From: Karl Cifius on

Hi,

I'm making a Facebook application that can generate images to user's
albums. To publish a story a thumbnail of this image is stored on my
server. Since this server currently is very limited I want to be able
to clean these thumbnails pretty often.

To not get broken links in older facebook stories the address to the
thumbnail is a php script that checks if the thumbnail is available
and returns it, or otherwise returns a default thumbnail.

I have solved this using the following code:

$tImage = $_GET['i'];
$tURL = "upload/$tImage.jpg";
if(!($fp=fopen($tURL,"rb"))){
header("Location: thumb.jpg");
}else{
header("Location: upload/$tImage.jpg");
fclose($fp);
}

My question is if it would be better to have a mysql database with
information about the thumbnail and check if the image is there,
instead of checking if the image file can be loaded? What is the most
optimized approach if I start to gain traffic?


Thanks,

/Karl

From: Ashley Sheridan on
On Fri, 2010-06-25 at 19:31 +0200, Karl Cifius wrote:

> Hi,
>
> I'm making a Facebook application that can generate images to user's
> albums. To publish a story a thumbnail of this image is stored on my
> server. Since this server currently is very limited I want to be able
> to clean these thumbnails pretty often.
>
> To not get broken links in older facebook stories the address to the
> thumbnail is a php script that checks if the thumbnail is available
> and returns it, or otherwise returns a default thumbnail.
>
> I have solved this using the following code:
>
> $tImage = $_GET['i'];
> $tURL = "upload/$tImage.jpg";
> if(!($fp=fopen($tURL,"rb"))){
> header("Location: thumb.jpg");
> }else{
> header("Location: upload/$tImage.jpg");
> fclose($fp);
> }
>
> My question is if it would be better to have a mysql database with
> information about the thumbnail and check if the image is there,
> instead of checking if the image file can be loaded? What is the most
> optimized approach if I start to gain traffic?
>
>
> Thanks,
>
> /Karl
>


I think checking for the existence of a file is probably going to be the
quicker approach. Unless you have a server with loads of RAM and your DB
is very small, it's unlikely your DB will exist entirely in memory, so
you will at some point have to access the files that the DB uses, even
though this is done by the server automatically.

On another note, I would try to sanitise that $_GET variable a bit, as
it could lead to issues down the line later. Maybe limit the string to
patterns you expect for an image URL.

Thanks,
Ash
http://www.ashleysheridan.co.uk


From: Peter Lind on
On 25 June 2010 19:35, Ashley Sheridan <ash(a)ashleysheridan.co.uk> wrote:
> On Fri, 2010-06-25 at 19:31 +0200, Karl Cifius wrote:
>
>> Hi,
>>
>> I'm making a Facebook application that can generate images to user's
>> albums. To publish a story a thumbnail of this image is stored on my
>> server. Since this server currently is very limited I want to be able
>> to clean these thumbnails pretty often.
>>
>> To not get broken links in older facebook stories the address to the
>> thumbnail is a php script that checks if the thumbnail is available
>> and returns it, or otherwise returns a default thumbnail.
>>
>> I have solved this using the following code:
>>
>> $tImage = $_GET['i'];
>> $tURL   = "upload/$tImage.jpg";
>> if(!($fp=fopen($tURL,"rb"))){
>>    header("Location: thumb.jpg");
>> }else{
>>    header("Location: upload/$tImage.jpg");
>>    fclose($fp);
>> }
>>
>> My question is if it would be better to have a mysql database with
>> information about the thumbnail and check if the image is there,
>> instead of checking if the image file can be loaded? What is the most
>> optimized approach if I start to gain traffic?
>>
>>
>> Thanks,
>>
>> /Karl
>>
>
>
> I think checking for the existence of a file is probably going to be the
> quicker approach. Unless you have a server with loads of RAM and your DB
> is very small, it's unlikely your DB will exist entirely in memory, so
> you will at some point have to access the files that the DB uses, even
> though this is done by the server automatically.
>
> On another note, I would try to sanitise that $_GET variable a bit, as
> it could lead to issues down the line later. Maybe limit the string to
> patterns you expect for an image URL.
>
> Thanks,
> Ash
> http://www.ashleysheridan.co.uk
>

Might be quicker to do with a .htaccess file - you can avoid loading php at all.

Regards
Peter


--
<hype>
WWW: http://plphp.dk / http://plind.dk
LinkedIn: http://www.linkedin.com/in/plind
BeWelcome/Couchsurfing: Fake51
Twitter: http://twitter.com/kafe15
</hype>
From: Ashley Sheridan on
On Fri, 2010-06-25 at 20:57 +0200, Peter Lind wrote:

> On 25 June 2010 19:35, Ashley Sheridan <ash(a)ashleysheridan.co.uk> wrote:
> > On Fri, 2010-06-25 at 19:31 +0200, Karl Cifius wrote:
> >
> >> Hi,
> >>
> >> I'm making a Facebook application that can generate images to user's
> >> albums. To publish a story a thumbnail of this image is stored on my
> >> server. Since this server currently is very limited I want to be able
> >> to clean these thumbnails pretty often.
> >>
> >> To not get broken links in older facebook stories the address to the
> >> thumbnail is a php script that checks if the thumbnail is available
> >> and returns it, or otherwise returns a default thumbnail.
> >>
> >> I have solved this using the following code:
> >>
> >> $tImage = $_GET['i'];
> >> $tURL = "upload/$tImage.jpg";
> >> if(!($fp=fopen($tURL,"rb"))){
> >> header("Location: thumb.jpg");
> >> }else{
> >> header("Location: upload/$tImage.jpg");
> >> fclose($fp);
> >> }
> >>
> >> My question is if it would be better to have a mysql database with
> >> information about the thumbnail and check if the image is there,
> >> instead of checking if the image file can be loaded? What is the most
> >> optimized approach if I start to gain traffic?
> >>
> >>
> >> Thanks,
> >>
> >> /Karl
> >>
> >
> >
> > I think checking for the existence of a file is probably going to be the
> > quicker approach. Unless you have a server with loads of RAM and your DB
> > is very small, it's unlikely your DB will exist entirely in memory, so
> > you will at some point have to access the files that the DB uses, even
> > though this is done by the server automatically.
> >
> > On another note, I would try to sanitise that $_GET variable a bit, as
> > it could lead to issues down the line later. Maybe limit the string to
> > patterns you expect for an image URL.
> >
> > Thanks,
> > Ash
> > http://www.ashleysheridan.co.uk
> >
>
> Might be quicker to do with a .htaccess file - you can avoid loading php at all.
>
> Regards
> Peter
>
>


PHP can do things that .htaccess can't, like verify a specific ID has
access to an image, etc.

Thanks,
Ash
http://www.ashleysheridan.co.uk


From: Peter Lind on
On 25 June 2010 20:59, Ashley Sheridan <ash(a)ashleysheridan.co.uk> wrote:
>
> On Fri, 2010-06-25 at 20:57 +0200, Peter Lind wrote:
>
> On 25 June 2010 19:35, Ashley Sheridan <ash(a)ashleysheridan.co.uk> wrote:
> > On Fri, 2010-06-25 at 19:31 +0200, Karl Cifius wrote:
> >
> >> Hi,
> >>
> >> I'm making a Facebook application that can generate images to user's
> >> albums. To publish a story a thumbnail of this image is stored on my
> >> server. Since this server currently is very limited I want to be able
> >> to clean these thumbnails pretty often.
> >>
> >> To not get broken links in older facebook stories the address to the
> >> thumbnail is a php script that checks if the thumbnail is available
> >> and returns it, or otherwise returns a default thumbnail.
> >>
> >> I have solved this using the following code:
> >>
> >> $tImage = $_GET['i'];
> >> $tURL   = "upload/$tImage.jpg";
> >> if(!($fp=fopen($tURL,"rb"))){
> >>    header("Location: thumb.jpg");
> >> }else{
> >>    header("Location: upload/$tImage.jpg");
> >>    fclose($fp);
> >> }
> >>
> >> My question is if it would be better to have a mysql database with
> >> information about the thumbnail and check if the image is there,
> >> instead of checking if the image file can be loaded? What is the most
> >> optimized approach if I start to gain traffic?
> >>
> >>
> >> Thanks,
> >>
> >> /Karl
> >>
> >
> >
> > I think checking for the existence of a file is probably going to be the
> > quicker approach. Unless you have a server with loads of RAM and your DB
> > is very small, it's unlikely your DB will exist entirely in memory, so
> > you will at some point have to access the files that the DB uses, even
> > though this is done by the server automatically.
> >
> > On another note, I would try to sanitise that $_GET variable a bit, as
> > it could lead to issues down the line later. Maybe limit the string to
> > patterns you expect for an image URL.
> >
> > Thanks,
> > Ash
> > http://www.ashleysheridan.co.uk
> >
>
> Might be quicker to do with a .htaccess file - you can avoid loading php at all.
>
> Regards
> Peter
>
>
>
> PHP can do things that .htaccess can't, like verify a specific ID has access to an image, etc.
>

I must've missed the part in the code where the ID was checked ...
Nope, still can't find it.

Regards
Peter


--
<hype>
WWW: http://plphp.dk / http://plind.dk
LinkedIn: http://www.linkedin.com/in/plind
BeWelcome/Couchsurfing: Fake51
Twitter: http://twitter.com/kafe15
</hype>