locale.html

unix 下的C开发手册,还用详细的例程。

enclosed between angle brackets
(<
and
>),
and must not duplicate
any symbolic name in the current charmap file (if any), or any other
symbolic name defined in this collation definition.
A
&lt;<i>collating-symbol</i>&gt;
defined via this
keyword is only recognised with the LC_COLLATE category.
<p>
<b>Example</b>:
<code>
<pre>
collating-symbol &lt;UPPER_CASE&gt;
collating-symbol &lt;HIGH&gt;
</code>
</pre>
<p>
The
<b>collating-symbol</b>
keyword defines a symbolic name
that can be associated with a relative position in
the character order sequence.
While such a symbolic
name does not represent any collating element, it
can be used as a weight.
<h5><a name = "tag_005_003_002_003">&nbsp;</a>The order_start Keyword</h5>
The
<b>order_start</b>
keyword must precede collation order entries and
also defines the number of weights for this collation sequence
definition and other collation rules.
<p>
The syntax of the
<b>order_start</b>
keyword is:
<p>
<dl compact><dt> <dd>
<tt>"order_start %s;%s;...;%s\n"</tt>, &lt;<i>sort-rules</i>&gt;,
&lt;<i>sort-rules</i>&gt;
</dl>
</p>
<p>
The operands to the
<b>order_start</b>
keyword are optional.
If present, the operands define rules to be applied when
strings are compared.
The number of operands define how
many weights each element is assigned;
if no operands
are present, one
<b>forward</b>
operand is assumed.
If present, the first operand defines rules to be applied
when comparing strings using the first (primary) weight;
the second when comparing strings using the second weight,
and so on.
Operands are separated by semicolons
(;).
Each operand consists of one or more collation directives,
separated by commas
(,).
If the number of operands exceeds the
{COLL_WEIGHTS_MAX}
limit, the utility will issue a warning message.
The following directives will be supported:
<dl compact>

<dt><b>forward</b><dd>Specifies that comparison operations for the
weight level proceed from start of
string towards the end of string.

<dt><b>backward</b><dd>Specifies that comparison operations for the
weight level proceed from end of
string towards the beginning of string.

<dt><b>position</b><dd>Specifies that comparison operations for the
weight level will consider the relative
position of
elements in the strings not subject to
<b>IGNORE</b>.
The string containing an element not subject to
<b>IGNORE</b>
after the fewest collating elements subject to
<b>IGNORE</b>
from the start of the compare
will collate first.
If both strings contain a character not subject to
<b>IGNORE</b>
in the same relative position, the
collating values assigned to the elements will
determine the ordering.
In case of equality, subsequent characters not subject to
<b>IGNORE</b>
are considered in the same manner.

