Finding unused selectors?

[Rhino]

What's the simplest way to determine which, if any, of my selectors are not
needed in a given stylesheet?

I have a small number of HTML pages that share two stylesheets; one
stylesheet is for display purposes and one is for printing. I suspect that
some of the selectors in my stylesheet are no longer needed due to deletions
of parts of the HTML pages and would like to remove those selectors from my
stylesheets. But I'd like to avoid a manual inspection process to determine
which selectors can be removed since it would be quite easy to mess that up.

I suspect that there is some sort of automated process - maybe a freeware
tool? - for finding all the unused selectors in a given CSS, given a list of
the names of the HTML files using it. If that is right, I'd like someone to
tell me how to accomplish that process.
--
Rhino

[Spartanicus]

"Rhino" <no************ ***********@nos pam.com> wrote:

I suspect that there is some sort of automated process - maybe a freeware
tool? - for finding all the unused selectors in a given CSS

Topstyle sort of[*] has a function to report unused id's and classes,
payware though.

* It cannot handle multiple classes in the HTML.

--
Spartanicus

[Eric Lindsay]

In article <vz************ *****@news20.be llglobal.com>,
"Rhino" <no************ ***********@nos pam.com> wrote:

What's the simplest way to determine which, if any, of my selectors are not
needed in a given stylesheet?

I will shortly have a similar problem, but with about 400 pages and
about a dozen style sheets, all totally out of control. While I don't
have a solution yet, it does occur to me that as a starting point,
extracting all the id and class names from all the HTML pages would be a
good starting point for comparing with those in the CSS. I just tried
this:

$ grep -h class=\" *.htm | awk -F "<|>" '{print $2}' | sort | uniq

and although it has a few spurious classes in it, that is doing most of
the HTML side of the extraction of classes from one directory of files.
If you happen to have Unix style command line tools on your computer (I
am using a Macintosh) that may be a worthwhile starting point.

--
http://www.ericlindsay.com

[Eric Lindsay]

In article <NO************ *************** *****@freenews. iinet.net.au>,
Eric Lindsay <NO**********@e riclindsay.com> wrote:

In article <vz************ *****@news20.be llglobal.com>,
"Rhino" <no************ ***********@nos pam.com> wrote:

What's the simplest way to determine which, if any, of my selectors are not
needed in a given stylesheet?

$ grep -h class=\" *.htm | awk -F "<|>" '{print $2}' | sort | uniq

on second thoughts, you don't need awk, and you can recurse through your
file tree.
$ grep -h -r --include=*.htm* -o id=\".*\" * | sort | uniq

--
http://www.ericlindsay.com

[Rhino]

"Eric Lindsay" <NO**********@e riclindsay.com> wrote in message
news:NO******** *************** *********@freen ews.iinet.net.a u...

In article <vz************ *****@news20.be llglobal.com>,
"Rhino" <no************ ***********@nos pam.com> wrote:

What's the simplest way to determine which, if any, of my selectors are
not
needed in a given stylesheet?

I will shortly have a similar problem, but with about 400 pages and
about a dozen style sheets, all totally out of control. While I don't
have a solution yet, it does occur to me that as a starting point,
extracting all the id and class names from all the HTML pages would be a
good starting point for comparing with those in the CSS. I just tried
this:

$ grep -h class=\" *.htm | awk -F "<|>" '{print $2}' | sort | uniq

and although it has a few spurious classes in it, that is doing most of
the HTML side of the extraction of classes from one directory of files.
If you happen to have Unix style command line tools on your computer (I
am using a Macintosh) that may be a worthwhile starting point.

As it happens, I'm using Windows XP. I have access to a Linux server and
could, in a pinch, transfer my pages over there in order to grep them but
I'm hoping someone can recommend a tool that will work in Windows before I
resort to transferring files.

Rhino

[Johannes Koch]

> "Eric Lindsay" <NO**********@e riclindsay.com> wrote in message

news:NO******** *************** *********@freen ews.iinet.net.a u...

$ grep -h class=\" *.htm | awk -F "<|>" '{print $2}' | sort | uniq

Rhino wrote: As it happens, I'm using Windows XP.

You caould install MinGW (and some more libraries) and then use the
stuff proposed by Eric.
--
Johannes Koch
In te domine speravi; non confundar in aeternum.
(Te Deum, 4th cent.)

[Andy Dingley]

On Sun, 8 Jan 2006 12:01:43 -0500, "Rhino"
<no************ ***********@nos pam.com> wrote:

What's the simplest way to determine which, if any, of my selectors are not
needed in a given stylesheet?

"Unused" are generally less of a problem than "duplicates " in this
context.