[Freeciv-Dev] [bug #21896] double free or corruption after Connect to network game
Follow-up Comment #5, bug #21896 (project freeciv): output window thread safety pacth committed, but as I have not reproduced the crash with any version, leaving ticket open for jtn to confirm if it helped. ___ Reply to this item at: http://gna.org/bugs/?21896 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4692] Lucky achievement
Update of patch #4692 (project freeciv): Status: Ready For Test = Done Assigned to:None = cazfi Open/Closed:Open = Closed ___ Reply to this item at: http://gna.org/patch/?4692 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #2, patch #4393 (project freeciv): Ooh, maybe I can use this to do the thing I've wanted to do for ages, of making the capital city obvious on the map. Should be, if you define three walls sets: 1 - Walls 2 - Capital 3 (1+2) - Capital with Walls ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #3, patch #4393 (project freeciv): - Updated against svn, and this time tested too (file #20754) ___ Additional Item Attachment: File name: MultiWallGfx-2.patch Size:10 KB ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #5, patch #4393 (project freeciv): Value 5 is completely arbitrary, just something I threw in for testing the implementation. You're right that 8 (or 7, if I recall correctly that non-walled city does not count toward NUM_WALL_TYPES) would be much more sensible. UINT8 stands for 8-bit unsigned integer so it can contain values 0-255. ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #6, patch #4393 (project freeciv): The non-boolean use-case I've created this for is that wall types act as upgrades to previous ones, so that only one is active at the time - EFT_VISIBLE_WALLS value is the level of the walls. ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #21896] double free or corruption after Connect to network game
Update of bug #21896 (project freeciv): Status: Ready For Test = Fixed Open/Closed:Open = Closed ___ Follow-up Comment #6: I was able to reproduce this particular crash symptom before the fix (r24891) but not afterwards (r24892) in 17 attempts. I'm content to declare this fixed. (Complicated by there being a much more frequent crash symptom now whose backtrace looks different and not obviously related. I'll raise another bug for that.) ___ Reply to this item at: http://gna.org/bugs/?21896 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #4, patch #4393 (project freeciv): This is brilliant, but could NUM_WALL_TYPES be 8 rather than 5, to support another boolean type, so that one could do something like: City Walls: EFT_VISIBLE_WALLS: 1 Palace: EFT_VISIBLE_WALLS: 2 Television Tower: EFT_VISIBLE_WALLS: 4 Using potential images ilke: 0: Regular city 1: Walled city 2: Regular city with Palace 3: Walled city with Palace 4: Regular city with Television Tower 5: Walled city with Television Tower 6: Regular city with Palace and Television Tower 7: Walled city with Palace and Television Tower Note that this would require restructuring the guard in packhand.c, as 8 wouldn't fit in UINT8. That said, if there is a non-boolean intended use of this which I fail to understand, please share, as I may agree that 5 is a better value. ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22047] Crash in gtk_text_layout_get_cursor_locations() on Connect to network game
URL:
http://gna.org/bugs/?22047
Summary: Crash in gtk_text_layout_get_cursor_locations() on
Connect to network game
Project: Freeciv
Submitted by: jtn
Submitted on: Sun 18 May 2014 12:58:01 BST
Category: client-gtk-2.0
Severity: 3 - Normal
Priority: 5 - Normal
Status: None
Assigned to: None
Originator Email:
Open/Closed: Open
Release: trunk r24892
Discussion Lock: Any
Operating System: GNU/Linux
Planned Release: 2.6.0
___
Details:
(Found while verifying fix for bug #21896.)
Trunk code on Ubuntu 12.04 amd64: start client/freeciv-gtk2; press Alt-O for
connect to network game. Crashed immediately in 8 of 17 trials.
The crash details differ (just a segfault, double free or corruption, etc),
but the function gtk_text_layout_get_cursor_locations() is always in the
backtrace (so may be the same as bug #21575).
The metaserver thread appears quiescent (in ppoll()) in all the coredumps, so
this isn't as obviously a thread-safety issue as bug #21896 was.
Not tried branches other than trunk.
Some backtraces (main thread):
Most common symptom: seen in 7 of 8 crashes.
#0 0x7fc22a4cc425 in __GI_raise (sig=optimised out)
at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
resultvar = 0
pid = optimised out
selftid = 6789
#1 0x7fc22a4cfb8b in __GI_abort () at abort.c:91
save_stage = 2
act = {__sigaction_handler = {sa_handler = 0x4, sa_sigaction = 0x4},
sa_mask = {__val = {5, 140733887709049, 19, 140471911392743, 3,
140733887697754, 6, 140471911392747, 2, 140733887697774, 2,
140471911383746, 1, 140471911392743, 3, 140733887697748}},
sa_flags = 12, sa_restorer = 0x7fc22a6125eb}
sigs = {__val = {32, 0 repeats 15 times}}
#2 0x7fc22a50a39e in __libc_message (do_abort=2,
fmt=0x7fc22a614748 *** glibc detected *** %s: %s: 0x%s ***\n)
at ../sysdeps/unix/sysv/linux/libc_fatal.c:201
ap = {{gp_offset = 40, fp_offset = 48,
overflow_arg_area = 0x7fff29625cd0,
reg_save_area = 0x7fff29625be0}}
ap_copy = {{gp_offset = 16, fp_offset = 48,
overflow_arg_area = 0x7fff29625cd0,
reg_save_area = 0x7fff29625be0}}
fd = 11
on_2 = optimised out
list = optimised out
nlist = optimised out
cp = optimised out
written = optimised out
#3 0x7fc22a514b96 in malloc_printerr (action=3,
str=0x7fc22a614938 double free or corruption (fasttop),
ptr=optimised out) at malloc.c:5039
buf = 045228c0
cp = optimised out
#4 0x7fc22b3a5176 in gtk_text_layout_get_cursor_locations ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#5 0x7fc22b3b0d40 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#6 0x7fc22b3b0d77 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#7 0x7fc22b3b0dd3 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#8 0x7fc22af2bd56 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgdk-x11-2.0.so.0
No symbol table info available.
#9 0x7fc22be85d13 in g_main_context_dispatch ()
from /lib/x86_64-linux-gnu/libglib-2.0.so.0
No symbol table info available.
#10 0x7fc22be86060 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
No symbol table info available.
#11 0x7fc22be8645a in g_main_loop_run ()
from /lib/x86_64-linux-gnu/libglib-2.0.so.0
No symbol table info available.
#12 0x7fc22b2f4397 in gtk_main ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#13 0x0044d1e9 in ui_main (argc=1, argv=0x7fff296263a8)
at gui_main.c:1677
home = optimised out
sig = optimised out
__FUNCTION__ = ui_main
#14 0x00476f58 in client_main (argc=1, argv=0x7fff296263a8)
at client_main.c:620
i = 1
loglevel = LOG_NORMAL
ui_options = optimised out
ui_separator = optimised out
option = optimised out
user_tileset = optimised out
fatal_assertions = -1
aii = optimised out
__FUNCTION__ = client_main
#15 0x7fc22a4b776d in __libc_start_main (main=0x449f40 main, argc=1,
ubp_av=0x7fff296263a8, init=optimised out, fini=optimised out,
rtld_fini=optimised out, stack_end=0x7fff29626398) at libc-start.c:226
result = optimised out
unwind_buf = {cancel_jmp_buf = {{jmp_buf = {0, -8850937098092952370,
4497224, 140733887701920, 0, 0, 8851109690452985038,
8840411090749211854}, mask_was_saved = 0}}, priv = {pad = {
0x0,
[Freeciv-Dev] [bug #21575] GTK2 client crash at startup
Follow-up Comment #1, bug #21575 (project freeciv): May be the same as bug #22047? Is this seen literally at client startup, or only after pressing the Connect to network game button? ___ Reply to this item at: http://gna.org/bugs/?21575 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22047] Crash in gtk_text_layout_get_cursor_locations() on Connect to network game
Follow-up Comment #1, bug #22047 (project freeciv): The metaserver thread appears quiescent Erm, what I've been assuming to be the metaserver thread -- it might be something else (there are no Freeciv symbols in the backtrace so it probably is something else, on reflection). ___ Reply to this item at: http://gna.org/bugs/?22047 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22048] Crashes and other trouble with repeated Connect to network game / Cancel
URL:
http://gna.org/bugs/?22048
Summary: Crashes and other trouble with repeated Connect to
network game / Cancel
Project: Freeciv
Submitted by: jtn
Submitted on: Sun 18 May 2014 13:18:00 BST
Category: client-gtk-2.0
Severity: 3 - Normal
Priority: 5 - Normal
Status: None
Assigned to: None
Originator Email:
Open/Closed: Open
Release: trunk r24892
Discussion Lock: Any
Operating System: GNU/Linux
Planned Release: 2.6.0
___
Details:
(Found while trying to verify fix for bug #21896.)
Xubuntu 12.04 amd64; start freeciv-gtk2; repeatedly Connect to network game
then Cancel (using keyboard shortcuts for speed: hold down Alt then
c[O]nnect, [C]ancel, [O], [C], ...).
Observe a variety of symptoms:
1. This verbose warning message from Gtk:
(freeciv-gtk2:10919): Gtk-WARNING **: Invalid text buffer iterator: either the
iterator is uninitialized, or the characters/pixbufs/widgets in the buffer
have been modified since the iterator was created.
You must use marks, character numbers, or line numbers to preserve a position
across buffer modifications.
You can apply tags and insert marks without invalidating your iterators,
but any mutation that affects 'indexable' buffer contents (contents that can
be referred to by character offset)
will invalidate all outstanding iterators
2. A crash like this in the metaserver thread (seen twice):
Program terminated with signal 11, Segmentation fault.
#0 0x7fab9b6f95f8 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#1 0x7fab9b6dd55c in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#2 0x7fab9b6e035f in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#3 0x7fab9abbaca2 in g_closure_invoke ()
from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
No symbol table info available.
#4 0x7fab9abcc339 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
No symbol table info available.
#5 0x7fab9abd4069 in g_signal_emit_valist ()
from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
No symbol table info available.
#6 0x7fab9abd4212 in g_signal_emit ()
from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
No symbol table info available.
#7 0x7fab9b6e1de8 in ?? ()
from /usr/lib/x86_64-linux-gnu/libgtk-x11-2.0.so.0
No symbol table info available.
