Search Results

Search found 9156 results on 367 pages for 'partial match'.

Page 65/367 | < Previous Page | 61 62 63 64 65 66 67 68 69 70 71 72  | Next Page >

  • Why always fires OnFailure when return View() to Ajax Form ?

    - by Wahid Bitar
    I'm trying to make a log-in log-off with Ajax supported. I made some logic in my controller to sign the user in and then return simple partial containing welcome message and log-Off ActionLink my Action method looks like this : public ActionResult LogOn(LogOnModel model, string returnUrl) { if (ModelState.IsValid) { if (MembershipService.ValidateUser(model.UserName, model.Password)) { FormsService.SignIn(model.UserName, model.RememberMe); if (Request.IsAjaxRequest()) { //HERE IS THE PROBLEM :( return View("LogedInForm"); } else { if (!String.IsNullOrEmpty(returnUrl)) return Redirect(returnUrl); else return RedirectToAction("Index", "Home"); } } else { ModelState.AddModelError("", "The user name or password provided is incorrect."); if (Request.IsAjaxRequest()) { return Content("There were an error !"); } } } return View(model); } and I'm trying to return this simple partial : Welcome <b><%= Html.Encode(Model.UserName)%></b>! <%= Html.ActionLink("Log Off", "LogOff", "Account") %> and of-course the two partial are strongly-typed to LogOnModel. But if i returned View("PartialName") i always get OnFailure with status code 500. While if i returned Content("My Message") everything is going right. so please tell me why i always get this "StatusCode = 500" ??. where is the big mistake ??. By the way in my Site MasterPage i rendered partial to show long-on simple form this partial looks like this : <script type="text/javascript"> function ShowErrorMessage(ajaxContext) { var response = ajaxContext.get_response(); var statusCode = response.get_statusCode(); alert("Sorry, the request failed with status code " + statusCode); } function ShowSuccessMessage() { alert("Hey everything is OK!"); } </script> <div id="logedInDiv"> </div> <% using (Ajax.BeginForm("LogOn", "Account", new AjaxOptions { UpdateTargetId = "logedInDiv", InsertionMode = InsertionMode.Replace, OnSuccess = "ShowSuccessMessage", OnFailure = "ShowErrorMessage" })) { %> <%= Html.TextBoxFor(m => m.UserName)%> <%= Html.PasswordFor(m => m.Password)%> <%= Html.CheckBoxFor(m => m.RememberMe)%> <input type="submit" value="Log On" /> < <% } %>

    Read the article

  • Cast IEnumerable<Inherited> To IEnumerable<Base>

    - by david2342
    I'm trying to cast an IEnumerable of an inherited type to IEnumerable of base class. Have tried following: var test = resultFromDb.Cast<BookedResource>(); return test.ToList(); But getting error: You cannot convert these types. Linq to Entities only supports conversion primitive EDM-types. The classes involved look like this: public partial class HistoryBookedResource : BookedResource { } public partial class HistoryBookedResource { public int ResourceId { get; set; } public string DateFrom { get; set; } public string TimeFrom { get; set; } public string TimeTo { get; set; } } public partial class BookedResource { public int ResourceId { get; set; } public string DateFrom { get; set; } public string TimeFrom { get; set; } public string TimeTo { get; set; } } [MetadataType(typeof(BookedResourceMetaData))] public partial class BookedResource { } public class BookedResourceMetaData { [Required(ErrorMessage = "Resource id is Required")] [Range(0, int.MaxValue, ErrorMessage = "Resource id is must be an number")] public object ResourceId { get; set; } [Required(ErrorMessage = "Date is Required")] public object DateFrom { get; set; } [Required(ErrorMessage = "Time From is Required")] public object TimeFrom { get; set; } [Required(ErrorMessage = "Time to is Required")] public object TimeTo { get; set; } } The problem I'm trying to solve is to get records from table HistoryBookedResource and have the result in an IEnumerable<BookedResource> using Entity Framework and LINQ. UPDATE: When using the following the cast seams to work but when trying to loop with a foreach the data is lost. resultFromDb.ToList() as IEnumerable<BookedResource>; UPDATE 2: Im using entity frameworks generated model, model (edmx) is created from database, edmx include classes that reprecent the database tables. In database i have a history table for old BookedResource and it can happen that the user want to look at these and to get the old data from the database entity framework uses classes with the same name as the tables to receive data from db. So i receive the data from table HistoryBookedResource in HistoryBookedResource class. Because entity framework generate the partial classes with the properties i dont know if i can make them virtual and override. Any suggestions will be greatly appreciated.

    Read the article

  • Unusual RJS error

    - by rrb
    Hi, I am getting the following error in my RoR application: RJS error: TypeError: element is null Element.update("notice", "Comment Posted"); Element.update("allcomments", "\n\n\n \n\n waht now?\n\n \n\n \n\n \n\n asdfasdfa\n \n\n \n\n asdfasdf\n \n\n\n\n\n"); But when I hit the refresh button, I can see my partial updated. Here's my code: show_comments View: <table> <% comments.each do |my_comment| %> <tr> <td><%=h my_comment.comment%></td> </tr> <% end %> </table> show View: <div class="wrapper"> <div class="rescale"> <div class="img-main"> <%= image_tag @deal.photo.url %> </div> </div> <div class="description"> <p class ="description_content"> <%=h @deal.description %> </p> </div> </div> <p> <b>Category:</b> <%=h @deal.category %> </p> <p> <b>Base price:</b> <%=h @deal.base_price %> </p> <%#*<p>%> <%#*<b>Discount:</b>%> <%#=h @deal.discount %> <%#*</p>%> <%= link_to 'Edit', edit_deal_path(@deal) %> | <%= link_to 'Back', deals_path %> <p> <%= render :partial=>'deal_comments', :locals=>{ :comments=>Comment.new(:deal_id=>@deal.id)} %> </p> <div id="allcomments"> <%= render :partial=>'show_comments', :locals=>{ :comments=>Comment.find(@deal.comments)} %> </div> Controller: def create @comment = Comment.new(params[:comment]) render :update do |page| if @comment.save page.replace_html 'notice', 'Comment Posted' else page.replace_html 'notice', 'Something went wrong' end page.replace_html 'allcomments', :partial=> 'deals/show_comments', :locals=>{:comments=> @comment.deal.comments} end end def show_comments @deal = Deal.find(params[:deal_id]) render :partial=> "deals/show_comments", :locals=>{:comments=>@deal.comments} end end

    Read the article

  • Ajax page.replace_html problems with partials in Rails

    - by Chris Power
    Hello, I am having a problem with a pretty simple AJAX call in rails. I have a blog-style application and each post has a "like" feature. I want to be able to increment the "like" on each post in the index using AJAX onclick. I got it to work; however, the DOM is a bit tricky here, because no matter what partial its looking at, it will only update the TOP partial. so if I click "like" on post #2, it will update and replace the "likes" on post #1 instead. Code for _post partial: <some code here...> <div id="postcontent"> Posted <%= post.created_at.strftime("%A, %b %d")%> <br /> </div> <div id="postlikes"> <%= link_to_remote 'Like', :url => {:controller => 'posts', :action => 'like_post', :id => post.id}%> <%= post.like %> </div> code for _postlikes partial: <div id="postlikes"> <%= link_to_remote 'Like', :url => {:controller => 'posts', :action => 'like_post', :id => @post.id}%> <%= @post.like %> </div> </div> like_post.rjs code: page.replace_html "postlikes", :partial => "postlikes", :object => @post page.visual_effect :highlight, "postlikes", :duration => 3 So this all works properly for the first "postcontent" div. But this is an index of posts, so if I wanted to updated the second "postcontent" div on the page, it will still replace the html of the first. I understand the problem, I just don't know how to fix it :) Thanks in advance!

    Read the article

  • Improving Partitioned Table Join Performance

    - by Paul White
    The query optimizer does not always choose an optimal strategy when joining partitioned tables. This post looks at an example, showing how a manual rewrite of the query can almost double performance, while reducing the memory grant to almost nothing. Test Data The two tables in this example use a common partitioning partition scheme. The partition function uses 41 equal-size partitions: CREATE PARTITION FUNCTION PFT (integer) AS RANGE RIGHT FOR VALUES ( 125000, 250000, 375000, 500000, 625000, 750000, 875000, 1000000, 1125000, 1250000, 1375000, 1500000, 1625000, 1750000, 1875000, 2000000, 2125000, 2250000, 2375000, 2500000, 2625000, 2750000, 2875000, 3000000, 3125000, 3250000, 3375000, 3500000, 3625000, 3750000, 3875000, 4000000, 4125000, 4250000, 4375000, 4500000, 4625000, 4750000, 4875000, 5000000 ); GO CREATE PARTITION SCHEME PST AS PARTITION PFT ALL TO ([PRIMARY]); There two tables are: CREATE TABLE dbo.T1 ( TID integer NOT NULL IDENTITY(0,1), Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T1 PRIMARY KEY CLUSTERED (TID) ON PST (TID) );   CREATE TABLE dbo.T2 ( TID integer NOT NULL, Column1 integer NOT NULL, Padding binary(100) NOT NULL DEFAULT 0x,   CONSTRAINT PK_T2 PRIMARY KEY CLUSTERED (TID, Column1) ON PST (TID) ); The next script loads 5 million rows into T1 with a pseudo-random value between 1 and 5 for Column1. The table is partitioned on the IDENTITY column TID: INSERT dbo.T1 WITH (TABLOCKX) (Column1) SELECT (ABS(CHECKSUM(NEWID())) % 5) + 1 FROM dbo.Numbers AS N WHERE n BETWEEN 1 AND 5000000; In case you don’t already have an auxiliary table of numbers lying around, here’s a script to create one with 10 million rows: CREATE TABLE dbo.Numbers (n bigint PRIMARY KEY);   WITH L0 AS(SELECT 1 AS c UNION ALL SELECT 1), L1 AS(SELECT 1 AS c FROM L0 AS A CROSS JOIN L0 AS B), L2 AS(SELECT 1 AS c FROM L1 AS A CROSS JOIN L1 AS B), L3 AS(SELECT 1 AS c FROM L2 AS A CROSS JOIN L2 AS B), L4 AS(SELECT 1 AS c FROM L3 AS A CROSS JOIN L3 AS B), L5 AS(SELECT 1 AS c FROM L4 AS A CROSS JOIN L4 AS B), Nums AS(SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) AS n FROM L5) INSERT dbo.Numbers WITH (TABLOCKX) SELECT TOP (10000000) n FROM Nums ORDER BY n OPTION (MAXDOP 1); Table T1 contains data like this: Next we load data into table T2. The relationship between the two tables is that table 2 contains ‘n’ rows for each row in table 1, where ‘n’ is determined by the value in Column1 of table T1. There is nothing particularly special about the data or distribution, by the way. INSERT dbo.T2 WITH (TABLOCKX) (TID, Column1) SELECT T.TID, N.n FROM dbo.T1 AS T JOIN dbo.Numbers AS N ON N.n >= 1 AND N.n <= T.Column1; Table T2 ends up containing about 15 million rows: The primary key for table T2 is a combination of TID and Column1. The data is partitioned according to the value in column TID alone. Partition Distribution The following query shows the number of rows in each partition of table T1: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T1 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are 40 partitions containing 125,000 rows (40 * 125k = 5m rows). The rightmost partition remains empty. The next query shows the distribution for table 2: SELECT PartitionID = CA1.P, NumRows = COUNT_BIG(*) FROM dbo.T2 AS T CROSS APPLY (VALUES ($PARTITION.PFT(TID))) AS CA1 (P) GROUP BY CA1.P ORDER BY CA1.P; There are roughly 375,000 rows in each partition (the rightmost partition is also empty): Ok, that’s the test data done. Test Query and Execution Plan The task is to count the rows resulting from joining tables 1 and 2 on the TID column: SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; The optimizer chooses a plan using parallel hash join, and partial aggregation: The Plan Explorer plan tree view shows accurate cardinality estimates and an even distribution of rows across threads (click to enlarge the image): With a warm data cache, the STATISTICS IO output shows that no physical I/O was needed, and all 41 partitions were touched: Running the query without actual execution plan or STATISTICS IO information for maximum performance, the query returns in around 2600ms. Execution Plan Analysis The first step toward improving on the execution plan produced by the query optimizer is to understand how it works, at least in outline. The two parallel Clustered Index Scans use multiple threads to read rows from tables T1 and T2. Parallel scan uses a demand-based scheme where threads are given page(s) to scan from the table as needed. This arrangement has certain important advantages, but does result in an unpredictable distribution of rows amongst threads. The point is that multiple threads cooperate to scan the whole table, but it is impossible to predict which rows end up on which threads. For correct results from the parallel hash join, the execution plan has to ensure that rows from T1 and T2 that might join are processed on the same thread. For example, if a row from T1 with join key value ‘1234’ is placed in thread 5’s hash table, the execution plan must guarantee that any rows from T2 that also have join key value ‘1234’ probe thread 5’s hash table for matches. The way this guarantee is enforced in this parallel hash join plan is by repartitioning rows to threads after each parallel scan. The two repartitioning exchanges route rows to threads using a hash function over the hash join keys. The two repartitioning exchanges use the same hash function so rows from T1 and T2 with the same join key must end up on the same hash join thread. Expensive Exchanges This business of repartitioning rows between threads can be very expensive, especially if a large number of rows is involved. The execution plan selected by the optimizer moves 5 million rows through one repartitioning exchange and around 15 million across the other. As a first step toward removing these exchanges, consider the execution plan selected by the optimizer if we join just one partition from each table, disallowing parallelism: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = 1 AND $PARTITION.PFT(T2.TID) = 1 OPTION (MAXDOP 1); The optimizer has chosen a (one-to-many) merge join instead of a hash join. The single-partition query completes in around 100ms. If everything scaled linearly, we would expect that extending this strategy to all 40 populated partitions would result in an execution time around 4000ms. Using parallelism could reduce that further, perhaps to be competitive with the parallel hash join chosen by the optimizer. This raises a question. If the most efficient way to join one partition from each of the tables is to use a merge join, why does the optimizer not choose a merge join for the full query? Forcing a Merge Join Let’s force the optimizer to use a merge join on the test query using a hint: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN); This is the execution plan selected by the optimizer: This plan results in the same number of logical reads reported previously, but instead of 2600ms the query takes 5000ms. The natural explanation for this drop in performance is that the merge join plan is only using a single thread, whereas the parallel hash join plan could use multiple threads. Parallel Merge Join We can get a parallel merge join plan using the same query hint as before, and adding trace flag 8649: SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (MERGE JOIN, QUERYTRACEON 8649); The execution plan is: This looks promising. It uses a similar strategy to distribute work across threads as seen for the parallel hash join. In practice though, performance is disappointing. On a typical run, the parallel merge plan runs for around 8400ms; slower than the single-threaded merge join plan (5000ms) and much worse than the 2600ms for the parallel hash join. We seem to be going backwards! The logical reads for the parallel merge are still exactly the same as before, with no physical IOs. The cardinality