BaseX-Talk

basex-talk@mailman.uni-konstanz.de

1 participants
5252 discussions

Root?
by Charles F. Munat 04 Jan '11

04 Jan '11

I have an embedded Base-X database in a web application. I have a huge XQuery that involves a lot of local functions, many of them recursive. In several of these functions, I need to query the database to look up referenced elements. I can't for the life of me figure out a way to access the database without doing one of two things: 1. My document root is /data. I can let $root := /data and then pass that into every function. Passing things into functions rather than setting global variable is certainly and FP way of doing things, but in this instance it seems to me that it should be the scope I'm working in. 2. If something else I pass into the function is a node in the database, then I can, for example, use $node/root() to back up to the root node. But sometimes I'm only passing in a string and I still need to query the whole database. Is there some way to access the root easily from within a function? I am very confused by the whole "context" idea. I've looked all over, but I must be searching using the wrong terms, or it simply isn't done. Anyone know a way? Thanks, Chas.

3 3

Fwd: Re: Root?
by Andreas Weiler 04 Jan '11

04 Jan '11

-------- Original-Nachricht -------- Betreff: Re: [basex-talk] Root? Datum: Tue, 04 Jan 2011 08:56:13 -0300 Von: Charles F. Munat <charles.munat(a)gmail.com> An: Andreas Weiler <andreas.weiler(a)uni-konstanz.de> Actually, this is perfect. Exactly what I was trying to do -- set a global variable. I must have had the syntax wrong, because it wasn't working. But this is great. One less thing to pass around. When you're query involves a couple dozen functions, it gets tiresome. Thanks much for all your help! This makes it a lot easier. Chas. On 01/04/2011 8:24 AM, Andreas Weiler wrote: > Hi Chas, > > thanks for you example, that made it easy to understand. > The database context isn't available in own declared functions and > variables. > So you could use the doc('database') function and run the following query: > > declare variable $data as node() := doc('database')/data; > > declare function local:getPersonFromID($id) { > $data/persons/person[@id=$id] > }; > > local:getPersonFromID('1') > > So it's almost the same than your workaround. > The doc('database') function gives you access to your database. > > I'm sorry there is no easier way, > kind regards, > Andreas > > Am 04.01.11 11:26, schrieb Charles F. Munat: >> Sorry, it's hard to explain these things since I have no idea what, if >> anything, I'm doing wrong. Here's a basic example. >> >> I create a database in code: >> >> (I'm using Scala, but hopefully this is close enough so you can see >> what I'm doing.) >> >> private val context = new Context() >> new org.basex.core.cmd.Set("dbpath", "/var/www/xxx/db").execute(context) >> >> try { >> new Open("data").execute(context) >> } catch { >> case _ => new CreateDB("data", "<data/>").execute(context) >> } >> >> So now I have a new database that contains only<data/>. I now load it: >> >> val file = "/var/www/xxx/content/persons.xml" >> new XQuery("insert node " + >> XML.loadFile(file).toString + >> " into /data").execute(context) >> >> Now my data is in the embedded database. Suppose the file looked like >> this: >> >> <persons> >> <person id="1"> >> <names> >> <given>John</given> >> <family>Lennon</family> >> </names> >> </person> >> <person> >> <names id="2"> >> <given>Paul</given> >> <family>McCartney</family> >> </names> >> </person> >> </persons> >> >> Now I can run an XQuery like this (really just XPath): >> >> /data/persons/person[@id='1'] >> >> And I'll get back: >> >> <person id="1"> >> <names> >> <given>John</given> >> <family>Lennon</family> >> </names> >> </person> >> >> Simple, right? >> >> But if I do this: >> >> declare function local:getPersonFromID($id) { >> /data/persons/person[@id=$id] >> }; >> >> local:getPersonFromID('1') >> >> Then I get this error: >> >> No context item set for 'root()'. >> >> The workaround I've been using is this: >> >> declare function local:getPersonFromID($data, $id) { >> $data/persons/person[@id=$id] >> }; >> >> let $data := /data >> return local:getPersonFromID('1') >> >> There must be a better way. >> >> Does this make more sense? >> >> Chas. >> >> On 01/04/2011 5:47 AM, Andreas Weiler wrote: >>> Hi Chas, >>> >>> i actually don't understand what your concrete problem is, but you could >>> have a look >>> in our xquery database functions: >>> http://www.inf.uni-konstanz.de/dbis/basex/xq-db >>> >>> If you have opened the database before running queries, the whole >>> database is in your context and so it is used >>> in the queries. >>> >>> Kind regards, >>> Andreas >>> >>> Am 04.01.11 00:54, schrieb Charles F. Munat: >>>> I have an embedded Base-X database in a web application. I have a huge >>>> XQuery that involves a lot of local functions, many of them recursive. >>>> In several of these functions, I need to query the database to look up >>>> referenced elements. I can't for the life of me figure out a way to >>>> access the database without doing one of two things: >>>> >>>> 1. My document root is /data. I can let $root := /data and then pass >>>> that into every function. Passing things into functions rather than >>>> setting global variable is certainly and FP way of doing things, but >>>> in this instance it seems to me that it should be the scope I'm >>>> working in. >>>> >>>> 2. If something else I pass into the function is a node in the >>>> database, then I can, for example, use $node/root() to back up to the >>>> root node. But sometimes I'm only passing in a string and I still need >>>> to query the whole database. >>>> >>>> Is there some way to access the root easily from within a function? I >>>> am very confused by the whole "context" idea. I've looked all over, >>>> but I must be searching using the wrong terms, or it simply isn't done. >>>> >>>> Anyone know a way? >>>> >>>> Thanks, >>>> >>>> Chas. >>>> _______________________________________________ >>>> BaseX-Talk mailing list >>>> BaseX-Talk(a)mailman.uni-konstanz.de >>>> https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk >>> >

1 0

Fwd: RE: Database creation from a directory tree
by Andreas Weiler 03 Jan '11

03 Jan '11

-------- Original-Nachricht -------- Betreff: RE: [basex-talk] Database creation from a directory tree Datum: Fri, 31 Dec 2010 17:59:34 +0100 Von: David leigh <david(a)leighweb.com> Antwort an: <david(a)leighweb.com> An: 'Andreas Weiler' <andreas.weiler(a)uni-konstanz.de> Wohoo! That works great! Thanks. You've provided the missing link for me! Have a great new year! David DavidEmailSig <mailto:david@leighweb.com> *From:*Andreas Weiler [mailto:andreas.weiler@uni-konstanz.de] *Sent:* Friday, December 31, 2010 04:41 PM *To:* david(a)leighweb.com *Subject:* Re: [basex-talk] Database creation from a directory tree Hi David, thanks a lot for your tests. I couldnt fix the gui bug yet...but i can help you with command line... There is a createfilter option...it can be set with: set createfilter * so all files in the folder are recognized for creating a database. I hope this works for now, Happy new year, Andreas Am 31.12.10 15:13, schrieb David leigh: Andreas, I just did an interesting test (using the version that the windows installer installs: 6.3.4). I copied my directory tree of files and in the new copy, I gave all my data files an extension of "xml" (previously they had NO extension). NOW, both with the command line and the GUI, the full database is created without any problem. To double check, I went back to the old directory with no extensions and created a new database using a filter of "*.*" and only 14 documents were created in the database (instead of 5000+). Conclusions: In the command line you must have input documents with an extension of "xml" since there is no visible way to provide a filter. The GUI interfaces differently than the command line with the basex core since it can receive a filter as well as a path in the database creation process The GUI filter functionality in 6.3.4 is different than in 6.1...for me it is "broken" vs. how it works in 6.1 So, I think for my project to work completely, I need new functionality: a file name filter that works correctly on the CREATE DB statement in the command line interface. Does this sound correct to you? Unfortunately, I don't have control over the extension names of my input files as they are created by another piece of software. Until I can do a create database in the command line, I'll have to do manual gui-based (6.1 version) refreshes of the data or some sort of long copy/rename process in the background. Thanks! David DavidEmailSig <mailto:david@leighweb.com> *From:*Andreas Weiler [mailto:andreas.weiler@uni-konstanz.de] *Sent:* Wednesday, December 29, 2010 11:26 AM *To:* david(a)leighweb.com <mailto:david@leighweb.com>; BaseX-Talk(a)mailman.uni-konstanz.de <mailto:BaseX-Talk@mailman.uni-konstanz.de> *Subject:* Re: [basex-talk] Database creation from a directory tree Hi David, the syntax of the create command changed between the releases of 6.1 and 6.3.4. In release 6.1 you have to use: create db C:\Users\David\Documents\OpenSong\songpacks\Songs songs Kind regards, Andreas Am 28.12.10 18:00, schrieb David leigh: Here are my command line results from 6.3.4: C:\Users\David\Documents\OpenSong\OpenSongSearch\BaseX>java -cp BaseX.jar org.basex.BaseXClient -Uadmin -Padmin BaseX 6.3.5 [Client] Try "help" to get more information. > drop db songs Database 'songs' was dropped. > create db song Database 'song' created in 1683.61 ms. > info index Tags - Structure: Hash - Entries: 0 Attributes - Structure: Hash - Entries: 0 Text Index - Structure: Binary tree - Size: 4 Bytes - Entries: 0 Attribute Index - Structure: Binary tree - Size: 4 Bytes - Entries: 0 Full-Text Index - Not available Path Summary doc() 1x > info table Stopped at line 1, column 10: Syntax: INFO ([DATABASE|INDEX|STORAGE]) Show information on current database. Shows information on the currently opened database: - no argument: show global information - DATABASE: shows database information - INDEX: shows index information - STORAGE [start end] | [query]: show internal database table > info storage PRE DIS SIZ ATS NS KIND