Hi,

I updated to the 6.x-1.x-dev (to test) because alpha 1 was not saving my field mapping.

The problem is now in the dev version, my fields from the csv file are not being distinguished from each other.

Here's my sample csv file:

id,name,area,option
123,"Name 1","Area 1","Option 1"
234,"Name 2","Area 2","Option 2"
345,"Name 3","Area 3","Option 3"
456,"Name 4","Area 4","Option 4"
567,"Name 5","Area 5","Option 5"

Here's the 'Feed item example' from the Map tab:

Array
(
    [title] => No title
    [description] => 
    [options] => Array
        (
            [timestamp] => 1233155945
            [guid] => 85db06b20164d666734e4f8f30a1a5a0
            [raw] => Array
                (
                    [id,name,area,option] => 123,Name 1,Area 1,Option 1
                )

        )

)

Where [options][raw][id,name,area,option] is now, there should be an array like 

[id] => 123
[name] => Name 1
[area] => Area 1
[option] => Option 1

Is there any other testing that would be helpful?

CommentFileSizeAuthor
#4 parser_csv.inc_.patch816 byteskarens

Comments

alex_b’s picture

alex_b commented
Status: Active » Postponed (maintainer needs more info)

What delimiter setting are you using? (check on the feed node)

Keep in mind that alpha version and latest dev are barely compatible due to changes in the parsing engine and the lack of a built in mapper in the dev version.

scottrigby’s picture

scottrigby
he/him
Primary language English
Location Brooklyn, NY
commented

hi alex_b,
in the above example I had a comma set as the delimiter.

next, i tried exporting that same file using a semicolon - and set semicolon as the delimiter - and it seems to work wit semicolon :)

Ok, so I just did another test, re-exported the csv with comma dilimiter - and set comma in the Parser settings on node/edit - but it still doesn't work with comma.
hope that narrows it down a bit more

BTW, regarding compatibility between alpha & dev...
I'm not worried about that personally. I want to use this module because node_import doesn't currently import cck nodes yet in the D6 version, so this is a very welcome solution!
The other benefit of this module is the ability to easily re-map fields from a csv file. This is huge... because in my case I have a csv with a lot of fields, but current;y not using very many of them on the nodes - but this module seems to allow the option to re-map csv fields to new cck fields - which adds a lot of flexibility.

So I'm happy to use alpha or dev, whichever works. Currently alpha doesn't seem to work, so I thought helping to test dev would be the best use of my time. But do you recommend testing the dev version? Or is that not helpful in it's current state?

karens’s picture

karens commented

I think this would be fixed by the patch at #350405: CSV Format - delimiting double quotes are not discarded. You could test that and report back.

karens’s picture

karens commented
Status: Postponed (maintainer needs more info) » Needs review
StatusFileSize
new parser_csv.inc_.patch816 bytes

Here's the patch that I think is needed to fix this problem. It is definitely needed, I just am not 100% sure it is the patch you need for this problem.

alex_b’s picture

alex_b commented

#4 committed.

alex_b’s picture

alex_b commented
Status: Needs review » Postponed (maintainer needs more info)

Setting this back to needs more info.