The Mana World Forums

After searching through git reflog, I have managed to get a working testing client and server on commits 3d93f9133541b35bbe2eb1391a5f4d6a3285d22a and 7db9f6fe36b737d2eec7c6070497035b0834def2 respectively, built using cmake and enet 1.2.

Both are the most recent commits I have of pre-merge testing. I believe the server commit here was the final commit done in testing before the merge, but my client testing history appears incomplete so I cannot say for sure whether this was the final commit made in client testing or not.


Using the now unsupported autotools to build the latest mainline server (and in doing so, building with enet 1.2 such that this older client could connect), I can connect using this older client, move around, talk, use/activate items, display attributes correctly (aside from display hp/weight bars, but that was broken on account of someone not maintaining this functionality in one of the mainline to testing merges and is a known issue. ), etc etc everything fundamental works fine.

Using the now unsupported autotools to build the latest mainline client (and in doing so, building with enet 1.2 to connect to the previously mentioned latest mainline server that was built in the same way), the client can go through the account server correctly but halts when connecting to the game-server.

I cannot build the 3d93f9133541b35bbe2eb1391a5f4d6a3285d22a client with enet 1.3 to check that there are no problems there, but there certainly does seem to be a problem with the client introduced somewhere between 3d93f9133541b35bbe2eb1391a5f4d6a3285d22a and HEAD.


All server compiles were done using sqlite and all server/client runs were done using the same testing data (with stats.xml/attributes.xml root tags changed from stats to attributes or vice versa depending on the version compiled)


Since my local client testing branch history seems incomplete, and Bertram deleted the remote client testing clone, perhaps he could upload the history somewhere so I (and anyone else interested in testing) can have a look to see where the client stopped working?


---Freeyorp

[Breaking point found.]

"StatChange" should have been "StateChange", a problem that would have never gone undetected if the event system had used enums instead of sending strings everywhere.

[Fix pushed], latest mainline client should now work with latest mainline server (tested locally using sqlite, cmake, and enet 1.3)


---Freeyorp

A BIG BIG extra thanks to Freeyorp

The server was just restarted and I'm now able to log in without counter thoughts.

Login in, walking, chatting is now possible.

Baby Step I COMPLETE!

Best regards.

Waiting for the client data to still sync to my request, but still: good job, guys.

Let's see if we can get the ball somewhat rolling again.

BS = Baby Step
TBA = To Be Announced

BS1: Walking, chatting, no playerset, no
BS2: BS1 + incomplete tilesets + complete maps
BS3: BS2 + TBA

This thread also just went [REQ] -> [WIP].

Hi,

Some News:

As requested:
- The map has been changed to the requested empty grassy square.
- The grass tileset is now following the guidelines, but still missing an author panel.
- The playerset has been changed to an invisible pixel.
- Most errors in the mana client log have been disabled by adding xml placeholders.

In progress:
Bertram:
- Addition of log rotation + zipping of old log files + bugfixes along thway.
- GUI fixes in order to display the different character's attributes nicely.
- Addition of custom listen ports possibility.
- Addition of parameters handling for the monitor.
- Addition of sending the max available character per accounts number at login, and making the client display it nicely.
- Addition of a 'low-admin' tool to permit accounts creation and little admin tasks while I find them along the way.
- Addition of a manaweb interface for dev.themanaworld.org

Freeyorp:
- Progress on the auto-attack system. (In order to get PvP running initially.)
- Will then follow the fixation of the monster loading and handling.

Crush:
- GM commands
- Special system continuation once the auto-attack system features are becoming stable.
- AoE (Area of Effect) lua scripts

I'm also requesting to make the tilesets validated one by one, and tagged with their author panel, so that mappers can know that validated tiles will then be frozen and reusable.
Speaking of art:

Irukard:
- Block walls (under heavy progress)

MerlinX420:
- Tests on given tilesets, needed fixes and the first map?

Best regards.

Bertram wrote:Irukard:
- Block walls (under heavy progress) :)

Version 0.1.0 (map ready) should be released somewhere between 2010.10.10 and 2010.10.17.
After that I will continue work (with weekly update schedule) on Block Walls Add-ons. Estimated time needed for that task is about 4 weeks. Those two sets should provide enough tiles for constructing quite diversified Academy walls.
For ground tile inside Academy I would prefer IvanMorves city pavement. It's fits those walls pretty nice, and I do like the shape of those tiny stones.

So if time permits in second part of November I would take on schedule two tasks:
1. Block Wall<-> Cliff integration
2. Port tiles: Gema II, some smaller boats, port add-ons.

And one more thing about "The Cliffs". I would be more than happy if we could use something like that:

Source: http://mistraal.wordpress.com/2008/04/1 ... l-english/
I did test with resized version (where tile is 64x64px instead of 128x128px) and it looks very good. Of course it needs some work, but it's really good starting point. Original tile is made by Tanja Borzel (http://kd21.de/index.php?cfx=22). Maybe we should ask her for permission for redrawing it, or even if it's possible license change for GPL2.

I've been testing the tiles and I think that the other "sets" really need some work.
The port town set doesn't tile correctly and The Small House by ax. and foth. isn't useable in it's current state.
I've been messing with all the CR1 tiles and wips. I like the pavement that Irukard recommends for the block wall.
The Other pavement type in that set fits as well, but I would need to make a grass/road transition tile as well. Nothing to hard there. The new grass bugs me thou, maybe I'm not understanding how it's meant to be used but it doesn't tile to well imo. I'm also not to fond of the particular shade of green used.

I'm making a 160x160 map for a placeholder. I figure that gives me about 140x140 workable space.
I think it should be big enough to accommodate the buildings required for the academy.
I should be done with the placeholder/concept map tonight or tomorrow.

MerlinX420 wrote: The port town set doesn't tile correctly and The Small House by ax. and foth. isn't useable in it's current state.

Could you be more specific please?

i wrote:Version 0.1.0 (map ready) should be released somewhere between 2010.10.10 and 2010.10.17.

I'm certain that 0.1.0 will be released on Sunday 2010.10.10. Actually it's done right now. Some of you may already seen it on Wave.

Hello mappers!!

How can I test this?

krazyjakee wrote:How can I test this?

First of all, you need Google Wave account. Then click this link:
https://wave.google.com/wave/waveref/go ... +SPuggCVFA

or

Wait until Sunday.

or

PM me with your e-mail address, and I send you tileset in PNG.

here's a ss of a building made in tiled with the stuff1 and stuff2 sets.



If you put the tileset into tile you should be able to see all the tileing errors with these layouts.
It's nothing more in some cases then being offset or not in the exact right spot.
Nothing I couldn't edit into a workable set. I figured you weren't done and those were just worksheets.

Thanks. I will redo theses craps soon. And maybe provide a frozen tileset.

Should I put the pavement on a different tileset in order to use it on the academy? (+ I have a road and a transition to grass in the work, somewhere in my worksheet)

What about the coastal house who "isn't usable in it's current state"? You need more tiles? Which ones?