From: Eero Saynatkari <eero.saynatkari@...>
Date: 2006-08-11T03:20:11+09:00
Subject: Re: using YAML?

Peter Bailey wrote:
> Eero Saynatkari wrote:
>> Peter Bailey wrote:
>>> I have a spreadsheet with 3 columns in it, and 174 rows. I need to 
>>> cross-reference the items in one column with their corresponding entry 
>>> in the other column. Here is a sample with the first 3 entries in this 
>>> table.
>>> 
>>> File Prefix	Service Code	Publication Title
>>> aacmc7p	        AACM00	        Affirmative Action Compliance Manual
>>> acm059p	        BACM00	        BNA/ACCA Compliance Manual
>>> acm061p	        BACM00	        BNA/ACCA Compliance Manual, Interim Index 
>>> Update
>>> 
>>> I need to look at filenames coming into a directory, see what their 
>>> "file prefix" is, based on column 1 above, and then notate in a separate 
>>> variable the contents of column 2. I'm needing to ascertain what the 
>>> company accounting code is for files being sent to our printer.
>>> 
>>> Should I do this with just an array, or could I use YAML?
>> 
>> Well, YAML is a serialisation format so it is really not
>> something you would use as runtime data storage. An Array
>> or a Hash would certainly work.
>> 
>> If I misunderstood you and you were looking for a file format,
>> then YAML would work great. When you YAML.load the file, you
>> will get your Array/Hash and you can also YAML.dump data into
>> the file. The file format would look like this:
>> 
>>  ---
>>  -
>>    - acmcp7p
>>    - AACM00
>>    - Affirmative # ...
>>  -
>>    -
>>  # ...
>> 
>> Or this:
>> 
>>  ---
>>  -
>>    prefix: acmcp7p
>>    code: AACM00
>>    title: Affirmative #...
>>  -
>>    prefix: # ...
>>  # ...
>> 
>>> Thanks a lot,
>>> Peter
> 
> 
> Thanks, Eero.
> No, I wasn't looking for a file format. I just need to make an 
> association between the items in the first column with their 
> corresponding entry in the second column. I guess a hash makes more 
> sense. I'd just read a bit about YAML, and, I was intrigued with the 
> fact that it's simply text based and I could "parse" to it in some way. 
> At least, that's how I understand it.
> 
> I've got filenames coming into a directory. I need to look at the first 
> 7 characters of those filenames and determine what the "charge" code is 
> for the publication that's publishing that file. This is all for our 
> company accounting system.

Right, OK. Then you want a Hash of file_prefix => service_code
mappings, let us say:

 maps = {"aacmc7p" => "AACM00", "acmp059p" => "BACM00", ...}
 first_seven = read_first_seven_from filename
 puts maps[first_seven]

That will, for a file starting with 'aacmc7p', produce 'AACM00'.

Regarding YAML, it is a serialisation format. If it helps, you
may consider it in some respects a replacement of XML and it
would be used for similar purposes. In your case, you would use
it to store the mappings (instead of using the spreadsheet you
would use YAML) and then load them into a runtime data structure
such as an Array or a Hash. The data format for it would look as
I posted above and the automated parsing of it (YAML.load/YAML.dump)
would produce the Array/Hash data structures notated in the second post.

See http://yaml4r.sourceforge.net

> Thanks again.


-- 
Posted via http://www.ruby-forum.com/.