The mainland China partial users are possibly unable to visit the Vicky hundred branches. If you can succeed the browsing, welcome toHereReport.
At present the numerous □□□□breakdown, □sends □multipurpose □, □□, the geographic name and so on □□□□, □each □□, □and so on awaits repairs □.

Unicode □□□

Table of contents

Unicode □□□ (Unicode-at-on) Is Taiwan □ □□user □□ is big five □ (Big-5) from appears on the market by □, the character □uses □□(□to lack □solution □plan which character □□) □takes (□to look big five □# shades □), enables □the people □□□ChineseOrDate □□station, □□□□□the writing, is not intact □□in □"Unicode □□□" □□□(□字元 □□looked字元 □source) . It is one □Free □□.

Must pay attention is"Unicode □□□" was not equal to Unicode. □you looked □some people □"my peaceful □Unicode", usually was he "□□□and Unicode does Unicode" mixes.

"Unicode □□□" is not □□shows with □Xie?? □□□□. □□□must have □the glyph (for example supports the entire □Unicode □character glyph), can in □□□telltaleSees to. Because □"Unicode □□□" has contained □□□□the table only, □not including glyph in □. But some days □□□□in Windows XP □□□□□, □useMicrosoft AppLocale□□□□and so on the formula makes □□□.

Background □principle

"Unicode □□□" the principle is is big five □makes the character □ □position, with Unicode □字元 does □to □□, by □to □□is borrowingOutside character collectionAlso can cause the big five □documents or □a use is short of □the character originally.

Because big five □□receives □13,,053 □□characters, □many users say □□are insufficient □, for example date □ fictitious name, the personal name, the branch □use the special character and so on all owes presents. □for a long time the □□□way all adds □each □outside character collection by □Xie??, for example□the flower □enters the law(Support date □fictitious name),Center □sea character collectionAnd so on. But the present world tidal current is by uses contains most characters □Unicode □item □. Outside the character collection fundamentally receives □the character □by to make □□the material to hand over □using, only if □side also peaceful □outside character collection.

In □□□□under, does □is □in the character □table, is big five □makes the character □is and "Unicode makes the character □" does □to □□, also is □□□□□takes to some □originally is falls makes the character □□□□, □□□is going □to take □its □□Unicode to make character □字元. □the fruit is, because is not each human of □□□Unicode makes the character □all to use identically makes the character □document, therefore □delivers the outside character collection with the character □other people □, □Fang Re? the peaceful □same outside character collection, cannot see □accommodates to □the surface.

"Unicode □□□" □□by revises does □is □the character □table way by Xie?? □. It is big five □makes character □字元 □□to □Unicode □□. □makes, "Unicode which the character is different □□□" □□字元 maintained □the counter current general characteristic. In □after □□on, □□字元 □is big five □□□□saves after Unicode, it □all □□□to □Unicode position in, afterwards even if is □in □has An?? the □□□□user, so long as his department □and the formula supports Unicode, in □takes □a writing □, completely □has □□.

Makes □the platform

"Unicode □□□" the revision does □is □the center character □table, □manages Unicode and the non- Unicode character □□□. "Unicode □□□" first is by a Unicode □□coreMicro □ Windows NT (includes Windows 2,000 sum Windows XP) does □is □on □□, afterwards has supported by ANSI □□core Windows 98 sum Windows Me.

Linux also has other user, □□Linux version □□□.

Palm also has other user, □□□□□□□:

Unicode □Windows

IncludingWindows 2,000、Windows XP、Windows Server 2,003, □Windows series.

Revision □document:

  • $sysdir\c_950.Nls Unicode <- -> Big-5 □according to table
  • If the user has peaceful □Microsoft Applocale, peaceful □formula □□$WINDIR\AppPatch\AppLoc.Tmp by a □of the same name blank only □□document substitution.
  • If the user □has peaceful □Microsoft Applocale, the peaceful □formula □is direct □lives a □blank only □□the document: $windir\apppatch\apploc.Tmp.

ANSI □Windows

IncludingWindows 98、Windows ME.

Revision □document:

  • $sysdir\cp_950.Nls IE (□document □tube) □shows □the use □□the table
  • $sysdir\unicode.Bin □□with □the document is □the operation □□□table (if you have □the interest to be possible □□to play: $sysdir\unicode.Bin has chopped, after is heavy □□, is □□uses most primitive □□CodePage:437 □shows, all □□shows in □sentiment □under all existences in □the name Chinese character by "__", □including major part tabletop Czechoslovakia □and □entire □"□beginning" 功能表)
  • $sysdir\gdi.Exe in □shows the font □, obtains the font □the document (□CodePage □□□□to contain □□table according to this □□to become unicode □, again □□center withdraws font by this unicode to the TrueType font)

□: $sysdir is represents someLu?One □□□, in Windows XP □□is C:\WINDOWS\system32, in Windows 2,000 □□is C:\WINNT\system32, in Windows 98 and Windows ME □□is C:\WINDOWS\system.

$windir also is □Lu?? □, in Windows 98, Windows ME and Windows XP □□is C:\WINDOWS, Windows 2,000 □□is C:\WINNT.

字元 □source

In "Unicode □□□" in 2nd edition, 字元 the basic □source is □illuminates "Center □sea character collection" In addition center □the sea character collection institute □leaks□□ChineseDate □The character but becomes.

