There's no need to keep all this infrastructure to run every frame for a barely used situation. This cleans it up a bit and does the position reset in the actual drawString function.
This range check is to ensure that the ship's hull is in a valid range, so it should therefore check that the hull is less than the max strength, not less than the current strength. This was inadvertently changed in 593be42e66 and a04e83ab01, but retail behavior is to set the exact hull from the previous mission.
Tested for both regular ships and delayed-arrival ships. Fixes#4472.
Sync the "mission save" and "error reporting" versions of the SEXP variable string. Also use `stuff_sexp_text_string` to print the "bad node" token rather than `Sexp_nodes[bad_node].text`, so that a human-readable token is displayed instead of a magic number like 26.
Related to #4226, as the generated error message is formatted differently enough to be confusing.
1) A comprehensive fix for the is_target_dying variable
2) Also do distance correctly for non-player target brackets by adding a check that we're dealing with the target before using the current target distance. Then also calc distance if it is not the player target.
3) Update comments and checks to be consistent
Many of these did not reflect how the variable was actually used, so update the comments.
* Fix AI Goal Display Strings
Okay so previously, if a target goal was not valid the code was
`strcpy(outbuf, XSTR("no orders", 495)); `
but then it was changed to
`outbuf += XSTR("no orders", 495);`
So it would instead append the "no orders" instead of properly replace the whole string. Looks just like a copy and paster oversight, since all the other text appends the name of the target with the order.
This PR correctly replaces, rather than appends that text. Fixes#4345
* appease clang
* use more simple =
Tests with standalone multiplayer builds revealed that the server would seg fault if a player tried to switch secondaries if the ship did not have secondaries. Tests with standalones showed this fixed the bug. Thanks to chief for the bug IDing and the fix code.
Another bug discovered by Durandal. The message sender was a variable containing an invalid value, and the game crashed. This handles it more gracefully.
Tested and verified.
This goal was missed in `ai_remove_goal_sexp_sub`, so add handling for it. Also make the error message clearer in case another goal is missed in the future.
The two `VIRTVAR`s for Orientation in subsystems and submodel instances do the same thing, but did not have the same code - in particular, the subsystem version did not update the subsystem angle and turret angle. This syncs them up.
Arrow, function, Ctrl, Alt ... and any other non-ascii keys
would previously add a blank space to the multi msg buffer.
This prevents that from happening, making sure that it's an
ascii character before appending it to the buffer.
(Retail bug??)
The previous way to clear a skybox was to set it to "default", and if the mission had no skybox then it would be cleared. The skybox code can correctly handle the lack of a skybox, but if you pass an empty string to the skybox preloader it fails with a file not found error. This fixes the preloader, and also adds "none" as an explicit no-skybox parameter.
Certain campaign-related functions would run even when the game was not in campaign mode (e.g. when playing in the tech room). Normally these would be benign, but with a new pilot or with a skipped mission, the current campaign mission would be -1 and a crash would occur.
Should fully fix#4412.
Goober5000