Search Results

Search found 17924 results on 717 pages for 'z order'.

Page 403/717 | < Previous Page | 399 400 401 402 403 404 405 406 407 408 409 410  | Next Page >

  • saving information in file in php

    - by Mac Taylor
    hey guys i want to write a tracking system and now i can save in my Mysql database . but saving information about each ip that visits is a huge work for mysql so i think if i could save the information in a file , then there is no discussion about database and its problems . but to begin this : i realy dont know how to save in a file in a way that i can read it with no problem and show the details this is what is used to insert into my database sql_query("insert into tracking (date_time, ip_address, hostname,referer, page , page_title) values ('".sql_quote($dt)."', '".sql_quote($ipaddr)."', '".sql_quote($hostnm)."','$referer', '".sql_quote($pg)."', '".sql_quote($pagetitle)."')", $dbi); i need to show information about all ips in rows , after saving in a file what should i do to save and show in row order ( table ) php/mysql

    Read the article

  • Suggest Cassandra data model for an existing schema

    - by Andriy Bohdan
    Hello guys! I hope there's someone who can help me suggest a suitable data model to be implemented using nosql database Apache Cassandra. More of than I need it to work under high loads and large amounts of data. Simplified I have 3 types of objects: Product Tag ProductTag Product: key - string key name - string .... - some other fields Tag: key - string key name - unique tag words ProductTag: product_key - foreign key referring to product tag_key - foreign key referring to tag rating - this is rating of tag for this product Each product may have 0 or many tags. Tag may be assigned to 1 or many products. Means relation between products and tags is many-to-many in terms of relational databases. Value of "rating" is updated "very" often. I need to be run the following queries Select objects by keys Select tags for product ordered by rating Select products by tag order by rating Update rating by product_key and tag_key The most important is to make these queries really fast on large amounts of data, considering that rating is constantly updated.

    Read the article

  • Problem with numbers

    - by StolePopov
    I am given a number N, and i must add some numbers from the array V so that they wil be equal. V is consisting of numbers that are all powers of 3: N = 17 S = 0 V = 1 3 9 27 81 .. I should add numbers from V to N and S in order to make them equal. The solution to the example above is : 17 + 1 + 9 = 27, 27, 1 and 9 are taken from V, a number from V can be taken only once, and when taken it's removed from V. I tried sorting V and then adding the biggest numbers from V to S until S has reached N, but it fails on some tests when it's like: N = 7 S = 0 V = 1 3 9 27 So the solution will be: 7 + 3 = 9 + 1 In examples like this i need to add numbers both to N and S, and also select them so they become equal. Any idea of solving this ? Thanks.

    Read the article

  • How to resolve this very heavy query that slows down the application?

    - by Juan Paredes
    Hi, We have a web application running in a production enviroment and at some point the client complained about how slow the application got. When we checked what was going on with the application and the database we discover this "precious" query that was being executed by several users at the same time (thus inflicting a extremely high load on the database server): SELECT NULL AS table_cat, o.owner AS table_schem, o.object_name AS table_name, o.object_type AS table_type, NULL AS remarks FROM all_objects o WHERE o.owner LIKE :1 ESCAPE :"SYS_B_0" AND o.object_name LIKE :2 ESCAPE :"SYS_B_1" AND o.object_type IN(:"SYS_B_2", :"SYS_B_3") ORDER BY table_type, table_schem, table_name Our application does not execute this query, I believe it is an Hibernate internal query. I've found little information on why Hibernate does this extremely heavy query, so any help is very much appreciated! The production enviroment information: Red Hat Enterprise Linux 5.3 (Tikanga), JDK 1.5, web container OC4J (whitin Oracle Application Server), Oracle Database 10g Release 10.0.0.1, JDBC Driver for JDK 1.2 and 1.3, Hibernate version 3.2.6.ga. Thank you.

    Read the article

  • accepts_nested_attributes_for and has_many :through relations.

