| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
|
|
Previous commit from MR !194 changed requiring password to
requiring char name for confirmation during deletion.
Text input can now be unmasked since user is
entering char name instead of password.
****
mana/plus!205
|
|
CMake way failed to build because main.h uses PACKAGE_NAME.
Reported by bjorn and meway.
Note: I like PROJECT_NAME better, but:
1) Project name is not something that changes at compile time, but
package name might.
2) PROGRAM_VERSION is already set in CMakeLists
NOTE2: earlier I made PACKAGE_EXTENDED_VERSION (used for useragent) and FULL_VERSION (used in GUI & logs) use PACKAGE_NAME, instead of hardcoding ManaVerse, for improved branding.
****
mana/plus!204
Approved-by: Thorbjørn Lindeijer <bjorn@lindeijer.nl>
|
|
****
mana/plus!203
|
|
Causes Launcher issues otherwise.
When playing from Launcher, the a/c password gets reset to some random string. Deleting character (specifically on TMW) requires password for confirmation, which player no longer has easy access to.
This changes character deletion to require confirming username instead of password.
Squashed with:
* Change to confirm character name instead of username when deleting
* change text 'password' to 'username' in error dialog.
****
mana/plus!194
|
|
Missed this part in mana/plus!186
|
|
only about 160 out of 1500 files that include localconsts.h need
to include performance.h
Saves 12% from compile times (with profiler ON) and 0.1% with profiler OFF.
****
mana/plus!189
|
|
The "move to target in attack range" option in the Quick Settings is not handled correctly. In its current state that option ends up becoming equivalent to "move to target in distance 10".
This is a fix for that.
****
mana/plus!195
|
|
call of overloaded 'toString(long long unsigned int)' is ambiguous, see !189
Squashed with:
* Apply perfo[r]mance typofixes that were missed in mana/plus!186
... because USE_PROFILER was off/undefined.
****
mana/plus!200
|
|
depends on !192
****
mana/plus!197
|
|
Better names. In case of hunger even corrects a bug as it decreases.
****
mana/plus!196
|
|
This brings it inline with how other models are implemented, and also
makes code so much clearer.
****
mana/plus!193
|
|
instead of hardcoding constants.
****
mana/plus!192
|
|
Removes the "ManaPlus" in version info on updater screen;
Screenshots now have ManaVerse in their file name, too
PS: a lot of instances of "ManaVerse" could use PACKAGE_NAME from
configure
PPS: CMake passes PROJECT_NAME. Would need a proxy header with ifdefs.
Split off by Fedja from: Rename the program from ManaPlus to ManaVerse. (!45)
****
mana/plus!198
|
|
- [x] Something needs to be done about that deletion as it may leak memory. Ditto for inputActionReplayListener (for EditDialog)
- [x] Okay, that M+ -> MV rename must be completed to make the linters pass.
Rename was completed and it works, proper cleanup happens in scheduleDelete, so that's not an issue either.
Squashed with:
* Actually fire homun (code was set to only change name, for testing of
replay stuff)
* Clarify my worries
****
mana/plus!109
|
|
(consequence: if localChatTab is nullptr, then the message will go to debug.)
****
mana/plus!187
|
|
Having the widget pop up in the middle of a fight when shooting arrows makes you go below 50% weight is kinda annoying. This changes the notification to a speech bubble on player which is less intrusive.
Squashed with:
* Move messages to top of branch (so they can be integrated to dialog with
fewer changes).
* Adjust overweight message.
* Fix linter issue. Shortened long line
****
mana/plus!184
Co-authored-by: Fedja Beader <fedja@protonmail.ch>
|
|
|
|
libxml2 deprecated __xmlParserVersion() and added a pointer to constant string version in commit
2023-09-20 "globals: Don't store xmlParserVersion in global state"
and added a compat function with it. This commit was made before v2.12.0 was released. Thus, any version *past* unreleased v2.12.0 is guaranteed to have the new API.
The compat function was removed at some point later, causing a compile failure.
****
mana/plus!190
|
|
****
mana/plus!186
|
|
in chat are disabled.
****
mana/plus!185
|
|
launcher users
Pre: 
Post: 
****
mana/plus!183
|
|
.. instead of manaplus's header.
& update copyrights to 2025.
Open problem: should it be "The ManaVerse Client", "the ManaVerse client" or something in between in "This file is part of ..." line? All caps is current situation.
Squashed with:
* update copyright dates for 2025
* Revert "Switch to branch of mplint that checks for manaverse copyright headers"
* Change back to uppercase The
Why? This line was changed 3 times in history and all had an uppercase
The.
PS: ManaVerse has no endorsment from the parent project, ManaPlus.
