November 2006 - users-list

[MonetDB-users] XQuery API and XLink/XUpdate support
by Alexandra Mart�nez 27 Jun '24

27 Jun '24

Hello all, I was wondering if there is any documentation about the API for XQuery (i.e., how to write xqueries from a Java program). Also, I am interested in using some features of XLink and XUpdate on top of my XML database. Is support for any of these in progress? Thanks in advance, Alexandra ______________________________________________________ Click here to donate to the Hurricane Katrina relief effort. http://store.yahoo.com/redcross-donate3/

4 3

Re: [MonetDB-users] Using MonetDB / XQuery with PHP
by Menno Holtkamp 13 Dec '06

13 Dec '06

>Unfortunately I don't have too much time this and next week, >so the fix will have to wait a bit. I'll try and see in the weekend if >I can easily fix it. Ok, in the mean time my work around is as follows: 1 - make a temporary .xq file somwhere on your file-system using PHP 2 - invoke this file using: $output = shell_exec('MapiClient.bat -l xquery tempXQFile.xq'); header("Content-Type: text/xml"); echo $output; But offcourse a proper database-connection is preferred. Looking forward to some kind of solution! With kind regards, Menno Holtkamp -----Original Message----- From: Fabian Groffen [mailto:Fabian.Groffen@cwi.nl] Sent: Thu 11/23/2006 2:35 PM To: monetdb-users(a)lists.sourceforge.net Cc: Subject: Re: [MonetDB-users] Using MonetDB / XQuery with PHP On 23-11-2006 14:21:44 +0100, Menno Holtkamp wrote: [snip] > Performing a query with: > > $query = 'count(doc("HelloWorld.xml")/doc)'; > $result = monetdb_query($query); > if(!$result)die("Query failed : " . monetdb_error() ."\n"); > else echo '<br>Performed query'; > > succeeds as well. > > But the content of the result set seems to remain 'empty'. > > print_r($result) -> Resource id #3 > > But > while($row = monetdb_fetch_array($result)) { > print_r($row); > } > > remains empty. I see. This is probably because the default XQuery output is a wee bit different. ;) I might be able to fix monetdb_php to support XQuery like JDBC does. Unfortunately I don't have too much time this and next week, so the fix will have to wait a bit. I'll try and see in the weekend if I can easily fix it. > Does somebody have a working example of performing an XQuery using PHP? No ;) > I wondered, is it required to connect to a specific database in the monetdb_connect command? (the XML-document) I think not, but hey! Nope; this only works on M5 which has support for multiple catalogs. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ MonetDB-users mailing list MonetDB-users(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

3 2

[MonetDB-users] SQL BLOB minor issue
by Ronald Oussoren 12 Dec '06

12 Dec '06

Hi, This is probably bad SQL-fu on my part, but you never know :-). I'm trying to insert a 0-byte blob into a table, but that doesn't work for me. The SQL code below fails on the second insert, the database complains about a bad blob syntax. Inserting an implicit NULL into a BLOB colum also fails (the 3th insert), even though inserting an explicit NULL (the fourth one) does succeed. CREATE TABLE testtable ( id INTEGER data BLOB ); INSERT INTO testtable (id, data) VALUES (0, BLOB '00'); INSERT INTO testtable (id, data) VALUES (1, BLOB ''); INSERT INTO testtable (id) VALUES (2); INSERT INTO testtable (id, data) VALUES (2, NULL); Regards, Ronald

2 2

Re: [MonetDB-users] Large transactions or large updates
by Ronald Oussoren 12 Dec '06

12 Dec '06

On 21 Nov 2006, at 2:57 PM, Ronald Oussoren wrote: > > I'll see if I can create a script that demonstrates the problem, I found the problem, my data was not as clean as I expected and I tried to insert a non-ascii value into a string column in the database. That still leaves the lame error reporting (raising RuntimeError (None)) in the MonetSQLdb wrappers. The MapiClient does report a useful error and I've tried to compare this with the Python bindings. It turns out that the error-checking code in the python binding has ample room for improvement. I debugged my code by adding calls to mapi_error, mapi_result and mapi_explain_query to the C code of the mapi_query wrapper, but that's obviously not the right solution. Ronald

2 1

[MonetDB-users] Bulk load
by JLP00993＠correo.aeat.es 30 Nov '06

30 Nov '06

5 6

Re: [MonetDB-users] About Static Type Analysis for XQuery in MonetDB/XQuery
by Jens Teubner 29 Nov '06

29 Nov '06

On Wed, Nov 29, 2006 at 08:40:59PM +0800, Wang Zhen (Selina) wrote: > As you said, your work "goes beyond the actual requirements in the W3C > specs", is it because in some aspect, the requirements in the W3C > specs is not enough? Or, is it because the implementation of W3C > specs are difficult or cause some problems? When the development of Pathfinder started, it had not yet been clear what kind of subtype relationship the XQuery language would use. For quite a while, the W3C drafts depended on /structural subtyping/, which (computationally) is much more expensive than the /named typing/ that you find now in the W3C documents. The XML structure leads to tree-structured types. The structural subtype relationship is now decided by looking at the full structure of both types and testing, whether each instance of the one type will always be an instance of the other type. Example: Consider the two types t1 := element a { element b { item* } } t2 := element a { element b { item* }, element c { item* }? } (that is, an element `a' with a single `b' child of arbitrary content, and an element `a' with a `b' and an optional `c' element in its content) In this example, each instance of type t1 will always be an instance of type t2, thus t1 <: t2. Testing this relationship can be quite expensive, which probably is why structural typing didn't make it into the current Candidate Recommendation of the specs. These specs now use named typing. Here, it is assumed that items are annotated with a type /name/ and that there is a hierarchy (the derivation hierarchy) between type names. To test a subtype relationship, one only has to look at the two names now, but /not/ at their full structure. The Pathfinder compiler includes a full implementation of a structural subtype algorithm (we use the one described in [1,2]). In some occasions, this allows the inference of a more precise type information. Our system does not yet include validation support, but we expect that the structural type information can help to avoid an expensive validation if we can statically derive that validation will succeed. > Besides, do you provide a formal document, or any tutorials about the > actual static typing you have achieved? Sorry, but we haven't. You find the subtype algorithm we use in the references below. Our inference rules more or less follow the W3C specs, though the implementation needs quite some work. Regards, Jens [1] Martin Kempa, Volker Linnemann. Type Checking in XOBE. BTW 2003. [2] Martin Kempa. Programmierung von XML-basierten Anwendungen unter Berücksichtigung der Sprachbeschreibung. PhD thesis, U Lübeck, Germany, 2003. -- Jens Teubner Technische Universitaet Muenchen, Department of Informatics D-85748 Garching, Germany Tel: +49 89 289-17259 Fax: +49 89 289-17263 No scientist is ever right, they just can't be proved wrong at the time! -- Richard Feynman

1 0

[MonetDB-users] appealed
by TFTP Edit 29 Nov '06

29 Nov '06

Do install searching functions outlined user guide pp? Copyrights or else heavy, fine responded selling. Los angeles riaa mpaa lawsuit is still. Fast called supernodes located sourced has. Networks australia and vanuatu, late march appeal reversed an. Own legal, action, massive copyright breaches! Amsterdam nederland ironically, forced our hosting provider uses, peertopeer. Company, modify within months. Bumastemra november ordered take steps prevent from violating copyrights? Official became more widespread developers used their ability? Greeted message important, notice download desktop permitted? Associate kevin bermeister knowingly. Bodies restrict use sharing. Up users morpheus program formerly, client musiccity however. Deliver collect system settings latest adds. Ordered take steps prevent. Boss nikki hemming associate kevin bermeister. Need central point management all you do. Supreme reached case mgm studios favor september. Responded selling mesh offshore companies.

1 0

[MonetDB-users] Connecting to iSQL-Viewer
by JLP00993＠correo.aeat.es 29 Nov '06

29 Nov '06

2 1

Re: [MonetDB-users] About Static Type Analysis for XQuery in MonetDB/XQuery
by Stefan Manegold 28 Nov '06

28 Nov '06

Dear Selina, Wang Zhen, thank you very much for your interest in MonetDB/XQuery. Assuming that your question can best be answered by our colleagues in Munich that take care of the Pathfinder XQuery to relational algebra compiler (cf., http://www.pathfinder-xquery.org/) I feel free to forward your email to the MonetDB-users list at SourceForge. Kind regards, Stefan On Tue, Nov 14, 2006 at 07:49:39PM +0800, Wang Zhen (Selina) wrote: > To Who It may Concern: > > I am conducting an XQuery implementation survey for my thesis study, and > find your MonetDB/XQuery link from W3C website. > > I'm very interested in MonetDB/XQuery, yet I still got some questions > and may need your help: > > 1. I found on your website that, MonetDB/XQuery "adheres to all the > typing rules prescribed in the W3C standard". > I would like to know which version do you support currently? > > 2. Are those typing rules encoded in MonetDB/XQuery? If so, since the > W3C standard is still changing, how do you keep up with > the change efficiently? > > Thank you. > Selina, Wang Zhen -- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4312 |

3 2

[MonetDB-users] Next Big market Winner!
by Jazz singer 27 Nov '06

27 Nov '06

1 0