You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
... and of course, mapnik::value_integer is defined in mapnik as a long long (2.2.0, 2.3.x)
I've found some advice from rice mailing list posts and docs that suggest we need to explicitly convert the fixnum into a long long, but I don't know actually how to do that i.e. what code to put where. Within _mapnik_colour.cpp there is an example of using NUM2INT, which might provide a clue, but I don't know where to go from here.
Suggestions welcome!
The text was updated successfully, but these errors were encountered:
I'm not sure how to help ruby work directly with long long. But we can fake it with int which I've done in 39e76e0 and that gets the test passing. The problem will be if a feature id value is greater than the size of a 32 bit integer - which is unlikely but feasible with OSM data. In that case the value would overflow... But I committed 39e76e0 knowing that defect because I can't see off hand how to work with long long properly.
When running the tests we receive the error:
The test is straighforward:
There are four identical failing tests, unsurprisingly! The constructor takes a value_integer type:
... and value_integer is defined based on the mapnik_version:
... and of course,
mapnik::value_integer
is defined in mapnik as along long
(2.2.0, 2.3.x)I've found some advice from rice mailing list posts and docs that suggest we need to explicitly convert the fixnum into a long long, but I don't know actually how to do that i.e. what code to put where. Within _mapnik_colour.cpp there is an example of using NUM2INT, which might provide a clue, but I don't know where to go from here.
Suggestions welcome!
The text was updated successfully, but these errors were encountered: