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
Using the shell script to build from commit 9bc0c2c, I tried to load the page en.wikipedia.org/wiki/Towson_University_buildings_and_structures (from wikitext) and get
I believe the red ellipse should have the save text as the blue ellipse
Not sure how that's happening
The text was updated successfully, but these errors were encountered:
So, this was a weird bug. It's probably been in there for years. The problem was that there was a byte array trim function which sometimes reused the same byte array.
More details are below. As always, thanks for the detailed report!
First, here is the simplest reproduction step
__TOC__
==0123456789012345==
==1234567890123456==
This would print 1234567890123456 twice.
Here is the problem:
The TOC calls a function called bfr.To_bry_and_clear_and_trim for each item which in turn calls Bry_.Trim
This function has logic that trims whitespace from the front and back of the byte array
However, if there is no whitespace, as an optimization, it returns back the same byte array
Due to how the bfr gets initialized, this bug only surfaces when strings are 16 length
The commit fixes the above by selectively skipping over the optimization. A few other files are changed because a new variable was added to the public
Using the shell script to build from commit 9bc0c2c, I tried to load the page en.wikipedia.org/wiki/Towson_University_buildings_and_structures (from wikitext) and get
I believe the red ellipse should have the save text as the blue ellipse
Not sure how that's happening
The text was updated successfully, but these errors were encountered: