Missing import fields

[leehack]

I also place an id in the first column. I then split the file into Product and Section by sorting on the headers. I then cut and paste these parts into other workbooks and then trim the number of field columns to the bare minimum that are needed.

When all done copy everything back into the first sheet, sort and you're ready to save as csv for importing.

Be very careful however if you have any fields with preceding zeros - make sure you rather open Excel and then import the data from the file rather than open the csv file directly - you can then set the field column data format for each field rather than let Excel choose for you.

Make sure you snapshot lots and always compact the database after importing.

Removal of the headers Duncan, is the one thing that scares the begeebers out of a lot of people. I guess a KB article on what the minimum headers you can have in order to update a product would be good or something along those lines.

Sean's tip has to go into the KB though.

[Duncan Rounding]

To update products you just need the prod ref, product short description and whatever other fields you require. Similar for sections.

Most importantly of all though is that the final structure must be spot on.

[Jan]

The format required to import the details you want as a hierarchical import file is very simple, here is an example :

Header:Product,Product reference, brand, ean
Product,"39", "Manufacturer for Product 20 Digital Camera", "Ean 39"
Product,"5", "Manufacturer for product 5 Photo Frame", "Ean 5"

Any user defined property that you add to your site can be included as a title in the product header line and then imported (you have to restart actinic after adding the variable to have it recognised by import). This is a simple and extensible way of importing data, much better than a fixed format flat file import.

If you have auto generate product reference switched on, you also have to include the product description.

Regards,

[Sean Williams]

Aw Shucks - I'm slightly embarrassed to receive this praise from such distinguished forum members

I'm really not trying to be clever when I say it just seemed the sensible thing to do and when we started doing it I did feel bit of a wimp in fact as I felt sure everyone else was far too experienced and manly to go to such lengths!

Cheers!

[Sean Williams]

Oh and Rookers - you might need to read thru this thread: http://community.actinic.com/showthread.php?t=49818

My post is at #6

[Jan]

I've added a new article for the Mole End knowledge-base, with examples of the format when autogenerate is on and when autogenerate is off.

http://moleend.helpserve.com/index.p...barticleid=206

This one gives the format for V7 :

http://moleend.helpserve.com/index.p...&kbarticleid=1

Regards,

[leehack]

Sean in that post you mentioned about the blank columns and how that can screw up the import, could the numbering column going far right instead solve two birds in one stone or do you find that the data has to be done in a certain way to avoid that issue and the numbering column would not do that successfully?

Reason i ask whenever i have had problems with the blank row, i have added a brand new row on far right in next available empty column and just filled it with a number from top to bottom, this has always fixed it for me, but i'm sure you've done more hierarchicals than me.

[Duncan Rounding]

You can certainly use the id field as the very right side column - any dummy column will do - but it's easier to work when it's on the very left (for me anyway).

[Duncan Rounding]

You also sometimes get an error with the empty row just after the headers because it gets filled with commas . To stop this simply remove remove the row - it's only cosmetic and not required for the import.

[nileshpa]

Jan,
Let me understand this correctly.
If you want to import NEW products via hierarchical format, the very basic / minimal fields required, are simply the Product Ref & Product description for each product. Is this correct?

Reason I'm asking is I need to bring in the component & attribute structure (i.e. Size&Colour) of all NEW products from our ePos via SQL whilst keeping real products hidden. See attached file that gets generated.
The SQL does get complicated when you have hundreds of fields to deal with. When a comma or double-quote drops off, it knocks off the entire import.

If what you are saying is true, Its simplify my sql. Please note that this very question was raised with support some time ago & was told all fields within the hierarchical format are required whether they are used or not. Unused fields on the every record have to be set to null.

Regards, Nilesh

[Jan]

Hi Nilespa,

This is the minimum hierarchical import file for new products :

Header:Product,Product reference, Short Description
Product,"1139", "New product 1139"
Product,"1115", "New Product 1115"

This creates products at the top level catalog section level. The product reference (if auto generate is switched off) and product short description are set up, all other fields are set to their default values.

If you want to put them into a new section then this is the simplest form :

Header:Section,Section Name
Header:Product,Product reference, Short Description
Section,"New Products"
Product,"11139", "New product 11139"
Product,"11115", "New Product 11115"
End,Section

Regards,

[nileshpa]

Thanks, Jan. Much appreciated.
Will review & simplify the sql soon.
Regards,
Nilesh

[Jan]

The other thing I do to simplify the file is to make the import a two stage process, the first to create the products and second to add components etc. I make a flat import file for the first stage and so can use a section tree to specify where I want it to go and to add basic details like price, description etc. Then the second is the simple hierarchical format shown above but can include user defined variables and you can also put component / attribute details below each product line. I have written SQL to produce a full hierarchical import file including nested sections but I find it easier to use the approach above.

Regards,

[Sean Williams]

Sean in that post you mentioned about the blank columns and how that can screw up the import, could the numbering column going far right instead solve two birds in one stone or do you find that the data has to be done in a certain way to avoid that issue and the numbering column would not do that successfully?

Reason i ask whenever i have had problems with the blank row, i have added a brand new row on far right in next available empty column and just filled it with a number from top to bottom, this has always fixed it for me, but i'm sure you've done more hierarchicals than me.

Sorry I missed this Lee - you must think me very rude!

We always add a column of cells filled with 'n'.
We don't have any particular reason for not using numbers except for the first one that came to mind - it might confuse Actinic! So we've just stuck with it. Not very scientific

The reason we insert the temporary column on the far left is that most of the data we need to see when editing tends to be on the left side, so it's easier to work with less scrolling left to see the line number.

HTH

Cheers