pipe.c

来自「HTML-XML-utils由一套小型C程序（过滤器）组成」· C语言 代码 · 共 66 行

/*
 * pipe - output HTML/XML in canonical ("sgmls" form
 *
 * Parse HTML/XML and output in approximate "nsgmls" format. Some of
 * the differences are that comments are also printed (see * below),
 * that implied attributes are not, and that entities are left
 * unexpanded. Use "unent" to expand entities to UTF-8.
 *
 * The program doesn't interpret the source in any way, and doesn't
 * read DTDs. That means that, e.g., end tags are not automatically
 * added. Pipe the source through normalize(1) first in order to
 * convert HTML to XML and infer missing tags.
 *
 * The possible command characters and arguments are as follows:
 *
 *     (gi
 *
 *	    The start of an element whose generic identifier is gi.
 *	    Any attributes for this element will have been speci- fied
 *	    with A commands.
 *
 *     )gi
 *
 *	    The end of an element whose generic identifier is gi.
 *
 *     |gi
 *
 *	    An empty element (an element whose tag in the source ended
 *	    with a slash). Any attributes will have been specified
 *	    with A commands. (Note that this distinguishes empty
 *	    elements from elements that happen to have no content,
 *	    even though XML doesn't.)
 *
 *     -data
 *
 *	    Data.
 *
 *     ?pi
 *
 *	    A processing instruction with data pi.
 *
 *     *comment
 *
 *	    A comment
 *
 *     Aname type val
 *
 *	    The next element to start has an attribute name with value
 *	    val and type type. Implied attribute are not shown. All
 *	    attributes are assumed to be of type CDATA, since pipe
 *	    doesn't read DTDs.
 *
 *     !root "fpi" url
 *     !root "fpi"
 *     !root "" url
 *
 *	    A document type declaration. The fpi (public identifier)
 *	    is a quoted string. If there is no fpi, the string is
 *	    empty: "". If there is no url, itis omitted.
 *
 *     Llineno
 *
 *	    Set the current line number. This will be output only if
 *	    the -l option has been given.
 *
 * Copyright