#8 0x004d51c7 in apply_text_tag (ptag=optimised out, buf=0x1a0b100,
text_start_offset=405, text=optimised out) at chatline.c:841
tag = 0x3f12840
foreground = {pixel = 2139528224, red = 61423, green = 32639, blue =
0}
background = {pixel = 27308288, red = 0, green = 0, blue = 10316}
initalized = true
start = {dummy1 = 0x1aaf200, dummy2 = 0x7fab78012bc0, dummy3 = -1,
dummy4 = 0, dummy5 = 405, dummy6 = -1, dummy7 = 595129065,
dummy8 = -1729242823, dummy9 = 0x7fab780144b0,
dummy10 = 0x7fab78012710, dummy11 = -1, dummy12 = 0, dummy13 = 0,
dummy14 = 0x7fab7f869c20}
stop = {dummy1 = 0x1aaf200, dummy2 = 0x7fab78012bc0, dummy3 = -1,
dummy4 = 54, dummy5 = 459, dummy6 = -1, dummy7 = 595129065,
dummy8 = -1729242823, dummy9 = 0x1aaf1a0, dummy10 = 0x1aaf4e0,
dummy11 = -1, dummy12 = 0, dummy13 = 68009248, dummy14 = 0x1a0b100}
#9 0x004d5597 in real_output_window_append (
astring=0x7fab7f869d40 There is no newer stable release of Freeciv
available., tags=optimised out, conn_id=optimised out) at chatline.c:916
ptag_iter = 0x0
ptag = 0x7fab78014240
buf = 0x1a0b100
iter = {dummy1 = 0x1aaf200, dummy2 = 0x7fab78012bc0, dummy3 = 54,
dummy4 = 54, dummy5 = -1, dummy6 = -1, dummy7 = 595129065,
dummy8 = -1729242823, dummy9 = 0x1aaf1a0, dummy10 = 0x1aaf4e0,
dummy11 = 0, dummy12 = 0, dummy13 = 68009248,
dummy14 = 0x7fab9b6f7679}
mark = 0x40dbd20
text_start_offset = 405
__FUNCTION__ = real_output_window_append
#10 0x00478480 in output_window_append (color=...,
featured_text=optimised out) at chatline_common.c:130
plain_text = There is no newer stable release of Freeciv available.,
'\000' repeats 11 times, \001, '\000' repeats 14 times\377,
\377\377\377\060\000\000\000
\241\206\177\253\177\000\000`\240\206\177\253\177\000\000\240/\272\232\253\177\000\000\000\000\000\000\000\000\000\000\n\000\000\000\253\177\000\000\003,
'\000' repeats 23 times, P, '\000' repeats 31 times,
t\212A\235\253\177, '\000' repeats 11 times,
\004\000\000\000\000\000\000\220\236\206\177\253\177\000\000\000\347`o\377\177\000\000\300Ɇ\177\253\177\000\000\000\263\247B\237++\260\003,
[Freeciv-Dev] [bug #21896] double free or corruption after Connect to network game
Follow-up Comment #7, bug #21896 (project freeciv): On the other hand, I managed to reproduce other trouble in this area while trying to break this, so there might be more work to do here. However, the trigger and symptom is sufficiently different that I raised a new ticket, bug #22048. ___ Reply to this item at: http://gna.org/bugs/?21896 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #21326] You already had a city called Helsinki. The city was renamed to Helsinki.
Update of bug #21326 (project freeciv): Status: Ready For Test = Fixed Open/Closed:Open = Closed ___ Reply to this item at: http://gna.org/bugs/?21326 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22049] Players can guess the number of units inside enemy cities
URL: http://gna.org/bugs/?22049 Summary: Players can guess the number of units inside enemy cities Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 03:50:53 PM CEST Category: general Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: Discussion Lock: Any Operating System: Any Planned Release: ___ Details: The server sends out of sight packets for every unit inside cities when a unit loses vision of it. It results that a hacker can easily know how many units are occupying a city. All branches affected (at least until S2_0). I think the server shouldn't send this packets if the unit wasn't visible before. ___ Reply to this item at: http://gna.org/bugs/?22049 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4393] Support multiple city graphics providing buildings
Follow-up Comment #7, patch #4393 (project freeciv): - Make maximum number of Wall Types 7. (file #20755) ___ Additional Item Attachment: File name: MultiWallGfx-3.patch Size:10 KB ___ Reply to this item at: http://gna.org/patch/?4393 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4699] Alien Ruleset to support different gfx for stronger Force Walls
Update of patch #4699 (project freeciv): Depends on: = patch #4393 ___ Reply to this item at: http://gna.org/patch/?4699 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4699] Alien Ruleset to support different gfx for stronger Force Walls
URL: http://gna.org/patch/?4699 Summary: Alien Ruleset to support different gfx for stronger Force Walls Project: Freeciv Submitted by: cazfi Submitted on: Sun 18 May 2014 05:48:27 PM EEST Category: rulesets Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: None Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: Provide ruleset side support for having distinct Force Walls gfx depending whether owner knows Strong Force, making walls more effective, or not. ___ File Attachments: --- Date: Sun 18 May 2014 05:48:27 PM EEST Name: ForceWallVisibleLevels.patch Size: 588B By: cazfi http://gna.org/patch/download.php?file_id=20756 ___ Reply to this item at: http://gna.org/patch/?4699 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22050] Recursive transport problems
URL:
http://gna.org/bugs/?22050
Summary: Recursive transport problems
Project: Freeciv
Submitted by: pepeto
Submitted on: Sun 18 May 2014 05:46:41 PM CEST
Category: general
Severity: 3 - Normal
Priority: 5 - Normal
Status: Need Info
Assigned to: None
Originator Email:
Open/Closed: Open
Release:
Discussion Lock: Any
Operating System: Any
Planned Release:
___
Details:
When I was investigating for bug #21899, I noticed some other related bugs or
strange behaviors. In patch #4694, I said I couldn't understand
unit_transport_check(). After looking to patch #2480 about recursive
transport, I now know why.
After a quick test, I noticed I was enabled to unload a land transport from
a sea transport (on land of course). I got an enigmatic server message about
tiles nativity I couldn't reproduce afterward.
But before making a fix, I have some questions I cannot answer alone.
Analysis:
In could_unit_load() (common/unit.c):
* The call to unit_transport_check() will never be effective:
/* Only top-level transporters may be loaded or loaded into. */
if (unit_transported(pcargo)
|| unit_transported(ptrans)) {
return FALSE;
}
/* Check iff this is a valid transport. */
if (unit_transported(pcargo)
!unit_transport_check(pcargo, ptrans)) {
return FALSE;
}
*
In unit_transport_check() (common/unit.c):
* The argument _ptrans_ is totally ignored. I don't think it is intentional.
* I don't like the usage of game_unit_by_number() just to drop the _const_
flag.
* It useless to use a _unit_list_ as long as units are already chained by
pcargo-transporter-transporter-etc.
In check_units() (server/sanitycheck.c):
SANITY_CHECK(unit_list_search(pcargos, plevel) != NULL);
This test duplicates the test just above (+ unit iteration). It will just do
the tests^2.
Questions:
0 May really only top-level transporters be loaded or loaded into?
0 unit_transport_check() assumes that a unit cannot be loaded to its direct
transport if it can be transported by any indirect transport. It is not what I
would expect. What do you think?
0 A full transport loaded into another counts only for 1 occupancy. It is not
what I would expect neither. But I would like your opinions.
___
Reply to this item at:
http://gna.org/bugs/?22050
___
Message sent via/by Gna!
http://gna.org/
___
Freeciv-dev mailing list
Freeciv-dev@gna.org
https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #20722] [metaticket] S2_4 unit transport issues
Update of bug #20722 (project freeciv): Depends on: = bugs #22050 ___ Reply to this item at: http://gna.org/bugs/?20722 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22050] Recursive transport problems
Follow-up Comment #1, bug #22050 (project freeciv): (Lost the second point of my analysis about could_unit_load()) * can_unit_exist_at_tile(ptrans) wouldn't be effective neither for the same reason. ___ Reply to this item at: http://gna.org/bugs/?22050 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22051] Missing route for a city on river
URL: http://gna.org/bugs/?22051 Summary: Missing route for a city on river Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 06:27:32 PM CEST Category: None Severity: 3 - Normal Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: S2_5,trunk Discussion Lock: Any Operating System: Any Planned Release: ___ Details: At start, building a city on river doesn't give a route. Earlier versions of Freeciv gave a free route, even on river. ___ Reply to this item at: http://gna.org/bugs/?22051 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #21161] human team player changes my reserch goal
Follow-up Comment #3, bug #21161 (project freeciv): I can't reproduce. Is there any extra option set. Could you provide a savegame? ___ Reply to this item at: http://gna.org/bugs/?21161 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22051] Missing route for a city on river
Follow-up Comment #1, bug #22051 (project freeciv): Feature. Roads do not have AlwaysOnCityCenter flag on classic ruleset. See patch #3007, patch #3154, and patch #3522 ___ Reply to this item at: http://gna.org/bugs/?22051 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22051] Missing route for a city on river
Follow-up Comment #2, bug #22051 (project freeciv): I was wondering if it was a feature. This looks strange for not to be able to connect cities with roads at start (because of the river). But if I am the only one complaining, I will live with... ___ Reply to this item at: http://gna.org/bugs/?22051 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22050] Recursive transport problems
