Ups, started this post, went to bed, finished this post - and missed
the php files from Dan coming in the meantime.
I will have a look at that right now.
Greetings, Martin
On 27 Nov., 11:12, Martin <[EMAIL PROTECTED]> wrote:
> Thank you all for supporting this idea!
>
> The basic concept would be to work the csv file as a collection of
> rows.
> This is well introduced by the fellow who made the php class mentioned
> above (who in fact thinks about a grid, but we do have fmt/template to
> return not only rows but columns too).
>
> His concept could be enhanced with Boltwire's ability to search.
>
> Imagine Boltwire's index file. It is not a well-formed csv file but
> kind of. There are different cases to work with such a file.
>
> 1. Doing a search, returns the matching row(s):
>
> [(csv page=some.page find="Bob||Alice")] A search for "Bob" or "Alice"
> would get me all rows with the matching values line by line.
>
> 2. Doing a search, work on defined values of the matching row(s):
>
> Some more parameters could be:
>
> "delimiter=": defines the colums. Maybe having a default value.
> Everything except of \n which defines the row. If nothing there is
> only one column. If anything is enclosed in " " the delimiter would be
> ignored (therefore you can handle text or "merge columns").
>
> "header=Name,Age,Skill": a custom header row which defines the columns
> starting from the first column. A "Name,Age,Skill" would define the
> first three columns as found using the delimiter. The remaining
> columns would be treated as additional single column. "header=true"
> could take the first row of the file as header row.
>
> Now, a search for "Bob" or "Alice" would get me all rows with "Bob" or
> "Alice". For each of these rows I would be able to address the defined
> 3+1 columns, like fmt="Name: {+name}, Age: {+age}, Skill: {+skill},
> Whatever: {+lastcolumn}".
>
> Or using a form in a template to work with single fields as defined in
> header.
>
> > And then we probably need a command that can save a specific entry in
> > the table, maybe something like:
>
> > [form]
> > [text name]
> > [text skill]
> > [submit]
> > [session csv "page='some.page' find='{=name}' field='{=skill}']
> > [form]
>
> > This would obviously update the csv file for say Bob to include Java now.
>
> 3. Knowing the number(s) of the row(s), work on defined values of the
> matching row(s):
>
> [(csv page=some.page row=7,11,33,-22)] returns one or more rows,
> exclude row 22. Maybe an additional parameter for the last row would
> be nice to have. Returns just these rows as collection of lines if no
> fmt or template defined.
>
> 4. Knowing the number of the column
>
> [(csv page=some.page delimiter="," header=Name,Age,Skill fmt="Name:
> {+name}, Skill: {+skill}")] Using delimiter and header or
> header="true" I would be able to define the columns I am looking for.
> fmt/template returns the values from each row.
>
> 5. Doing a search or having the numbers of rows, deleting this/these
> row(s)
>
> [(csv page=some.page find="Bob" template=csv.delete)] Maybe this could
> be handled by a template which returns the matching rows line by line
> checkboxed, delete rows if checked and purge the csv file. Or action
> (?)
>
> 6. Appending a row
>
> [(log page=some.page)] adds another line at the end of the file.
>
> 7. Get the whole file formatted as a table
>
> [(csv page= header="true" template=csv.table)] shows whole page,
> formatted as a table, perhaps integrated with sorttable plugin. Or use
> custom header="Name,Age,Skill...". Template or action(?)
>
> > To edit the entire table we could just use the regular page edit
> > function I think, or we might need a option for that within the csv
> > command.
>
> Agreed.
>
> > Questions: how do you like the syntax?
> > And second, should we escape the delimiter automatically when entered into
> > a field value? Seems
> > reasonable.
> > Third, can you think of any other functionality that we might need, that we
> > can get when integrating this into BoltWire--with
> > all it's other tools?
> > Fourth, do we want a special csv folder for these kinds of files or should
> > we use regular in-wiki pages?
>
> I think the regular ones would be good.
>
> > Let me know what you think. I'll take a look at putting something
> > together. Shouldn't take an hour--once we settle on exactly what we
> > want. I'll enjoy it!
>
> Thank you all again. I am looking forward to import my contact
> database (45 columns, 4000 rows) into Boltwire.
>
> Greetings, Martin
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"BoltWire" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/boltwire?hl=en
-~----------~----~----~----~------~----~------~--~---
