aboutsummaryrefslogtreecommitdiffstats
path: root/README-maintainer
diff options
context:
space:
mode:
authorJay Berkenbilt <ejb@ql.org>2022-10-06 14:05:48 +0200
committerJay Berkenbilt <ejb@ql.org>2022-10-06 14:40:08 +0200
commit5c5b4e640e2df248192ff7aa518fb0d345884a33 (patch)
tree1f2a331ee8f0aa8cba685b0192d093238a919e7e /README-maintainer
parentdb6598b449f4edbb69a581ecc8bc6791e24ad3ea (diff)
downloadqpdf-5c5b4e640e2df248192ff7aa518fb0d345884a33.tar.zst
Be more conservative about QPDF_DLL_CLASS with mingw (fixes #799)
* Define it even though previous experiments have shown it to be unnecessary since it seems like it may be necessary after all * Add QPDF_DLL_CLASS to QPDFObjectHelper and QPDFDocumentHelper in case there's some future unknown reason why someone may want to have them and/or in case it helps with the weird QPDFNameTreeObjectHelper problem.
Diffstat (limited to 'README-maintainer')
-rw-r--r--README-maintainer6
1 files changed, 4 insertions, 2 deletions
diff --git a/README-maintainer b/README-maintainer
index 545612c4..3658a347 100644
--- a/README-maintainer
+++ b/README-maintainer
@@ -179,9 +179,11 @@ CODING RULES
the shared object boundary (or "shared library boundary" -- we may
use either term in comments and documentation). In particular,
anything new derived from Pipeline or InputSource should be marked
- with QPDF_DLL_CLASS, but we don't need to do it for QPDFObjectHelper
+ with QPDF_DLL_CLASS. We shouldn't need to do it for QPDFObjectHelper
or QPDFDocumentHelper subclasses since there's no reason to use
- dynamic_cast with those.
+ dynamic_cast with those, but doing it anyway may help with some
+ strange cases for mingw or with some code generators that may
+ systematically do this for other reasons.
IMPORTANT NOTE ABOUT QPDF_DLL_CLASS: On mingw, the vtable for a
class with some virtual methods and no pure virtual methods seems