* Change to "This file is part of the ManaVerse Client"
poppet says 'the' instead of 'The'
find . \( -name "*.h" -o -name "*.cpp" -o -name "*.cc" -o -name "*.inc"
\) -execdir sed -i -e 's/This file is part of The ManaPlus Client/This
file is part of the ManaVerse Client/' {} \+
* Change 'The ManaPlus Client' to 'The ManaVerse Client' in headers
find . \( -name "*.h" -o -name "*.cpp" -o -name "*.cc" -o -name "*.inc"
\) -execdir sed -i -e '2s/ManaPlus/ManaVerse/' {} \+
* Switch to branch of mplint that checks for manaverse copyright headers
.. instead of manaplus's
****
mana/plus!179
|
|
introduced in 8ad99cc3 / "Add profiler lines into some functions."
****
mana/plus!181
|
|
Squashed with:
* Remove TODOs. Repeats and the other two fields are implemented in latest herc, but
they set beingID to 0 when repeats are requested... what...
Repeats need more in-client support to implement, anyways
****
mana/plus!180
|
|
counterparts.
|
|
Some if-s were changed into else if-s. They cannot happen if the
branches above them happen, anyway.
****
mana/plus!177
|
|
****
mana/plus!176
|
|
****
mana/plus!175
|
|
Dead players have to be returned so you can still see
their equipment, do whisper or trade (yes you can trade
with the dead :D) or use any other option in the menu.
****
mana/plus!174
|
|
Squashed with:
* Fix some english in mail-related notification messages
****
mana/plus!169
|
|
+ shift one argument's position in changeMode(...)
Addendum to !171.
****
mana/plus!172
|
|
+ slight reformat
****
mana/plus!171
|
|
Rationale: By the standard, cstdint guarantees these basic types
in std::, but *may* put them into the global namespace.
By including stdint.h directly, we guarantee that these types are
in the global namespace. stdint.h is
also a C++ standard header guaranteed as part of the C compatibility
support, thus toolchains should support this.
The proper(tm) way of going about this would mean either prefixing
[u]int[0-9]*_t with std:: everywhere where they are used (a chore)
or having a proxy header that includes cstdint and around 50
using statements. The latter sounds to me like a good approach for
toolchains that do not provide stdint.h and do not pollute global
namespace. If such a toolchain exists in the wild.
See discussion at
https://stackoverflow.com/questions/13642827/cstdint-vs-stdint-h
In practice, my toolchain's (gcc 14 on GNU) cstdint includes stdint.h
then drags those types into std:: as well.
****
mana/plus!170
|
|
Squashed with:
* Rename - it's member shift, not printf shift
* So it seems that CILK PLUS either ignores the this argument of member
functions or already considers it on its own.
Besides.. is CILK PLUS even relevant today? Whatever..
Plus, deduplicate some more functions
..should this attribute be moved to its own file?
* Correct indents
* logger.h: de-duplicate __attribute__(__format__(..))
****
mana/plus!163
|
|
blocks in pickup search functions
This block repeats many times.. but cannot be de-duplicated without that
pesky search for empty string. perhaps I should make an (inline) function
that takes allowAll and the item or a scoped functor that does the lookup in
constructor.
****
mana/plus!166
|
|
Why 6? 2*2 + 1*1 is still < 6, so this allowed you to pickup items 2
tiles away. A bug.
Test in ML: can pickup in 5x5 area with Quick option selector set
to 3x3.
Test in TMW: can only pickup in 3x3 area regardless of this setting.
****
mana/plus!167
|
|
Addendum to 437258c4 / "Remove -enablecommandlinepassword necessity"
****
mana/plus!156
|
|
Git blame for this says that Hercules does this on its own, but this is
only true if server side attack is enabled. Using that setting results in
archers being warped into targets, so its use is discouraged.
End of unit_attack_timer_sub in Hercules:src/map/unit.c has unit
direction updates gated on ud->state.attack_continue.
****
mana/plus!160
|
|
Prolific use of immediate constants strikes again.
Side note: why are these limits not in clientdata?
****
mana/plus!168
|
|
pathJoin indent is probably going to annoy a linter, got a patch ready
if that happens. Not ideal but better than now.
****
mana/plus!162
|
|
Linters! Wake up!
****
mana/plus!161
|
|
.. this will prevent the shopping list from going above your free
carry weight, and thus the buy action failing.
Note: I don't particularly like this implemention due to code repetition. But perfection is enemy of progress.
TODO:
1) add the same for total price (other MR?)
2) maybe do a small refactor + add comments where necessary (other MR?)
3) ~~free weight display~~ done
Squashed with:
* buydialog: Add free weight display to money label
+ fix free weight going below 0
****
mana/plus!152
|
|
Squashed with:
* More docs for validateBeing
* addendum documentation for validateBeing
****
mana/plus!154
|
|
bb= bounding box
**** mana/plus!145
|
|
seems to skip beings with info if stopattack is active or if they are
marked as unselectable (like hammocks in ML)... huh?
**** mana/plus!145
|
|
as without filtering, priority and closestPriority are always the same
**** mana/plus!145
|
|
making special handling before first suitable being is found unnecessary.
**** mana/plus!145
|
|
from 3x)
Note: the filter used priorityMobs (instead of priorityMobsMap).
The former seems to be the std::set variant of the latter, but I can't
find where either are filled. It does work in practice, however...
**** mana/plus!145
|
|
**** mana/plus!145
|