    - by antiarchitect
    I want to make a simple examing application on RoR 2.3. The problem area is to make an exam_session in a one form with only one submit action. For the exam session there are selected some number of questions from the question pool in random order. For these questions there are selected some number of alternatives (to check is this a single answer question or multi answer question I use the number of correct alternatives: if only 1 - single, 1 - multi. Radiobuttons or checkboxes in form to answer depends on it). I have models: Questions ---< Alternative and ExamSession. I think there must be has_many :through relations between ExamSession and Questions and has_many :through relation between the intermediate table (for example QuestionsExamSession) and Alternative to point what alternatives are answers of the student on this Question. So the questions are: Is this scheme is too complicated and there is a way to do it simple and clear? Is there any way to organize models in such a way to make the form I want to work?

    Read the article

  • What is the correct way to backup ZODB blobs?

    - by joeforker
    I am using plone.app.blob to store large ZODB objects in a blobstorage directory. This reduces size pressure on Data.fs but I have not been able to find any advice on backing up this data. I am already backing up Data.fs by pointing a network backup tool at a directory of repozo backups. Should I simply point that tool at the blobstorage directory to backup my blobs? What if the database is being repacked or blobs are being added and deleted while the copy is taking place? Are there files in the blobstorage directory that must be copied over in a certain order?

    Read the article

  • best way to authenticate and consume web service using phonegap (html5/javascript)

    - by Raiss
    I am going to develop a phonegap application which is pretty simple. I need to implement an authentication and some simple data transfer back and forth to the phone and server. I prefer to use ASP.NET as a web service and our database is MS SQL but I am not sure what approach should I take to create a secure communication between Phonegap App and webservice. The problem with a simple AJAX request is limitation in cross-domain and I’m not sure if JSONP is a good option. I was wondering if someone can tell me what technology I should use in order to make a semi secure connection which works with PhoneGap (html5, javascript ) and .Net webservice. I understand that it’s a general question but I need to know what technology is the best in such a case. thanks

    Read the article

  • WPF: "Items collection must be empty before using ItemsSource."

    - by Zack Peterson
    I'm trying to get images to display in a WPF ListView styled like a WrapPanel as described in this old ATC Avalon Team article: How to Create a Custom View. When I try to populate the ListView with a LINQ-to-Entities queried collection of ADO.NET Entity Framework objects I get the following exception: Exception Items collection must be empty before using ItemsSource. My code… Visual Basic Private Sub Window1_Loaded(...) Handles MyBase.Loaded ListViewImages.ItemsSource = From g In db.Graphic _ Order By g.DateAdded Ascending _ Select g End Sub XAML <ListView Name="ListViewImages" SelectionMode="Single" ItemsSource="{Binding}"> <local:ImageView /> </ListView> I put a breakpoint on that line. ListViewImages.ItemsSource is Nothing just before the LINQ assignment.

    Read the article

  • Successful business model of small development tool

    - by Vladimir
    I have an idea on my mind to develop and sell small helpful developer tools like the tool asked in this question http://stackoverflow.com/questions/2562378/visual-objects-editor. I am looking for your answers which will shortly describe "How" and "What" to sell in order to be successful at that. I imagine different answers here from concrete to more general. Examples: "Sell your visual objects editor and I will buy it :-)" "Tbe only way to sell developer tool is to integrate it with known Java IDE" "Developers won't pay for the tools they use. Make it free and be famous" I encourage others to vote for the answers they like the most.

