在 Debian Jessie GNU/Linux 上重现的步骤。
检查xmllint
版本:
$ xmllint --version
xmllint: using libxml version 20901
compiled with: Threads Tree Output Push Reader Patterns Writer SAXv1 FTP HTTP DTDValid HTML Legacy C14N Catalog XPath XPointer XInclude Iconv ISO8859X Unicode Regexps Automata Expr Schemas Schematron Modules Debug Zlib Lzma
通过将其保存为来制作一个 XHTML 1.0 过渡文件example.xhtml
:
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8"/>
<title>A title</title>
</head>
<body>
Some content
</body>
</html>
注意 将 example.xhtml 的内容粘贴到W3C 验证器会产生“此文档已成功检查为 XHTML 1.0 Transitional!”,因此在使用xmllint
.
xmllint 在线验证
尽管计算机可以访问 Internet,但此操作失败:
$ xmllint --noout --valid example.xhtml
example.xhtml:1: warning: failed to load external entity "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
^
example.xhtml:2: validity error : Validation failed: no DTD found !
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
^
xmllint 离线验证
$ wget -qO- https://www.w3.org/TR/xhtml1/xhtml1.tgz | tar xvz
xhtml1-20020801/
xhtml1-20020801/W3C-REC.css
xhtml1-20020801/xhtml.css
xhtml1-20020801/logo-REC.png
xhtml1-20020801/w3c_home.png
xhtml1-20020801/wcag1AAA.png
xhtml1-20020801/acks.html
xhtml1-20020801/Cover.html
xhtml1-20020801/definitions.html
xhtml1-20020801/diffs.html
xhtml1-20020801/dtds.html
xhtml1-20020801/guidelines.html
xhtml1-20020801/introduction.html
xhtml1-20020801/issues.html
xhtml1-20020801/normative.html
xhtml1-20020801/Overview.html
xhtml1-20020801/prohibitions.html
xhtml1-20020801/references.html
xhtml1-20020801/xhtml1-diff.html
xhtml1-20020801/DTD/
xhtml1-20020801/DTD/xhtml-lat1.ent
xhtml1-20020801/DTD/xhtml-special.ent
xhtml1-20020801/DTD/xhtml-symbol.ent
xhtml1-20020801/DTD/xhtml.soc
xhtml1-20020801/DTD/xhtml1-frameset.dtd
xhtml1-20020801/DTD/xhtml1-strict.dtd
xhtml1-20020801/DTD/xhtml1-transitional.dtd
xhtml1-20020801/DTD/xhtml1.dcl
xhtml1-20020801/xhtml1.ps
xhtml1-20020801/xhtml1.pdf
仍然失败:
$ xmllint --noout --dtdvalid xhtml1-20020801/DTD/xhtml1-transitional.dtd example.xhtml
example.xhtml:1: warning: failed to load external entity "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
^
同样,如果使用--nonet
选项:
$ xmllint --noout --nonet --dtdvalid xhtml1-20020801/DTD/xhtml1-transitional.dtd example.xhtml
I/O error : Attempt to load network entity http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd
example.xhtml:1: warning: failed to load external entity "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"
^
问题
我有两个问题:
- 为什么这些验证尝试都没有成功?
- 第二个似乎失败了,因为尽管使用了该
--dtdvalid
选项,xmllint
但仍然尝试访问http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd
,因为它在example.xhtml
. 有什么方法可以告诉xmllint
忽略该引用,而是使用本地 DTD(例如,已经存储在xhtml1-20020801/DTD/xhtml1-transitional.dtd
?