[wp-trac] [WordPress Trac] #15149: image_get_intermediate_size partly ignores 'path' and 'url' fields and has inconsistent return values

[WordPress Trac]

#15149: image_get_intermediate_size partly ignores 'path' and 'url' fields and has
inconsistent return values
-------------------------+--------------------------------------------------
 Reporter:  whoismanu    |       Owner:                 
     Type:  enhancement  |      Status:  new            
 Priority:  normal       |   Milestone:  Awaiting Review
Component:  Media        |     Version:  3.1            
 Severity:  normal       |    Keywords:  has-patch      
-------------------------+--------------------------------------------------

Comment(by whoismanu):

 the more i think about this, the more i am convinced that in the end there
 is a more fundamental underlying question that should be answered:

 does one want to give plugin authors the freedom to choose the location of
 intermediate image sizes or not?

 right now this is not the case (mainly because of what i detailed in my
 posts above) and the file/directory structure is the following:
 {{{
 uploaddir/yyyy/mm/image.jpg
 uploaddir/yyyy/mm/image-size1.jpg
 uploaddir/yyyy/mm/image-size2.jpg
 }}}
 where image.jpg is the original image, the other two are resized versions
 of it. finally, size1 and size2 usually have the form {width}x{height},
 but they could also be set to something more readable like "thumbnail" via
 filters. in the attachment metadata "sizes" field only "width", "height"
 and "file" are stored and "file" is the filename of the resized image. the
 path to the resized image is constructed as detailed above, by taking the
 path of the original image and exchanging the filename with the one found
 in "file".

 when dealing with images, i think, however, that other structures can make
 a lot of sense and it would therefore be good to have some flexibility in
 this respect. in particular i am thinking about a structure like this:
 {{{
 uploaddir/original/yyyy/mm/image.jpg
 uploaddir/thumbnail/yyyy/mm/image.jpg
 uploaddir/medium/yyyy/mm/image.jpg
 }}}
 why is this in some cases a better suited structure? well, it allows you
 to immediately find all files that belong to a certain size. for example,
 it makes it very easy to delete all medium images. just delete the
 corresponding directory. or what about renaming a size? just rename the
 corresponding directory. with the structure wordpress imposes right now, i
 have to traverse all of the year-month directories to find the
 corresponding files and treat them one by one, based on their filename.

 so, for some plugins it certainly makes sense to have a different
 directory structure. therefore i think wordpress should offer this
 flexibility. further, this could be quite easily achieved by storing the
 "path" and "url" information in addition to the "file", "width" and
 "height" information inside the metadata. it looks like someone already
 thought about this, otherwise why would there be references to a "path"
 and an "url" field inside image_get_intermediate_size().

 whatever the decision is on this, i think that in any case the intention
 needs to be made clearer. if sizes are to be distinguished only based on
 their filename, references to fields like "path" or "url" that suggest the
 opposite should go away. if paths are to be flexible, the above patch is
 probably not enough, but it would be better to make sure that all sizes
 have the corresponding fields and that the core as well as plugins rely on
 these fields to find the location of a size, rather than computing it from
 the "file" field.

 i am looking forward to hear your thoughts on this.

-- 
Ticket URL: <http://core.trac.wordpress.org/ticket/15149#comment:3>
WordPress Trac <http://core.trac.wordpress.org/>
WordPress blogging software