estimates and thread distribution are also still very good (click to enlarge): A big clue to the reason for the poor performance is shown in the wait statistics (captured by Plan Explorer Pro): CXPACKET waits require careful interpretation, and are most often benign, but in this case excessive waiting occurs at the repartitioning exchanges. Unlike the parallel hash join, the repartitioning exchanges in this plan are order-preserving ‘merging’ exchanges (because merge join requires ordered inputs): Parallelism works best when threads can just grab any available unit of work and get on with processing it. Preserving order introduces inter-thread dependencies that can easily lead to significant waits occurring. In extreme cases, these dependencies can result in an intra-query deadlock, though the details of that will have to wait for another time to explore in detail. The potential for waits and deadlocks leads the query optimizer to cost parallel merge join relatively highly, especially as the degree of parallelism (DOP) increases. This high costing resulted in the optimizer choosing a serial merge join rather than parallel in this case. The test results certainly confirm its reasoning. Collocated Joins In SQL Server 2008 and later, the optimizer has another available strategy when joining tables that share a common partition scheme. This strategy is a collocated join, also known as as a per-partition join. It can be applied in both serial and parallel execution plans, though it is limited to 2-way joins in the current optimizer. Whether the optimizer chooses a collocated join or not depends on cost estimation. The primary benefits of a collocated join are that it eliminates an exchange and requires less memory, as we will see next. Costing and Plan Selection The query optimizer did consider a collocated join for our original query, but it was rejected on cost grounds. The parallel hash join with repartitioning exchanges appeared to be a cheaper option. There is no query hint to force a collocated join, so we have to mess with the costing framework to produce one for our test query. Pretending that IOs cost 50 times more than usual is enough to convince the optimizer to use collocated join with our test query: -- Pretend IOs are 50x cost temporarily DBCC SETIOWEIGHT(50);   -- Co-located hash join SELECT COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID OPTION (RECOMPILE);   -- Reset IO costing DBCC SETIOWEIGHT(1); Collocated Join Plan The estimated execution plan for the collocated join is: The Constant Scan contains one row for each partition of the shared partitioning scheme, from 1 to 41. The hash repartitioning exchanges seen previously are replaced by a single Distribute Streams exchange using Demand partitioning. Demand partitioning means that the next partition id is given to the next parallel thread that asks for one. My test machine has eight logical processors, and all are available for SQL Server to use. As a result, there are eight threads in the single parallel branch in this plan, each processing one partition from each table at a time. Once a thread finishes processing a partition, it grabs a new partition number from the Distribute Streams exchange…and so on until all partitions have been processed. It is important to understand that the parallel scans in this plan are different from the parallel hash join plan. Although the scans have the same parallelism icon, tables T1 and T2 are not being co-operatively scanned by multiple threads in the same way. Each thread reads a single partition of T1 and performs a hash match join with the same partition from table T2. The properties of the two Clustered Index Scans show a Seek Predicate (unusual for a scan!) limiting the rows to a single partition: The crucial point is that the join between T1 and T2 is on TID, and TID is the partitioning column for both tables. A thread that processes partition ‘n’ is guaranteed to see all rows that can possibly join on TID for that partition. In addition, no other thread will see rows from that partition, so this removes the need for repartitioning exchanges. CPU and Memory Efficiency Improvements The collocated join has removed two expensive repartitioning exchanges and added a single exchange processing 41 rows (one for each partition id). Remember, the parallel hash join plan exchanges had to process 5 million and 15 million rows. The amount of processor time spent on exchanges will be much lower in the collocated join plan. In addition, the collocated join plan has a maximum of 8 threads processing single partitions at any one time. The 41 partitions will all be processed eventually, but a new partition is not started until a thread asks for it. Threads can reuse hash table memory for the new partition. The parallel hash join plan also had 8 hash tables, but with all 5,000,000 build rows loaded at the same time. The collocated plan needs memory for only 8 * 125,000 = 1,000,000 rows at any one time. Collocated Hash Join Performance The collated join plan has disappointing performance in this case. The query runs for around 25,300ms despite the same IO statistics as usual. This is much the worst result so far, so what went wrong? It turns out that cardinality estimation for the single partition scans of table T1 is slightly low. The properties of the Clustered Index Scan of T1 (graphic immediately above) show the estimation was for 121,951 rows. This is a small shortfall compared with the 125,000 rows actually encountered, but it was enough to cause the hash join to spill to physical tempdb: A level 1 spill doesn’t sound too bad, until you realize that the spill to tempdb probably occurs for each of the 41 partitions. As a side note, the cardinality estimation error is a little surprising because the system tables accurately show there are 125,000 rows in every partition of T1. Unfortunately, the optimizer uses regular column and index statistics to derive cardinality estimates here rather than system table information (e.g. sys.partitions). Collocated Merge Join We will never know how well the collocated parallel hash join plan might have worked without the cardinality estimation error (and the resulting 41 spills to tempdb) but we do know: Merge join does not require a memory grant; and Merge join was the optimizer’s preferred join option for a single partition join Putting this all together, what we would really like to see is the same collocated join strategy, but using merge join instead of hash join. Unfortunately, the current query optimizer cannot produce a collocated merge join; it only knows how to do collocated hash join. So where does this leave us? CROSS APPLY sys.partitions We can try to write our own collocated join query. We can use sys.partitions to find the partition numbers, and CROSS APPLY to get a count per partition, with a final step to sum the partial counts. The following query implements this idea: SELECT row_count = SUM(Subtotals.cnt) FROM ( -- Partition numbers SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1 ) AS P CROSS APPLY ( -- Count per collocated join SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals; The estimated plan is: The cardinality estimates aren’t all that good here, especially the estimate for the scan of the system table underlying the sys.partitions view. Nevertheless, the plan shape is heading toward where we would like to be. Each partition number from the system table results in a per-partition scan of T1 and T2, a one-to-many Merge Join, and a Stream Aggregate to compute the partial counts. The final Stream Aggregate just sums the partial counts. Execution time for this query is around 3,500ms, with the same IO statistics as always. This compares favourably with 5,000ms for the serial plan produced by the optimizer with the OPTION (MERGE JOIN) hint. This is another case of the sum of the parts being less than the whole – summing 41 partial counts from 41 single-partition merge joins is faster than a single merge join and count over all partitions. Even so, this single-threaded collocated merge join is not as quick as the original parallel hash join plan, which executed in 2,600ms. On the positive side, our collocated merge join uses only one logical processor and requires no memory grant. The parallel hash join plan used 16 threads and reserved 569 MB of memory:   Using a Temporary Table Our collocated merge join plan should benefit from parallelism. The reason parallelism is not being used is that the query references a system table. We can work around that by writing the partition numbers to a temporary table (or table variable): SET STATISTICS IO ON; DECLARE @s datetime2 = SYSUTCDATETIME();   CREATE TABLE #P ( partition_number integer PRIMARY KEY);   INSERT #P (partition_number) SELECT p.partition_number FROM sys.partitions AS p WHERE p.