To write headers once with different values in separate row in CSV

[Steven D'Aprano]

On Wed, 24 Jun 2015 09:37 pm, kbtyo wrote:

> On Tuesday, June 23, 2015 at 9:50:50 PM UTC-4, Steven D'Aprano wrote:
>> On Wed, 24 Jun 2015 03:15 am, Sahlusar wrote:
>> 
>> > That is not the underlying issue. Any thoughts or suggestions would be
>> > very helpful.
>> 
>> 
>> Thank you for spending over 100 lines to tell us what is NOT the
>> underlying issue. I will therefore tell you what is NOT the solution to
>> your problem (whatever it is, since I can't tell). The solution is NOT to
>> squeeze lemon juice into your keyboard.
>> 
>> If someday you feel like telling us what the issue actually IS, instead
>> of what it IS NOT, then perhaps we will have a chance to help you find a
>> solution.
>> 
>> 
>> 
>> --
>> Steven
> 
> Curious - what should I have provided? 

To start with, you should tell us what is the problem you are having. You
gave us some code, and then said "That is not the underlying issue". Okay,
so what is the underlying issue? What is the problem you want help solving?

In another post, you responded to John Gordon's question:

    # John
    Have you tried creating some dummy data by hand and seeing 
    how makerows() handles it?


by answering:

    Yes I did do this.


Okay. What was the result? Do you want us to guess what result you got?


John also suggested that you provide sample data, and an implementation of
flatten_dict, and your answer is:

    Yes, unfortunately, due to NDA protocols I cannot share this.


You don't have to provide your *actual* data. You can provide *sample* data,
that does not contain any of your actual confidential values. If your XML
file looks like this:

<?xml version="1.0"?>
<catalog>
   <book id="bk101">
      <author>Gambardella, Matthew</author>
      <title>XML Developer's Guide</title>
      <genre>Computer</genre>
      <price>44.95</price>
      <publish_date>2000-10-01</publish_date>
      <description>An in-depth look at creating applications 
      with XML.</description>
   </book>
</catalog>


you can replace the data:

<?xml version="1.0"?>
<catalog>
   <book id="1111">
      <author>Smith, John</author>
      <title>ABCDEF</title>
      <genre>Widgets</genre>
      <price>9999.99</price>
      <publish_date>1900-01-01</publish_date>
      <description>blah blah blah blah</description>
   </book>
</catalog>


You can even change the tags:


<?xml version="1.0"?>
<whatzit>
   <spam id="1111">
      <a>Smith, John</a>
      <b>ABCDEF</b>
      <c>Widgets</c>
      <d>9999.99</d>
      <e>1900-01-01</e>
      <f>blah blah blah blah</f>
   </spam>
</whatzit>


If you're still worried that the sample XML has the same structure as your
real data, you can remove some fields and add new ones:

<?xml version="1.0"?>
<whatzit>
   <spam id="1111">
      <b>ABCDEF</b>
      <d>9999.99</d>
      <e>1900-01-01</e>
      <z>fe fi fo fum</z>
      <f>blah blah blah blah</f>
   </spam>
</whatzit>


If you can't share the flatten_dict() function, either: 

(1) get permission to share it from your manager or project leader.
flatten_dict is not a trade secret or valuable in any way, and
half-competent Python programmer can probably come up with two or three
different ways to flatten a dict in five minutes. They're all going to look
more or less the same, because there's only so many ways to flatten a dict.

(2) Or accept that we can't help you, and deal with it on your own.



> Detailed and constructive feedback 
> (like your reply to my post regarding importing functions) is more useful
> than to "squeeze lemon juice" into one's keyboard.

Of course. That is why I said it was NOT the solution. Don't waste your time
squeezing lemon juice over your keyboard, it won't solve your problem.

But you can't expect us to guess what your problem is, or debug code we
can't see, or read your mind and understand your data.

Before you ask any more questions, please read this:

http://sscce.org/



-- 
Steven