[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: About the glibc.pot file.

On Tue, Dec 26, 2000 at 03:37:11PM +0800, Mendel L Chan wrote:
>     I coverted the zh_TW.po of glibc into GB and modified
> some translations in it. But I found some problems there

Good! Please note however, the zh_TW.po was based on a very early glibc 2.2
beta. Part of it is seriously out of sync with final 2.2 release. You probably
want to run "msgmerge" with the latest source, and then remove the obsolete
entries and translated the new ones.

> In file malloc/mcheck.c
>    msgid "memory clobbered before allocated block\n"
>    I'm not sure the translation in zh_TW.po and it made me confused. :(
> We need to discuss. I read the source and I think it mean system
> finds a fetal error when checking memory, and the error locates before
> the allocated block. Maybe I made mistakes. 

"clobbered" here means "overwritten" or "modified". So the message says:
"data located in memory addresses immediately before the allocated block 
have been overwritten or modified". In another words, mcheck has found 
illegal memory access trying to write to out-of-boundary memory addresses.
This is one of the common mistakes when dealing with dynamically allocated

> In file locale/programs/charmap.c:
> msgid "upper limit in range is not higher then lower limit"
> It's "then" but not "than". Is it a type? How about asking GCC Team?
> I just subscribe the gcc-bugs maillist for this reason. If someone 
> has done that, would you like to ask that? Thank you.

I guaranty you it's a typo!

> In file elf/dl-reloc.c:
> msgid "can't restore segment prot after reloc"
> The "prot" here is a problem. Is it "port" or "prot"? A typo or not.

"prot" is NOT a typo! Original zh_TW.po translation was mistaken on this.
"prot" is an abbreviation for "protection mode". So the message says:
"can not restore original segment protection mode after relocation."

I read through the first few hundred lines of your translation. The followings
are some points I want to raise for discussion. The indented parts are my
comments. They are simply my personal opinions. You and others might not agree.

#: nis/nis_print.c:294
msgid "\tEntry data of type %s\n"
msgstr "\t蝐餃?銝?%s ???????娉n"

  Here "entry" roughly means "item". So I think the original zh_TW.po
  translation 憿寧? is much better a choice. Most (if not all) of the
  appearances of the word "entry" in this po file have the same meaning.

#: nscd/nscd_stat.c:154
"%s cache:\n"
"%s 蝻??:\n"

    I'd prefer "蝻??" for all the occurrences of "cache".

"%15ld  cache hits on positive entries\n"
"%15ld  cache misses on positive entries\n"
"%15ld  甇????葉?曉?蝻???豹n"
"%15ld  甇????葉???????脫?\n"

    Other than "entry" and "cache" the two problems mentioned above, it
    also reversed object and subject. The translation should be:

#: timezone/zic.c:984
msgid "\"Zone %s\" line and -l option are mutually exclusive
msgstr "\"?箸挾 %s\" 銵?? -l ??★?臭??亦?"

    All the occurrences of "zone" in timezone/*.c should be translated to
    "?嗅?" instead of "?箸挾".

#: timezone/zic.c:2229
msgid "%s: %d did not sign extend correctly\n"
msgstr "%s: %d ???甇?&?啣辣撅?縑?愧n"

    "sign" != "signal"! The original po translation was incorrect. 
    Here "sign" has the same meaning as in "plus/minus sign".

Best regard,

| This message was re-posted from debian-chinese-gb@lists.debian.org
| and converted from gb2312 to big5 by an automatic gateway.

Reply to: