git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH 0/8] character encoding portability updates

From
Brandon Casey <casey@nrlssc.navy.mil>
Date
May 18, 2009, 23:44 UTC
Message-ID
<KfeKNR3Jm6YKvPIcYGBMpCbbOxxIaBRphS4qzgN2W9mXIvp7Hl8SgBQzbAhZRV4A4Q6X14JTLR4@cipher.nrlssc.navy.mil>

I got a clue from Alex Riesen's "Use UTF-8 instead of utf8 for backward compatibility" commit 39d404d1 that some of the iconv issues on Solaris and IRIX could be caused by character code naming issues. It seems that these platforms do not understand all of the alternate names that modern platforms do.

Character encodings are not a strong point of mine, so do comment if you see something wrong.

In particular, notice 4/8 that changes the default fallback encoding to ISO8859-1 instead of latin1. My understanding is that these two are equivalent.

With this series applied, Solaris 10 must skip only 4 tests for me (plus two more if I use Korn shell). The remaining ones that are skipped have to do with iconv but the character code conversions being tested are not installed.

Solaris 7 must skip 30 (32 with Korn shell), down from 52 (54), and IRIX 6.5 must skip only 13, down from 44.

-brandon
Brandon Casey (8):
  t3900: use ancient iconv names for backward compatibility
  Use 'UTF-8' rather than 'utf-8' everywhere for backward compatibility
  builtin-mailinfo.c: compare character encodings case insensitively
  builtin-mailinfo.c: use "ISO8859-1" instead of "latin1" as fallback
    encoding
  t3901: avoid negation on right hand side of '|'
  t3901: Use ISO8859-1 instead of ISO-8859-1 for backward compatibility
  t9301: use ISO8859-1 rather than ISO-8859-1
  t5100: use ancient encoding syntax for backwards compatibility
 builtin-log.c                             |    2 +-
 builtin-mailinfo.c                        |   10 +++----
 builtin-revert.c                          |    4 +-
 pretty.c                                  |    4 +-
 t/t3900-i18n-commit.sh                    |   28 ++++++++++----------
 t/t3900/{ISO-8859-1.txt => ISO8859-1.txt} |    0
 t/t3900/{EUCJP.txt => eucJP.txt}          |    0
 t/t3901-i18n-patch.sh                     |   40 ++++++++++++++--------------
 t/t5100/rfc2047-samples.mbox              |   32 +++++++++++-----------
 t/t5100/sample.mbox                       |    8 +++---
 t/t9129-git-svn-i18n-commitencoding.sh    |   10 +++---
 t/t9301-fast-export.sh                    |    2 +-
 t/t9500-gitweb-standalone-no-errors.sh    |    2 +-
 13 files changed, 70 insertions(+), 72 deletions(-)
 rename t/t3900/{ISO-8859-1.txt => ISO8859-1.txt} (100%)
 rename t/t3900/{EUCJP.txt => eucJP.txt} (100%)
Next: Brandon Casey
Message 1 of 9 in “character encoding portability updates”
  1. 0/8 character encoding portability updatesBrandon Casey, May 18, 2009
  2. 1/8 t3900: use ancient iconv names for backward compatibilityBrandon Casey, May 18, 2009
  3. 2/8 Use 'UTF-8' rather than 'utf-8' everywhere for backward compatibilityBrandon Casey, May 18, 2009
  4. 3/8 builtin-mailinfo.c: compare character encodings case insensitivelyBrandon Casey, May 18, 2009
  5. 4/8 builtin-mailinfo.c: use "ISO8859-1" instead of "latin1" as fallback encodingBrandon Casey, May 18, 2009
  6. 5/8 t3901: avoid negation on right hand side of '|'Brandon Casey, May 18, 2009
  7. 6/8 t3901: Use ISO8859-1 instead of ISO-8859-1 for backward compatibilityBrandon Casey, May 18, 2009
  8. 7/8 t9301: use ISO8859-1 rather than ISO-8859-1Brandon Casey, May 18, 2009
  9. 8/8 t5100: use ancient encoding syntax for backwards compatibilityBrandon Casey, May 18, 2009

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.