ifhub/inform7
1
0
Fork 0
mirror of https://github.com/ganelson/inform.git synced 2024-07-03 07:24:58 +03:00
Code Issues Projects Releases Wiki Activity

Fix and test case for Mantis bug 1978

Browse source
This commit is contained in:
Graham Nelson 2022-04-10 16:09:40 +01:00
parent 76ae1a9083
commit 06272f08e0
3 changed files with 45 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
inform7/Tests/Test Problems/PM_AmbiguitiesTooDeep.txt Normal file
View file

@ -0,0 +1,3 @@
Foo is a room.
When play begins: if a room encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room that encloses a room, do nothing.

23
inform7/Tests/Test Problems/_Results_Ideal/PM_AmbiguitiesTooDeep.txt Normal file
View file

@ -0,0 +1,23 @@
Inform 7 v10.1.0 has started.
I've now read your source text, which is 55 words long.
I've also read Basic Inform by Graham Nelson, which is 7691 words long.
I've also read English Language by Graham Nelson, which is 2328 words long.
I've also read Standard Rules by Graham Nelson, which is 32092 words long.
Problem__ PM_AmbiguitiesTooDeep
>--> The phrase a room encloses a room that encloses a room that encloses a
room that encloses a room that encloses a room that encloses a room that
encloses a room that encloses a room that encloses a room that encloses a
room that encloses a room is too complicated for me to disentangle without
running very, very slowly as I check many ambiguities in it. There ought to
be some way to simplify things for me?
I was trying to match this phrase:
if (a room encloses a room that encloses a room that encloses a room that
encloses a room that encloses a room that encloses a room that encloses a
room that encloses a room that encloses a room that encloses a room that
encloses a room - a condition):
I recognised:
a room encloses a room that encloses a room that encloses a room that
encloses a room that encloses a room that encloses a room that encloses a
room that encloses a room that encloses a room that encloses a room that
encloses a room = a condition
Inform 7 has finished.

20
inform7/values-module/Chapter 5/Dash.w
View file

@ -828,8 +828,26 @@ later rules can override earlier ones but still make use of them.
for (int i=0; i<no_of_possible_readings; i++)
LOGIF(MATCHING, "Possibility (P%d) $e\n", i, list_of_possible_readings[i]);
@ In general, it's not great for typecheckers in compilers to put an upper bound
on complexity, because although human-written code seldom hits such maxima, there's
always the possibility of mechanically-generated code which does. On the other hand,
the result of that doctrine is that a lot of modern compilers (Swift, for example)
slow to a painful crawl and allocate gigabytes of memory trying to understand
strange type constraints in two or three lines of code. So, for now at least,
let's be pragmatic.
@<Add this reading to the list of test cases@> =
if (no_of_possible_readings >= MAX_INVOCATIONS_PER_PHRASE) internal_error("overrun");
if (no_of_possible_readings >= MAX_INVOCATIONS_PER_PHRASE) {
THIS_IS_AN_ORDINARY_PROBLEM;
Problems::quote_wording(1, Node::get_text(p));
StandardProblems::handmade_problem(Task::syntax_tree(), _p_(PM_AmbiguitiesTooDeep));
Problems::issue_problem_segment(
"The phrase %1 is too complicated for me to disentangle without "
"running very, very slowly as I check many ambiguities in it. There "
"ought to be some way to simplify things for me?");
Problems::issue_problem_end();
return NEVER_MATCH;
}
list_of_possible_readings[no_of_possible_readings++] = alt;
Dash::clear_flags(alt);