CONTENT ------------------------------------- 0 1 1 1 0 DOC song > info General Information Database Path: C:\Users\David\Documents\OpenSong\OpenSongSearch\BaseX\data Used Main Memory: 895 KB Database Creation Whitespace Chopping: ON Entity Parsing: OFF Indexes Path Summary: ON Text Index: ON Attribute Index: ON Full-Text Index: OFF > So, for 6.3.4, it creates from the command line, but it's empty. The root of that path has NO files in it. The 5861 files are in all the subdirectories. Then, when I try the same syntax with 6.1, I get the following results: > create db songs C:\Users\David\Documents\OpenSong\songpacks\Songs Stopped at line 1, column 17: Syntax: CREATE [DB|COLL|FS|INDEX|USER] [...] Create database, index or user. Creates a new database, index or user: - DB [path] [name?]: creates the database [name] for the file or directory [path] - COLL [name]: creates an empty database [name] - INDEX [TEXT|ATTRIBUTE|FULLTEXT|PATH]: creates the specified index - FS [path] [name] ([mountpoint] [backingstore]): creates filesystem database [name] for [path] (using [mountpoint] and [backingstore] if FUSE is activated) - USER [name] [password?]: creates the specified user Summary: 6.1 GUI -- works as I want it 6.1 command line -- can't get it to create 6.3.4 GUI -- creates but with only 14 entries 6.3.4 command line -- creates, but it's empty I don't mind staying with 6.1, but I DO want to do everything with the command line. Is the ADD syntax the same, in that I can simply specify the path and it will recurse the whole directory tree? Thanks!!! David DavidEmailSig <mailto:david@leighweb.com> *From:*Andreas Weiler [mailto:andreas.weiler@uni-konstanz.de] *Sent:* Tuesday, December 28, 2010 01:29 PM *To:* david(a)leighweb.com <mailto:david@leighweb.com> *Cc:* basex-talk(a)mailman.uni-konstanz.de <mailto:basex-talk@mailman.uni-konstanz.de> *Subject:* Re: [basex-talk] Database creation from a directory tree Additional question: That was in the GUI. I'd like to do the same thing using the command line with the .JAR file, but somewhere my syntax is not right. Additionally, all the XML files have NO EXTENSION so I don't know how to specify that either in the command line. So, any help in that area would also be appreciated. Could you provide your syntax? The syntax should be: create db dbname path/to/folder The missing extension shouldnt be a problem, i just tested it. In the GUI in 6.1, I can create a database from a whole directory hierarchy of 5,681 individual XML files. When I point to the same source file directory and using the 6.3.4 GUI (which I installed with a pre-release version of the installer), it only processes 14 files. No error message (that I saw) was produced. The behavior that I'm wanting is what 6.1 does. Glad to help in solving the problem. We will check for modifications between the releases. As quick workaround you could try to create an empty database and then add the documents with the Database/Add documents menu. Kind regards, Andreas Am 28.12.10 12:55, schrieb David leigh: I've just installed 6.3.4 and I've noticed a difference from version 6.1. In the GUI in 6.1, I can create a database from a whole directory hierarchy of 5,681 individual XML files. When I point to the same source file directory and using the 6.3.4 GUI (which I installed with a pre-release version of the installer), it only processes 14 files. No error message (that I saw) was produced. The behavior that I'm wanting is what 6.1 does. Glad to help in solving the problem. Additional question: That was in the GUI. I'd like to do the same thing using the command line with the .JAR file, but somewhere my syntax is not right. Additionally, all the XML files have NO EXTENSION so I don't know how to specify that either in the command line. So, any help in that area would also be appreciated. Thank you, David DavidEmailSig <mailto:david@leighweb.com> _______________________________________________ BaseX-Talk mailing list BaseX-Talk(a)mailman.uni-konstanz.de <mailto:BaseX-Talk@mailman.uni-konstanz.de> https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk

1 1

Working with embedded databases
by Charles F. Munat 02 Jan '11

02 Jan '11

I have a project using an embedded Base-X db that I access through application code. I want to view this information using the Base-X GUI. Currently, I use a REST interface to output the entire database as one large XML file. Then I import that file into Base-X using the GUI so that I can run queries on it with the benefits the GUI brings. But this process is cumbersome and is getting downright unwieldy as the size of the db grows. Is there a way to open the .basex files directly in the GUI to avoid this intermediate step? Thanks! Chas.

2 2

Free file hosting offer for BaseX (new features)
by DownloadRoute.com 02 Jan '11

02 Jan '11

January 2, 2011 ------------------------------- Dear Freeware author, only for freeware authors, we offer on our site hosting for your files for free. You can easy upload your BaseX file to our server which is located in US (Dallas, Texas). If you are interested just click on this URL and follow instructions on this page: http://files.downloadroute.com/reciprocal-url/BaseX-DBIS-Group-University-o… We have only limits that your file can't exceed 50MB and must be Freeware, but the traffic is unlimited. Sincerely, George www.DownloadRoute.com

1 0

Fwd: Re: Database creation from a directory tree
by Andreas Weiler 31 Dec '10

31 Dec '10

Forwarding to mailing-list...maybe interesting for others. -------- Original-Nachricht -------- Betreff: Re: [basex-talk] Database creation from a directory tree Datum: Fri, 31 Dec 2010 16:41:05 +0100 Von: Andreas Weiler <andreas.weiler(a)uni-konstanz.de> An: david(a)leighweb.com Hi David, thanks a lot for your tests. I couldnt fix the gui bug yet...but i can help you with command line... There is a createfilter option...it can be set with: set createfilter * so all files in the folder are recognized for creating a database. I hope this works for now, Happy new year, Andreas Am 31.12.10 15:13, schrieb David leigh: > > Andreas, > > I just did an interesting test (using the version that the windows > installer installs: 6.3.4). > > I copied my directory tree of files and in the new copy, I gave all my > data files an extension of "xml" (previously they had NO extension). > NOW, both with the command line and the GUI, the full database is > created without any problem. To double check, I went back to the old > directory with no extensions and created a new database using a filter > of "*.*" and only 14 documents were created in the database (instead > of 5000+). > > Conclusions: > > 1.In the command line you must have input documents with an extension > of "xml" since there is no visible way to provide a filter. > > 2.The GUI interfaces differently than the command line with the basex > core since it can receive a filter as well as a path in the database > creation process > > 3.The GUI filter functionality in 6.3.4 is different than in 6.1...for > me it is "broken" vs. how it works in 6.1 > > So, I think for my project to work completely, I need new > functionality: a file name filter that works correctly on the CREATE > DB statement in the command line interface. Does this sound correct > to you? > > Unfortunately, I don't have control over the extension names of my > input files as they are created by another piece of software. Until > I can do a create database in the command line, I'll have to do manual > gui-based (6.1 version) refreshes of the data or some sort of long > copy/rename process in the background. > > Thanks! > > David > > DavidEmailSig <mailto:david@leighweb.com> > > *From:*Andreas Weiler [mailto:andreas.weiler@uni-konstanz.de] > *Sent:* Wednesday, December 29, 2010 11:26 AM > *To:* david(a)leighweb.com; BaseX-Talk(a)mailman.uni-konstanz.de > *Subject:* Re: [basex-talk] Database creation from a directory tree > > Hi David, > > the syntax of the create command changed between the releases of 6.1 > and 6.3.4. > > In release 6.1 you have to use: > > create db C:\Users\David\Documents\OpenSong\songpacks\Songs songs > > Kind regards, > Andreas > > Am 28.12.10 18:00, schrieb David leigh: > > Here are my command line results from 6.3.4: > > C:\Users\David\Documents\OpenSong\OpenSongSearch\BaseX>java -cp > BaseX.jar org.basex.BaseXClient -Uadmin -Padmin > > BaseX 6.3.5 [Client] > > Try "help" to get more information. > > > drop db songs > > Database 'songs' was dropped. > > > create db song > > Database 'song' created in 1683.61 ms. > > > info index > > Tags > > - Structure: Hash > > - Entries: 0 > > Attributes > > - Structure: Hash > > - Entries: 0 > > Text Index > > - Structure: Binary tree > > - Size: 4 Bytes > > - Entries: 0 > > Attribute Index > > - Structure: Binary tree > > - Size: 4 Bytes > > - Entries: 0 > > Full-Text Index > > - Not available > > Path Summary > > doc() 1x > > > info table > > Stopped at line 1, column 10: > > Syntax: INFO ([DATABASE|INDEX|STORAGE]) > > Show information on current database. > > Shows information on the currently opened database: > > - no argument: show global information > > - DATABASE: shows database information > > - INDEX: shows index information > > - STORAGE [start end] | [query]: show internal database table > > > info storage > > PRE DIS SIZ ATS NS KIND CONTENT > > ------------------------------------- > > 0 1 1 1 0 DOC song > > > info > > General Information > > Database Path: C:\Users\David\Documents\OpenSong\OpenSongSearch\BaseX\data > > Used Main Memory: 895 KB > > Database Creation > > Whitespace Chopping: ON > > Entity Parsing: OFF > > Indexes > > Path Summary: ON > > Text Index: ON > > Attribute Index: ON > > Full-Text Index: OFF > > > > > So, for 6.3.4, it creates from the command line, but it's empty. The > root of that path has NO files in it. The 5861 files are in all the > subdirectories. > > Then, when I try the same syntax with 6.1, I get the following results: > > > create db songs C:\Users\David\Documents\OpenSong\songpacks\Songs > > Stopped at line 1, column 17: > > Syntax: CREATE [DB|COLL|FS|INDEX|USER] [...] > > Create database, index or user. > > Creates a new database, index or user: > > - DB [path] [name?]: > > creates the database [name] for the file or directory [path] > > - COLL [name]: > > creates an empty database [name] > > - INDEX [TEXT|ATTRIBUTE|FULLTEXT|PATH]: > > creates the specified index > > - FS [path] [name] ([mountpoint] [backingstore]): > > creates filesystem database [name] for [path] > > (using [mountpoint] and [backingstore] if FUSE is activated) > > - USER [name] [password?]: > > creates the specified user > > Summary: > > 6.1 GUI -- works as I want it > > 6.1 command line -- can't get it to create > > 6.3.4 GUI -- creates but with only 14 entries > > 6.3.4 command line -- creates, but it's empty > > I don't mind staying with 6.1, but I DO want to do everything with the > command line. > > Is the ADD syntax the same, in that I can simply specify the path and > it will recurse the whole directory tree? > > Thanks!!! > > David > > DavidEmailSig <mailto:david@leighweb.com> > > *From:*Andreas Weiler [mailto:andreas.weiler@uni-konstanz.de] > *Sent:* Tuesday, December 28, 2010 01:29 PM > *To:* david(a)leighweb.com <mailto:david@leighweb.com> > *Cc:* basex-talk(a)mailman.uni-konstanz.de > <mailto:basex-talk@mailman.uni-konstanz.de> > *Subject:* Re: [basex-talk] Database creation from a directory tree > > Additional question: That was in the GUI. I'd like to do the same > thing using the command line with the .JAR file, but somewhere my > syntax is not right. Additionally, all the XML files have NO > EXTENSION so I don't know how to specify that either in the > command line. So, any help in that area would also be appreciated. > > > Could you provide your syntax? > > The syntax should be: > create db dbname path/to/folder > > The missing extension shouldnt be a problem, i just tested it. > > > > In the GUI in 6.1, I can create a database from a whole directory > hierarchy of 5,681 individual XML files. When I point to the same > source file directory and using the 6.3.4 GUI (which I installed with > a pre-release version of the installer), it only processes 14 files. > No error message (that I saw) was produced. The behavior that I'm > wanting is what 6.1 does. Glad to help in solving the problem. > > We will check for modifications between the releases. As quick > workaround you could try to create an empty database and then add the > documents with the Database/Add documents menu. > > Kind regards, > Andreas > > Am 28.12.10 12:55, schrieb David leigh: > > I've just installed 6.3.4 and I've noticed a difference from version 6.1. > > In the GUI in 6.1, I can create a database from a whole directory > hierarchy of 5,681 individual XML files. When I point to the same > source file directory and using the 6.3.4 GUI (which I installed with > a pre-release version of the installer), it only processes 14 files. > No error message (that I saw) was produced. The behavior that I'm > wanting is what 6.1 does. Glad to help in solving the problem. > > Additional question: That was in the GUI. I'd like to do the same > thing using the command line with the .JAR file, but somewhere my > syntax is not right. Additionally, all the XML files have NO > EXTENSION so I don't know how to specify that either in the command > line. So, any help in that area would also be appreciated. > > Thank you, > > David > > DavidEmailSig <mailto:david@leighweb.com> > > > > _______________________________________________ > BaseX-Talk mailing list > BaseX-Talk(a)mailman.uni-konstanz.de <mailto:BaseX-Talk@mailman.uni-konstanz.de> > https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk >

1 0

Fwd: Re: How to retrieve HTTP request parameters and use them in XQuery with BaseX
by Andreas Weiler 31 Dec '10

31 Dec '10

Forwarding to the mailing-list... -------- Original-Nachricht -------- Betreff: Re: [basex-talk] How to retrieve HTTP request parameters and use them in XQuery with BaseX Datum: Thu, 30 Dec 2010 13:59:57 +0000 Von: software developer <computer.software.developer(a)gmail.com> An: Andreas Weiler <andreas.weiler(a)uni-konstanz.de> Well, then the problem changes to 'how to pass dynamic parameters from an HTTP client to the java class in question - java class is running on a different machine' - that bridge was to be the HTTP request parameters - what else can serve to pass dynamic parameters to the xquery? I invoke BaseX via jax-rx and xquery is stored on server-side (cannot put that in the URL itself) - xquery is invoked using run param to jax-rx invocation. Your solution has xquery passing the param to the java class, in my case an external HTTP client is to send the parameters...so, how would the java class receive the HTTP request param or alternatively, how would the xquery receive request param directly via jax-rx? On Thu, Dec 30, 2010 at 1:18 PM, Andreas Weiler <andreas.weiler(a)uni-konstanz.de <mailto:andreas.weiler@uni-konstanz.de>> wrote: Hi, passing parameters to xquery dynamically can be done with the xquery java bindings. So the parameter comes directly from your java class.

3 3

Cannot execute xxforms:get-request-parameter in XPL - XSLT processor
by software developer 31 Dec '10

31 Dec '10

Hi, I get the error: XPath syntax error at ... in {...get-request-parameter('quer...}: Cannot find a matching 1-argument function named { http://orbeon.org/oxf/xml/xforms}get-request-parameter(). *Note: external function calls have been disabled* when I attempt to execute a pipeline with: <p:processor name="oxf:xslt"> <p:input name="config"> <xsl:stylesheet version="2.0"> <xsl:template match="/"> <TargetURL> <xsl:variable name="location" select="/Configuration/XMLDB/Location/text()"/> <xsl:variable name="name" select="/Configuration/XMLDB/Name/text()"/> <xsl:variable name="query" select="xxforms:get-request-parameter('query')"/> <xsl:value-of select="fn:concat($location,'/',$name,'?',$query)"/> </TargetURL> </xsl:template> </xsl:stylesheet> </p:input> <p:input name="data" href="#configuration"/> <p:output name="data" id="Target"/> </p:processor> Is XPL not the correct location to retrieve the HTTP request params (should I be doing it in page-flow.xml instead?)

1 1

Multiple producers/writers for one collection and concurrency
by Jan Vlčinský (CAD) 31 Dec '10

31 Dec '10

Hi This topic was mentioned few times before, but it is key aspect of my application, so I will try anyway. *Question*: What is recommended approach for application, where exists multiple document producers which shall store/modify documents in one, share collection. *Background* - Having *feeds*, which might deliver some new messages. - *Multiple preprocessing processes* - preprocessing the messages before storing them in collection (Python, celery and RabbitMQ to be ready for processing bigger amount of incoming messages in parallel) - Having *single process for write access to my BaseX collection*(arranging this is not very easy for me so far) - There are some read only queries for stored messages, there are no problems with this as it shall be easy to run multiple read-only queries concurrently. *Assumptions* If I am wrong here, please, correct me. - one collection can be written/modified only by one process at given moment, other attempts to write are rejected and these attempts will fail. In other words - one collection can be opened only by one process (or are multiple opening clients allowed as long as the do not modify database concurrently?) - read only queries are easy as they can be run concurrently without blocking/rejecting *Current approach* 1. Multiple preprocessing processes are concurrently sending documents into single basex-storage queue in RabbitMQ 2. There is single one-process worker, consuming documents from queue basex-storage which does all the work with adding documents for given collection. *Wishes* Managing single write process adds quite complexity to whole solution. It would be nice, if I can write to single collection by multiple processes at once and could expect success. Any comments are welcome Best regards Jan

2 4

Big recursive XQueries
by Charles F. Munat 31 Dec '10

31 Dec '10

I have a BaseX backed website which stores all the pieces of the page in the DB. There is a significant amount of processing to be done. Pretty much any piece of the page can have effective or expiration dates (so that page parts can come and go on schedule). Some pages have dynamic elements, such as a Twitter feed. The page is build from objects which are wrapped in frames (for adding temporal and styling parameters), which are inserted into columns, which go into rows, which make up a page (which itself is wrapped in a template). So there is a lot of recursion. I do the recursion using local functions. It works well, but the page-building query is getting enormous, most of which is all these functions to insert parts of the page, often recursively. I am wondering what the best practice would be for this. I'm working in Scala, and I could break the process up into repeated passes hitting the db with a different query for each, but it seems to me one big query will almost certainly be faster. But including all the local functions in the query makes it huge. Is there some way to compile and load the functions once on startup, and then just run a small XQuery that calls functions which call functions, etc.? Any ideas (or resources) for ways to optimize the query? Everything gets reused, so there is very little nesting -- mostly references are passed. For example: <pages> <page> <id>1</id> <rows> <row id="2"/> <row id="3"/> </rows> </page> </pages> <rows> <row> <id>2</id> <columns> <column width="18"> <frames> <frame id="4"/> <frame id="5"/> </frames> </column> <column width="6"> <frames> <frame id="6/> </frames> </column> </columns> </row> </rows> <frames> <frame> <id>4</id> <contents> <content id="7"/> </contents> </frame> </frames> <contents> <content type="TEXT"/> <id>7</id> <body> <p>Some text here.</p> </body> </content> <content type="WIDGET"/> <id>7</id>  </content> </contents> Thanks! Chas.

6 11

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

BaseX-Talk