Interesting corner case here. The first page is make up of 8 content streams. The first content stream ends with:
-0.0102 Tc -0.0218 Tw 27.0001 0 0 27.0001 54.0001 724.2301 Tm
[
Where the last byte is the '[' or start array token. The next stream contains the rest of the array declaration:
(T)118(r)18(aditional IRA C)9(ustodial Ac)9(c)9(ount Agr)18(eement)]TJ ...
We have code in the Lexer to move to the next content stream when the end is reach but had an old corner case where if the '[' was the last token we'd just return an empty array. If memory serves me correctly this was for a malformed content stream that didn't have an end array ']'. I've made a small change to the lexer to take into consideration both cases and the PDF renders correctly.
Will need a full QA test as this is a fairly high risk change.
Bug
Major
Patrick Corless
added a comment - Marking as fixed.
Interesting corner case here. The first page is make up of 8 content streams. The first content stream ends with:
-0.0102 Tc -0.0218 Tw 27.0001 0 0 27.0001 54.0001 724.2301 Tm
[
Where the last byte is the '[' or start array token. The next stream contains the rest of the array declaration:
(T)118(r)18(aditional IRA C)9(ustodial Ac)9(c)9(ount Agr)18(eement)]TJ ...
We have code in the Lexer to move to the next content stream when the end is reach but had an old corner case where if the '[' was the last token we'd just return an empty array. If memory serves me correctly this was for a malformed content stream that didn't have an end array ']'. I've made a small change to the lexer to take into consideration both cases and the PDF renders correctly.
Will need a full QA test as this is a fairly high risk change.