window
I have the option to hide the tabbar when there's only one tab
I talk to someone, then the message log gets full and appears that scroll thing on the right side...
then I open another tab, with another person
then I turn back to that tab I was using later....

the log won't get scrolled down, obviously because the tabbar occupied some space on the window when it appeard...
#8
Test release |
January 09, 2005, 14:20:14
"neooffs" said:
I'll try to explain it again... hope I don't repeat myself...

ok
I have tabs on the bottom of the window
I have the option to hide the tabbar when there's only one tab
I talk to someone, then the message log gets full and appears that scroll thing on the right side...
then I open another tab, with another person
then I turn back to that tab I was using later....

the log won't get scrolled down, obviously because the tabbar occupied some space on the window when it appeard...
Thats exactly what I tried. It scrolls nicely, i tried with both bottom and top tabs.

Does it autoscroll when you resize an open window with multiple tabs?
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#9
Test release |
January 10, 2005, 14:06:21
Hi people.
I don't undenstand. I download a lates version of tabsrmm, but I don't see version 0.9.9.0. This version I see in file changelog. But in miranda I see version 0.9.5.0. What is problem. Please explain me.
Thanks.
#10
Test release |
January 10, 2005, 14:21:11
"Punisher" said:
Hi people.
I don't undenstand. I download a lates version of tabsrmm, but I don't see version 0.9.9.0. This version I see in file changelog. But in miranda I see version 0.9.5.0. What is problem. Please explain me.
Thanks.
0.9.9.0 will be the "official" release which is planned to be released in a few days.

The versions you can download here are development snapshots.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#11
Test release |
January 10, 2005, 14:22:40
"Nightwish" said:
"neooffs" said:
I'll try to explain it again... hope I don't repeat myself...

ok
I have tabs on the bottom of the window
I have the option to hide the tabbar when there's only one tab
I talk to someone, then the message log gets full and appears that scroll thing on the right side...
then I open another tab, with another person
then I turn back to that tab I was using later....

the log won't get scrolled down, obviously because the tabbar occupied some space on the window when it appeard...

Thats exactly what I tried. It scrolls nicely, i tried with both bottom and top tabs.

Does it autoscroll when you resize an open window with multiple tabs?


I think you still dont understand me...

the point is the tabbar auto-hide feature... it appears and then it occupie a window space... so the log stays where it is... I need to scroll it down by the same size the tabbar is...
#12
Test release |
January 10, 2005, 15:09:50
"neooffs" said:
"Nightwish" said:
"neooffs" said:
I'll try to explain it again... hope I don't repeat myself...

ok
I have tabs on the bottom of the window
I have the option to hide the tabbar when there's only one tab
I talk to someone, then the message log gets full and appears that scroll thing on the right side...
then I open another tab, with another person
then I turn back to that tab I was using later....

the log won't get scrolled down, obviously because the tabbar occupied some space on the window when it appeard...
Thats exactly what I tried. It scrolls nicely, i tried with both bottom and top tabs.

Does it autoscroll when you resize an open window with multiple tabs?

I think you still dont understand me...

the point is the tabbar auto-hide feature... it appears and then it occupie a window space... so the log stays where it is... I need to scroll it down by the same size the tabbar is...
I know what you're saying.

I cannot reproduce it though, so I asked if the automatic scrolling *on resize* basically works...

If you have a container with multiple tabs opened, and you attempt to make it smaller, it still needs to scroll down the message log while you're changing the container size.

It is basically the same code which reacts to geometry changes of the container. This code doesn't even know why it needs to scroll the message log - it does it whenever the size of the message log control has changed. So if scrolling works when you manually resize containers but fails when the geometry changes automatically (because of the tabbar), then it would be rather strange.

Also...

* do you use smileyadd?
* how is the message log format configured (icons on/off etc..)
* container layout (menubar, statusbar, title bar, toolbar and so on..)

I need this info to reproduce it eventually, because all these settings have an impact on how the layouting code works.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#13
Test release |
January 10, 2005, 20:15:31
I made 5 screenshots for the no-scrolling-bug, that I receive:

1. One tab open with tab-bar hidden:


2. Open a second tab in the same container (tab-bar comes up):


3. Switching back to the first tab, the message-log doesn't scroll, it seems to be "locked" on the top of the messagelog:


4. Scrolling manually down:


5. Closing the second tab, the tab-bar hides, and the message-log has empty space at the bottom:


I hope this helps you a little bit.

Btw: Resizing a container works fine, the log is scrolled right
__
Rettet dem Genitiv sein Bestehen.
#14
Test release |
January 10, 2005, 21:54:26
you are right there is a problem with the msg log

1. open a tab window, it's the only tab in the container
2. write enough lines so a scroll bar is needed
3. open another (empty) tab in that container
4. go back to the first tab, and the scroll bar won't be scrolled all the way down (last line isn't visible)
#15
Test release |
January 10, 2005, 22:17:38
If you can see this bug, please try the following:

Set the "Tab Control Border" (Options, message log options) to at least 1 pixel. The bug seems to only appear when this is set to 0. Pretty weird actually, but if its really the case, then I'll probably able to fix it.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#16
Test release |
January 10, 2005, 22:30:55
"Nightwish" said:
If you can see this bug, please try the following:

Set the "Tab Control Border" (Options, message log options) to at least 1 pixel. The bug seems to only appear when this is set to 0. Pretty weird actually, but if its really the case, then I'll probably able to fix it.


Yes! changing the Tab Control Border (in Message Tabs options BTW) to anything <> 0 fixes this annoying thing entirely :) and i can hardly even see the difference, let alone decide what is nicer ;)
#17
Test release |
January 10, 2005, 22:36:59
miranda (0.4) is freezing after typing: "?" (same for "!" and other symbols)

*ticked "send on 'enter'" option, and still freezing
#18
Test release |
January 10, 2005, 22:43:24
"xmith" said:
miranda (0.4) is freezing after typing: "?" (same for "!" and other symbols)

*ticked "send on 'enter'" option, and still freezing
Do you have the autoreplacer plugin installed?

If yes, then uninstall it. Its incompatible with the rich edit input area.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#19
Test release |
January 10, 2005, 23:57:36
Binary updated...

The scrolling bug *may* be fixed (i just tried a method to get around that problem when the tab control border is set to 0).

Also, there is a slightly modified NewEventNotify included which can use the OSD plugin instead of popups for announcing messages. It can be enabled in the NewEventNotify options (make sure that you DISABLE "Announce messages" in the OSD setup though, otherwise you'll get everything twice).

The metacontact support has been updated. Forcing a protocol, including automatic revert to auto selection mode if the forced protocol goes offline, works now, but requires the latest metacontacts build.

Other changes are in the changelog.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#20
Test release |
January 11, 2005, 03:03:36
If you have downloaded the latest release, please check the .dll timestamps...

They should be: 2005/01/11  :  00:56

For some reason, the old file was still in place after I uploaded the new release for some hours...

If you have .dlls dated 01/08, please re-download.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#21
Test release |
January 11, 2005, 17:46:00
thanks, it's now fixed
#22
Test release |
January 11, 2005, 20:09:24
The bug is still there, when you do the following (even with Tab-control-border <> 0):
1. Open a tab (and have enough text for scrolling)
2. Open a second tab in the same container
3. Switch to the first tab (looks fine)
4. close the first tab
5. In the second tab (now the only one in the container) has empty space at the bottom like in the fifth screenshot in my other post

So, if the "last" tab is not the first (the one at the left) in the tab-bar, there is the bug I descriped.
__
Rettet dem Genitiv sein Bestehen.
#23
Test release |
January 11, 2005, 20:21:05
"Oemmes" said:
The bug is still there, when you do the following (even with Tab-control-border <> 0):
1. Open a tab (and have enough text for scrolling)
2. Open a second tab in the same container
3. Switch to the first tab (looks fine)
4. close the first tab
5. In the second tab (now the only one in the container) has empty space at the bottom like in the fifth screenshot in my other post

So, if the "last" tab is not the first (the one at the left) in the tab-bar, there is the bug I descriped.
This will stay as it is, and as long as the scrolling does not render a message unreadable, it doesn't matter...
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork
#24
Test release |
January 12, 2005, 06:25:57
the unicode bug return

chinese cant display in tab windows...
#25
Test release |
January 12, 2005, 13:56:54
Updated with a few minor fixes.

Also, the archive now includes the first version of the translation template. Most of the strings should now be considered "final", so translators should no longer have to worry about changes which may break the language packs in the future.

The template maybe not 100% complete yet, but contains all the strings from dialog boxes, option pages and menus.
__
Every program has at least one bug and can be shortened by at least one instruction -- from which, by induction, one can deduce that every program can be reduced to a single instruction that doesn't work.
My SMF-based forum fork