Profile: shoecake

That works perfectly now. Thank you so very much. I've already left a review for FeedManager a few months ago, but I'll see if it lets me add another.

OK, I can successfully upload to ebay with only 2 minor changes.

The upload does not like quotes round the first column's header (it doesn't read the first header so errors in determining the file version) so I suggest removing quotes round all of the headers (top row of file). I can't imagine a situation where any csv file would have commas in a header.

Secondly, if a field does not have a value, then it does not like the "" in these fields as it seems to expect there to be some data there. I did a search/replace to remove all "" in the outputted file.

Could you make these changes? I am eternally grateful for all you have done on this.

That works except the code is now removing the " from the data. The best option is to replace a " with a "" this way it will still be a valid csv and will keep the quote.

Also, my code is replacing any line feeds with a space so the html (description field) is on one line of the csv file (as per the eBay specification). I think this would be a good addition as any line feeds would make any csv file invalid.

I have been manually editing the csv file before uploading to eBay so I can confirm they do support field enclosure via the "file exchange" method of uploading.

They certainly do for the description field, so I assume they also will for all the other fields.

I am exporting using the Ebay File Exchange feed which outputs a CSV. I have noticed it is stripping out commas in my description.

Ideally, it should put quotes round the text so the commas will not be seen as the start of a new column.

I have already put the quotes round using the new plugin functionality, is there a way of stopping it removing the commas?


The feeds.xml entry is:

<FeedNetWork>
    <Id>173</Id>
    <Name>Ebay File Exchange Template UK (File Exchange Center)</Name>
    <Image>ebay.png</Image>
    <Schema>ebayFileExchangeTemplateUK.xml</Schema>
    <ExportMethodId>172</ExportMethodId>
    <DisplayOrder>26</DisplayOrder>
    <HasMerchantId>true</HasMerchantId>
    <ExportFileType>CSV</ExportFileType>
    <Documentations>
      <DocLinkValue>
        <Message>Products Feed Specification</Message>
        <HtmlUrl>https://securepics.ebaystatic.com/aw/pics/uk/pdf/file_exchange/File_Exchange_Basic_Template_Mini_Guide.pdf</HtmlUrl>
      </DocLinkValue>
      <DocLinkValue>
        <Message>Ebay Category Tree</Message>
        <HtmlUrl></HtmlUrl>
      </DocLinkValue>
      <DocLinkValue>
        <Message>File Exchange - Instructions and Resources</Message>
        <HtmlUrl>https://securepics.ebaystatic.com/aw/pics/uk/pdf/file_exchange/File_Exchange_Advanced_Instructions.pdf</HtmlUrl>
      </DocLinkValue>
    </Documentations>
  </FeedNetWork>  


Can I do this by changing ExportMethodId or ExportFileType ?

Thanks a lot for this, I have already used it in a number of places. Very usefull.

Yes, that is right

When I am creating default values for categories on my site, it would be great if the system looked at values from parent categories. At the moment it only takes values from the lowest category in the tree.

Fore example, I sell fabric of various types and sizes, my category tree looks something like:


Fabric
Fabric > Aida
Fabric > Aida > 14 Count


For an eBay feed for example, I wand to define the "Category" field at the "Fabric" level, the "Fabric Type" field at the "Fabric > Aida" level and the "Fabric Count" field at the "Fabric > Aida > 14 Count" level.

At the moment I have to define all of this at the "Fabric > Aida > 14 Count" level and repeat it for all categories in the Fabric section.

One other thought I had, this subplugin function for the whole row should run before the "post process action" on each individual field.

I think this because the post process actions are about formatting the data for the output file and so should happen after the subplugin has made its changes.

I think an external function is the best option and also will work in many other situations.

One way would be to have one function for the whole row of a feed. So code would send productid and an IList (or Dictionary) of KeyValue Pairs. The subplugin can then modify this collection and return it.

This would be better as one field may rely on calculations of data in other fields in the row.

Yes a badly coded subplugin could slow performance, but that is the subplugin authors responsibility. Also, generating a feed is a background task so if it runs a few seconds slower it does not really matter.