[object_id] = OBJECT_ID(N'T1', N'U') AND p.index_id = 1;   SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals;   DROP TABLE #P;   SELECT DATEDIFF(Millisecond, @s, SYSUTCDATETIME()); SET STATISTICS IO OFF; Using the temporary table adds a few logical reads, but the overall execution time is still around 3500ms, indistinguishable from the same query without the temporary table. The problem is that the query optimizer still doesn’t choose a parallel plan for this query, though the removal of the system table reference means that it could if it chose to: In fact the optimizer did enter the parallel plan phase of query optimization (running search 1 for a second time): Unfortunately, the parallel plan found seemed to be more expensive than the serial plan. This is a crazy result, caused by the optimizer’s cost model not reducing operator CPU costs on the inner side of a nested loops join. Don’t get me started on that, we’ll be here all night. In this plan, everything expensive happens on the inner side of a nested loops join. Without a CPU cost reduction to compensate for the added cost of exchange operators, candidate parallel plans always look more expensive to the optimizer than the equivalent serial plan. Parallel Collocated Merge Join We can produce the desired parallel plan using trace flag 8649 again: SELECT row_count = SUM(Subtotals.cnt) FROM #P AS p CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: One difference between this plan and the collocated hash join plan is that a Repartition Streams exchange operator is used instead of Distribute Streams. The effect is similar, though not quite identical. The Repartition uses round-robin partitioning, meaning the next partition id is pushed to the next thread in sequence. The Distribute Streams exchange seen earlier used Demand partitioning, meaning the next partition id is pulled across the exchange by the next thread that is ready for more work. There are subtle performance implications for each partitioning option, but going into that would again take us too far off the main point of this post. Performance The important thing is the performance of this parallel collocated merge join – just 1350ms on a typical run. The list below shows all the alternatives from this post (all timings include creation, population, and deletion of the temporary table where appropriate) from quickest to slowest: Collocated parallel merge join: 1350ms Parallel hash join: 2600ms Collocated serial merge join: 3500ms Serial merge join: 5000ms Parallel merge join: 8400ms Collated parallel hash join: 25,300ms (hash spill per partition) The parallel collocated merge join requires no memory grant (aside from a paltry 1.2MB used for exchange buffers). This plan uses 16 threads at DOP 8; but 8 of those are (rather pointlessly) allocated to the parallel scan of the temporary table. These are minor concerns, but it turns out there is a way to address them if it bothers you. Parallel Collocated Merge Join with Demand Partitioning This final tweak replaces the temporary table with a hard-coded list of partition ids (dynamic SQL could be used to generate this query from sys.partitions): SELECT row_count = SUM(Subtotals.cnt) FROM ( VALUES (1),(2),(3),(4),(5),(6),(7),(8),(9),(10), (11),(12),(13),(14),(15),(16),(17),(18),(19),(20), (21),(22),(23),(24),(25),(26),(27),(28),(29),(30), (31),(32),(33),(34),(35),(36),(37),(38),(39),(40),(41) ) AS P (partition_number) CROSS APPLY ( SELECT cnt = COUNT_BIG(*) FROM dbo.T1 AS T1 JOIN dbo.T2 AS T2 ON T2.TID = T1.TID WHERE $PARTITION.PFT(T1.TID) = p.partition_number AND $PARTITION.PFT(T2.TID) = p.partition_number ) AS SubTotals OPTION (QUERYTRACEON 8649); The actual execution plan is: The parallel collocated hash join plan is reproduced below for comparison: The manual rewrite has another advantage that has not been mentioned so far: the partial counts (per partition) can be computed earlier than the partial counts (per thread) in the optimizer’s collocated join plan. The earlier aggregation is performed by the extra Stream Aggregate under the nested loops join. The performance of the parallel collocated merge join is unchanged at around 1350ms. Final Words It is a shame that the current query optimizer does not consider a collocated merge join (Connect item closed as Won’t Fix). The example used in this post showed an improvement in execution time from 2600ms to 1350ms using a modestly-sized data set and limited parallelism. In addition, the memory requirement for the query was almost completely eliminated  – down from 569MB to 1.2MB. The problem with the parallel hash join selected by the optimizer is that it attempts to process the full data set all at once (albeit using eight threads). It requires a large memory grant to hold all 5 million rows from table T1 across the eight hash tables, and does not take advantage of the divide-and-conquer opportunity offered by the common partitioning. The great thing about the collocated join strategies is that each parallel thread works on a single partition from both tables, reading rows, performing the join, and computing a per-partition subtotal, before moving on to a new partition. From a thread’s point of view… If you have trouble visualizing what is happening from just looking at the parallel collocated merge join execution plan, let’s look at it again, but from the point of view of just one thread operating between the two Parallelism (exchange) operators. Our thread picks up a single partition id from the Distribute Streams exchange, and starts a merge join using ordered rows from partition 1 of table T1 and partition 1 of table T2. By definition, this is all happening on a single thread. As rows join, they are added to a (per-partition) count in the Stream Aggregate immediately above the Merge Join. Eventually, either T1 (partition 1) or T2 (partition 1) runs out of rows and the merge join stops. The per-partition count from the aggregate passes on through the Nested Loops join to another Stream Aggregate, which is maintaining a per-thread subtotal. Our same thread now picks up a new partition id from the exchange (say it gets id 9 this time). The count in the per-partition aggregate is reset to zero, and the processing of partition 9 of both tables proceeds just as it did for partition 1, and on the same thread. Each thread picks up a single partition id and processes all the data for that partition, completely independently from other threads working on other partitions. One thread might eventually process partitions (1, 9, 17, 25, 33, 41) while another is concurrently processing partitions (2, 10, 18, 26, 34) and so on for the other six threads at DOP 8. The point is that all 8 threads can execute independently and concurrently, continuing to process new partitions until the wider job (of which the thread has no knowledge!) is done. This divide-and-conquer technique can be much more efficient than simply splitting the entire workload across eight threads all at once. Related Reading Understanding and Using Parallelism in SQL Server Parallel Execution Plans Suck © 2013 Paul White – All Rights Reserved Twitter: @SQL_Kiwi

    Read the article

  • Regular Expression Transformation

    The regular expression transformation exposes the power of regular expression matching within the pipeline. One or more columns can be selected, and for each column an individual expression can be applied. The way multiple columns are handled can be set on the options page. The AND option means all columns must match, whilst the OR option means only one column has to match. If rows pass their tests then rows are passed down the successful match output. Rows that fail are directed down the alternate output. This transformation is ideal for validating data through the use of regular expressions. You can enter any expression you like, or select a pre-configured expression within the editor. You can expand the list of pre-configured expressions yourself. These are stored in a Xml file, %ProgramFiles%\Microsoft SQL Server\nnn\DTS\PipelineComponents\RegExTransform.xml, where nnn represents the folder version, 90 for 2005, 100 for 2008 and 110 for 2012. If you want to use regular expressions to manipulate data, rather than just validating it, try the RegexClean Transformation. The component is provided as an MSI file, however for 2005/200 you will have to add the transformation to the Visual Studio toolbox by hand. This process has been described in detail in the related FAQ entry for How do I install a task or transform component?, just select Regular Expression Transformation in the Choose Toolbox Items window. Downloads The Regular Expression Transformation is available for SQL Server 2005, SQL Server 2008 (includes R2) and SQL Server 2012. Please choose the version to match your SQL Server version, or you can install multiple versions and use them side by side if you have more than one version of SQL Server installed. Regular Expression Transformation for SQL Server 2005 Regular Expression Transformation for SQL Server 2008 Regular Expression Transformation for SQL Server 2012 Version History SQL Server 2012Version 2.0.0.87 - SQL Server 2012 release. Includes upgrade support for both 2005 and 2008 packages to 2012. (5 Jun 2012) SQL Server 2008Version 2.0.0.87 - Release for SQL Server 2008 Integration Services. (10 Oct 2008) SQL Server 2005 Version 1.1.0.93 - Added option for you to choose AND or OR logic when multiple columns have been selected. Previously behaviour was OR only. (31 Jul 2008) Version 1.0.0.76 - Installer update and improved exception handling. (28 Jan 2008) Version 1.0.0.41 - Update for user interface stability fixes. (2 Aug 2006) Version 1.0.0.24 - SQL Server 2005 RTM Refresh. SP1 Compatibility Testing. (12 Jun 2006) Version 1.0.0.9 - Public Release for SQL Server 2005 IDW 15 June CTP (29 Aug 2005) Screenshots  

    Read the article

  • What does the program cclive do?

    - by Pili Garcia
    I run: $ cclive "http://www.youtube.com/watch?v=NfLTl1yBzpg" and receive this error message: Checking ... ........ libquvi: error: /usr/share/quvi/lua/website/youtube.lua:117: no match: fmt_url_map I run: $ cclive "http://www.youtube.com/watch?v=0uuFeQXyODA" and receive this error message: Checking ... ....... libquvi: error: /usr/share/quvi/lua/website/youtube.lua:117: no match: fmt_url_map Etc., etc., etc.

    Read the article

  • Using T4 to generate Configuration classes

    - by Justin Hoffman
    I wanted to try to use T4 to read a web.config and generate all of the appSettings and connectionStrings as properties of a class.  I elected in this template only to output appSettings and connectionStrings but you can see it would be easily adapted for app specific settings, bindings etc.  This allows for quick access to config values as well as removing the potential for typo's when accessing values from the ConfigurationManager. One caveat: a developer would need to remember to run the .tt file after adding an entry to the web.config.  However, one would quickly notice when trying to access the property from the generated class (it wouldn't be there).  Additionally, there are other options as noted here. The first step was to create the .tt file.  Note that this is a basic example, it could be extended even further I'm sure.  In this example I just manually input the path to the web.config file. <#@ template debug="false" hostspecific="true" language="C#" #><#@ output extension=".cs" #><#@ assembly Name="System.Configuration" #><#@ assembly name="System.Xml" #><#@ assembly name="System.Xml.Linq" #><#@ assembly name="System.Net" #><#@ assembly name="System" #><#@ import namespace="System.Configuration" #><#@ import namespace="System.Xml" #><#@ import namespace="System.Net" #><#@ import namespace="Microsoft.VisualStudio.TextTemplating" #><#@ import namespace="System.Xml.Linq" #>using System;using System.Configuration;using System.Xml;using System.Xml.Linq;using System.Linq;namespace MyProject.Web { public partial class Configurator { <# var xDocument = XDocument.Load(@"G:\MySolution\MyProject\Web.config"); var results = xDocument.Descendants("appSettings"); const string key = "key"; const string name = "name"; foreach (var xElement in results.Descendants()) {#> public string <#= xElement.Attribute(key).Value#>{get {return ConfigurationManager.AppSettings[<#= string.Format("{0}{1}{2}","\"" , xElement.Attribute(key).Value, "\"")#>];}} <#}#> <# var connectionStrings = xDocument.Descendants("connectionStrings"); foreach(var connString in connectionStrings.Descendants()) {#> public string <#= connString.Attribute(name).Value#>{get {return ConfigurationManager.ConnectionStrings[<#= string.Format("{0}{1}{2}","\"" , connString.Attribute(name).Value, "\"")#>].ConnectionString;}} <#} #> }} The resulting .cs file: using System;using System.Configuration;using System.Xml;using System.Xml.Linq;using System.Linq;namespace MyProject.Web { public partial class Configurator { public string ClientValidationEnabled{get {return ConfigurationManager.AppSettings["ClientValidationEnabled"];}} public string UnobtrusiveJavaScriptEnabled{get {return ConfigurationManager.AppSettings["UnobtrusiveJavaScriptEnabled"];}} public string ServiceUri{get {return ConfigurationManager.AppSettings["ServiceUri"];}} public string TestConnection{get {return ConfigurationManager.ConnectionStrings["TestConnection"].ConnectionString;}} public string SecondTestConnection{get {return ConfigurationManager.ConnectionStrings["SecondTestConnection"].ConnectionString;}} }} Next, I extended the partial class for easy access to the Configuration. However, you could just use the generated class file itself. using System;using System.Linq;using System.Xml.Linq;namespace MyProject.Web{ public partial class Configurator { private static readonly Configurator Instance = new Configurator(); public static Configurator For { get { return Instance; } } }} Finally, in my example, I used the Configurator class like so: [TestMethod] public void Test_Web_Config() { var result = Configurator.For.ServiceUri; Assert.AreEqual(result, "http://localhost:30237/Service1/"); }

    Read the article

  • Confused as to which Prototype helper to use

    - by user284194
    After reading http://api.rubyonrails.org/classes/ActionView/Helpers/PrototypeHelper.html I just can't seem to find what I'm looking for. I have a simplistic model that deletes the oldest message after the list of messages reaches 24, the model is this simple: class Message < ActiveRecord::Base after_create :destroy_old_messages protected def destroy_old_messages messages = Message.all(:order => 'updated_at DESC') messages[24..-1].each {|p| p.destroy } if messages.size >= 24 end end There is a message form below the list of messages which is used to add new messages. I'm using Prototype/RJS to add new messages to the top of the list. create.rjs: page.insert_html :top, :messages, :partial => @message page[@message].visual_effect :grow #page[dom_id(@messages)].replace :partial => @message page[:message_form].reset My index.html.erb is very simple: <div id="messages"> <%= render :partial => @messages %> </div> <%= render :partial => "message_form" %> When new messages are added they appear just fine, but when the 24 message limit has been reached it just keeps adding messages and doesn't remove the old ones. Ideally I'd like them to fade out as the new ones are added, but they can just disappear. The commented line in create.rjs actually works, it removes the expired message but I lose the visual effect when adding a new message. Does anyone have a suggestion on how to accomplish adding and removing messages from this simple list with effects for both? Help would be greatly appreciated. Thanks for reading. P.S.: would periodically_call_remote be helpful in this situation?

