Search Results

Search found 9254 results on 371 pages for 'approach'.

Page 345/371 | < Previous Page | 341 342 343 344 345 346 347 348 349 350 351 352  | Next Page >

• Dealing with personal failure

  - by codeelegance

  A while ago I was given the task of updating and extending the functionality of a software project. I was given a year to make the needed changes working solo. A month into development I came to the conclusion that it would take longer to change the existing product than to rewrite it from the ground up. I'd never attempted a complete rewrite so I talked with my boss about it and he was thrilled with the idea. I'm a fan of agile development but had never had the opportunity to take advantage of all of the prescribed practices so when I set to work I tried to incorporate as many as I could. I didn't have direct access to the customer and my coworkers (non-programmers) knew the business domain but were already so busy they didn't really have time to participate in design meetings so I resigned to working in the dark and occasionally calling one of them over to my desk to get feedback on my progress. I used TDD and refactored mercilessly and even tried taking a domain driven design approach. Things went well for a while. As the deadline came closer and the complexity of the project grew my productivity start slipping. I found myself cutting corners and ignoring the practices I had established as the pressure increased to meet the deadline. I also started working late nights and weekends to keep up with the load. In the end it made little difference how hard I worked. The project missed its deadline and what was completed wasn't enough to give to the customer. I had failed. Not only had I not finished on time but the previous version had sat untouched for almost a year so it wouldn't be of any help. Luckily we had another product that offered some of the same functionality. My boss decided to cancel the project entirely and moved all our orphaned customers to the other product. I spent weeks (along with everyone else at the company) manning the phones providing technical support for those customers. After it was all over, my boss was gracious enough not to fire me for nearly ruining the company. I was moved to the other product and have been trying to redeem myself ever since. Where did I go wrong? Has anyone else had to deal with this kind of defeat? How did you recover?

  Read the article

• How can I refactor this JavaScript code to avoid making functions in a loop?

  - by Bungle

  I wrote the following code for a project that I'm working on: var clicky_tracking = [ ['related-searches', 'Related Searches'], ['related-stories', 'Related Stories'], ['more-videos', 'More Videos'], ['web-headlines', 'Publication'] ]; for (var x = 0, length_x = clicky_tracking.length; x < length_x; x++) { links = document.getElementById(clicky_tracking[x][0]) .getElementsByTagName('a'); for (var y = 0, length_y = links.length; y < length_y; y++) { links[y].onclick = (function(name, url) { return function() { clicky.log(url, name, 'outbound'); }; }(clicky_tracking[x][1], links[y].href)); } } What I'm trying to do is: define a two-dimensional array, with each instance the inner arrays containing two elements: an id attribute value (e.g., "related-searches") and a corresponding description (e.g., "Related Searches"); for each of the inner arrays, find the element in the document with the corresponding id attribute, and then gather a collection of all <a> elements (hyperlinks) within it; loop through that collection and attach an onclick handler to each hyperlink, which should call clicky.log, passing in as parameters the description that corresponds to the id (e.g., "Related Searches" for the id "related-searches") and the value of the href attribute for the <a> element that was clicked. Hopefully that wasn't thoroughly confusing! The code may be more self-explanatory than that. I believe that what I've implemented here is a closure, but JSLint complains: http://img.skitch.com/20100526-k1trfr6tpj64iamm8r4jf5rbru.png So, my questions are: How can I refactor this code to make JSLint agreeable? Or, better yet, is there a best-practices way to do this that I'm missing, regardless of what JSLint thinks? Should I rely on event delegation instead? That is, attaching onclick event handlers to the document elements with the id attributes in my arrays, and then looking at event.target? I've done that once before and understand the theory, but I'm very hazy on the details, and would appreciate some guidance on what that would look like - assuming this is a viable approach. Thanks very much for any help!

  Read the article

• Matching array elements to get an array element at another index

  - by bccarlso

  I have a PHP array that I'm using to generate an HTML form. The PHP array is this: <?php $vdb = array ( array( "Alabama", 275), array( "Alaska", 197), array( "Arizona", 3322)); ?> The PHP to generate the HTML form is below. I need to have the value be the name of the state, because there is some AJAX I'm using to display which states a user has chosen. <?php echo "<table border='1'><thead><tr><th></th><th>State</th><th>Contacts</th><th>Email</th></tr></thead>"; for ($row = 0; $row < 42; $row++) { echo "<tr><td class='input_button'><input type='checkbox' name='vdb[]' value='".$vdb[$row][0]."' title='".$vdb[$row][1]."' /></td>"; echo "<td>".$vdb[$row][0]."</td>"; echo "<td>".$vdb[$row][1]."</td>"; } echo "</table>"; ?> What I'm trying to do is, on submission of the form, with the states the user selected, loop through the PHP array and total the numbers from the selected states. So if I checked Alabama and Alaska, I'd want to add 275 + 197. This is what I thought would have worked, but it's not: <?php $vendors = array(); if (isset($_POST["vdb"])) { $vendors = $_POST["vdb"]; } $ven_i = 0; $ven_j = 0; $ven_total = 0; foreach ($vendors as $value) { foreach ($vdb as $vdb_value) { if ($vendors[$ven_i] == $vdb[$ven_j][0]) { $ven_total += $vdb[$ven_j][1]; } $ven_j++; } $ven_i++; } ?> and then $ven_total should be the total I'm looking for. However, $ven_total just ends up being the first checkbox selected, and it ignores the rest. I am doing this correctly with the AJAX, displaying the total on the front end, but I don't know how to pass that on to the form submission. I'd rather not using GET and URL variables, because a user could type something into the URL and modify the count. Any idea what I'm doing wrong, or a better way to approach this that I would be able to understand? (Very much a novice programmer.)

  Read the article

• .NET: Is there a way to finagle a default namespace in an XPath 1.0 query?

  - by Cheeso

  I'm building a tool that performs xpath 1.0 queries on XHTML documents. The requirement to use a namespace prefix in the query is killing me. The query looks like this: html/body/div[@class='contents']/div[@class='body']/ div[@class='pgdbbyauthor']/h2[a[@name][starts-with(.,'Quick')]]/ following-sibling::ul[1]/li/a (all on one line) ...which is bad enough, except because it's xpath 1.0, I need to use an explicit namespace prefix on each QName, so it looks like this: ns1:html/ns1:body/ns1:div[@class='contents']/ns1:div[@class='body']/ ns1:div[@class='pgdbbyauthor']/ns1:h2[ns1:a[@name][starts-with(.,'Quick')]]/ following-sibling::ns1:ul[1]/ns1:li/ns1:a To set up the query, I do something like this: var xpathDoc = new XPathDocument(new StringReader(theText)); var nav = xpathDoc.CreateNavigator(); var xmlns = new XmlNamespaceManager(nav.NameTable); foreach (string prefix in xmlNamespaces.Keys) xmlns.AddNamespace(prefix, xmlNamespaces[prefix]); XPathNodeIterator selection = nav.Select(xpathExpression, xmlns); But what I want is for the xpathExpression to use the implicit default namespace. Is there a way for me to transform the unadorned xpath expression, after it's been written, to inject a namespace prefix for each element name in the query? I'm thinking, anything between two slashes, I could inject a prefix there. Excepting of course axis names like "parent::" and "preceding-sibling::" . And wildcards. That's what I mean by "finagle a default namespace". Is this hack gonna work? Addendum Here's what I mean. suppose I have an xpath expression, and before passing it to nav.Select(), I transform it. Something like this: string FixupWithDefaultNamespace(string expr) { string s = expr; s = Regex.Replace(s, "^(?!::)([^/:]+)(?=/)", "ns1:$1"); // beginning s = Regex.Replace(s, "/([^/:]+)(?=/)", "/ns1:$1"); // stanza s = Regex.Replace(s, "::([A-Za-z][^/:*]*)(?=/)", "::ns1:$1"); // axis specifier s = Regex.Replace(s, "\\[([A-Za-z][^/:*\\(]*)(?=[\\[\\]])", "[ns1:$1"); // predicate s = Regex.Replace(s, "/([A-Za-z][^/:]*)(?!<::)$", "/ns1:$1"); // end s = Regex.Replace(s, "^([A-Za-z][^/:]*)$", "ns1:$1"); // edge case s = Regex.Replace(s, "([-A-Za-z]+)\\(([^/:\\.,\\)]+)(?=[,\\)])", "$1(ns1:$2"); // xpath functions return s; } This actually works for simple cases I tried. To use the example from above - if the input is the first xpath expression, the output I get is the 2nd one, with all the ns1 prefixes. The real question is, is it hopeless to expect this Regex.Replace approach to work, as the xpath expressions get more complicated?

