0.4.5 release procedure

Discussion about the project in general, organization, website, or any other details that aren't directly about the game.
Message
Author
User avatar
Sloth
Content Scripter
Posts: 685
Joined: Sat Mar 17, 2007 12:28 am

Re: 0.4.5 release procedure

#31 Post by Sloth »

MatGB wrote:Just recompiled my release branch and ran quickstart, I get the exact same layout as last week when I tested, it's quite a good one for Elliptical, plenty of planets that you can colonise nearby including interesting /useful specials that might spawn a guard.
I've also just tested Elliptical and it's a good map. Should i post a screenshot so we can compare that we are really testing the same thing?
MatGB wrote:When I tested Spiral-2 last week it was also good, plenty of space to expand in but I got stuck behind a maintaenance ship for a bit which was sat on the only starlane out of the spiral point you start in, 4 planets you can colonise on your side but might be a bad start, given they're not a guaranteed spawn it's not necessarily a huge issue, perhaps worth someone else trying spiral-2?
I've also tested Spiral-2 and it was also good (started right in the middle).
All released under the GNU GPL 2.0 and Creative Commons Attribution-ShareAlike 3.0 licences.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: 0.4.5 release procedure

#32 Post by MatGB »

Sloth wrote:
MatGB wrote:Just recompiled my release branch and ran quickstart, I get the exact same layout as last week when I tested, it's quite a good one for Elliptical, plenty of planets that you can colonise nearby including interesting /useful specials that might spawn a guard.
I've also just tested Elliptical and it's a good map. Should i post a screenshot so we can compare that we are really testing the same thing?
Good plan, especially given below.
MatGB wrote:When I tested Spiral-2 last week it was also good, plenty of space to expand in but I got stuck behind a maintaenance ship for a bit which was sat on the only starlane out of the spiral point you start in, 4 planets you can colonise on your side but might be a bad start, given they're not a guaranteed spawn it's not necessarily a huge issue, perhaps worth someone else trying spiral-2?
I've also tested Spiral-2 and it was also good (started right in the middle).
I wasn't in the middle, I was right out at the point of the bottom spiral, about three systems from the extreme tip. It was a good start overall, but different, so definitely worth checking.

Screenshot is from a clean config file compiled earlier today from current source, the hash# is mine because I've got some work on the predefined ships in the folder I compiled from but it's identical to last week so it hasn't affected things.
Attachments
elliptical-seed0.png
elliptical-seed0.png (348.54 KiB) Viewed 1581 times
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
Sloth
Content Scripter
Posts: 685
Joined: Sat Mar 17, 2007 12:28 am

Re: 0.4.5 release procedure

#33 Post by Sloth »

Interesting. The map is the same, but the stars are different (and i guess mostly everything else also).
Attachments
ellipse.png
ellipse.png (113.52 KiB) Viewed 1579 times
All released under the GNU GPL 2.0 and Creative Commons Attribution-ShareAlike 3.0 licences.

wheals
Space Squid
Posts: 88
Joined: Sun Mar 24, 2013 3:56 pm

Re: 0.4.5 release procedure

#34 Post by wheals »

Sounds like maybe the C++ side is using the seed, but Python isn't getting seeded properly?
All my code and content provided herein or on GitHub is released under the GPL 2.0 and/or CC-BY-SA 3.0, as appropriate.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: 0.4.5 release procedure

#35 Post by MatGB »

Just checked out the release branch completely cleanly and did a recompile, got exactly the same galaxies for both elliptical and spiral2 as I got last week. I'll check on Windows later to see what I get with them.

I have zero clue where to even start looking at why Linux and Windows would be giving different outputs from the same seed.
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

User avatar
adrian_broher
Programmer
Posts: 1156
Joined: Fri Mar 01, 2013 9:52 am
Location: Germany

Re: 0.4.5 release procedure

#36 Post by adrian_broher »

Vezzra wrote:1) @Marcel: What's the status of your translation work? Do you think you'll have this ready before Monday, or do you need more time?
I'm very sorry but I won't be able to provide any translation until a possible release for reasons I don't want to announce in the public part of the FO forum (I will write an dedicated thread in the SSAF). Please continue with the RC without waiting for the completion any contributions made by me.
Resident code gremlin
Attached patches are released under GPL 2.0 or later.
Git author: Marcel Metz

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: 0.4.5 release procedure

#37 Post by Vezzra »

Todays builds will again be pre-release test builds, not RC1, because:

1) The issue that apparently the same seeds produce different maps on Windows and Linux. For my test builds I only do test runs on Windows and OSX, and AFAICT there the maps are identical for the same seed. This issue needs to be sorted out. While we are at it: can someone open an issue on github for that?

2) I've been not at home the last few days and didn't get around to do as much FO stuff as I wanted to. So I didn't update the changelog, and also didn't dig through the settings to be changed for 0.4.5 thread to commit the settings proposed there to the release branch. I'll try get around to that ASAP this week.

