Mailing List Archive

To rewrite all *.htm to *.html just use:

^/(.*)\.htm$ => /$1.html


Greetings,
Jędrzej Nowak



On Mon, Oct 31, 2011 at 6:53 PM, Etienne Desautels <tiit@sympatico.ca> wrote:
>
> Hi,
>
> How I'm supposed to do simple page redirections in Cherokee.
>
> For example if I have some old pages/URL that no longer exits and I want the redirect theses URL to other existing URL like:
>
> /home.htm -> /index.html
> /about_us.htm -> about.html
> ...
>
> I have like 20 or more redirections like this to do. The only way I found right now is to create a new Regex Rule for each URL with a redirection Regex Handler. But I found this solution a bit heavy.
>
> Thanks
>
> Etienne
> _______________________________________________
> Cherokee mailing list
> Cherokee@lists.octality.com
> http://lists.octality.com/listinfo/cherokee
>
_______________________________________________
Cherokee mailing list
Cherokee@lists.octality.com
http://lists.octality.com/listinfo/cherokee

2011/10/31 Jędrzej Nowak <me@pigmej.eu>

> To rewrite all *.htm to *.html just use:
>
> ^/(.*)\.htm$ => /$1.html
>

He's not doing redirects that can be easily merged into one regular
expression:

/home.htm -> /index.html
> /about_us.htm -> about.html


Hence the question if there's a way to just specify a 'from' and 'to'
field, without having to mess around with regexes at all.

OP, I don't have a Cherokee-admin available to me at the moment, but if
it's not in the list of ways to do redirects, then there is no simpler
method for redirects.

One simple way I can think of, I didn't test this though:

- Create a final "File Exists" behaviour using "List & Send" handler to
serve all your pages
- Change the default behaviour to use the "Redirection" handler"
- Add all your pages as regular expressions (eg. /home\.html) to this
behaviour

Another option is to create a "Not file exists" rule (by creating a File
Exists rule and then clicking the "NOT" button). But I assume one of these
is what you're already doing. By "bit heavy" did you mean it was causing
performance issues, or are you just worried about potential performance
impacts?

Another option is to create a script (for example, a PHP script) and
redirect all 404s to it, and have the script try to work out where to
redirect to. This would essentially move all the redirect logic into a
single script, and might end up more efficient than using a large number of
regular expressions (as you can do simple string matching in your script).
This can easily be done by using the "custom redirections" option on the
"error handler" tab and adding an internal 404 redirect to /404.php (or
whatever you call the script).

- Daniel

On Tue, Nov 1, 2011 at 4:53 AM, Etienne Desautels <tiit@sympatico.ca> wrote:

>
> Hi,
>
> How I'm supposed to do simple page redirections in Cherokee.
>
> For example if I have some old pages/URL that no longer exits and I want
> the redirect theses URL to other existing URL like:
>
> /home.htm -> /index.html
> /about_us.htm -> about.html
> ...
>
> I have like 20 or more redirections like this to do. The only way I found
> right now is to create a new Regex Rule for each URL with a redirection
> Regex Handler. But I found this solution a bit heavy.
>
> Thanks
>
> Etienne
> _______________________________________________
> Cherokee mailing list
> Cherokee@lists.octality.com
> http://lists.octality.com/listinfo/cherokee
>

Thanks for answering.

On 2011-11-01, at 02:53, Daniel Lo Nigro wrote:

> One simple way I can think of, I didn't test this though:
> Create a final "File Exists" behaviour using "List & Send" handler to serve all your pages
> Change the default behaviour to use the "Redirection" handler"
> Add all your pages as regular expressions (eg. /home\.html) to this behaviour
If I understand correctly, this will not work because the file doesn't exists! So the rule will not catch 'home.htm' that should be redirected to 'index.html' ('home.htm' doesn't exists).

> Another option is to create a "Not file exists" rule (by creating a File Exists rule and then clicking the "NOT" button). But I assume one of these is what you're already doing.

Unfortunately it doesn't work either because it will by applied to all files that don't exists, but in general these requests should be handle by my Django app.

> By "bit heavy" did you mean it was causing performance issues, or are you just worried about potential performance impacts?

No, I mean "A bit heavy to write and maintain". If I could do simple list of matches and redirections, it will be more light!

> Another option is to create a script (for example, a PHP script) and redirect all 404s to it, and have the script try to work out where to redirect to. This would essentially move all the redirect logic into a single script, and might end up more efficient than using a large number of regular expressions (as you can do simple string matching in your script). This can easily be done by using the "custom redirections" option on the "error handler" tab and adding an internal 404 redirect to /404.php (or whatever you call the script).

I know I can do this kind of approach with my Django app, but I find it stupid to handle that at the application level when is just a list of matches with redirections. It's a nice job for a web server.

I found a new solution that is a little bit better. I'm using the Full Path Rule, adding one path for each bad URL, and adding a new Regex for each file in the Redirection Handler. This way I only have one Rule. It's mostly a list of matching URL with redirections but I have to copy the bad URL 2 times (in the rule and then in the handler) and in the handler I need to write it as a Regex. I also need to be careful that a Regex will not match another URL in the list.

In the Rule tab:

Full Path

- /home.htm
- /about_us.htm
- ...


And in the Handler tab:

Redirection

- External | /home\.htm$ | index.html
- External | /about_us\.htm$ | about.html
- ...


Etienne

> - Daniel
>
> On Tue, Nov 1, 2011 at 4:53 AM, Etienne Desautels <tiit@sympatico.ca> wrote:
>
> Hi,
>
> How I'm supposed to do simple page redirections in Cherokee.
>
> For example if I have some old pages/URL that no longer exits and I want the redirect theses URL to other existing URL like:
>
> /home.htm -> /index.html
> /about_us.htm -> about.html
> ...
>
> I have like 20 or more redirections like this to do. The only way I found right now is to create a new Regex Rule for each URL with a redirection Regex Handler. But I found this solution a bit heavy.
>
> Thanks
>
> Etienne
> _______________________________________________
> Cherokee mailing list
> Cherokee@lists.octality.com
> http://lists.octality.com/listinfo/cherokee
>

Does all your "old" urls have .htm on the end ??and you don't have any
existing .htm file ??
If yes just use the approach that I suggested with my first answer.

If you have any existing .htm file just add "file exists rule" and place it
above that redirection one, set the handler to "static".
01-11-2011 19:56 użytkownik "Etienne Desautels" <tiit@sympatico.ca> napisał:

>
> Thanks for answering.
>
> On 2011-11-01, at 02:53, Daniel Lo Nigro wrote:
>
> One simple way I can think of, I didn't test this though:
>
> - Create a final "File Exists" behaviour using "List & Send" handler
> to serve all your pages
> - Change the default behaviour to use the "Redirection" handler"
> - Add all your pages as regular expressions (eg. /home\.html) to this
> behaviour
>
> If I understand correctly, this will not work because the file doesn't
> exists! So the rule will not catch 'home.htm' that should be redirected to
> 'index.html' ('home.htm' doesn't exists).
>
> Another option is to create a "Not file exists" rule (by creating a File
> Exists rule and then clicking the "NOT" button). But I assume one of these
> is what you're already doing.
>
>
> Unfortunately it doesn't work either because it will by applied to all
> files that don't exists, but in general these requests should be handle by
> my Django app.
>
> By "bit heavy" did you mean it was causing performance issues, or are you
> just worried about potential performance impacts?
>
>
> No, I mean "A bit heavy to write and maintain". If I could do simple list
> of matches and redirections, it will be more light!
>
> Another option is to create a script (for example, a PHP script) and
> redirect all 404s to it, and have the script try to work out where to
> redirect to. This would essentially move all the redirect logic into a
> single script, and might end up more efficient than using a large number of
> regular expressions (as you can do simple string matching in your script).
> This can easily be done by using the "custom redirections" option on the
> "error handler" tab and adding an internal 404 redirect to /404.php (or
> whatever you call the script).
>
>
> I know I can do this kind of approach with my Django app, but I find it
> stupid to handle that at the application level when is just a list of
> matches with redirections. It's a nice job for a web server.
>
> I found a new solution that is a little bit better. I'm using the Full
> Path Rule, adding one path for each bad URL, and adding a new Regex for
> each file in the Redirection Handler. This way I only have one Rule. It's
> mostly a list of matching URL with redirections but I have to copy the bad
> URL 2 times (in the rule and then in the handler) and in the handler I need
> to write it as a Regex. I also need to be careful that a Regex will not
> match another URL in the list.
>
> In the Rule tab:
>
> Full Path
>
> - /home.htm
> - /about_us.htm
> - ...
>
>
> And in the Handler tab:
>
> Redirection
>
> - External | /home\.htm$ | index.html
> - External | /about_us\.htm$ | about.html
> - ...
>
>
> Etienne
>
> - Daniel
>
> On Tue, Nov 1, 2011 at 4:53 AM, Etienne Desautels <tiit@sympatico.ca>wrote:
>
>>
>> Hi,
>>
>> How I'm supposed to do simple page redirections in Cherokee.
>>
>> For example if I have some old pages/URL that no longer exits and I want
>> the redirect theses URL to other existing URL like:
>>
>> /home.htm -> /index.html
>> /about_us.htm -> about.html
>> ...
>>
>> I have like 20 or more redirections like this to do. The only way I found
>> right now is to create a new Regex Rule for each URL with a redirection
>> Regex Handler. But I found this solution a bit heavy.
>>
>> Thanks
>>
>> Etienne
>> _______________________________________________
>> Cherokee mailing list
>> Cherokee@lists.octality.com
>> http://lists.octality.com/listinfo/cherokee
>>
>
>
>
> _______________________________________________
> Cherokee mailing list
> Cherokee@lists.octality.com
> http://lists.octality.com/listinfo/cherokee
>
>