    Read the article

  • ASP.NET MVC - PartialView not refreshing

    - by Billy Logan
    Hello Everyone, I have a view that uses a javascript callback to reload a partial view. For whatever reason the contents of the partial class do not refresh even though i can step through the entire process and see the page being recalled and populated. Any reason why the page would not display? Code is as follows: <div id="big_image_content"> <% Html.RenderPartial("ZoomImage", Model); %> </div> This link should reload the div above: <a href="javascript:void(0)" onclick="$('#big_image_content').load('/ShopDetai/ZoomImage);" title="<%= shape.Shape %>" alt="<%= shape.Shape %>"> <img src="http://images.rugs-direct.com/<%= shape.Image.ToLower() %>" width="40" alt="<%= shape.Shape %>"> </a> partial view(ZoomImage.ascx) simplified for now, but still doesn't load: <%@ Control Language="C#" Inherits="System.Web.Mvc.ViewUserControl<RugsDirect.Data.ItemDetailsModel>" %> <%= Model.Category.ToLower()%> And finally the controller side of things: public ActionResult ZoomImage() { try { ItemDetailsModel model = GetMainImageContentModel(); return PartialView("ZoomImage", model); } catch (Exception ex) { //send the error email ExceptionPolicy.HandleException(ex, "Exception Policy"); //redirect to the error page return RedirectToAction("ViewError", "Shop"); } } Again, i can step through this entire process and all seems to be working accept for the page not reloading. I can even break on the <%= Model.Category.ToLower()% of the partial view, but it will not be displayed. Thanks in advance, Billy

    Read the article

  • Passing an instance variable through RJS?

    - by Elliot
    Hey guys here is my code (roughly): books.html.erb <% @books.each do |book| %> <% @bookid = book.id %> <div id="enter_stuff"> <%= render "input", :bookid => @bookid %> </div> <%end%> _input.html.erb <% @book = Book.find_by_id(@bookid) %> <strong>your book is: <%=h @book.name %></strong> create.rjs page.replace_html :enter_stuff, :partial => 'input', :object => @bookid The problem here is that only create.js doesn't seem to work (though, if instead of passing the partial I passed "..." it does work, so I know its that there are instance variables in the partial that aren't being reset. Any ideas?) So the final question, is how do I pass an instance variable to a partial through the create.rjs file? p.s. I know I will have duplicate div IDs, I'm not worrying about that for now though. Best, Elliot

    Read the article

  • View Models (ViewData), UserControls/Partials and Global variables - best practice?

    - by elado
    Hi I'm trying to figure out a good way to have 'global' members (such as CurrentUser, Theme etc.) in all of my partials as well as in my views. I don't want to have a logic class that can return this data (like BL.CurrentUser) I do think it needs to be a part of the Model in my views So I tried inheriting from BaseViewData with these members. In my controllers, in this way or another (a filter or base method in my BaseController), I create an instance of the inheriting class and pass it as a view data. Everything's perfect till this point, cause then I have my view data available on the main View with the base members. But what about partials? If I have a simple partial that needs to display a blog post then it looks like this: <%@ Control Language="C#" AutoEventWireup="true" Inherits="ViewUserControl<Post>" %> and simple code to render this partial in my view (that its model.Posts is IEnumerable<Post>): <%foreach (Post p in this.Model.Posts) {%> <%Html.RenderPartial("Post",p); %> <%}%> Since the partial's Model isn't BaseViewData, I don't have access to those properties. Hence, I tried to make a class named PostViewData which inherits from BaseViewData, but then my containing views will have a code to actually create the PostViewData in them in order to pass it to the partial: <%Html.RenderPartial("Post",new PostViewData { Post=p,CurrentUser=Model.CurrentUser,... }); %> Or I could use a copy constructor <%Html.RenderPartial("Post",new PostViewData(Model) { Post=p }); %> I just wonder if there's any other way to implement this before I move on. Any suggestions? Thanks!

    Read the article

  • Validation L2S question

    - by user158020
    This may be a bit winded because I am new to wpf. I have created a partial class for an entity in my L2S class that is primarily used for validation. It implements the onchanging and onvalidate methods. I am trying to use the MVVM pattern, and in a window/view I have set the datacontext in the xaml: <Window.DataContext> <vm:StartViewModel /> </Window.DataContext> when a user leaves a required field in the view blank, the onchanging event of the partial class is fired when I close the form, not when I save the data. So, if a user leaves the textbox blank, the old value is retained and the onchaging method is fired, but I have no idea how to alert the user of the resulting error. here is my onchanging code in the partial class: partial void Ondocument_titleChanging(string value) { if (value.Length == 0) throw new Exception("Document title is required."); if (value.Length > 256) throw new Exception("Document title cannot be longer than 256 characters."); } throwing an exception doesn't notify the user of the error. it just allows the form to close and rejects the changes to the textbox. hope this makes sense... edit: this example was taken from Scott Guthries article here: http://aspalliance.com/1427_LINQ_to_SQL_Part_5__Binding_UI_using_the_ASPLinqDataSource_Control.5

    Read the article

  • RegisterClientScriptInclude doesn't work for some reason...

    - by Andrew
    Hey, I've spent at least 2 days trying anything and googling this...but for some reason I can't get RegisterClientScriptInclude to work the way everyone else has it working? First off, I am usting .NET 3.5 Ajax,...and I am including javascript in my partial page refreshes...using this code: ScriptManager.RegisterClientScriptBlock(this, typeof(Page), "MyClientCode", script, true); It works perfectly, my javascript code contained in the script variable is included every partial refresh. The javascript in script is actually quite extensive though, and I would like to store it in a .js file,..so logically I make a .js file and try to include it using RegisterClientScriptInclude ...however i can't for the life of my get this to work. here's the exact code: ScriptManager.RegisterClientScriptInclude(this, typeof(Page), "mytestscript", "/js/testscript.js"); the testscript.js file is only included in FULL page refreshes...ie. when I load the page, or do a full postback....i can't get the file to be included in partial refreshes...have no idea why..when viewing the ajax POST in firebug I don't see a difference whether I include the file or not.... both of the ScriptManager Includes are being ran from the exact same place in "Page_Load"...so they should execute every partial refresh (but only the ScriptBlock does). anyways,..any help or ideas,..or further ways I can trouble shoot this problem, would be appreciated. Thanks, Andrew

    Read the article

  • AngularJS: How to make angular load script inside ng-include?

    - by Ranjith R
    Hey I am building a web page with angular. The problem is that there are somethings already build without angular and I have to include them as well The problem is this. I have something like this in my main.html: <ngInclude src="partial.html"> </ngInclude> And my partial.html has something like this <h2> heading 1 <h2> <script type="text/javascript" src="static/js/partial.js"> </script> And my partial.js has nothing to do with angularjs. nginclude works and I can see the html, but I can not see the javascript file being loaded at all. I know how to use firebug/ chrome-dev-tool, but I can not even see the network request being made. What am I doing wrong? I knwo angular has some special meaning to script tag. Can I override it?