3) This issue which has been opened on github. Although it looks like something that shouldn't hold up the release I guess - Dilvish, Geoff?

Once those are solved/addressed, I'll go ahead and produce RC1 (I won't necessarily wait for next Monday).

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.4.5 release procedure

#38 Post by Geoff the Medio »

Vezzra wrote:3) This issue which has been opened on github. Although it looks like something that shouldn't hold up the release I guess - Dilvish, Geoff?
It's a few missing strings... would take a few min to add them.

Edit: https://github.com/freeorion/freeorion/ ... 914809c582

Could be clearer, but avoid the error messages.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: 0.4.5 release procedure

#39 Post by Vezzra »

Vezzra wrote:Todays builds will again be pre-release test builds, not RC1
Sorry guys, I continue the get caught up in RL issues that eat too much of my time (last week was particularly hectic, had to do a complete reinstall of my system, among other things...), so no significant progress made. Apparently most of you didn't fare much better judging by the lack of activity last week ;) Consequently, tomorrow test builds will again be pre-release test builds, if any, because there haven't been any commits to the release branch since the last test builds, so no point in making new ones.

If we want to get this release out, we need to get back to business. These issues still remain:
1) The issue that apparently the same seeds produce different maps on Windows and Linux. For my test builds I only do test runs on Windows and OSX, and AFAICT there the maps are identical for the same seed. This issue needs to be sorted out. While we are at it: can someone open an issue on github for that?
I went ahead and opened an issue myself. Please, we need to address this issue ASAP, continue the discussion there.
MatGB wrote:Just checked out the release branch completely cleanly and did a recompile, got exactly the same galaxies for both elliptical and spiral2 as I got last week. I'll check on Windows later to see what I get with them.
Did you get around to that?
Vezzra wrote:2) I've been not at home the last few days and didn't get around to do as much FO stuff as I wanted to. So I didn't update the changelog, and also didn't dig through the settings to be changed for 0.4.5 thread to commit the settings proposed there to the release branch. I'll try get around to that ASAP this week.
Yeah, well, see above. Will try again this week...
Once those are solved/addressed, I'll go ahead and produce RC1 (I won't necessarily wait for next Monday).
Still stands, provided we get those issues sorted out...

User avatar
Dilvish
AI Lead and Programmer Emeritus
Posts: 4768
Joined: Sat Sep 22, 2012 6:25 pm

Re: 0.4.5 release procedure

#40 Post by Dilvish »

Vezzra wrote:
1) The issue that apparently the same seeds produce different maps on Windows and Linux. For my test builds I only do test runs on Windows and OSX, and AFAICT there the maps are identical for the same seed. This issue needs to be sorted out. While we are at it: can someone open an issue on github for that?
I went ahead and opened an issue myself. Please, we need to address this issue ASAP, continue the discussion there.
That's the best place for the substantive discussion, yes, but I'll note here that I don't consider this at all to be a 'must fix' issue-- it's not really that important and seems likely to me to be tricky-to-impossible to sort out.

Note this old stack overflow question on exactly this issue-- there was one slightly interesting suggestion about a straightforward way to check if this is because the seed values are getting hashed differently, but other than that no help that could be applicable to us.

On the hash issue, if I go into python and check hash("freeorion") I get the following:

Code: Select all

>>> hash("freeorion")
-1909360091210629068
What do windows and OSX give?
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.4.5 release procedure

#41 Post by Geoff the Medio »

Code: Select all

Python 2.7.9 (default, Apr 18 2015, 13:30:31) [MSC v.1800 32 bit (Intel)] on win32
>>> hash('freeorion')
-15384524

User avatar
Dilvish
AI Lead and Programmer Emeritus
Posts: 4768
Joined: Sat Sep 22, 2012 6:25 pm

Re: 0.4.5 release procedure

#42 Post by Dilvish »

ok, thanks much.

Here is a thread specific to the python hash differences between platforms. It has some interesting discussion and perhaps even some potential solutions for us.
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: 0.4.5 release procedure

#43 Post by Geoff the Medio »

Maybe:

Code: Select all

>>> import hashlib
>>> m = hashlib.md5()
>>> m.update("freeorion")
>>> hexval = m.hexdigest()
>>> decval = int(hexval, 16)
>>> decval
77260143718384987102530862731060942158L

User avatar
Dilvish
AI Lead and Programmer Emeritus
Posts: 4768
Joined: Sat Sep 22, 2012 6:25 pm

Re: 0.4.5 release procedure

#44 Post by Dilvish »

Geoff the Medio wrote:Maybe:...
Yes, ok, I get the same value from that:

Code: Select all

77260143718384987102530862731060942158L
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: 0.4.5 release procedure

#45 Post by Vezzra »

Dilvish wrote:On the hash issue, if I go into python and check hash("freeorion") I get the following:

Code: Select all

>>> hash("freeorion")
-1909360091210629068
What do windows and OSX give?
Hm, interesting... on OSX I get the same value as you on Linux... funny...

Post Reply