From s.anderson.au at gmail.com Wed Sep 4 06:45:58 2019 From: s.anderson.au at gmail.com (Steve) Date: Tue, 3 Sep 2019 23:45:58 -0700 (PDT) Subject: [Info-ingres] set unicode_substitution command Message-ID: <51b50821-6146-4654-8307-952728adfcb4@googlegroups.com> select dbmsinfo('charset') \g Executing . . . ?????????????????????????????????????????????????????????????????? ?col1 ? ?????????????????????????????????????????????????????????????????? ?ISO88591 ? ?????????????????????????????????????????????????????????????????? set unicode_substitution 'X' \g Executing . . . * select u&'\062E'; * \g Executing . . . ???????? ?col1 ? ???????? Wed Sep 4 16:40:17 2019 E_AD5016_NOUNIMAP_FOUND Unicode code point 0000062E cannot be mapped to local character set. ?? ???????? (1 row) Why am I getting error E_AD5016 here? I expected Ingres to return 'X', what am I missing? u&'\062E' = Arabic letter Khah Thanks Steve From martin.bowes at ndph.ox.ac.uk Thu Sep 5 14:34:31 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Thu, 5 Sep 2019 14:34:31 +0000 Subject: [Info-ingres] set unicode_substitution command In-Reply-To: <51b50821-6146-4654-8307-952728adfcb4@googlegroups.com> References: <51b50821-6146-4654-8307-952728adfcb4@googlegroups.com> Message-ID: <102bc2e5cfca45ec87bc0dbddc1613d6@EXCHANGE2.ndph.ox.ac.uk> Hi Steve, I'm seeing the same thing on II 11.0.0 (a64.lnx/100) + p15360 and II 11.1.0 (a64.lnx/100) + p15445. This appears to be a bug. Please report it. Martin Bowes -----Original Message----- From: Steve [mailto:s.anderson.au at gmail.com] Sent: 04 September 2019 07:46 To: info-ingres at lists.planetingres.org Subject: [Info-ingres] set unicode_substitution command select dbmsinfo('charset') \g Executing . . . ?????????????????????????????????????????????????????????????????? ?col1 ? ?????????????????????????????????????????????????????????????????? ?ISO88591 ? ?????????????????????????????????????????????????????????????????? set unicode_substitution 'X' \g Executing . . . * select u&'\062E'; * \g Executing . . . ???????? ?col1 ? ???????? Wed Sep 4 16:40:17 2019 E_AD5016_NOUNIMAP_FOUND Unicode code point 0000062E cannot be mapped to local character set. ?? ???????? (1 row) Why am I getting error E_AD5016 here? I expected Ingres to return 'X', what am I missing? u&'\062E' = Arabic letter Khah Thanks Steve _______________________________________________ Info-ingres mailing list Info-ingres at lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres From services.info at dhl.com Thu Sep 5 09:33:18 2019 From: services.info at dhl.com (DHL EXPRESS) Date: Thu, 05 Sep 2019 04:33:18 -0500 Subject: [Info-ingres] DHL EXPRESS COURIER SERVICES- DELIVERY NOTIFICATION Message-ID:

-------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 101006 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Bill Of Lading & AWB.gz Type: application/octet-stream Size: 553372 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Document.zip Type: application/zip Size: 553984 bytes Desc: not available URL: From martin.bowes at ndph.ox.ac.uk Fri Sep 6 10:00:09 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Fri, 6 Sep 2019 10:00:09 +0000 Subject: [Info-ingres] Why can't I leave my window function ajar? Message-ID: <448393d1e5eb442b8017bea459434657@EXCHANGE2.ndph.ox.ac.uk> Hi All, II 11.0.0 (a64.lnx/100) + p15426 It appears the first N feature isn't working with windowing functions like row_number() over () This query was supposed to select the first 1000 rows from the derived table ...so 12050912 was a bit of a shock. DECLARE GLOBAL TEMPORARY TABLE session.my_gp_clinical_p1 AS SELECT FIRST 1000 s.* FROM ( SELECT *, row_number() over ( partition by reg_id, data_provider, event_dt, read_2, read_3, map_icd10, value1, value2, value3 ) as row_number FROM actual_gp_clinical WHERE mod(reg_id, 10) = 0 ) s WHERE s.row_number = 1 ON COMMIT PRESERVE ROWS WITH NORECOVERY, PAGE_SIZE = 32768 Executing . . . (12050912 rows in 452.269470 secs) Martin Bowes -------------- next part -------------- An HTML attachment was scrubbed... URL: From nikos.vaggalis at gmail.com Fri Sep 6 20:05:38 2019 From: nikos.vaggalis at gmail.com (nikosv) Date: Fri, 6 Sep 2019 13:05:38 -0700 (PDT) Subject: [Info-ingres] Weird error on reconnect from perl program In-Reply-To: References: Message-ID: Have you checked with DBI_TRACE level 5 and above ? https://www.effectiveperlprogramming.com/2010/04/use-dbi_trace-to-follow-dbis-work/ From martin.bowes at ndph.ox.ac.uk Sat Sep 7 08:05:03 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Sat, 7 Sep 2019 08:05:03 +0000 Subject: [Info-ingres] Weird error on reconnect from perl program In-Reply-To: References:

Message-ID: <6abfff88f6324912acfb1be3f02d7afb@EXCHANGE2.ndph.ox.ac.uk> Hi Nikos Yes indeed I have. Sadly it wasn't as enlightening as I hoped. Even at trace level 2. However, I've managed to work out that the problem was caused by a combination of a few things. The main one being that one of the 5 database connections was to a host outside our firewall and the network dropped the connection after an hour of inactivity. My job started a database query, and apparently the waiting for it to complete looks inactivity. The network manager was unwilling to extend the hour. I've had to program around it by splitting the job into slices that will complete within the hour...I'm now up to ten slices on one task. Marty -----Original Message----- From: nikosv [mailto:nikos.vaggalis at gmail.com] Sent: 06 September 2019 21:06 To: info-ingres at lists.planetingres.org Subject: Re: [Info-ingres] Weird error on reconnect from perl program Have you checked with DBI_TRACE level 5 and above ? https://www.effectiveperlprogramming.com/2010/04/use-dbi_trace-to-follow-dbis-work/ _______________________________________________ Info-ingres mailing list Info-ingres at lists.planetingres.org https://lists.planetingres.org/mailman/listinfo/info-ingres From martin.bowes at ndph.ox.ac.uk Sat Sep 7 08:19:10 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Sat, 7 Sep 2019 08:19:10 +0000 Subject: [Info-ingres] E_AD5015 what? Message-ID: <0a70951089a24570a36d07976c24ebbf@EXCHANGE2.ndph.ox.ac.uk> Hi All, I'm processing some data which instead of an empty string has been provided with a C-like end of string marker x'00'. I'm not allowed to alter this data. So I've allowed for this with a view which does: CASE WHEN g.quantity_v <> 2 THEN NULL WHEN g.quantity = X'00' THEN '' ELSE g.quantity END AS quantity When I use the view to select the data I get greeted with: E_AD5015 000000B7, is not a valid byte sequence for server codepage. If I remove the X'00' part of the statement then it runs without error, but I get the crappy data. Any Ideas? Martin Bowes -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.anderson.au at gmail.com Tue Sep 10 06:36:08 2019 From: s.anderson.au at gmail.com (Steve) Date: Mon, 9 Sep 2019 23:36:08 -0700 (PDT) Subject: [Info-ingres] E_AD5015 what? In-Reply-To: References: Message-ID: <38859d49-6b46-4451-97b7-6e74d56d52f3@googlegroups.com> On Saturday, 7 September 2019 18:20:18 UTC+10, Martin Bowes wrote: > Hi All, > > ? > > I?m processing some data which instead of an empty string has been provided with a C-like end of string marker x?00?. I?m not allowed to alter this data. > > ? > > So I?ve allowed for this with a view which does: > > ? > > ??? CASE WHEN g.quantity_v <> 2 THEN NULL > > ??? ?????WHEN g.quantity = X'00' THEN '' > > ???????? ELSE g.quantity > > ??? END AS quantity > > ? > > When I use the view to select the data I get greeted with: > > E_AD5015 000000B7, is not a valid byte sequence for server codepage. > > ? > > If I remove the X?00? part of the statement then it runs without error, but I get the crappy data. > > ? > > Any Ideas? > > ? > > Martin Bowes Hi Martin I came across the following from the Ingres 11 SQL guide when looking for something else the other day. I wonder if it might apply to your situation? "Be careful when using hex constants for strings because the hex values depend on the used character set. For example, the euro sign ? has a hex value of x'A4' when using ISO8859-15, but x'80' in WINDOWS-1252. Moreover, in UTF8 the euro sign takes three bytes: x'E2', x'82', x'AC'. You would produce invalid UTF8 data when using one of the other representations above. Instead, it is better to use Unicode literals for the euro sign: u&'\20AC'." Steve From martin.bowes at ndph.ox.ac.uk Fri Sep 13 08:45:54 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Fri, 13 Sep 2019 08:45:54 +0000 Subject: [Info-ingres] Why x100 rocks Message-ID: Hi All, If you haven't played with x100 tables in 11.1 then it's time you looked at it. We gave up on this query after watching it for over 3 hours grinding away on standard ingres table structures... SELECT primary_sub.eid, primary_sub.ins_index, primary_sub.primary_level, primary_sub.primary_diagnosis, secondary.level AS secondary_level, secondary.diag_icd10 AS secondary_diagnosis, hesin_oper.oper4 AS operation_code, hesin_oper.opdate AS operation_date FROM ( SELECT hesin.eid, hesin.ins_index, primary.diag_icd10 AS primary_diagnosis, primary.level AS primary_level FROM hesin INNER JOIN hesin_diag AS primary on hesin.eid = primary.eid AND hesin.ins_index = primary.ins_index WHERE primary.level = 1 AND primary.diag_icd10 = 'Z966' ) AS primary_sub INNER JOIN hesin_diag AS secondary on primary_sub.eid = secondary.eid AND primary_sub.ins_index = secondary.ins_index INNER JOIN hesin_oper on primary_sub.eid = hesin_oper.eid AND primary_sub.ins_index = hesin_oper.ins_index WHERE secondary.level = 2 UNION SELECT secondary_sub.eid, secondary_sub.ins_index, primary.level AS primary_level, primary.diag_icd10 AS primary_diagnosis, secondary_sub.secondary_level, secondary_sub.secondary_diagnosis, hesin_oper.oper4 AS operation_code, hesin_oper.opdate AS operation_date FROM ( SELECT hesin.eid, hesin.ins_index, secondary.diag_icd10 AS secondary_diagnosis, secondary.level AS secondary_level FROM hesin INNER JOIN hesin_diag AS secondary on hesin.eid = secondary.eid AND hesin.ins_index = secondary.ins_index WHERE secondary.level = 2 AND secondary.diag_icd10 = 'Z966' ) AS secondary_sub INNER JOIN hesin_diag AS primary on secondary_sub.eid = primary.eid AND secondary_sub.ins_index = primary.ins_index INNER JOIN hesin_oper on secondary_sub.eid = hesin_oper.eid AND secondary_sub.ins_index = hesin_oper.ins_index WHERE primary.level = 1; So after a fair deal of work we converted the tables in question (which are essentially static) to be x100 tables and reran the query ... 72seconds! X100! Not just a Fuji camera, or a rust inhibitor. Marty -------------- next part -------------- An HTML attachment was scrubbed... URL: From specially at processed.almost.meat Fri Sep 13 09:41:31 2019 From: specially at processed.almost.meat (Roy Hann) Date: Fri, 13 Sep 2019 09:41:31 +0000 (UTC) Subject: [Info-ingres] Duplicate sequence values Message-ID: We've got a database that is cloned overnight from another system. In iisequences I see that next_value is (e.g.) 1200. However when I select next value for my_seq; I get a value like 889, which usually turns out to be a duplicate. The cached state of the sequence seems to be retained even though the database was closed for the cloning operation. I know we're fooling about with checkpoints in a not-totally sanctioned way, but surely when the database is opened by the first user the cached value should be refreshed? I can call it a bug right? Roy PS: It's just a nuisance, not a show-stopper. As a workaround I do alter sequence my_seq increment by 1; From nikos.vaggalis at gmail.com Sat Sep 14 09:54:14 2019 From: nikos.vaggalis at gmail.com (nikosv) Date: Sat, 14 Sep 2019 02:54:14 -0700 (PDT) Subject: [Info-ingres] Issue with characters being inserted as garbabge Message-ID: <80c3eda8-f609-4446-8d60-8c5253c1e0ee@googlegroups.com> I'm sending a soap XML message to an Apache server which runs a Perl CGI script which in turn acts as the soap server. The script interacts with Ingres, reading and inserting data.Perl and Ingres are tuned to understand iso8859/7 Greek. While I can retrieve correct Greek from the database,when I attempt to insert data and retrieve it back I get garbage .Even running the script locally at the machine that hosts Apache the same happens. For example for Greek character A I get a sequence of ?\201 .I don't know what that means.In other cases I get some other weird characters. The thing is that in a previous installation on another machine everything worked out of the box,so I'm wondering whether there is something wrong with the configuration. It would be of great help in deciphering it if anyone could understand why Ingres uses that sequence of characters described above. Thanks From nksvg1 at gmail.com Mon Sep 16 13:24:03 2019 From: nksvg1 at gmail.com (nksvg1 at gmail.com) Date: Mon, 16 Sep 2019 06:24:03 -0700 (PDT) Subject: [Info-ingres] Issue with characters being inserted as garbabge In-Reply-To: <80c3eda8-f609-4446-8d60-8c5253c1e0ee@googlegroups.com> References: <80c3eda8-f609-4446-8d60-8c5253c1e0ee@googlegroups.com> Message-ID: OK I figured it out. The issue is too lengthy to describe in a forum's post,so I'm going to write about it elsewhere. If anyone is interested in it, just ping me or leave a note here and I'll let you know when it's ready From nikos.vaggalis at gmail.com Mon Sep 16 18:53:19 2019 From: nikos.vaggalis at gmail.com (nikosv) Date: Mon, 16 Sep 2019 11:53:19 -0700 (PDT) Subject: [Info-ingres] Issue with characters being inserted as garbabge In-Reply-To: <80c3eda8-f609-4446-8d60-8c5253c1e0ee@googlegroups.com> References: <80c3eda8-f609-4446-8d60-8c5253c1e0ee@googlegroups.com> Message-ID: OK I figured it out. The issue is too lengthy to describe in a forum's post,so I'm going to write about it elsewhere. If anyone is interested in it, just ping me or leave a note here and I'll let you know when it's ready From Darren.Harvey at fusion5.com.au Wed Sep 25 06:45:08 2019 From: Darren.Harvey at fusion5.com.au (Darren Harvey) Date: Wed, 25 Sep 2019 06:45:08 +0000 Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Message-ID: Hi I am attempting to export data from an Ingres Database using the COPY TABLE statement, but having problems with fields defined as 'decimal' with negative values. The screen shot below shows part of 2 records being exported into a text output file. The 1st record has a decimal value of 78.00000 and displays correctly. However the 2nd record has a negative decimal value of -0.40000 and the output places an '8' in front of the value, which I assume is to indicate the value has 8 individual characters (-0.40000). The format in the COPY TABLE statement for this field is 'VARCHAR(0)' [cid:image003.png at 01D573BC.3FB155D0] When we try to import this data into a non-Ingres database table it fails because of the character '8' at the start of the data for that field. I have tried outputting this field with a Data Type of 'Decimal(0)', 'Decimal(11,5)' and 'Float(0)' in the COPY TABLE statement, but the output is in a Binary format which again is unusable for the Import process. [cid:image001.png at 01D573BE.8ECF2BE0] I have tested the Import we are attempting to use with the field in the text file just displaying as '-0.40000' instead of '8-0.40000', and the import works fine. Does anyone know if there is a way to output 'negative' values without that leading 'length' value? Thanks in Advance Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: info-ingres-bounces at lists.planetingres.org On Behalf Of Darren Harvey Sent: Monday, 29 July 2019 9:29 AM To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Performance of Primary Key v Index Hi All If I have a table that has a Primary Key with 2 columns in the key, does the Primary Key provide the same performance for selecting data on those 2 columns as an Index would with the same 2 columns? Thanks Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] [cid:image005.jpg at 01D573BC.3FB155D0] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy Message protected by MailGuard: e-mail anti-virus, anti-spam and content filtering. http://www.mailguard.com.au/mg Report this message as spam -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 2713 bytes Desc: image003.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image005.jpg Type: image/jpeg Size: 43812 bytes Desc: image005.jpg URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 2542 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.png Type: image/png Size: 900 bytes Desc: image002.png URL: From martin.bowes at ndph.ox.ac.uk Wed Sep 25 06:56:47 2019 From: martin.bowes at ndph.ox.ac.uk (Martin Bowes) Date: Wed, 25 Sep 2019 06:56:47 +0000 Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values In-Reply-To: References: Message-ID: <2a439d3dc6134b51b218c228880010e6@EXCHANGE1.ndph.ox.ac.uk> Hi Darren, The varchar(0) format prefixes the length to the data. It did that on all lines so your 78.00000 is actually 8.00000 it's 7 characters long. Change the varchar(0) to a c0 and you should be OK. Martin Bowes From: Darren Harvey [mailto:Darren.Harvey at fusion5.com.au] Sent: 25 September 2019 07:45 To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi I am attempting to export data from an Ingres Database using the COPY TABLE statement, but having problems with fields defined as 'decimal' with negative values. The screen shot below shows part of 2 records being exported into a text output file. The 1st record has a decimal value of 78.00000 and displays correctly. However the 2nd record has a negative decimal value of -0.40000 and the output places an '8' in front of the value, which I assume is to indicate the value has 8 individual characters (-0.40000). The format in the COPY TABLE statement for this field is 'VARCHAR(0)' [cid:image003.png at 01D57376.C7A0B170] When we try to import this data into a non-Ingres database table it fails because of the character '8' at the start of the data for that field. I have tried outputting this field with a Data Type of 'Decimal(0)', 'Decimal(11,5)' and 'Float(0)' in the COPY TABLE statement, but the output is in a Binary format which again is unusable for the Import process. [cid:image004.png at 01D57376.C7A0B170] I have tested the Import we are attempting to use with the field in the text file just displaying as '-0.40000' instead of '8-0.40000', and the import works fine. Does anyone know if there is a way to output 'negative' values without that leading 'length' value? Thanks in Advance Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: info-ingres-bounces at lists.planetingres.org On Behalf Of Darren Harvey Sent: Monday, 29 July 2019 9:29 AM To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Performance of Primary Key v Index Hi All If I have a table that has a Primary Key with 2 columns in the key, does the Primary Key provide the same performance for selecting data on those 2 columns as an Index would with the same 2 columns? Thanks Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] [cid:image007.jpg at 01D57376.C7A0B170] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy Message protected by MailGuard: e-mail anti-virus, anti-spam and content filtering. http://www.mailguard.com.au/mg Report this message as spam -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 2713 bytes Desc: image003.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image004.png Type: image/png Size: 2542 bytes Desc: image004.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image006.png Type: image/png Size: 900 bytes Desc: image006.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image007.jpg Type: image/jpeg Size: 43812 bytes Desc: image007.jpg URL: From adrian.williamson at rationalcommerce.com Wed Sep 25 07:11:07 2019 From: adrian.williamson at rationalcommerce.com (Adrian Williamson) Date: Wed, 25 Sep 2019 08:11:07 +0100 Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values In-Reply-To: References: Message-ID: <001d01d57370$673f22e0$35bd68a0$@rationalcommerce.com> Hi Darren, When I am moving things around systems I use the "copydb -c" command to give me the 'ASCII' i.e. non Binary syntax for extracting data. e.g. copydb -c -uowner mydatabase mytable This gives me a copy.out and in which I cannibalise. Apologies if you have already tried this, but the phrase "but the output is in a Binary format" made me think other wise. Good luck Adrian From: info-ingres-bounces at lists.planetingres.org On Behalf Of Darren Harvey Sent: 25 September 2019 07:45 To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi I am attempting to export data from an Ingres Database using the COPY TABLE statement, but having problems with fields defined as 'decimal' with negative values. The screen shot below shows part of 2 records being exported into a text output file. The 1st record has a decimal value of 78.00000 and displays correctly. However the 2nd record has a negative decimal value of -0.40000 and the output places an '8' in front of the value, which I assume is to indicate the value has 8 individual characters (-0.40000). The format in the COPY TABLE statement for this field is 'VARCHAR(0)' When we try to import this data into a non-Ingres database table it fails because of the character '8' at the start of the data for that field. I have tried outputting this field with a Data Type of 'Decimal(0)', 'Decimal(11,5)' and 'Float(0)' in the COPY TABLE statement, but the output is in a Binary format which again is unusable for the Import process. I have tested the Import we are attempting to use with the field in the text file just displaying as '-0.40000' instead of '8-0.40000', and the import works fine. Does anyone know if there is a way to output 'negative' values without that leading 'length' value? Thanks in Advance Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: info-ingres-bounces at lists.planetingres.org

> On Behalf Of Darren Harvey Sent: Monday, 29 July 2019 9:29 AM To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Performance of Primary Key v Index Hi All If I have a table that has a Primary Key with 2 columns in the key, does the Primary Key provide the same performance for selecting data on those 2 columns as an Index would with the same 2 columns? Thanks Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy Message protected by MailGuard: e-mail anti-virus, anti-spam and content filtering. http://www.mailguard.com.au/mg Report this message as spam -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 2713 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image004.png Type: image/png Size: 2542 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image006.png Type: image/png Size: 900 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image007.jpg Type: image/jpeg Size: 43812 bytes Desc: not available URL: From Darren.Harvey at fusion5.com.au Thu Sep 26 00:10:26 2019 From: Darren.Harvey at fusion5.com.au (Darren Harvey) Date: Thu, 26 Sep 2019 00:10:26 +0000 Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Message-ID: Hi Martin Using 'c0' worked perfectly. Thank you very much. Cheers Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: Martin Bowes Sent: Wednesday, 25 September 2019 4:57 PM To: Darren Harvey ; info-ingres at lists.planetingres.org Subject: RE: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi Darren, The varchar(0) format prefixes the length to the data. It did that on all lines so your 78.00000 is actually 8.00000 it's 7 characters long. Change the varchar(0) to a c0 and you should be OK. Martin Bowes From: Darren Harvey [mailto:Darren.Harvey at fusion5.com.au] Sent: 25 September 2019 07:45 To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi I am attempting to export data from an Ingres Database using the COPY TABLE statement, but having problems with fields defined as 'decimal' with negative values. The screen shot below shows part of 2 records being exported into a text output file. The 1st record has a decimal value of 78.00000 and displays correctly. However the 2nd record has a negative decimal value of -0.40000 and the output places an '8' in front of the value, which I assume is to indicate the value has 8 individual characters (-0.40000). The format in the COPY TABLE statement for this field is 'VARCHAR(0)' [cid:image002.png at 01D57452.9CBD4E50] When we try to import this data into a non-Ingres database table it fails because of the character '8' at the start of the data for that field. I have tried outputting this field with a Data Type of 'Decimal(0)', 'Decimal(11,5)' and 'Float(0)' in the COPY TABLE statement, but the output is in a Binary format which again is unusable for the Import process. [cid:image003.png at 01D57452.9CBD4E50] I have tested the Import we are attempting to use with the field in the text file just displaying as '-0.40000' instead of '8-0.40000', and the import works fine. Does anyone know if there is a way to output 'negative' values without that leading 'length' value? Thanks in Advance Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: info-ingres-bounces at lists.planetingres.org

> On Behalf Of Darren Harvey Sent: Monday, 29 July 2019 9:29 AM To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Performance of Primary Key v Index Hi All If I have a table that has a Primary Key with 2 columns in the key, does the Primary Key provide the same performance for selecting data on those 2 columns as an Index would with the same 2 columns? Thanks Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] [cid:image004.jpg at 01D57452.9CBD4E50] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy Message protected by MailGuard: e-mail anti-virus, anti-spam and content filtering. http://www.mailguard.com.au/mg Report this message as spam Message protected by MailGuard: e-mail anti-virus, anti-spam and content filtering. http://www.mailguard.com.au/mg Report this message as spam -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 900 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.png Type: image/png Size: 2713 bytes Desc: image002.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 2542 bytes Desc: image003.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image004.jpg Type: image/jpeg Size: 43812 bytes Desc: image004.jpg URL: From Darren.Harvey at fusion5.com.au Thu Sep 26 00:12:36 2019 From: Darren.Harvey at fusion5.com.au (Darren Harvey) Date: Thu, 26 Sep 2019 00:12:36 +0000 Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values In-Reply-To: <001d01d57370$673f22e0$35bd68a0$@rationalcommerce.com> References: <001d01d57370$673f22e0$35bd68a0$@rationalcommerce.com> Message-ID: Hi Adrian Thanks very much for your response, which would have worked. But Martin's suggestion to just use 'c0' was a simple change and worked. Cheers Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: Adrian Williamson Sent: Wednesday, 25 September 2019 5:11 PM To: Darren Harvey ; info-ingres at lists.planetingres.org Subject: RE: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi Darren, When I am moving things around systems I use the "copydb -c" command to give me the 'ASCII' i.e. non Binary syntax for extracting data. e.g. copydb -c -uowner mydatabase mytable This gives me a copy.out and in which I cannibalise. Apologies if you have already tried this, but the phrase "but the output is in a Binary format" made me think other wise. Good luck Adrian From: info-ingres-bounces at lists.planetingres.org

> On Behalf Of Darren Harvey Sent: 25 September 2019 07:45 To: info-ingres at lists.planetingres.org Subject: [Info-ingres] Issues with COPY TABLE statement with negative DECIMAL Values Hi I am attempting to export data from an Ingres Database using the COPY TABLE statement, but having problems with fields defined as 'decimal' with negative values. The screen shot below shows part of 2 records being exported into a text output file. The 1st record has a decimal value of 78.00000 and displays correctly. However the 2nd record has a negative decimal value of -0.40000 and the output places an '8' in front of the value, which I assume is to indicate the value has 8 individual characters (-0.40000). The format in the COPY TABLE statement for this field is 'VARCHAR(0)' [cid:image002.png at 01D57452.EA488860] When we try to import this data into a non-Ingres database table it fails because of the character '8' at the start of the data for that field. I have tried outputting this field with a Data Type of 'Decimal(0)', 'Decimal(11,5)' and 'Float(0)' in the COPY TABLE statement, but the output is in a Binary format which again is unusable for the Import process. [cid:image003.png at 01D57452.EA488860] I have tested the Import we are attempting to use with the field in the text file just displaying as '-0.40000' instead of '8-0.40000', and the import works fine. Does anyone know if there is a way to output 'negative' values without that leading 'length' value? Thanks in Advance Darren Darren Harvey Senior Developer M +61 400 398 188 P +61 3 9922 5519 Skype for Business ID darren.harvey at fusion5.com.au www.fusion5.com.au [Follow us on LinkedIn] Fusion5 | Level 27, 570 Bourke Street, Melbourne, 3000, Australia This email and any attachments are confidential and intended exclusively for the person to whom the email is addressed. Please see our Privacy Policy From: info-ingres-bounces at lists.planetingres.org