    Read the article

  • Rails syntax for comments in templates: is this bug understood?

    - by brahn
    Using rails 2.3.2 I have a partial _foo.rhtml that begins with a comment as follows: <% # here is a comment %> <li><%= foo %></li> When I render the partial from a view in the traditional way, e.g. <% some_numbers = [1, 2, 3, 4, 5] %> <ul> <%= render :partial => "foo", :collection => some_numbers %> </ul> I found that the <li> and </li> tags are ommitted in the output -- i.e. the resulting HTML is <ul> 1 2 3 4 5 </ul> However, I can solve this problem by fixing _foo.rhtml to eliminate the space between the <% and the # so that the partial now reads: <%# here is a comment %> <li><%= foo %></li> My question: what's going on here? E.g., is <% # comment %> simply incorrect syntax for including comments in a template? Or is the problem more subtle? Thanks!

    Read the article

  • Good Replacement for User Control?

    - by David Lively
    I found user controls to be incredibly useful when working with ASP.NET webforms. By encapsulating the code required for displaying a control with the markup, creation of reusable components was very straightforward and very, very useful. While MVC provides convenient separation of concerns, this seems to break encapsulation (ie, you can add a control without adding or using its supporting code, leading to runtime errors). Having to modify a controller every time I add a control to a view seems to me to integrate concerns, not separate them. I'd rather break the purist MVC ideology than give up the benefits of reusable, packaged controls. I need to be able to include components similar to webforms user controls throughout a site, but not for the entire site, and not at a level that belongs in a master page. These components should have their own code not just markup (to interact with the business layer), and it would be great if the page controller didn't need to know about the control. Since MVC user controls don't have codebehind, I can't see a good way to do this. I've searched previous SO questions, and have yet to find a good answer. Options so far In an attempt to avoid turning the comments section into a discussion... RenderAction This allows the view to call another controller, which will be responsible for interacting with the BLL and whatever data is necessary to its corresponding view. The calling view needs to be aware of the sub controller. This seems to provide a nice way to encapsulate partial views and controls, without having to modify the calling controller. RenderPartial The calling controller is still responsible for executing whatever code is associated with the partial view, and making sure that the model passed to the partial view contains the data it expects. Effectively, modifying the partial view potentially means modifying the calling controller. Annoying especially if this is used in multiple places. Portable Areas Place each control in its own project/area?

    Read the article

  • Django Save Incomplete Progress on Form

    - by jimbob
    I have a django webapp with multiple users logging in and fill in a form. Some users may start filling in a form and lack some required data (e.g., a grant #) needed to validate the form (and before we can start working on it). I want them to be able to fill out the form and have an option to save the partial info (so another day they can log back in and complete it) or submit the full info undergoing validation. Currently I'm using ModelForm for all the forms I use, and the Model has constraints to ensure valid data (e.g., the grant # has to be unique). However, I want them to be able to save this intermediary data without undergoing any validation. The solution I've thought of seems rather inelegant and un-django-ey: create a "Save Partial Form" button that saves the POST dictionary converts it to a shelf file and create a "SavedPartialForm" model connecting the user to partial forms saved in the shelf. Does this seem sensible? Is there a better way to save the POST dict directly into the db? Or is an add-on module that does this partial-save of a form (which seems to be a fairly common activity with webforms)? My biggest concern with my method is I want to eventually be able to do this form-autosave automatically (say every 10 minutes) in some ajax/jquery method without actually pressing a button and sending the POST request (e.g., so the user isn't redirected off the page when autosave is triggered). I'm not that familiar with jquery and am wondering if it would be possible to do this.

    Read the article

  • ASP.net MVC Linq-To-SQL Extended Class Field Binding

    - by user336858
    Hi there, The short version of this question is "Is there a way to get automatic View Object binding for fields defined in a partial class for a Linq-To-SQL generated class?" Apologies if it's been asked before. Example Suppose I have a typical MVC setup with the tables: Posts {PostID, ...} Categories {CategoryID, ...} A post can have more than one category, and a category can identify more than one post. Thus suppose further that I need an extra table: PostCategories {PostID, CategoryID, ...} This handles the many-to-many relationship between posts and categories. As far as I know, there's no way to do this in Linq-to-SQL right now so I have to shoehorn it in by adding a partial Postclass to the project to add that functionality. Something like: public partial class Post { public IEnumerable<Category> Categories{ get { ... } set { ... } } } So here's my question: If a user is accessing my MVC application front-end and begins editing a Post object, they might enter an invalid category. When the server recognizes the invalid input, the usual practice is to return the faulty object to the original view for re-editing along with some error messages. The fields in the edit page are re-populated with the provided values. However I don't know how to get this mechanism to work with the properties I created with the partial class as shown above. Any terminology, links, or tips you can provide would be tremendously helpful!

    Read the article

  • In Rails 3, how does one render HTML within a JSON response?

    - by ylg
    I'm porting an application from Merb 1.1 / 1.8.7 to Rails 3 (beta) / 1.9.1 that uses JSON responses containing HTML fragments, e.g., a JSON container specifying an update, on a user record, and the updated user row looks like . In Merb, since whatever a controller method returns is given to the client, one can put together a Hash, assign a rendered partial to one of the keys and return hash.to_json (though that certainly may not be the best way.) In Rails, it seems that to get data back to the client one must use render and render can only be called once, so rendering the hash to json won't work because of the partial render. From reading around, it seems one could put that data into a JSON .erb view file, with <%= render partial % in and render that. Is there a Rails-way of solving this problem (return JSON containing one or more HTML fragments) other than that? In Merb: only_provides :json ... self.status = 204 # or appropriate if not async return { 'action' => 'update', 'type' => 'user', 'id' => @user.id, 'html' => partial('user_row', format: :html, user: @user) }.to_json In Rails?

    Read the article

  • Bash Script - Traffic Shaping

    - by Craig-Aaron
    hey all, I was wondering if you could have a look at my script and help me add a few things to it, How do I get it to find how many active ethernet ports I have? and how do I filter more than 1 ethernet port How I get this to do a range of IP address? Once I have a few ethenet ports I need to add traffic control to each one #!/bin/bash # Name of the traffic control command. TC=/sbin/tc # The network interface we're planning on limiting bandwidth. IF=eth0 # Network card interface # Download limit (in mega bits) DNLD=10mbit # DOWNLOAD Limit # Upload limit (in mega bits) UPLD=1mbit # UPLOAD Limit # IP address range of the machine we are controlling IP=192.168.0.1 # Host IP # Filter options for limiting the intended interface. U32="$TC filter add dev $IF protocol ip parent 1:0 prio 1 u32" start() { # Hierarchical Token Bucket (HTB) to shape bandwidth $TC qdisc add dev $IF root handle 1: htb default 30 #Creates the root schedlar $TC class add dev $IF parent 1: classid 1:1 htb rate $DNLD #Creates a child schedlar to shape download $TC class add dev $IF parent 1: classid 1:2 htb rate $UPLD #Creates a child schedlar to shape upload $U32 match ip dst $IP/24 flowid 1:1 #Filter to match the interface, limit download speed $U32 match ip src $IP/24 flowid 1:2 #Filter to match the interface, limit upload speed } stop() { # Stop the bandwidth shaping. $TC qdisc del dev $IF root } restart() { # Self-explanatory. stop sleep 1 start } show() { # Display status of traffic control status. $TC -s qdisc ls dev $IF } case "$1" in start) echo -n "Starting bandwidth shaping: " start echo "done" ;; stop) echo -n "Stopping bandwidth shaping: " stop echo "done" ;; restart) echo -n "Restarting bandwidth shaping: " restart echo "done" ;; show) echo "Bandwidth shaping status for $IF:" show echo "" ;; *) pwd=$(pwd) echo "Usage: tc.bash {start|stop|restart|show}" ;; esac exit 0 thanks

