> (Side note - it would be great if such functionality were available as a
> MARCedit add-on - put multiple MARC records side by side (all for the same
> biblio), tic which fields you want to keep from which sources and merge
> them all into one - sorry I'm not a coder nor have the resources to sponsor,
> just putting the idea out there (as if it's a new idea I'm sure 8- and I realize it
> wouldn't be that easy to code, and maybe not many people would want to
> do this)

Actually, there is one.  In newer versions of MarcEdit, there is a Merge Records tool.  The tool works with either a control number for matching or not -- but essentially, you can merge data from one set of MARC records (you select the fields to merge) into another set.  My guess is that this probably would work much better than doing all this in Excel.


> I *am* surprised however that once I've used MARCedit to convert
> *from* MARC to tab-delimited, and the latter contains the field/subfield tags
> in the first row, the Delimited Text tool to convert it *back* to MARC doesn't
> "just work" - I have to re-define the field-to-tag mappings all over again.

There is a good reason for this.  In the mappings, you can join fields, add constant data and set indicators and punctuation for a field.  If the headers were used, this information wouldn't be easily represented.

> This next one is more serious, in that it requires a lot more trouble and skill to
> overcome than the above. This is the fact that the MARC-to-CSV direction
> concatenates all the multiple-field values (e.g.
> 700$a, 500$a) into a single column separated by semicolons, which when
> going back in the other direction results, in a single such field with everything
> - potentially completely different categories of information - all squished in
> together, much less readable/useful for my library's patrons than keeping
> them separate.

Because MarcEdit doesn't know how many columns would be necessary without first reading the entire file and pre-mapping all the fields to columns.  The program actually makes use of a function in Excel that lets you build columns that can have and be read as having multiple delimiters.  

> Next-best possibility is that someone suggests a more elegant workaround
> solution than the kludges outlined above.

I still think that your best solution is to probably see if the merge tool will work for you.

--TR


> -----Original Message-----
> From: MarcEdit support in technical and instructional matters
> [mailto:[log in to unmask]] On Behalf Of Hans BKK
> Sent: Sunday, February 27, 2011 12:14 AM
> To: [log in to unmask]
> Subject: [MARCEDIT-L] Round-tripping MARC <--> CSV
> 
> Hi all, noob here, both to the list and to MARCedit. Let me start by thanking
> (mostly but not only) Terry for this wonderful tool, as well as those of you
> that have been contributing to the development process over the years.
> 
> My current issues:
> 
> Due to the limitations of my ILS' MARC-editing functionality, I'm exporting a
> batch of my ILS' records (biblio level only) in order to merge in enhancing
> data from matching MARC records from other sources, and them bringing
> them back into the ILS using the Delimited Text add-in.
> 
> I'd like to do the "data massaging" in a spreadsheet (I find OpenOffice better
> at text-handling than Excel). I put all the MARC records for a given title
> (including my ILS' "old" one) into one worksheet, so it's easy to see the
> different fields and their contents, cut and paste, delete unused fields,
> consolidate everything into one row and then convert it back to MARC.
> 
> (Side note - it would be great if such functionality were available as a
> MARCedit add-on - put multiple MARC records side by side (all for the same
> biblio), tic which fields you want to keep from which sources and merge
> them all into one - sorry I'm not a coder nor have the resources to sponsor,
> just putting the idea out there (as if it's a new idea I'm sure 8- and I realize it
> wouldn't be that easy to code, and maybe not many people would want to
> do this)
> 
> I *am* surprised however that once I've used MARCedit to convert
> *from* MARC to tab-delimited, and the latter contains the field/subfield tags
> in the first row, the Delimited Text tool to convert it *back* to MARC doesn't
> "just work" - I have to re-define the field-to-tag mappings all over again. Not
> that big a deal, but one would think the files resulting from going in one
> direction could be easily re-purposed for going back in the other. Manually
> adding lines to the profile list for any new mappings would of course be trivial
> and pretty much self-explanatory.
> 
> So that's IMO an area that could use improvement - easy round-tripping back
> and forth between MARC and CSV - but as I said NBD just an inconvenience.
> 
> 
> This next one is more serious, in that it requires a lot more trouble and skill to
> overcome than the above. This is the fact that the MARC-to-CSV direction
> concatenates all the multiple-field values (e.g.
> 700$a, 500$a) into a single column separated by semicolons, which when
> going back in the other direction results, in a single such field with everything
> - potentially completely different categories of information - all squished in
> together, much less readable/useful for my library's patrons than keeping
> them separate.
> 
> I'm looking at two possible workarounds - use the spreadsheet tool to do a
> "text-to-column" style splitting of the data manually, or use two separate
> target files for the MARC-to-CSV conversion, both containing the control key
> field, but one for single-fields and the other for repeating ones. This will
> allow them to be brought into the spreadsheet with different delimiters
> spec'd, so the repeating fields stay separate. However this would then mean
> massaging their data in a separate step and then also an added step to merge
> them back later, either before or while bringing them back into the ILS.
> 
> Both are major kludges that IMO shouldn't be necessary.
> 
> I'm hoping (best case) I'm missing something here in my use of the MARCedit
> tools and that someone will enlighten me on how to keep these repeating
> fields separate during the MARC-to-CSV step.
> 
> Next-best possibility is that someone suggests a more elegant workaround
> solution than the kludges outlined above.
> 
> And obviously it would be great if future improvement to MARCedit would
> make these processes easier.
> 
> Thanks for your time and consideration, and in advance for any and all
> responses.
> 
> __________________________________________________________
> ______________
> 
> This message comes to you via MARCEDIT-L, a Listserv(R) list for technical and
> instructional support in MarcEdit.  If you wish to communicate directly with
> the list owners, write to [log in to unmask] To
> unsubscribe, send a message "SIGNOFF MARCEDIT-L" to
> [log in to unmask]

________________________________________________________________________

This message comes to you via MARCEDIT-L, a Listserv(R) list for technical and instructional support in MarcEdit.  If you wish to communicate directly with the list owners, write to [log in to unmask] To unsubscribe, send a message "SIGNOFF MARCEDIT-L" to [log in to unmask]