    Read the article

  • Sorting list of URLs by length in Jython

    - by Eef
    Hi, I am writing a Jython script to sort a list of URLs. I have a list that looks like this: http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/|,1 http://www.domain.com/folder1/folder2/folder3/|,1 http://www.domain.com/folder1/|,1 http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/folder2/folder3/|,1 The pipe and the comma separates the path from the amount of files that are under that path. Is it possible some how use Jython to order the URLs by length, so it would end up look like the below list: http://www.domain.com/folder1/|,1 http://www.domain.com/folder1/|,1 http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/folder2/|,1 http://www.domain.com/folder1/folder2/folder3/|,1 http://www.domain.com/folder1/folder2/folder3/|,1 Hope you guys get what I mean, any help would be appreciated. Cheers

    Read the article

  • create a simple pdf report from html

    - by opensas
    I'm looking for a way to generate pdf files from html In order to make simple tabular reports I would need the following features table rendering variable page size repeating headers / footers on every page calculated page number / total page css support would be nice I know there have been many similar questions in stackoverflow, but I don't know if there's a product that supports the aforementioned features... Ideally, the source would be a plain and simple well built html with css, (I'm building the html files, so I can adapt to the products needs, that is, it won't have to render every piece of html crap you can throw at a browser) and with some custom tags to configure headings, footer, page size, etc... then I would run a command line to convert it from html to pdf. I think http://www.allcolor.org/YaHPConverter/ does something like that

    Read the article

  • How can I detect client-side when a page load is the result of an AJAX history point?

    - by Nick
    I'm trying to prevent a "flicker" effect that is occurring on my ASP.NET page which occurs when a user navigates to the page via the browser back button after having navigated away from it. The reason for the flicker is that I'm using an Update Panel which has some content in there on the initial page-load. As a result, when the page is loaded via a back button that initial content is shown very briefly before it is updated with the correct History-aware data. In order to overcome this I am intending on having the updatepanel hidden (display: none) on inital page load and then show it as long as we don't have any history to deal with. The problem is that I can't find out what to check to determine if there's any history. I can see that the Sys.Application has a _history member but when I'm checking it on page init it is null each time. Does anyone know what I should be checking to determine if there's history to deal with for a page load client-side? And at what point to do it?

    Read the article

  • Tomcat Exception-Type Ignoring Specific Exception for More General

    - by David Marks
    For one type of exception, IOException, I want to display one page. For all other exceptions I have a default error page. In my web.xml I have things setup like this: java.io.IOException /queryException.jsp java.lang.Exception /error.jsp The problem is the error.jsp is the only page that ever shows, even if an IOException is thrown. The order the tags appear in doesn't matter; if I remove the java.lang.Exception tag though, I can get queryException to show for IOExceptions. What is the solution here? How can I keep a general error page for all exceptions EXCEPT for those with specific pages?

    Read the article

  • Custom Sorting on Custom Field in Django

    - by RotaJota
    In my app, I have defined a custom field to represent a physical quantity using the quantities package. class AmountField(models.CharField): def __init__(self, *args, **kwargs): ... def to_python(self, value): create_quantities_value(value) Essentially the way it works is it extends CharField to store the value as a string in the database "12 min" and represents it as a quantities object when using the field in a model array(12) * min Then in a model it is used as such: class MyModel(models.Model): group = models.CharField() amount = AmountField() class Meta: ordering = ['group', 'amount'] My issue is that these fields do not seem to sort by the quantity, but instead by the string. So if I have some objects that contain something like {"group":"A", "amount":"12 min"} {"group":"A", "amount":"20 min"} {"group":"A", "amount":"2 min"} {"group":"B", "amount":"20 min"} {"group":"B", "amount":"1 hr"} they end up sorted something like this: >>> MyModel.objects.all() [{A, 12 min}, {A, 2 min}, {A, 20 min}, {B, 1 hr}, {B, 20 min}] essentially alphabetical order. Can I give my custom AmountField a comparison function so that it will compare by the python value instead of the DB value?

