|
|
 | | From: | Robin Read | | Subject: | Re: OLH | | Date: | Mon, 29 Nov 2004 14:12:24 -0400 |
|
|
 | Thanks for humoring me/my post.
Now that you've replied, I guess I'll have to make
it serious, ;-)
I have found the wish-list group and I'll post
this there with your
great suggestions....I too have encountered some
of those same
problems you detailed.
Here goes...
COPY TO
[]
[FOR ]
[WHILE ]
[FIELDS ]
[[TYPE] DBASE | DBMEMO3 | PARADOX | SDF |
*********************************
DELIMITED [WITH bracket delimted var...(var), or a
macro...&cMacro. Default is double quote.> |
BLANK
SEPARATED [WITH bracket delimted var...(var), or a
macro...&cMacro. Default is comma.> | BLANK ]
EOL [WITH delimted var...(var), or a macro...&cMacro.
Default is the combination of chr(13)+chr(10)> |
NOBLANKLINE ]
] ] |
*********************************
[[WITH] PRODUCTION]
The NOBLANKLINE means
that the last line of data has no chr(13)+chr(10)
appended to the end of it.
That is, it'll have the EOF marker at that point.
Without NOBLANKLINE, the default behavior will
persist.
Also, notice that all previous "WITH have been replaced with
"WITH there a string of any length
as long as it is a variable in brackets of a macro
substituted string.
That should take care of your "chr(9)" issue.
Enjoy,
Robin
"Ivar B. Jessen" wrote in
message
news:rjueq0pt25ktd1f0ba2rlfs080h637m6aj@4ax.com...
> Fri, 26 Nov 2004 13:24:35 -0400 chr(10)
dbase.bug-reports chr(10)Re:
> OLH chr(10) "Robin Read"
chr(10)
>
>
> >This is what I'd like to see...
> >...From the OLH and with the parts that I've
> >modified and added between the lines of
> >asterisks...
> >COPY TO
> >[]
> >[FOR ]
> >[WHILE ]
> >[FIELDS ]
> >[[TYPE] DBASE | DBMEMO3 | PARADOX | SDF |
> >*********************************
> > DELIMITED [WITH > >bracket delimted var...(var), or a
> >macro...&cMacro. Default is double quote.> |
> >BLANK
> >SEPARATED [WITH > >bracket delimted var...(var), or a
> >macro...&cMacro. Default is comma.> |
> >BLANK ] ] ] |
> >*********************************
> >[[WITH] PRODUCTION]
> >
> >Notice that the defaults stay in place...as
they
> >must to maintain backward compatibility.
> >This sort of idea should be aplied to APPEND
FROM
> >also.
> >
> >I was gonna' post this idea to the Wish List,
> >but where did it go?
>
> Robin,
>
> When you post the idea in the wish list could
you include the
> following,
>
> Stress that the separating character can be a
TAB chr(9).
>
> There should be an optional parameter for the
line endings.
> Presently they are always 0x0d0A, Linux requires
something
> different.
>
> There should be an optional parameter for the
last (text) line.
> At the present COPY Fish.dbf to Fishtail.txt
DELIMITED results in a
> last (text) line ending in 0x0D0A followed by a
line consisting of the
> end of file marker 0x1A. That last line ending
0x0D0A is not needed at
> all and the end of file marker 0x1A is even less
needed, it just leads
> to endless problems.
>
>
> Ivar B. Jessen
|
|
| | From: | Ivar B. Jessen | | Subject: | Re: OLH | | Date: | Mon, 29 Nov 2004 22:41:42 +0100 |
|
|
| Mon, 29 Nov 2004 14:12:24 -0400 chr(10) dbase.wishlist chr(10)Re: OLH
chr(10) "Robin Read" chr(10)
>Thanks for humoring me/my post.
>Now that you've replied, I guess I'll have to make
>it serious, ;-)
>I have found the wish-list group and I'll post
>this there with your
>great suggestions....I too have encountered some
>of those same
>problems you detailed.
Great! While you are waiting for the whish to come true you could try
the code below. Watch for word wrap.
Ivar B. Jessen
//-----
#define adCmdText 1
#define adClipString 2
con=new oleautoclient( "ADODB.Connection")
con.open("Driver={Microsoft dBase Driver
(*.dbf)};DBQ=D:\programmer\dbase\plus221\samples;DriverID=533")
rst = new oleautoclient( "ADODB.Recordset" )
rst.open('select ID, Name, Species, "Length CM" from Fish', con,
adCmdText)
? rst.getstring(adClipString, rst.recordCount, ";", ";'<--- Line ends
here'"+ chr(13))
release object rst; release rst
release object con; release con
//-----
|
|
|
