W3C home > Mailing lists > Public > public-xml-processing-model-comments@w3.org > October 2011

Re: Charsets, encodings, http-request, unescape-markup, and convenience, oh my!

From: Norman Walsh <ndw@nwalsh.com>
Date: Tue, 11 Oct 2011 10:23:04 -0400
To: public-xml-processing-model-comments@w3.org
Message-ID: <m2d3e3ppyv.fsf@nwalsh.com>
Florent Georges <fgeorges@fgeorges.org> writes:
> On 6 October 2011 21:45, Norman Walsh wrote:
>>      <c:request method="get" href="http://tests.xproc.org/tests/doc/html-utf8.data"/>
>> [...]
>> It returns a base64 encoded document:
>
>   In my humble opinion, I think those problems wouldn't happen if HTML
> content was parsed as a document node directly by the http-request
> step.

I was thinking about an extension to support this when it occurred to me that
it can be done in standard XProc. I think. Anyone see any problems with this:

<p:declare-step xmlns:p="http://www.w3.org/ns/xproc" version="1.0"
		xmlns:c="http://www.w3.org/ns/xproc-step"
                xmlns:l="http://xproc.org/library"
                type="l:http-get">
  <p:output port="result"/>
  <p:option name="href" required="true"/>
  <p:option name="username"/>
  <p:option name="password"/>
  <p:option name="auth-method" select="'Basic'"/>
  <p:option name="send-authorization" select="'false'"/>
  <p:option name="override-content-type"/>

  <p:choose>
    <p:when test="p:value-available('username')">
      <p:template>
        <p:input port="template">
          <p:inline>
            <c:request method="get" href="{$href}" detailed="false" status-only="false"
                       username="{$username}" password="{$password}" auth-method="{$auth-method}"
                       send-authorization="{$send-authorization}"/>
          </p:inline>
        </p:input>
        <p:input port="source"><p:empty/></p:input>
        <p:with-param name="href" select="$href"/>
        <p:with-param name="username" select="$username"/>
        <p:with-param name="password" select="$password"/>
        <p:with-param name="auth-method" select="$auth-method"/>
        <p:with-param name="send-authorization" select="$send-authorization"/>
      </p:template>
    </p:when>
    <p:otherwise>
      <p:template>
        <p:input port="template">
          <p:inline>
            <c:request method="get" href="{$href}" detailed="false" status-only="false"/>
          </p:inline>
        </p:input>
        <p:input port="source"><p:empty/></p:input>
        <p:with-param name="href" select="$href"/>
      </p:template>
    </p:otherwise>
  </p:choose>

  <p:choose>
    <p:when test="p:value-available('override-content-type')">
      <p:add-attribute match="/c:request" attribute-name="override-content-type">
        <p:with-option name="attribute-value" select="$override-content-type"/>
      </p:add-attribute>
    </p:when>
    <p:otherwise>
      <p:identity/>
    </p:otherwise>
  </p:choose>

  <p:http-request/>

  <p:choose>
    <p:when test="starts-with(/c:body/@content-type,'text/html') and /c:body/@encoding = 'base64'
                  and contains(/c:body/@content-type, 'charset')">
      <p:unescape-markup content-type="text/html" encoding="base64"/>
      <p:unwrap match="/c:body"/>
    </p:when>
    <p:when test="starts-with(/c:body/@content-type,'text/html') and /c:body/@encoding = 'base64'">
      <!-- Per RFC 2616, the default charset for text/* types is ISO 8859-1 -->
      <p:unescape-markup content-type="text/html" encoding="base64" charset="iso-8859-1"/>
      <p:unwrap match="/c:body"/>
    </p:when>
    <p:when test="starts-with(/c:body/@content-type,'text/html')">
      <p:unescape-markup content-type="text/html"/>
      <p:unwrap match="/c:body"/>
    </p:when>
    <p:when test="/c:body/@content-type = 'application/json'">
      <p:unescape-markup content-type="application/json" encoding="base64" charset="utf-8"/>
      <p:unwrap match="/c:body"/>
    </p:when>
    <p:otherwise>
      <p:identity/>
    </p:otherwise>
  </p:choose>

</p:declare-step>

                                        Be seeing you,
                                          norm

-- 
Norman Walsh
Lead Engineer
MarkLogic Corporation
Phone: +1 413 624 6676
www.marklogic.com

Received on Tuesday, 11 October 2011 14:23:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 11 October 2011 14:23:52 GMT