    Read the article

  • Java ByteBuffer to VB .NET

    - by user1797258
    I have this java code : ByteBuffer p = ByteBuffer.allocate(packet.length - 10 + 14); p.order(ByteOrder.LITTLE_ENDIAN); p.putInt(packet.length); p.putInt(packet.request_id); p.putInt(packet.type); p.put(packet.paylod); p.put((byte) 0); p.put((byte) 0); And i want to translate it to VB .NET I started with this : Dim p(packet.length - 10 + 14) As Byte But after i dont know how to put these variables in the right way

    Read the article

  • Automated fake mailbox

    - by Bernabé Panarello
    Hello, i'm about to start the development of a new automated-email application. The idea is that customers (or other external users) send emails to a mailbox and then an automated process will read them, extract their information and insert it into some database. It's a requirement that the emails have an standard format in order to be parsed (standard subject, etc.). The obvious thing to do would be to set up a process that periodically pools an ordinary mailbox, through pop-3 for example, processing the messages it finds. However, it would be for me much nicer to be able to process the emails as they arrive. I was wondering then, is there any way to set up a process that acts as a fake email-box? do you know about any open-source implementation of something like that I can extend? I would prefer something already written in c#. Thanks in advance for your help, bernabé

    Read the article

  • mod_rewrite to find missing /img/foo.jpg in /img/f/

    - by Ambrose
    I've got a folder of images which is reaching a critical mass after a few years. I want to move images into alphabetical folders, so that /img/foo.jpg goes into /img/f/foo.jpg and /img/bar.jpg goes into /img/b/bar.jpg and so on. In order to make the transition smooth, and to allow the manual uploaders to put stuff into the top level, I'd like to use mod_rewrite to do this: if /img/foo.jpg exists, serve it up, if not look for it in /img/f/foo.jpg thanks for any suggestions. For the record, no, I don't think we need to go /img/f/fo/foo.jpg just yet.

    Read the article

  • Nested While Loop for mysql_fetch_array not working as I expected

    - by Ayush Saraf
    I m trying to make feed system for my website, in which i have got i data from the database using mysql_fetch_array system and created a while loop for mysql_fetch_array and inside tht loop i have repeated the same thing again another while loop, but the problem is that the nested while loop only execute once and dont repeat the rows.... here is the code i have used some function and OOP but u will get wht is the point! i thought of an alternative but not yet tried vaz i wanna the way out this way only i.e. using a for loopo insted of while loop that mite wrk... public function get_feeds_from_my_friends(){ global $Friends; global $User; $friend_list = $Friends->create_friend_list_ids(); $sql = "SELECT feeds.id, feeds.user_id, feeds.post, feeds.date FROM feeds WHERE feeds.user_id IN (". $friend_list. ") ORDER BY feeds.id DESC"; $result = mysql_query($sql); while ($rows = mysql_fetch_array($result)) { $id = $rows['user_id']; $dp = $User->user_detail_by_id($id, "dp"); $user_name = $User->full_name_by_id($id); $post_id = $rows['id']; $final_result = "<div class=\"sharedItem\">"; $final_result .= "<div class=\"item\">"; $final_result .= "<div class=\"imageHolder\"><img src=\"". $dp ."\" class=\"profilePicture\" /></div>"; $final_result .= "<div class=\"txtHolder\">"; $final_result .= "<div class=\"username\"> " . $user_name . "</div>"; $final_result .= "<div class=\"userfeed\"> " . $rows['post'] . "</div>"; $final_result .= "<div class=\"details\">" . $rows['date'] . "</div>"; $final_result .= "</div></div>"; $final_result .= $this->get_comments_for_feed($post_id); $final_result .= "</div>"; echo $final_result; } } public function get_comments_for_feed($feed_id){ global $User; $sql = "SELECT feeds.comments FROM feeds WHERE feeds.id = " . $feed_id . " LIMIT 1"; $result = mysql_query($sql); $result = mysql_fetch_array($result); $result = $result['comments']; $comment_list = get_array_string_from_feild($result); $sql = "SELECT comment_feed.user_id, comment_feed.comment, comment_feed.date FROM comment_feed "; $sql .= "WHERE comment_feed.id IN(" . $comment_list . ") ORDER BY comment_feed.date DESC"; $result = mysql_query($sql); if(empty($result)){ return "";} else { while ($rows = mysql_fetch_array($result)) { $id = $rows['user_id']; $dp = $User->user_detail_by_id($id, "dp"); $user_name = $User->full_name_by_id($id); return "<div class=\"comments\"><div class=\"imageHolder\"><img src=\"". $dp ."\" class=\"profilePicture\" /></div> <div class=\"txtHolder\"> <div class=\"username\"> " . $user_name . "</div> <div class=\"userfeed\"> " . $rows['comment'] . "</div> <div class=\"details\">" . $rows['date'] . "</div> </div></div>"; } } }