    Read the article

  • "Unable to initialize module" fileinfo php-pecl-Fileinfo.x86_64

    - by Myers Network
    I have a brand new server server that I am trying to get setup up. This is a 64 bit machine that I can not install "fileinfo" or "memcache". I have uninstalled these and reinstalled them using yum and pecl with no luck. Yum install fine "no error" but then get error when running php. pecl from what I can tell is only installing 32bit. Does not put anything in the lib64 directory. Here is my output from php -v: PHP Warning: PHP Startup: fileinfo: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP Warning: PHP Startup: memcache: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP 5.2.14 (cli) (built: Aug 12 2010 16:03:48) Copyright (c) 1997-2010 The PHP Group Zend Engine v2.2.0, Copyright (c) 1998-2010 Zend Technologies Here is some other system info incase you need it uname: Linux server.actham.us 2.6.18-194.26.1.el5 #1 SMP Tue Nov 9 12:54:20 EST 2010 x86_64 x86_64 x86_64 GNU/Linux php -m: PHP Warning: PHP Startup: fileinfo: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP Warning: PHP Startup: memcache: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 [PHP Modules] bz2 calendar ctype curl date dbase dom exif filter ftp gd gettext gmp hash iconv imap json ldap libxml mbstring mcrypt mysql mysqli openssl pcntl pcre PDO pdo_mysql pdo_sqlite readline Reflection session shmop SimpleXML sockets SPL standard tokenizer wddx xml xmlreader xmlrpc xmlwriter xsl zip zlib [Zend Modules] Any help would be greatly appreciated, thanks....

    Read the article

  • "Unable to initialize module" fileinfo php-pecl-Fileinfo.x86_64

    - by Myers Network
    I have a brand new server server that I am trying to get setup up. This is a 64 bit machine that I can not install "fileinfo" or "memcache". I have uninstalled these and reinstalled them using yum and pecl with no luck. Yum install fine "no error" but then get error when running php. pecl from what I can tell is only installing 32bit. Does not put anything in the lib64 directory. Here is my output from php -v: PHP Warning: PHP Startup: fileinfo: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP Warning: PHP Startup: memcache: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP 5.2.14 (cli) (built: Aug 12 2010 16:03:48) Copyright (c) 1997-2010 The PHP Group Zend Engine v2.2.0, Copyright (c) 1998-2010 Zend Technologies Here is some other system info incase you need it uname: Linux server.actham.us 2.6.18-194.26.1.el5 #1 SMP Tue Nov 9 12:54:20 EST 2010 x86_64 x86_64 x86_64 GNU/Linux php -m: PHP Warning: PHP Startup: fileinfo: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 PHP Warning: PHP Startup: memcache: Unable to initialize module Module compiled with module API=20050922, debug=0, thread-safety=0 PHP compiled with module API=20060613, debug=0, thread-safety=0 These options need to match in Unknown on line 0 [PHP Modules] bz2 calendar ctype curl date dbase dom exif filter ftp gd gettext gmp hash iconv imap json ldap libxml mbstring mcrypt mysql mysqli openssl pcntl pcre PDO pdo_mysql pdo_sqlite readline Reflection session shmop SimpleXML sockets SPL standard tokenizer wddx xml xmlreader xmlrpc xmlwriter xsl zip zlib [Zend Modules] Any help would be greatly appreciated, thanks....

    Read the article

  • Cannot 301 redirect with IIS URL Rewrite Module

    - by Justin
    I am trying to troubleshoot my issue with the URL Rewrite Module on IIS 7. I migrated a Wordpress blog over to BlogEngine.net. There were only about 5 entries that I wanted to use 301 redirects to the new blog, so I wanted to simply create 5 exact match redirect rules using the rewrite module. For some reason the exact match rule never seems to take effect, I always get a 404 error when the original url is navigated to. I verified that my exact match pattern matched the existing backlinks and it does. I then tried a simple test and got the same behavior, no redirection. I created a page, test.html, on my site, I then created a second page, test2.html. So my exact match pattern is: "http://www.mydomain.com/test.html" And the rule is supposed to do a 301 redirect to "http://www.mydomain.com/test2.html " The redirect never happens. I created the steps for the rule based on the instructions in this page: http://learn.iis.net/page.aspx/461/creating-rewrite-rules-for-the-url-rewrite-module/ I don't see that I left out a step. After I apply the rule I've even gone as far as doing an IISReset to make sure it would be in effect but still no luck. Any thoughts on what I might have left out? (Note: my rewrite rules dont include the " " around them but I had to add since serverfault thinks I am trying to spam the system with multiple urls.)

    Read the article

  • Multiple Set Peer for VPN Failover

    - by Kyle Brandt
    I will have two Cisco routers at Location A serving the same internal networks, and one router in location B. Currently, I have one router in each location with a IPSec site-to-site tunnel connecting them. It looks something like: Location A: crypto map crypto-map-1 1 ipsec-isakmp description Tunnel to Location B set peer 12.12.12.12 set transform-set ESP-3DES-SHA match address internal-ips Location B: crypto map crypto-map-1 1 ipsec-isakmp description Tunnel to Location A set peer 11.11.11.11 set transform-set ESP-3DES-SHA match address internal-ips Can I achieve fail over by simply adding another set peer at location B?: Location A (New secondary Router, configuration on previous router stays the same): crypto map crypto-map-1 1 ipsec-isakmp description Tunnel to Location B set peer 12.12.12.12 set transform-set ESP-3DES-SHA match address internal-ips Location B (Configuration Changed): crypto map crypto-map-1 1 ipsec-isakmp description Tunnel to Location A set peer 11.11.11.11 ! 11.11.11.100 is the ip of the new second router at location A set peer 11.11.11.100 set transform-set ESP-3DES-SHA match address internal-ips Cisco Says: For crypto map entries created with the crypto map map-name seq-num ipsec-isakmp command, you can specify multiple peers by repeating this command. The peer that packets are actually sent to is determined by the last peer that the router heard from (received either traffic or a negotiation request from) for a given data flow. If the attempt fails with the first peer, Internet Key Exchange (IKE) tries the next peer on the crypto map list. But I don't fully understand that in the context of a failover scenerio (One of the routers as Location A blowing up).

    Read the article

< Previous Page | 61 62 63 64 65 66 67 68 69 70 71 72  | Next Page >