Follow-up Comment #2, bug #22050 (project freeciv): On questions, I have some opinions, as follows: 1a) For interface simplicity, it makes sense to me that only top-level transporters can be used as a load target, especially as the user has no way to indicate into which transporter a unit should be loaded. Using the external transport also makes narrative sense: it might be hard to load a tank only a boat already placed in a helicopter cargo net. 1b) There's no interface rationale to restrict unloading to the outermost transport, but it gains the same narrative support as 1a above. That said, this should be enforced at the client interaction level, as otherwise things get very odd (e.g. alliance ends, with loaded tank in boat in helicopter cargo net: should the tank not be bounced?). 2) While the intent (mentioned in the comment in unit_transport_check()) makes sense, I think this is actively annoying. For example consider the possiblility of a missle-carrying mech.inf. being loaded into a cargo plane, which then wants to land on a carrier: should this only be permitted if there are no missiles currently loaded? A more sensible check should probably be just to ensure that none of the containing transports are of the unit type of the unit to be transported. 3) I think this is correct, to support the same scenario of the missile-carrying Mech. Inf. in a cargo plane on a carrier: the carrier can carry the same number of planes, whether they have units loaded, and the plane doesn't care if the Mech. Inf. happens to have missiles loaded. It is up to ruleset authors to ensure that the capacities of various transport are narratively sensible. That said, I'm not sure that all units are the same size: it might be interesting to declare, for example, that the aforementioned plane can take 1 Mech. Inf. or 4 Paratroopers (although this depends on the narrative for the units: if 1 Paratrooper Unit actually represents a group of 15-20 effectives, whereas the Mech. Inf. is 6 effectives who can all fit inside the vehicle, they may require similar space), but this is probably a separate feature (units would have a size parameter, which would indicate how many cargo slots they occupied). ___ Reply to this item at: http://gna.org/bugs/?22050 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22051] Missing route for a city on river
Follow-up Comment #4, bug #22051 (project freeciv): I'd be tempted to set AlwaysOnCityCenter in classic, and AutoOnCityCenter in experimental, as it represents a noticeable change. Perhaps some of the other features that were enabled in experimental in 2.4 can be moved to classic in 2.5 instead? Do we have a list of these, or some other plan for when things move from experimental to classic? ___ Reply to this item at: http://gna.org/bugs/?22051 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4700] Revolentype
URL: http://gna.org/patch/?4700 Summary: Revolentype Project: Freeciv Submitted by: cazfi Submitted on: Sun 18 May 2014 11:59:18 PM EEST Category: general Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: None Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: Currently length of anarchy period is controlled by single 'revolen' setting. It's value tells the length of the revolution, except that special value 0 means random value between 1 and 5. This patch adds separate 'revolentype' setting. Currently it can have either value 'FIXED', which means that 'revolen' is the length of the revolution, or 'RANDOM' which means that revolution length is of random length up to 'revolen̈́'. 'revolen' itself can no longer take value 0 at all. This change already makes it possible to define maximum length of the random length revolution to be something else than hardcoded 5. ___ File Attachments: --- Date: Sun 18 May 2014 11:59:18 PM EEST Name: Revolentype.patch Size: 9kB By: cazfi http://gna.org/patch/download.php?file_id=20758 ___ Reply to this item at: http://gna.org/patch/?4700 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22053] Impossible to attack cities with pathfinding
URL: http://gna.org/bugs/?22053 Summary: Impossible to attack cities with pathfinding Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 11:06:40 PM CEST Category: ai Severity: 3 - Normal Priority: 5 - Normal Status: Ready For Test Assigned to: pepeto Originator Email: Open/Closed: Open Release: trunk r24881 Discussion Lock: Any Operating System: Any Planned Release: 2.6.0 ___ Details: First bug detected in patch #3901. Fix attached. ___ File Attachments: --- Date: Sun 18 May 2014 11:06:40 PM CEST Name: pf_able_to_attack.patch Size: 744B By: pepeto http://gna.org/bugs/download.php?file_id=20759 ___ Reply to this item at: http://gna.org/bugs/?22053 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4701] Faster priority queue
URL: http://gna.org/patch/?4701 Summary: Faster priority queue Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 11:09:22 PM CEST Category: general Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: pepeto Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: The execution time of the function pq_remove(), pq_insert() and pq_replace() should be a bit better (something like 5%). ___ File Attachments: --- Date: Sun 18 May 2014 11:09:22 PM CEST Name: faster_pqueue.patch Size: 5kB By: pepeto http://gna.org/patch/download.php?file_id=20760 ___ Reply to this item at: http://gna.org/patch/?4701 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4702] Reduce the number of calls to priority queue functions
URL: http://gna.org/patch/?4702 Summary: Reduce the number of calls to priority queue functions Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 11:15:32 PM CEST Category: general Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: pepeto Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: Try to prefer pq_replace() to reduce the much calls to pq_remove(). After the patch applied, the tiles index should be only listed once. Note that it can modify autogames because the order in the priority queue could be slightly different (for paths of same cost). ___ File Attachments: --- Date: Sun 18 May 2014 11:15:32 PM CEST Name: pf_pq_replace.patch Size: 10kB By: pepeto http://gna.org/patch/download.php?file_id=20761 ___ Reply to this item at: http://gna.org/patch/?4702 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
URL: http://gna.org/patch/?4703 Summary: Remove calls to same_pos() in pathfinding Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 11:17:09 PM CEST Category: ai Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: pepeto Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: Comparing pointer should do the same, and faster. ___ File Attachments: --- Date: Sun 18 May 2014 11:17:09 PM CEST Name: pf_remove_same_pos.patch Size: 6kB By: pepeto http://gna.org/patch/download.php?file_id=20762 ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4704] Zoc handling correction and optimization in pathfinding
URL: http://gna.org/patch/?4704 Summary: Zoc handling correction and optimization in pathfinding Project: Freeciv Submitted by: pepeto Submitted on: Sun 18 May 2014 11:23:37 PM CEST Category: ai Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: pepeto Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: Make ZOC_MINE the default, remove the duplicate of the city test and prefer terrain_has_flag(TER_NO_ZOC) to hardcoded terrain_type_terrain_class() == TC_OCEAN (see patch #4597). ___ Reply to this item at: http://gna.org/patch/?4704 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #21988] Genhash collision handling very inefficient
Update of bug #21988 (project freeciv): Assigned to:None = pepeto ___ Reply to this item at: http://gna.org/bugs/?21988 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
Follow-up Comment #1, patch #4703 (project freeciv): Are you sure than no tile compared can ever be virtual one (having distinct pointer from real tile)? ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
Follow-up Comment #2, patch #4703 (project freeciv): I am sure no virtual tile is used in pathfinding. I also noticed that virtual tiles are no more handled in same_pos() (I didn't investigate since when). ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
Follow-up Comment #3, patch #4703 (project freeciv): same_pos() only compares tile pointer for about 10 years (trunk r8387, rt PR#10047). I doubt this function is still useful, it probably can be removed completely. ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