    Read the article

  • Issue with mouse move event

    - by Vinjamuri
    I have a WPF control1 (has a moving control) that is hosted through elementhost on a windows form. My aim is to capture the mouse move events for the elementhost. I found out from the following link that MouseMove fires when Control moves under mouse while mouse stands still. http://social.msdn.microsoft.com/Forums/en/wpf/thread/56e7b331-ac6f-4d62-a83b-c09009b79fa0 I am getting fake mouse move events for elementhost. In order to fix this issue, I added a button on top of elementhost and set its Visible property to Hidden. Still I get fake mouse move events.. How to fix this issue? Is there any workaround? Appreciate your help...

    Read the article

  • Symfony app - how to add calculated fields to Propel objects?

    - by Thomas Kohl
    What is the best way of working with calculated fields of Propel objects? Say I have an object "Customer" that has a corresponding table "customers" and each column corresponds to an attribute of my object. What I would like to do is: add a calculated attribute "Number of completed orders" to my object when using it on View A but not on Views B and C. The calculated attribute is a COUNT() of "Order" objects linked to my "Customer" object via ID. What I can do now is to first select all Customer objects, then iteratively count Orders for all of them, but I'd think doing it in a single query would improve performance. But I cannot properly "hydrate" my Propel object since it does not contain the definition of the calculated field(s). How would you approach it?

    Read the article

  • Sinatra Variable Scope

    - by Ethan Turkeltaub
    Take the following code: ### Dependencies require 'rubygems' require 'sinatra' require 'datamapper' ### Configuration config = YAML::load(File.read('config.yml')) name = config['config']['name'] description = config['config']['description'] username = config['config']['username'] password = config['config']['password'] theme = config['config']['theme'] set :public, 'views/themes/#{theme}/static' ### Models DataMapper.setup(:default, "sqlite3://#{Dir.pwd}/marvin.db") class Post include DataMapper::Resource property :id, Serial property :name, String property :body, Text property :created_at, DateTime property :slug, String end class Page include DataMapper::Resource property :id, Serial property :name, String property :body, Text property :slug, String end DataMapper.auto_migrate! ### Controllers get '/' do @posts = Post.get(:order => [ :id_desc ]) haml :"themes/#{theme}/index" end get '/:year/:month/:day/:slug' do year = params[:year] month = params[:month] day = params[:day] slug = params[:slug] haml :"themes/#{theme}/post.haml" end get '/:slug' do haml :"themes/#{theme}/page.haml" end get '/admin' do haml :"admin/index.haml" end I want to make name, and all those variables available to the entire script, as well as the views. I tried making them global variables, but no dice.

    Read the article

  • the query is breaking. How to skip rows PHP

    - by Victor
    Hello all. I have a question probably lame but it made me stuck I have the a db query $query_Recordset10 = "SELECT * FROM products WHERE razdel='mix' AND ID='$ID+1' AND litraj='$litri' ORDER BY ID ASC"; $Recordset10 = mysql_query($query_Recordset10, $victor) or die(mysql_error()); $row_Recordset10 = mysql_fetch_array($Recordset10); $totalRows_Recordset10 = mysql_num_rows($Recordset10); This is the query for the next product in the line based in the ID of the current product thats on the page. But if the next product matching the criteria in the query is 2 or more ID's ahead my cycle breaks. So is there a way for skipping this rows and get the next ID matching the criteria. Thank you very much.

    Read the article

  • Storing index values in FAST-ESP without modifications

    - by Abel Morelos
    Hi, first of all I'm totally new to FAST but I already have a couple of issues I need to solve, so I'm sorry if my questions are very basic =) Well, the problem is that I have a field in the FAST index which in the source document is something like "ABC 12345" (please note the intentional whitespaces) but when stored in the index is in the form "ABC 123456" (please note that now there is a single space). If I retrieve all the document values then this specific value is OK (with all the whitespaces), my only problem is with the way the value is stored in the index since I need to retrieve and display it to my user just like it appears in the original document, and I don't want to go to the full document just for this value, I want the value that I already have in the index. I think I need to update one of the FAST XML configuration files but I don't have enough documentation at hand in order to decide where to perform the change, index_profile.xml? in the XMLMapper file?

    Read the article

  • nmake.exe: is there a way to exclude a file from a set of files specified in a macro?

    - by Cheeso
    I'm looking for something like the Exclude filter for msbuild, but I Want it in a makefile processed by nmake. Is that possible? Suppose I have a makefile that defines this macro: SOURCES=xxx.c yyy.c zzz.c and I invoke it with nmake OLD=xxx.c NEW=bbb.c ...can I produce, within the makefile, a macro with a value like: yyy.c zzz.c bbb.c ...basically substituting bbb.c for xxx.c ? The files can appear in any order. This would be pretty easy if the string substitution that is possible in nmake macros, allowed for evaluation of macros. In other words, I can do sources=xxx.c yyy.c zzz.c objs=$(sources:.c=.o) and the value of $(objs) is xxx.o yyy.o zzz.o But nmake does not allow a macro for the value of either argument to that substitution. I cannot do this: new=.o sources=xxx.c yyy.c zzz.c objs=$(sources:.c=$(new))