"Unicode □□□" 2.40 Alpha 3 edition, besides big five □original Fu? and □the character, altogether received □4,,916 □□characters and □the character lateral part of character, the date □ half shape及全形假名、俄語西里爾字母等,涵蓋了在GB 2312(不是 GBK 或 GB 18030)、Shift_JIS之中出現的所有漢字,和香港增補字符集之中,Unicode碼落在 U+4E00~U+9FA5 (即 Unicode 1.1版定義) 的漢字。(因編碼空間不足的關係,並不收錄在香港增補字符集的 Unicode 擴充漢字)

使用上的問題

「Unicode補完計畫」原先的立意是避免利用造字,以達成擴充Big-5的目的:但由於Unicode環境尚未成熟,以及使用者的誤用,「Unicode補完計畫」有時反而為使用者本身──甚至其他使用者──帶來了其他的麻煩。

網頁交換

這是「Unicode補完計畫」使用者可能會影響到其他使用者的最大問題。

一般的網頁瀏覽器電子郵件客戶端,在使用者打出了非該軟體顯示畫面預設編碼(例如寫一封用 Big-5 編碼的信件)的字元時,軟體會自動把這些字元轉換成Unicode 參照碼,例如「堃」會被自動轉換成「堃」;然而在補完後的電腦,由於即使打出的是原非大五碼預設的字元,也會被認為是大五碼的字元(裝了「Unicode補完計畫」的系統,在對照字碼表後,發現當中有這個字),因此就不會被轉換了。這也就是說,其他使用者不一定能看見該使用者所打的字──除非他們也有裝「Unicode補完計畫」。於是這反而與「Unicode補完計畫」的創立宗旨背道而馳:原先避免以外字集解決缺字的「Unicode補完計畫」,反而變成了另一套外字集。

解決方案:當使用者安裝「Unicode補完計畫」時,會獲得一份「HTML外字相容轉換器」(另外也有網路版),可以直接把原本不屬於大五碼的字元轉換為參照碼;接下來只要使用這份夾雜着參照碼的文件,就能讓其他使用者也見到這些字元。另一種做法是,像推廣櫻花輸入法一樣推廣「Unicode補完計畫」,使其也變成一種人人使用的非官方標準。然而最終的解決方案是放棄Big-5,將網頁或信件直接改用Unicode編碼。

註:但是在原本就只有ANSI的網路環境下──例如Telnet BBS──Unicode補完計畫還是只能當外字集用。

此時使用者可以選擇有內建特殊字碼表的Telnet程式,如PieTTY,即無須安裝「Unicode補完計畫」。

新舊檔名

這是原櫻花輸入法使用者會面臨的問題。

在未安裝「Unicode補完計畫」的電腦,當使用者使用櫻花輸入法為檔案命名時,儲存在電腦內的檔名雖然是 Unicode 編碼,卻是在「Unicode造字區」內的字碼;而補完後的電腦,由於字碼表已被修改,這些檔案名稱在 Unicode 架構的程式的檢視下,就會變成空白;而對於 ANSI 架構的程式(例如 ACDSee、Winamp 等),甚至會變成無法存取的亂碼。這就表示甚至連 ANSI 架構的作業系統(例如整個 Windows 98)都無法存取該檔案。另外還有一個問題:若電腦是未安裝「Unicode補完計畫」的 ANSI 架構作業系統,以上的狀況就會剛好相反(以 Unicode 架構的作業系統則無此問題)。

解決方案:Unicode補完計畫內附一個檔名轉換程式,可以把造字區檔名和 Unicode 檔名互換,但僅能在 Unicode 架構的作業系統下使用;ANSI 架構作業系統的使用者必須要手動改檔名。另一個比較有效率的作法:如果是雙系統的使用者,可以直接在 Unicode 架構的作業系統下使用改檔名程式,讓 ANSI 架構的作業系統使用。

其他問題

  • Unicode補完計畫的編碼不相容於香港增補字符集的編碼,兩者只能擇其一。簡單來說,即使Unicode補完計畫與香港增補字符集均有某一個字,但因這個字在大五碼中的位置不相同,故那個字只能在 Unicode 的環境下交換,不能直接透過大五碼交換。
  • Unicode補完計畫會使用到使用者造字區;也就是說如果使用者有自造字,這些字可能會不見。
  • 有人表示補完計畫會讓 Internet Explorer 的自動選擇網頁編碼準確度下降,但是無法證實。
  • 在安裝補完計畫後,Microsoft FrontPage 2003 會運作不正常,但是如果 FrontPage 2003 是以 Unicode 儲存網頁,就能正常運作。
  • 對於不是使用系統字碼表的軟體──例如 Mozilla Firefox 等跨平台瀏覽器或 Java 軟體──補完計畫會無效,這些軟體需要「個別補完」(例如有些私編版的 Firefox,就有將補完後的字碼表編譯進去)。
  • Windows 98 的細明體字形 (mingliu.ttc) 比起 Windows Me / 2000 / XP 的版本來得舊。舊版字形有不少缺字情況;使用者必須在新版 Windows 中,複製細明體去更新 Windows 98 的字形檔案。
  • Windows 98 / Me 的使用者需要再安裝「中國海字集」,才能正常顯示純文字檔案。
  • 由於 Windows XP Service Pack 2 改進了系統檔案保護的能力,而補完計畫需要變更一個系統檔案,故當出現一個視窗,詢問是否要還原此系統檔案時,使用者必須選擇「不還原」此檔案,才能安裝成功。

參看

其他條目

參考文件

外部連結

 

  > Chinese to English > zh.wikipedia.org (Machine translated into English)