Ruby Issue Tracking System

Today

08:21 AM Ruby Revision 9f3a5c22 (git): YJIT: Use rb_len_t instead of long for lengths

The length fields of String and Array are rb_len_t, which is not necessarily long. The Fixnum range checks stay based on long. The bindings still declare rb_len_t as c_long, so the build now fails if rb_len_t gets wider without them, ins... hsbt (Hiroshi SHIBATA)

08:21 AM Ruby Revision 05917fd1 (git): Return rb_len_t from rb_RSTRING_LEN

unsigned long can be narrower than rb_len_t, which would truncate the length that YJIT and ZJIT read through this function.
https://bugs.ruby-lang.org/issues/22400
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision 9511e08d (git): Compute the size of Array#repeated_permutation with rb_int_pow

rb_int_positive_pow takes the base as long and the exponent as unsigned long, so read the count with NUM2LEN and raise the Integer instead.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision 4ef08fc1 (git): Use rb_len_t for finalizer counts and tables in the GCs

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision db8a99e7 (git): Use rb_len_t for the argument of long2int in rb_parser_config_t

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision c9dd247d (git): Use rb_len_t in the bulk Hash builders and the ZJIT helpers

The ZJIT bindings are regenerated with make zjit-bindgen.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision d2ec641f (git): Use rb_len_t for String and Array lengths in ext/socket

The lengths passed to recv(2), send(2) and gethostname(2), and the lengths of sockaddr, are left as they are.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision 6b3b0b80 (git): Use rb_len_t for String and Array lengths in ext/coverage and ext/objspace

ObjectSpace.dump writes the lengths with dump_append_sizet, because dump_append_ld takes a long and is also used for Fixnum values. The lengths are never negative, so the output does not change.
Co-Authored-By: Claude Opus 5.5 <noreply@... hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision e6b82f03 (git): Use rb_len_t for lengths in the C API test extensions

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)

07:45 AM Ruby Revision 939f20c7 (git): Use rb_len_t for String lengths in cgi/escape

It falls back to long on Ruby versions that do not define HAVE_RB_LEN_T, so the change also builds there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> hsbt (Hiroshi SHIBATA)