    Read the article

  • Basics of Join Factorization

    - by Hong Su
    We continue our series on optimizer transformations with a post that describes the Join Factorization transformation. The Join Factorization transformation was introduced in Oracle 11g Release 2 and applies to UNION ALL queries. Union all queries are commonly used in database applications, especially in data integration applications. In many scenarios the branches in a UNION All query share a common processing, i.e, refer to the same tables. In the current Oracle execution strategy, each branch of a UNION ALL query is evaluated independently, which leads to repetitive processing, including data access and join. The join factorization transformation offers an opportunity to share the common computations across the UNION ALL branches. Currently, join factorization only factorizes common references to base tables only, i.e, not views. Consider a simple example of query Q1. Q1:    select t1.c1, t2.c2    from t1, t2, t3    where t1.c1 = t2.c1 and t1.c1 > 1 and t2.c2 = 2 and t2.c2 = t3.c2   union all    select t1.c1, t2.c2    from t1, t2, t4    where t1.c1 = t2.c1 and t1.c1 > 1 and t2.c3 = t4.c3; Table t1 appears in both the branches. As does the filter predicates on t1 (t1.c1 > 1) and the join predicates involving t1 (t1.c1 = t2.c1). Nevertheless, without any transformation, the scan (and the filtering) on t1 has to be done twice, once per branch. Such a query may benefit from join factorization which can transform Q1 into Q2 as follows: Q2:    select t1.c1, VW_JF_1.item_2    from t1, (select t2.c1 item_1, t2.c2 item_2                   from t2, t3                    where t2.c2 = t3.c2 and t2.c2 = 2                                  union all                   select t2.c1 item_1, t2.c2 item_2                   from t2, t4                    where t2.c3 = t4.c3) VW_JF_1    where t1.c1 = VW_JF_1.item_1 and t1.c1 > 1; In Q2, t1 is "factorized" and thus the table scan and the filtering on t1 is done only once (it's shared). If t1 is large, then avoiding one extra scan of t1 can lead to a huge performance improvement. Another benefit of join factorization is that it can open up more join orders. Let's look at query Q3. Q3:    select *    from t5, (select t1.c1, t2.c2                  from t1, t2, t3                  where t1.c1 = t2.c1 and t1.c1 > 1 and t2.c2 = 2 and t2.c2 = t3.c2                 union all                  select t1.c1, t2.c2                  from t1, t2, t4                  where t1.c1 = t2.c1 and t1.c1 > 1 and t2.c3 = t4.c3) V;   where t5.c1 = V.c1 In Q3, view V is same as Q1. Before join factorization, t1, t2 and t3 must be joined first before they can be joined with t5. But if join factorization factorizes t1 from view V, t1 can then be joined with t5. This opens up new join orders. That being said, join factorization imposes certain join orders. For example, in Q2, t2 and t3 appear in the first branch of the UNION ALL query in view VW_JF_1. T2 must be joined with t3 before it can be joined with t1 which is outside of the VW_JF_1 view. The imposed join order may not necessarily be the best join order. For this reason, join factorization is performed under cost-based transformation framework; this means that we cost the plans with and without join factorization and choose the cheapest plan. Note that if the branches in UNION ALL have DISTINCT clauses, join factorization is not valid. For example, Q4 is NOT semantically equivalent to Q5.   Q4:     select distinct t1.*      from t1, t2      where t1.c1 = t2.c1  union all      select distinct t1.*      from t1, t2      where t1.c1 = t2.c1 Q5:    select distinct t1.*     from t1, (select t2.c1 item_1                   from t2                union all                   select t2.c1 item_1                  from t2) VW_JF_1     where t1.c1 = VW_JF_1.item_1 Q4 might return more rows than Q5. Q5's results are guaranteed to be duplicate free because of the DISTINCT key word at the top level while Q4's results might contain duplicates.   The examples given so far involve inner joins only. Join factorization is also supported in outer join, anti join and semi join. But only the right tables of outer join, anti join and semi joins can be factorized. It is not semantically correct to factorize the left table of outer join, anti join or semi join. For example, Q6 is NOT semantically equivalent to Q7. Q6:     select t1.c1, t2.c2    from t1, t2    where t1.c1 = t2.c1(+) and t2.c2 (+) = 2  union all    select t1.c1, t2.c2    from t1, t2      where t1.c1 = t2.c1(+) and t2.c2 (+) = 3 Q7:     select t1.c1, VW_JF_1.item_2    from t1, (select t2.c1 item_1, t2.c2 item_2                  from t2                  where t2.c2 = 2                union all                  select t2.c1 item_1, t2.c2 item_2                  from t2                                                                                                    where t2.c2 = 3) VW_JF_1       where t1.c1 = VW_JF_1.item_1(+)                                                                  However, the right side of an outer join can be factorized. For example, join factorization can transform Q8 to Q9 by factorizing t2, which is the right table of an outer join. Q8:    select t1.c2, t2.c2    from t1, t2      where t1.c1 = t2.c1 (+) and t1.c1 = 1 union all    select t1.c2, t2.c2    from t1, t2    where t1.c1 = t2.c1(+) and t1.c1 = 2 Q9:   select VW_JF_1.item_2, t2.c2   from t2,             (select t1.c1 item_1, t1.c2 item_2            from t1            where t1.c1 = 1           union all            select t1.c1 item_1, t1.c2 item_2            from t1            where t1.c1 = 2) VW_JF_1   where VW_JF_1.item_1 = t2.c1(+) All of the examples in this blog show factorizing a single table from two branches. This is just for ease of illustration. Join factorization can factorize multiple tables and from more than two UNION ALL branches.  SummaryJoin factorization is a cost-based transformation. It can factorize common computations from branches in a UNION ALL query which can lead to huge performance improvement. 

    Read the article

< Previous Page | 399 400 401 402 403 404 405 406 407 408 409 410  | Next Page >