Talk:Server Side Includes

Progress

Re-did the first 3 paragraphs, hope that helps somewhat.

Whatever you did, it seems to be effective. I certainly don't see a need for the cleanup template anymore. If anything, it's refreshing how accessible this article is to a layman when compared to some of the other articles on scripting languages. Here's a vote for removing the template. skeeJay 01:37, 25 September 2006 (UTC)

SHTML

SHTML = Scripted HTML using scripting languages via Server Side Includes. Server Side Includes are used to extend the HTML markup language via scripting languages that are executed on the web server. The resultant package of HTML text is delivered to the web client for rendering by the web browser software on the client machine.

SHTML used to contain the above. Edited out. 222.99.205.117 14:22, 17 July 2006 (UTC)

is it just me or whenvever you read that, do you read it as "shit mail"?

Doesn't SHTML just mean "server-parsed HTML"? At least that's how the Apache handler is called. Thomas 01:11, 6 November 2007 (UTC)

The article needs to state that more than just #includes can be accomplished. A complete list of Apache-supported commands would help. 71.220.218.164 (talk) 06:39, 3 August 2010 (UTC)

History

What is the history here? Did Apache introduce the .shtml and <!--#directive parameter=value parameter=value--> syntax? 209.92.136.131 16:50, 17 July 2006 (UTC)

See Apache_HTTP_Server#History_and_name, perhaps the first versions of the mod_include was 1995. In the 1.3 version, released before 2000, all the SSI main features was there. See http://httpd.apache.org/docs/1.3/mod/mod_include.html ... There are also citations about the "Original NCSA HTTPd SSI", then, the first SSI was before 1995. --187.66.93.138 (talk) 19:26, 6 July 2010 (UTC)

Isn't this more or less an obsolete technology, given the rise of complex server-side languages like PHP and so on? Perhaps that should be pointed out. Equinox ◑ 18:03, 1 November 2016 (UTC)

ambiguous wording here

The following sentence is ambiguous to me:

"The file parameter defines the included file as relative to the document path; the virtual parameter defines the included file as relative to the document root."

Does "document" refer to the included file itself or the file that will accept the included file? THis is not clear to the layman. The wording should be either:

"The file parameter defines the included file as relative to the included file path; the virtual parameter defines the included file as relative to the included file root."

or

"The file parameter defines the included file as relative to the path of the document that will receive the included file; the virtual parameter defines the included file as relative to the document root."

Should there be a discussion about configuring the Apache server to enable SSIs?

StevesGroup 21:48, 2 March 2007 (UTC)

Client Side Includes?

Seems counter-intuitive to have what is essentially the opposite of server-side includes as a sub-section of the server-side includes page. Perhaps a separate page should be made for client side includes, or both should be moved to a more general page such as HTML Includes. I'd do this myself right now, but I'd like some feedback before making such a major change. --216.99.114.157 20:16, 21 September 2007 (UTC)

I agree fully, it is irrelevant. If at all then there could be a reference to a more general topic or overview, e.g. "For other technologies pertaining to dynamic web content, see ...". Thomas 01:04, 6 November 2007 (UTC) —Preceding unsigned comment added by Tkoeppe (talk • contribs)

Request removal of section

The section "The conventional view" is written in very poor style (at least for an encyclopedic article) and irrelevant and subjective. I suggest that it be removed, or heavily reworded.

Thomas 01:01, 6 November 2007 (UTC)

"include virtual" vs. "exec cgi" mix up?

This page recommended that you use "include virtual" over "exec cgi" to get access to CGI variables like QUERY_STRING, however, after testing this first-hand in Apache 2.2.8 and 2.2.12, I've found the opposite be the true!

So, I modified the page to reflect my actual first-hand results: recommend "exec cgi" over "include virtual" for access to CGI vars. After doing this, I had another look at Apache's documentation and both 1.3 and 2.2 suggest using "include virtual" over "exec cgi".

This inconsistency may be a bug in Apache, I'm not sure yet. What is the best way to reflect this conflict of information on the wiki? Maybe just say, "try either method, one should work, YMMV"?

Adamrmcd (talk) 21:03, 18 April 2010 (UTC)

Isn't this the same as {{Transcluding}}...?

I think CSIs and SSIs are the same idea as templates in MediaWiki. Is that worth noting? 68.173.113.106 (talk) 21:10, 22 February 2012 (UTC)

Standardization?

Apache, nginx, lighttpd and IIS are the four major web servers that support this language.

Is there a standardization body for SSI that takes care that different implementations don't drift off from a common specification base?`--Abdull (talk) 09:36, 18 April 2012 (UTC)

"Direct"?

The "Most common directives" table says that the include directive has parameters of "include file, direct or virtual" but it doesn't define what "direct" means. — Preceding unsigned comment added by 192.206.154.2 (talk) 02:08, 15 October 2012 (UTC)

Article is outdated

Apache versions 2.4 and later do not accept the syntax <!--#if expr="${Sec_Nav}" --> unless SSILegacyExprParser is set. The new syntax is <!--#if expr="%{Sec_Nav}" --> (% instead of $). See https://httpd.apache.org/docs/current/expr.html . The semantics also appear to have changed: before 2.4, if the variable (Sec_Nav in the above example) was undefined, the effect was the same as if it were the empty string. Now, an undefined variable in that context makes Apache 2.4 barf with the error message:

"syntax error, unexpected $end: Variable 'Sec_Nav' does not exist".

I don't know whether the new Apache syntax is compatible with the other servers because I only have Apache, but someone should check this and update the article. This is a backwards-incompatible change that has bitten many website administrators. Sayitclearly (talk) 09:29, 13 November 2016 (UTC)