Shouldn't your ,0 (replace every instance) be within your SRPL quotes and do you have too many entries in that command? Also, shouldn't your LISTOF second parenthesis be after RNameL?
Claudine :) -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of David M. Blocker Sent: Monday, August 26, 2002 1:55 PM To: [EMAIL PROTECTED] Subject: Limit to LISTOF Function? Hi folks, I'm writing an application to prepare a scheduling system for a company that transports special needs kids to school systems. The output desired should show the name of each child, the time and place they are picked up, then the "drop off" at the school, listing the kids with the school name, like this: Name Place Time Ken Richards 11 Jones Ave 6:45AM Christopher Raymond 11 Woodland Dr. 6:50AM Christopher McDonald 12 Robert Ave 6:55AM Sarah Gardner-Dickson 110 Friend St. 7:00AM Ken Richards, Christopher Triton Regional 7:10AM Raymond, Christopher High School McDonnald, Sarah Gardner-Dickson I'm accomplishing this quite well with a series of SELECTs unioned together. I'm having one minor problem with the SELECT that produces the "school stop" row. I'm using this to get the list of names: SELECT (SRPL (LISTOF( T1.RNameF & T1.RNameL),',',', ',0)) FROM ...... This works great, placing a space between the names after each comma, UNLESS the student's name (combined First + space + last) exceeds 20 characters, as in the last name above, Sarah. For each name, I get no more than 20 characters, with the rest truncated. What I get is this: Ken Richards, Christopher Triton Regional 7:10AM Raymond, Christopher High School McDonnald, Sarah Gardner-Dickso It doesn't matter where Sarah's name falls in the list, the "N" at the end of her name does not print: the cut off appears to be 20 characters. There is nothing at the RSyntax spot about a limitation of each item in a LISTOF cutting off at 20 characters, nor in the TXT file that came with that patch. Is there a limit? If not, can I do something to force R:Base to give me the full name, no matter how long it is? David Blocker ----- Original Message ----- From: "Troy Sosamon" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, August 26, 2002 9:42 AM Subject: RE: Insert Problem > Paul, > > Take a look at your rules. Do you have any rules on those large > tables your > are inserting into? Rules and constraints on large tables can cause > some serious performance problems. Steve already addressed below the > overhead of > indexes. > > Troy Sosamon > > ===== Original Message from [EMAIL PROTECTED] at 8/26/02 7:59 am > >Paul, I think I can help w/part of your problem. I often have to do INSERTs > >using SELECT or LOADs fm mainframe extracts, a goodly number at the > >scale you describe, and have encountered that "... the process has > >slowed considerably". The cause is this, assuming that INDEXes are > >defined f/the > >table that is being INSERTed INTO, when RBase performs an > >INSERT/LOAD, it builds the INDEX "on the fly", which greatly > >multiplies the disk I/O. This > >really bogs down the performance. > > > >So, in order to get a more satisfactory cycle-time f/such large > >loads, I do > >a couple of things, some of which you might be able to try. Firstly, > >I create a command file to do my INSERTs/LOADs. Yea', so what? > >Well, and secondarily, the what is this, I include, in addition to my > >INSERT/LOAD statement, a set of DROP INDEX statements and, more > >rarely, a DELETE ROWS statement. Each of these is placed prior to > >the INSERT/LOAD statement. I > >also issue a SET RULES off command. It might also be better to do > >this with > >SET MULTI off, as well - I'm sure there are many other folks on this > >list who can go more deeply than I into the nuance of these > >statements. Now, lastly, I issue a set of (controverting) CREATE > >INDEX statements which re-build those INDEXes DROPped prior to the > >INSERT/LOAD. > > > >Now, this might seem redundant, but trust me, all else being equal, > >this will reduce aggregate cycle-time by a "fer piece". I go back a > >long way w/RB and I recall some programmer fm Colorado who sold his > >own RB indexer utility, the source code of which, according to my > >recollection, was subsequently licensed by M'rim. At this point, I > >learned that, f/large INSERT/LOAD operations, it was more time > >efficient to explicitly issue CREATE INDEX statements rather than let > >RBase build the indices implicitly, > >according the DDL/metadata, that is, what would be defined with the TABLE. > >Believe me, (re-)building an index explicitly is WAY FAST! (This is > >probably true f/most decent DBMS packages.) I'm sure you could come > >up w/some simple tests to do comparisons. > > > >So, in short : > > > >-- DATE & TIME FORMAT/SEQUENCE settings according to extract > >definition (as > >necessary &&|| applicable) > >-- DROP INDEX statements > >-- DELETE ROWS (as necessary &&|| applicable) > >-- LOAD/INSERT statements > >-- CREATE INDEX statements > >RETURN > > > >As f/the workstation lock-up, can't offer any ideas there, short of > >some sort of local/server-side timeout which is due to the excessive > >length of time f/the INSERT/LOAD operation to complete. > > > >HTH, > >Steve in Memphis > > > > > >----- Original Message ----- > >From: <[EMAIL PROTECTED]> > >To: <[EMAIL PROTECTED]> > >Sent: Saturday, August 24, 2002 9:45 PM > >Subject: Insert Problem > > > > > >> > >> > >> I am having a problem with a 4.5++ database when doing inserts > >> to > >tables. > >> It started with workstation lock ups when inserting rows using > >> "insert > >into > >> ... select ...". Now when I am running an old program that inserts rows > >into > >> a table, the process has slowed considerably. I recently did a > >> pack and > >> reload without a problem. However, there are several tables containing > >> nearly a million rows and one with nearly 2.5 million rows. Any > >> ideas? > >> > >> Paul Hill > >> ================================================ > >> TO SEE MESSAGE POSTING GUIDELINES: > >> Send a plain text email to [EMAIL PROTECTED] > >> In the message body, put just two words: INTRO rbase-l > >> ================================================ > >> TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] > >> In the message body, put just two words: UNSUBSCRIBE rbase-l > >> ================================================ > >> TO SEARCH ARCHIVES: > >> http://www.mail-archive.com/rbase-l%40sonetmail.com/ > > > >================================================ > >TO SEE MESSAGE POSTING GUIDELINES: > >Send a plain text email to [EMAIL PROTECTED] > >In the message body, put just two words: INTRO rbase-l > >================================================ > >TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] In > >the message body, put just two words: UNSUBSCRIBE rbase-l > >================================================ > >TO SEARCH ARCHIVES: > >http://www.mail-archive.com/rbase-l%40sonetmail.com/ > > ================================================ > TO SEE MESSAGE POSTING GUIDELINES: > Send a plain text email to [EMAIL PROTECTED] > In the message body, put just two words: INTRO rbase-l > ================================================ > TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] In > the message body, put just two words: UNSUBSCRIBE rbase-l > ================================================ > TO SEARCH ARCHIVES: > http://www.mail-archive.com/rbase-l%40sonetmail.com/ > ================================================ TO SEE MESSAGE POSTING GUIDELINES: Send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: INTRO rbase-l ================================================ TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: UNSUBSCRIBE rbase-l ================================================ TO SEARCH ARCHIVES: http://www.mail-archive.com/rbase-l%40sonetmail.com/ ================================================ TO SEE MESSAGE POSTING GUIDELINES: Send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: INTRO rbase-l ================================================ TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED] In the message body, put just two words: UNSUBSCRIBE rbase-l ================================================ TO SEARCH ARCHIVES: http://www.mail-archive.com/rbase-l%40sonetmail.com/
