It turns out that checking the room height causes verbs not to be
redrawn after reading books. Let's just limit it to room 80, and hope
there aren't any other cases.
There is at least one scene in the game where inactive verbs are drawn,
and then the scene background is drawn over them. This should mean the
verbs are no longer visible, but the way we implement high-resolution
text on a scaled low-resolution background all text is assumed to have a
text mask on the text surface.
We work around this by only drawin verbs in standard rooms, where
"standard" is defined as being 128 pixels tall.
curRect.left is assigned in the script, and it is not modified. right is
assigned with the screen width - original left.
This results in bad highlighting of the verbs all over the line width,
instead of being limited to the actual string.
Due to that, sometimes the selection range of the up/down arrows overlaps
with some of the verbs, and then these verbs cannot be selected.
Solve by storing the original left value, and using it as initial x
position for the string (the actual right-to-left manipulation is done in
drawString()), and modify the value of curRect.left to match the string
that was actually drawn.
This bug is similar to the one that was fixed in 58e921eb87, but the
solution that was done there for v7 and v8 cannot work here, because the
string logic is much more complicated.
The value of left is applied from the script all the time, and was
replaced in drawVerb.
The problem with this approach is that redrawVerbs maps the mouse
location to verb index *before* calling drawVerb, so the rectangles it
compares against are invalid in X-axis (right is always _screenWidth -
1, and left is always 5).
This caused several bugs:
* Each verb was clickable all over the screen's width, although it was
not highlighted.
* If the mouse was between 2 verbs, long above short, the Y-axis has 6
overlapping pixels (e.g. 1: 330-360, 2: 354-384). Scanning is done
bottom-up, so 1 was highlighted, but 2 was selected because of the
previous bullet.
* The text on Hebrew was aligned to the right without any padding. Other
languages have left = 5.
Fixed by setting right instead of left when applying the script, and
adjusting left in drawVerb. Other languages are not affected.
Another issue, unrelated to language selection: A verb that was split to
2 lines was clickable all over the screen's width. That's because
curRect.right was set beyond the _screenWidth, and it was not trimmed to
the first (longer) line's length. On Hebrew, curRect.left became
negative. This is also fixed in this commit.
command in v0
If for instance an object necessary for the sentence command is not reachable or pickupable (try to use faucet (object 55) with jar with water in microwave (object 50), the pick-up script of the jar will tell the actor to pickup object 99 (jar not in microwave)) the actor will try to pick-up the jar infinitely.
This is fixed by counting the amount of nested scripts the sentence command has called (directly or indirectly) so far and aborts it if there have been too many.
- handle mode switching correctly
- do not freeze scripts in cutscene mode (mode 0), as some scripts are freezed in mode 0 that should not be freezed
- kModeNoNewKid (mode 2) needs the same userState as mode 3
- rename o_cursorCommand to o_setMode as it is not really cursor specific
- handle actorHiding correctly (do not set costume to 0 as the previous costume cannot be reverted after hiding)
- add drawSentence
- document meanings for actor misc flags
- fix actor names for "new kid" if the radiation suit is used (all kids are set to 0 then with actor 0 name " ")
- cleanup actor switching routine
- _userPut is not used anymore in v0
- MM C64 uses command stack (SentenceTab, doSentence()) now
- _cmdObject... added for current SentenceTab. The _active... variables are only used to build a sentence in the inventory but never by a script.
-> many routines are not needed anymore and are removed
- removed complicated and unnecessary _v0ObjectIndex, _v0ObjectInInventory, _v0ObjectFlag vars
- started to merge object id and type into one object value (type<<8|id)
- verb preposition ids do not dependent on language -> remove from VerbSettings
Note:
- objects with type=0 are foreground objects. They have a state, an owner and a bg overlay image.
- objects with type=1 are bg objects. They do not have a state or owner and are already contained in the bg image. The do not have an entry in objectState/OwnerTable
Note: the transition is not completed yet. The code compiles but is probably not runnable as not every occurrence of _activeInventory has been properly replaced.
The usage of _v0ObjectIndex and _v0ObjectInInventory should be revised too and both variables should be replaced by another mechanism (maybe by using a single variable "obj = (type << 8) | id").
- moved v0 only vars _activeInventory, _activeObject, _activeVerb from ScummEngine_v2 to ScummEngine_v0
- removed _activeActor, _activeInvExecute, _activeObject2Inv and _activeInventory. They are handled by _activeObject/_activeObjectType and _activeObject2/_activeObject2Type now.
- removed _activeObject(2)Index as they only bloat the code without any benefit (?)
- merge prep-name tables from ScummEngine_v2::drawPreposition() and ScummEngine_v0::drawSentenceWord() by introducing ScummEngine_v2::drawPreposition()
- rename ObjectData.flags -> obj_type (quick-fix only, needs review! Maybe obj_nr and obj_type can be merged into one var: obj_nr = (obj_type << 8) | obj_nr)
- o_unknown2 is negation of o_ifActiveObject (o_ifNotEqualActiveObject2)
- renamed o_ifActiveObject -> o_ifEqualActiveObject2 as it acts only on _activeObject2
- renamed ScummEngine_v0::drawSentenceWord() -> ScummEngine_v0::getObjectName()
This tries to make our code a bit more compliant with our code formatting
conventions. For future use, this is the command I used:
git ls-files "*.cpp" "*.h" | xargs sed -i -e 's/[ \t]*$//'
