ALiepinieks:
(no wonder you left this little chestnut alone)
LOL. Well,as I've been saying, the more we delve into the problem and issues, the more complex it actually becomes.
The 'ideal' solution really isn't considered to be a 'holy grail' without reason.
ALiepinieks:
i would imagine that i could return multiple quotes through a loop using a callback function to wait for each quote to return successfully, and once complete, the quote would then be merged with the parcel types itemised simply for clarity on the vendors behalf.
That is the solution I first envisioned too.
ALiepinieks:
unfortunately this would bring with it the caveat that multiple quotes use the same transport method (e.g. express, insured etc.), and this i haven't really got a clue on yet.
Not to mention the situation where 'leftovers' of the 'main' parcel could in some cases be small enough to be considered as a 'letter' or perhaps even suitable to fit into a prepaid satchel, where different rules apply.
ALiepinieks:
i'm also not sure the end-user needs to know the how (they don't really care about whether it's in a satchel or a box, they only care if its registered, insured, express etc. and of course, how much it's going to cost). this then brings another problem,
Well, more than one problem actually...
ALiepinieks:
regarding display. i had some really nice looking pics for each delivery method, and if i was to look into this they would have to be revised as there are simply too many possibilities to cater for every scenario.
Oh, I dunno, the current module only caters for around 30 different methods/scenarios, so what would 2 parcels give us ? 30x30? .. what about 3 parcels.. 30x30x30 ? Surely we can manage that? <grin>
The problem I was thinking about wasn't so much the end user, but moreso the merchant.. Assuming we end up with just 2 dissimilar parcels for any given order, the system needs to record what items were calculated to fit into what parcel so that the merchant can pack the parcels accordingly, otherwise they could end up with a totally different size/weight distribution that will throw the quotes way off.
ALiepinieks:
perhaps they could just be trimmed to parcel post, registerd, insured etc. with appropriate pics, but with no comment on the physical container.
Also, that won't solve the merchant packing issue.
ALiepinieks:
to be honest, i'm not even at that point yet anyway;
I do realise this, and I'm really not trying to make things even more complicated for you - I know it is a one step at a time process.
I am however 'tossing' these ideas your way in the hopes that you may think of ways to avoid or circumvent such stumbling blocks before you actually hit them (having been there and done that).
ALiepinieks:
i'm still dumping variables onto screen and debugging, and considering i've spent about 8hrs on this already i will prbably have to leave it for other work that is beckoning right now.
perhaps i'll be able to revisit it in a few days.
<grin> OK, I'll let you off the hook in having it completed in 7 days.
I'll give you 10 days instead LOL.
ALiepinieks:
in any case, in the event i do embark on parcel splitting (i'm still not sure i'll persist, it's driving me crazy already)
I sure hope others are reading this thread and are now really starting to appreciate the complexity of what first appears should be an 'easy' thing to do.
I've seriously lost count at the number of 'complaints' I've recieved over the years about the parcel splitting 'not working properly' and asking what can be done to 'fix it'.
This is also why when I re-did the ozpost module I didn't even try. My (proposed) solution to the problem is to add additional carriers (courier companys) to the module, so when the Australia Post limits are exceeded it provides quotes from a carrier that WILL accept them.
Due to several reasons (time being one of them, lack of information from the carriers being another) this solution is still to be implemented.
However, thanks to information received from another zencart participant recently I'm hoping to have quotes from TNT avaialble sometime before the new year. :-)
I wasn't going to make this information public until I had something functional, but I figured I'd mention it to you now just in case it will make a difference as to which way you should head.
ALiepinieks:
presently i'm storing the information in two arrays:
the first keeps info on every product id, weights dimensions etc. this array however has an additional field called 'boxid', which if set to '0' (the default before packing starts) means the item was not able to be packed. if set to 1,2,3... it simply means this is which box the product is in.
These arrays will ultimately need to be stored back into the database (more headaches).
ALiepinieks:
so i don't think i'll have a problem getting the info there, it's really just a case of customising it to suit every application
Even this is an issue.. I know for a fact that some 'ozpost' users are selling feathers, and others are selling kitchen sinks, etc. The range to consider is huge.
ALiepinieks:
and then deciding what should be user customisable and what should not.
I know many will disagree with me on this, but I find the less 'customisation' required, the easier it is for everyone concerned.
ALiepinieks:
seeing as i hadn't given the varying parcel sizes much thought, any other ideas you may have on potential issues are welcomed (albeit with some disgust when i realise how much more work is needed!).
I think I/we have now covered all of the issues that I know of (bearing in mind that I started backing out of the entire parcel splitting approach a long time ago).
Having better stacking code is always going to be good (and beneficial), but I do honestly believe that the 'ideal packing solution' AKA 'The holy Grail of shipping modules' that comes about with good stacking AND parcel splitting suitable to all possible item combinations has so many potential problems associated with it that a working solution may never be found. The 'best' we can hope for is something that is workable and usable for any given client's products within a limited range of options.
As I've said many times, I'm not stating any of this to discourage you in any way whatsoever, I'm discussing it with you in the hopes that it gives other people a far greater insight into the extent of the problem, and that the 'solution' really isn't as easy as it first appears.
Anyway, enjoy your few days rest, we'll be expecting a prototype from you before the new year :yes:
Cheers
Rod