</dl>
<p>
The directives
<b>forward</b>
and
<b>backward</b>
are mutually exclusive.
<p>
<b>Example</b>:
<code>
<pre>
order_start    forward;backward
</code>
</pre>
<p>
If no operands are specified, a single
<b>forward</b>
operand is assumed.
<p>
The character (and collating element) order is defined by
the order in which characters and elements are specified
between the
<b>order_start</b>
and
<b>order_end</b>
keywords.
This character order is used in range expressions in regular expressions (see
<xref href=re><a href="re.html#tag_007">
Regular Expressions
</a></xref>).
Weights assigned to the characters and
elements define the collation sequence;
in the absence
of weights, the character order is also the collation sequence.
<p>
The
<b>position</b>
keyword provides the capability to
consider, in a compare, the relative position of
characters not subject to
<b>IGNORE</b>.
As an example, consider the two strings &quot;o-ring&quot; and &quot;or-ing&quot;.
Assuming the hyphen is subject to
<b>IGNORE</b>
on the first pass, the two strings will
compare equal, and the position of the hyphen is immaterial.
On second pass, all characters except the hyphen are subject to
<b>IGNORE</b>,
and in the normal case the two strings would again compare equal.
By taking position into account, the first collates before the second.
<h5><a name = "tag_005_003_002_004">&nbsp;</a>Collation Order</h5>
<xref type="5" name="collorder"></xref>
The
<b>order_start</b>
keyword is followed by collating identifier entries.
The syntax for the collating element entries is:
<p>
<dl compact><dt> <dd>
<tt>"%s %s;%s;...;%s\n"</tt>, &lt;<i>collating-identifier</i>&gt;,
&lt;<i>weight</i>&gt;,
&lt;<i>weight</i>&gt;,
...
</dl>
</p>
<p>
Each
<i>collating-identifier</i>
consists of either a character
(in any of the forms defined in
<xref href=locldefn><a href="#tag_005_003">
Locale Definition
</a></xref>),
a
&lt;<i>collating-element</i>&gt;,
a
&lt;<i>collating-symbol</i>&gt;,
an ellipsis or the special symbol
<b>UNDEFINED</b>.
The order in which collating elements are specified
determines the character
order sequence, such that each
collating element compares less than the elements following it.
The NUL character compares lower than any other character.
<p>
A
&lt;<i>collating-element</i>&gt;
is used to specify multi-character
collating elements, and indicates that the character sequence
specified via the
&lt;<i>collating-element</i>&gt;
is to be collated as
a unit and in the relative order specified by its place.
<p>
A
&lt;<i>collating-symbol</i>&gt;
is used to define a position in
the relative order for use in weights.
No weights are specified with a
&lt;<i>collating-symbol</i>&gt;.
<p>
The ellipsis symbol specifies that a sequence of
characters will collate according to their encoded character values.
It is interpreted as indicating that all characters with
a coded character set value higher than the value of the
character in the preceding line, and lower than the
coded character set value for the character in the following
line, in the current coded character set, will be placed in
the character collation order between the previous and the
following character in ascending order according to their coded
character set values.
An initial ellipsis is
interpreted as if the preceding line specified the NUL
character, and a trailing ellipsis as if the following line
specified the highest coded character set value in the
current coded character set.
An ellipsis is treated
as invalid if the preceding or following lines do not
specify characters in the current coded character set.
The use of the ellipsis symbol ties the definition to a
specific coded character set and may preclude the definition
from being portable between implementations.
<p>
The symbol
<b>UNDEFINED</b>
is interpreted as including all coded character set values
not specified explicitly or via the ellipsis symbol.
Such characters are inserted in the character collation
order at the point indicated by the symbol, and in
ascending order according to their coded character set values.
If no
<b>UNDEFINED</b>
symbol is specified, and the current coded character set
contains characters not specified in this section, the
utility will issue a warning message and place such
characters at the end of the character collation order.
<p>
The optional
operands for each collation-element are used to define the primary,
secondary, or subsequent weights for the collating element.
The first operand specifies the relative primary weight, the second
the relative secondary weight, and so on.
Two or more collation-elements can be assigned the same weight;
they belong to the same
if they have the same primary weight.
Collation behaves as if, for each weight level,
elements subject to
<b>IGNORE</b>
are removed,
unless the
<b>position</b>
collation directive is specified for