Follow-up Comment #4, patch #4703 (project freeciv): So do we have any function to check if two tiles (of which at least one might be virtual) have same map coordinates? ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4703] Remove calls to same_pos() in pathfinding
Follow-up Comment #5, patch #4703 (project freeciv): I thought same_pos() was doing it. I don't know any other function able to do that. ___ Reply to this item at: http://gna.org/patch/?4703 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [patch #4705] Get rid of last function names with prefix my_
URL: http://gna.org/patch/?4705 Summary: Get rid of last function names with prefix my_ Project: Freeciv Submitted by: cazfi Submitted on: Mon 19 May 2014 01:24:07 AM EEST Category: general Priority: 5 - Normal Status: Ready For Test Privacy: Public Assigned to: None Originator Email: Open/Closed: Open Discussion Lock: Any Planned Release: 2.6.0 ___ Details: CodingStyle prohibits function name prefix my_. Attached patch gets rid of the last remaining ones. ___ File Attachments: --- Date: Mon 19 May 2014 01:24:07 AM EEST Name: NoMy.patch Size: 13kB By: cazfi http://gna.org/patch/download.php?file_id=20763 ___ Reply to this item at: http://gna.org/patch/?4705 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22054] Arrival of a caravan/freight by sea does not trigger traderoute dialog
URL: http://gna.org/bugs/?22054 Summary: Arrival of a caravan/freight by sea does not trigger traderoute dialog Project: Freeciv Submitted by: sds Submitted on: Mon 19 May 2014 01:53:27 AM UTC Category: general Severity: 4 - Important Priority: 5 - Normal Status: None Assigned to: None Originator Email: Open/Closed: Open Release: 2.4.2 Discussion Lock: Any Operating System: GNU/Linux Planned Release: ___ Details: When I bring a caravan/freight to a city by sea, I do not get the traderoute dialog. The caravan/freight must arrive by land. When I unload the caravan and move it in and out of the city, I do get the traderoute dialog. This used to work in a previous version. The problem means that I cannot establish a traderoute between two 1x1 islands. I run freeciv using this command: `ssh -X linux-box /usr/games/freeciv-gtk2` on a mac os x box. ___ Reply to this item at: http://gna.org/bugs/?22054 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22048] Crashes and other trouble with repeated Connect to network game / Cancel
Follow-up Comment #1, bug #22048 (project freeciv): I could reproduce the crash a couple of times without the attached patch, but not once with it. It initializes meta server scan result before new scan. Presumably it used to get SCAN_STATUS_DONE from the previous scan, causing attempt to handle the scan results even before they are created. Do not overwrite SCAN_STATUS_ABORT with SCAN_STATUS_DONE. Also get rid of scan-meta.state that was never used, only set at one place. (file #20766) ___ Additional Item Attachment: File name: InitMetaScanStatus.patch Size:1 KB ___ Reply to this item at: http://gna.org/bugs/?22048 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22048] Crashes and other trouble with repeated Connect to network game / Cancel
Update of bug #22048 (project freeciv): Status:None = Ready For Test ___ Reply to this item at: http://gna.org/bugs/?22048 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22040] AI creates too many triremes
Follow-up Comment #3, bug #22040 (project freeciv): Looked at the attachment screenshot, and there's sails (like I have experienced), *not* triremes as original submission claimed. ___ Reply to this item at: http://gna.org/bugs/?22040 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
[Freeciv-Dev] [bug #22040] AI creates too many triremes
Update of bug #22040 (project freeciv): Status:None = Ready For Test Planned Release: = 2.4.3, 2.5.0, 2.6.0 ___ Follow-up Comment #4: Untested patch to address one problem I noticed - death of potential passenger does not free the ferry but leaves it forever allocated. (file #20767, file #20768) ___ Additional Item Attachment: File name: PassengerDeath.patch Size:4 KB File name: PassengerDeath-S2_4.patch Size:1 KB ___ Reply to this item at: http://gna.org/bugs/?22040 ___ Message sent via/by Gna! http://gna.org/ ___ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev
