Back to bug 1961109

Who When What Removed Added
Red Hat Bugzilla 2021-05-17 09:34:17 UTC Pool ID sst_platform_tools_rhel_8
Red Hat One Jira (issues.redhat.com) 2021-05-17 09:37:04 UTC Link ID Red Hat Issue Tracker RHELPLAN-79603
Tulio Magno Quites Machado Filho 2021-05-17 20:06:34 UTC CC tulioqm
Florian Weimer 2021-05-18 04:24:21 UTC Link ID Red Hat Bugzilla 1591268
Florian Weimer 2021-05-18 04:32:21 UTC Link ID Sourceware 27882
Florian Weimer 2021-05-18 05:35:04 UTC Keywords Bugfix, Patch, Triaged
Doc Text Cause:
The IBM256, IBM277, IBM278, IBM280, IBM284, IBM297, IBM424 character sets encoded the EBCDIC codepoint 0xBC as the Unicode character U+203E OVERLINE.

Consequence:
When using the `iconv' program provided by glibc, converting text in those cahracter sets which contain the 0xBC codepoint fails for non-Unicode character sets such as ISO-8859-1 because they cannot encode the U+203E OVERLINE character.

Fix:
The mapping for the 0xBC codepoint for those IBM character sets is changed to U+00AF MACRON.

Result:
Input in the IBM277, IBM278, IBM280, IBM284, IBM297 character sets can now be converted to ISO-8859-1 in all cases. For the IBM256 and IBM424 character sets, conversion no longer fails if the input text contains the 0xBC codepoint, and U+00AF MACRON is used in the output.
Doc Type If docs needed, set a value Bug Fix
Denis Volkov 2021-05-24 07:37:37 UTC Flags needinfo?(fweimer)
Florian Weimer 2021-05-27 12:13:14 UTC Flags needinfo?(fweimer)
Carlos O'Donell 2021-06-18 13:17:15 UTC Flags needinfo?(dvolkov)
Denis Volkov 2021-06-23 07:26:15 UTC Flags needinfo?(codonell)
Denis Volkov 2021-07-08 07:32:27 UTC Flags needinfo?(codonell)
Florian Weimer 2021-07-08 07:37:35 UTC Flags needinfo?(codonell) needinfo?(codonell)
Denis Volkov 2021-08-17 09:03:09 UTC Flags needinfo?(dvolkov)
Carlos O'Donell 2021-08-17 21:25:25 UTC Resolution --- UPSTREAM
Status NEW CLOSED
Last Closed 2021-08-17 21:25:25 UTC
Carlos O'Donell 2021-08-17 21:26:19 UTC Summary iconv: missing macron (unicode 0xAF) in EBCDIC-CP-ES (IBM284) glibc: iconv: missing macron (unicode 0xAF) in EBCDIC-CP-ES (IBM284)
Pavel Najman 2021-09-17 12:24:26 UTC Pool ID sst_platform_tools_rhel_8 sst_pt_pcp_rhel_8
Pavel Najman 2021-09-17 12:34:06 UTC Pool ID sst_pt_pcp_rhel_8 sst_pt_gcc_glibc_rhel_8
Florian Weimer 2022-05-05 12:26:44 UTC Resolution UPSTREAM ---
CC dbodnarc
Status CLOSED NEW
Keywords Reopened
Keywords Reopened
Martin Cermak 2022-05-05 12:30:07 UTC QA Contact qe-baseos-tools-bugs skolosov
Carlos O'Donell 2022-05-05 14:32:20 UTC Flags needinfo?(fweimer)
Severity unspecified high
Priority unspecified high
Flags needinfo?(dvolkov)
Carlos O'Donell 2022-05-06 13:48:31 UTC Flags needinfo?(fweimer)
Flags needinfo?(dvolkov)
CC alanm
Flags needinfo?(alanm)
Assignee glibc-bugzilla ashankar
Martin Coufal 2022-05-10 15:10:31 UTC CC mcoufal
Martin Coufal 2022-05-10 15:12:28 UTC QA Contact skolosov mcoufal
CC skolosov
Arjun Shankar 2022-05-10 18:18:57 UTC Fixed In Version glibc-2.28-201.el8
Status NEW MODIFIED
Alan Matsuoka 2022-05-11 13:41:56 UTC Flags needinfo?(alanm)
RHEL Program Management Team 2022-05-12 12:39:26 UTC Blocks 2084564
RHEL Program Management Team 2022-05-12 12:39:30 UTC Keywords ZStream
errata-xmlrpc 2022-05-17 09:34:15 UTC Status MODIFIED ON_QA
Martin Coufal 2022-05-26 12:42:05 UTC Status ON_QA VERIFIED
Jacob Taylor Valdez 2022-10-05 09:34:45 UTC CC jvaldez
Docs Contact jvaldez
Jacob Taylor Valdez 2022-10-05 10:58:03 UTC Doc Text Cause:
The IBM256, IBM277, IBM278, IBM280, IBM284, IBM297, IBM424 character sets encoded the EBCDIC codepoint 0xBC as the Unicode character U+203E OVERLINE.

Consequence:
When using the `iconv' program provided by glibc, converting text in those cahracter sets which contain the 0xBC codepoint fails for non-Unicode character sets such as ISO-8859-1 because they cannot encode the U+203E OVERLINE character.

Fix:
The mapping for the 0xBC codepoint for those IBM character sets is changed to U+00AF MACRON.

Result:
Input in the IBM277, IBM278, IBM280, IBM284, IBM297 character sets can now be converted to ISO-8859-1 in all cases. For the IBM256 and IBM424 character sets, conversion no longer fails if the input text contains the 0xBC codepoint, and U+00AF MACRON is used in the output.
.The mapping for the `0xBC` code point for some IBM character sets is now `U+00AF MACRON`

Previously, the `IBM256`, `IBM277`, `IBM278`, `IBM280`, `IBM284`, `IBM297`, and `IBM424` character sets encoded the `EBCDIC` code point `0xBC` as the Unicode character `U+203E OVERLINE`. As a result, when using the `iconv` program provided by `glibc`, converting text in those character sets containing the `0xBC` code point failed for non-Unicode character sets such as `ISO-8859-1` because they could not encode the `U+203E OVERLINE` character.

With this update, the bug has been fixed. As a result, input in the `IBM277`, `IBM278`, `IBM280`, `IBM284`, and `IBM297` character sets can be converted to `ISO-8859-1` in all cases. For the `IBM256` and `IBM424` character sets, conversion no longer fails if the input text contains the 0xBC code point and the respective output is `U+00AF MACRON`.
Flags needinfo?(ashankar)
Arjun Shankar 2022-10-23 21:36:23 UTC Flags needinfo?(ashankar)
errata-xmlrpc 2022-11-08 00:18:34 UTC Status VERIFIED RELEASE_PENDING
errata-xmlrpc 2022-11-08 10:43:11 UTC Resolution --- ERRATA
Status RELEASE_PENDING CLOSED
Last Closed 2021-08-17 21:25:25 UTC 2022-11-08 10:43:11 UTC
errata-xmlrpc 2022-11-08 10:43:28 UTC Link ID Red Hat Product Errata RHBA-2022:7684
Mark O'Brien 2023-07-18 14:30:35 UTC Pool ID sst_pt_glibc_rhel_8 sst_pt_libraries_rhel_8

Back to bug 1961109