the corresponding level with the
<b>order_start</b>
keyword.
Then each successive pair of elements is compared according to the
relative weights for the elements.
If the two strings compare
equal, the process is repeated for the next weight level,
up to the limit
{COLL_WEIGHTS_MAX}.
<p>
Weights are expressed as characters (in any of the forms
specified in
<xref href=locldefn><a href="#tag_005_003">
Locale Definition
</a></xref>),
&lt;<i>collating-symbol</i>&gt;s,
&lt;<i>collating-element</i>&gt;s,
an ellipsis, or the special symbol
<b>IGNORE</b>.
A single character, a
&lt;<i>collating-symbol</i>&gt;
or a
&lt;<i>collating-element</i>&gt;
represent
the relative
position in the character collating sequence of the
character or symbol, rather than the character or characters themselves.
Thus, rather than assigning absolute values
to weights, a particular weight is expressed using the relative
order value assigned to a collating element based on its order
in the character collation sequence.
<p>
One-to-many mapping is indicated by specifying two or more
concatenated characters or symbolic names.
For example, if the character
&lt;eszet&gt;
is given the string
&lt;s&gt;&lt;s&gt;
as a weight,
comparisons are performed as if all occurrences of
the character
&lt;eszet&gt;
are replaced by
&lt;s&gt;&lt;s&gt;
(assuming that
&lt;s&gt;
has the collating weight
&lt;s&gt;).
If it is necessary to define
&lt;eszet&gt;
and
&lt;s&gt;&lt;s&gt;
as an equivalence class, then a
collating element must be defined for the string
ss.
<p>
All characters specified via an ellipsis will by default
be assigned unique weights, equal to the relative order of characters.
Characters specified via an explicit or implicit
<b>UNDEFINED</b>
special symbol will by default be assigned the same
primary weight (that is, belong to the same equivalence class).
An ellipsis symbol as a weight is interpreted to mean that
each character in the sequence has unique weights, equal
to the relative order of their character in the character collation sequence.
The use of the ellipsis as a weight is treated
as an error if the collating element is neither an ellipsis
nor the special symbol
<b>UNDEFINED</b>.
<p>
The special keyword
<b>IGNORE</b>
as a weight indicates that
when strings are compared using the weights at the level
where
<b>IGNORE</b>
is specified, the collating element is ignored; that is,
as if the string did not contain the collating element.
In regular expressions and pattern matching, all
characters that are subject to
<b>IGNORE</b>
in their primary weight form
an equivalence class.
<p>
An empty operand is interpreted as the collating element itself.
<p>
For example, the order statement:
<code>
<pre>
&lt;a&gt;    &lt;a&gt;;&lt;a&gt;
</code>
</pre>
is equal to:
<code>
<pre>
&lt;a&gt;
</code>
</pre>
<p>
An ellipsis can be used as an operand if the collating element was
an ellipsis, and is interpreted as the value of each character
defined by the ellipsis.
<p>
The collation order as defined in this section defines the
interpretation of bracket expressions in regular expressions
(see
<xref href=rebrack><a href="re.html#tag_007_003_005">
RE Bracket Expression
</a></xref>).
<br>
<p>
<b>Example</b>:
<pre>
<dl compact><dt> <dd>
<table <tr valign=top><td align=left>order_start
<td align=left>forward;backward
<tr valign=top><td align=left>UNDEFINED
<td align=left>IGNORE;IGNORE
<tr valign=top><td align=left>&lt;LOW&gt;
<tr valign=top><td align=left>&lt;space&gt;
<td align=left>&lt;LOW&gt;;&lt;space&gt;
<tr valign=top><td align=left>...
<td align=left>&lt;LOW&gt;;...
<tr valign=top><td align=left>&lt;a&gt;
<td align=left>&lt;a&gt;;&lt;a&gt;
<tr valign=top><td align=left>&lt;a-acute&gt;
<td align=left>&lt;a&gt;;&lt;a-acute&gt;
<tr valign=top><td align=left>&lt;a-grave&gt;
<td align=left>&lt;a&gt;;&lt;a-grave&gt;
<tr valign=top><td align=left>&lt;A&gt;
<td align=left>&lt;a&gt;;&lt;A&gt;
<tr valign=top><td align=left>&lt;A-acute&gt;
<td align=left>&lt;a&gt;;&lt;A-acute&gt;
<tr valign=top><td align=left>&lt;A-grave&gt;
<td align=left>&lt;a&gt;;&lt;A-grave&gt;
<tr valign=top><td align=left>&lt;ch&gt;
<td align=left>&lt;ch&gt;;&lt;ch&gt;
<tr valign=top><td align=left>&lt;Ch&gt;
<td align=left>&lt;ch&gt;;&lt;Ch&gt;
<tr valign=top><td align=left>&lt;s&gt;
<td align=left>&lt;s&gt;;&lt;s&gt;
<tr valign=top><td align=left>&lt;eszet&gt;
<td align=left>"&lt;s&gt;&lt;s&gt;";"&lt;eszet&gt;&lt;eszet&gt;"
<tr valign=top><td align=left>order_end
</table>
</dl>
</pre>
<p>
This example is interpreted as follows:
<ol>
<p>
<li>
The
<b>UNDEFINED</b>
means that all characters not specified in
this definition (explicitly or via the ellipsis) are
ignored for collation purposes; for regular expression
purposes they are ordered first.
<p>
<li>
All characters between
&lt;space&gt;
and
&lt;a&gt;
have the same primary equivalence class and individual secondary
weights based on