This is a little trickier now. The simple solution would be to treat it as a
list delimited by line breaks, but it is not as you have some fields that
contains breaks in it, and no "safe" end field markers, so this needs to be
a bit manual (as in loop line by line and see if it is a new field or not).
So, you loop through record by record (what you already have), and then line
by line within each record constructing the fields, and then doing something
with them. It may be best to turn the data into a query, so for each record
you add a row, and then for each field you set a cell.
Here is code that shows this:
<!--- Get all records --->
<CFSET records=REFindAll("(BOOS\|).*?(BOOE\|)", myData)>
<!--- Create a query --->
<CFSET
myQuery=QueryNew("BOOS,BOOK,TRAN,ANAM,TNAM,PBLS,SUB1,DESC,PRIC,UPDT,UPTM,BCA
T,ABLE,UBID,STAT,PPRC,PDSC,BOOE")>
<!--- Loop through them --->
<CFLOOP INDEX="i" FROM="1" TO="#ArrayLen(records.pos)#">
<!--- Add a row to the query --->
<CFSET QueryAddRow(myQuery)>
<!--- Loop through records --->
<CFSET record=Mid(myData, records.pos[i], records.len[i])>
<!--- Initialize field --->
<CFSET field="">
<!--- Loop through lines in this record --->
<CFLOOP INDEX="line" LIST="#record#" DELIMITERS="#Chr(10)##Chr(13)#">
<!--- If has a | in it, then treat it as start of new field --->
<CFIF line CONTAINS "|">
<!--- New field, save to "field" --->
<CFSET field=line>
<CFELSE>
<!--- Next line of field, append to "field" --->
<CFSET field=field&line>
</CFIF>
<!--- Now that we have the field, split it, before | is name, after is
value --->
<CFSET field_name=ListFirst(field, "|")>
<CFSET field_value=ListRest(field, "|")>
<!--- And save it to the query --->
<CFSET QuerySetCell(myQuery, field_name, field_value)>
</CFLOOP>
</CFLOOP>
Use this with the function I gave you earlier. You'll then end up with a
query named "myQuery" (change the name to something more descriptive). The
query will have the fields in your record (same names), and can be used like
any other ColdFusion query. To display it you could do this:
<!--- Show query contents --->
<CFDUMP VAR="#myQuery#">
The next step would be to turn this all into a function, pass it a file (or
file contents) and get back a query.
And you should be all set.
--- Ben
PS Thanks for asking, little challenges like this are good practice. :-)
_____
From: Seamus Campbell [mailto:[EMAIL PROTECTED]
Sent: Sunday, May 30, 2004 6:20 PM
To: CF-Talk
Subject: [Reply To] RE: parsing help for a slow learner
Thanks Ben. That's a wonderful piece of code (most of which I'm
still trying to understand!)
I feel really stupid now but I'm still stuck.
I added your code and can now get the below.
I can't see how to get each field out of each record to input
into the database. I can see each record is a list with delimiters
of BOOK|, TRAN| ANAM| etc but I can't see how to use these
delimiters to get each field (does that make sense?)
Any help or clues would be great.
++++++++++++++++++++++++
BOOS| BOOK|000004 TRAN|D ANAM|ROTH, WALTER E. TNAM|NORTH QUEENSLAND
ETHNOGRAPHY Bulletin No.4 : GAMES, SPORTS AND AMUSEMENTS PBLS|Home
Secretary's Department Brisbane, 1902 SUB1|ABORIGINAL CULTURE,
AUSTRALIA HISTORY, NATIVE MONOGRAPHS, 19TH CENTURY SPORTS
DESC|Good. 1st. 13.25 x 8.25. a Monograph of 39pp, illustrated by
photograph and drawings, the endpapers have browned and the
exceedingly plain covers have surface tarnish but a bright
interior on superior paper. Walter E.Roth was Queensland's
"Northern Protector of Aboriginals" previously he was "Demy of
Magdalen College, Oxford". PRIC|35 UPDT|10/05/2004 UPTM|12:44:38
PM BCAT|Sold ABLE|1 UBID|1 STAT|Sold PPRC|0 PDSC|QLD BOOE|
_____
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings]