  Read the article

• Cepstral Analysis for pitch detection

  - by Ohmu

  Hi! I'm looking to extract pitches from a sound signal. Someone on IRC just explain to me how taking a double FFT achieves this. Specifically: take FFT take log of square of absolute value (can be done with lookup table) take another FFT take absolute value I am attempting this using vDSP I can't understand how I didn't come across this technique earlier. I did a lot of hunting and asking questions; several weeks worth. More to the point, I can't understand why I didn't think of it. I am attempting to achieve this with vDSP library. it looks as though it has functions to handle all of these tasks. However, I'm wondering about the accuracy of the final result. I have previously used a technique which scours the frequency bins of a single FFT for local maxima. when it encounters one, it uses a cunning technique (the change in phase since the last FFT) to more accurately place the actual peak within the bin. I am worried that this precision will be lost with this technique I'm presenting here. I guess the technique could be used after the second FFT to get the fundamental accurately. But it kind of looks like the information is lost in step 2. as this is a potentially tricky process, could someone with some experience just look over what I'm doing and check it for sanity? also, I've heard there is an alternative technique involving fitting a quadratic over neighbouring bins. Is this of comparable accuracy? if so, I would favour it, as it doesn't involve remembering bin phases. so questions: does this approach makes sense? Can it be improved? I'm a bit worried about And the log square component; there seems to be a vDSP function to do exactly that: vDSP_vdbcon however, there is no indication it precalculates a log-table -- I assume it doesn't, as the FFT function requires an explicit pre-calculation function to be called and passed into it. and this function doesn't. Is there some danger of harmonics being picked up? is there any cunning way of making vDSP pull out the maxima, biggest first? Can anyone point me towards some research or literature on this technique? the main question: is it accurate enough? Can the accuracy be improved? I have just been told by an expert that the accuracy IS INDEED not sufficient. Is this the end of the line? Pi PS I get SO annoyed (npi) when I want to create tags, but cannot. :| I have suggested to the maintainers that SO keep track of attempted tags, but I'm sure I was ignored. we need tags for vDSP, accelerate framework, cepstral analysis

  Read the article

• Using fft2 with reshaping for an RGB filter

  - by Mahmoud Aladdin

  I want to apply a filter on an image, for example, blurring filter [[1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0]]. Also, I'd like to use the approach that convolution in Spatial domain is equivalent to multiplication in Frequency domain. So, my algorithm will be like. Load Image. Create Filter. convert both Filter & Image to Frequency domains. multiply both. reconvert the output to Spatial Domain and that should be the required output. The following is the basic code I use, the image is loaded and displayed as cv.cvmat object. Image is a class of my creation, it has a member image which is an object of scipy.matrix and toFrequencyDomain(size = None) uses spf.fftshift(spf.fft2(self.image, size)) where spf is scipy.fftpack and dotMultiply(img) uses scipy.multiply(self.image, image) f = Image.fromMatrix([[1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0]]) lena = Image.fromFile("Test/images/lena.jpg") print lena.image.shape lenaf = lena.toFrequencyDomain(lena.image.shape) ff = f.toFrequencyDomain(lena.image.shape) lenafm = lenaf.dotMultiplyImage(ff) lenaff = lenafm.toTimeDomain() lena.display() lenaff.display() So, the previous code works pretty well, if I told OpenCV to load the image via GRAY_SCALE. However, if I let the image to be loaded in color ... lena.image.shape will be (512, 512, 3) .. so, it gives me an error when using scipy.fttpack.ftt2 saying "When given, Shape and Axes should be of same length". What I tried next was converted my filter to 3-D .. as [[[1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0]], [[1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0]], [[1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0], [1/9.0, 1/9.0, 1/9.0]]] And, not knowing what the axes argument do, I added it with random numbers as (-2, -1, -1), (-1, -1, -2), .. etc. until it gave me the correct filter output shape for the dotMultiply to work. But, of course it wasn't the correct value. Things were totally worse. My final trial, was using fft2 function on each of the components 2-D matrices, and then re-making the 3-D one, using the following code. # Spiltting the 3-D matrix to three 2-D matrices. for i, row in enumerate(self.image): r.append(list()) g.append(list()) b.append(list()) for pixel in row: r[i].append(pixel[0]) g[i].append(pixel[1]) b[i].append(pixel[2]) rfft = spf.fftshift(spf.fft2(r, size)) gfft = spf.fftshift(spf.fft2(g, size)) bfft = spf.fftshift(spf.fft2(b, size)) newImage.image = sp.asarray([[[rfft[i][j], gfft[i][j], bfft[i][j]] for j in xrange(len(rfft[i]))] for i in xrange(len(rfft))] ) return newImage Any help on what I made wrong, or how can I achieve that for both GreyScale and Coloured pictures.

  Read the article

• SQL Server CTE referred in self joins slow

  - by Kharlos Dominguez

  Hello, I have written a table-valued UDF that starts by a CTE to return a subset of the rows from a large table. There are several joins in the CTE. A couple of inner and one left join to other tables, which don't contain a lot of rows. The CTE has a where clause that returns the rows within a date range, in order to return only the rows needed. I'm then referencing this CTE in 4 self left joins, in order to build subtotals using different criterias. The query is quite complex but here is a simplified pseudo-version of it WITH DataCTE as ( SELECT [columns] FROM table INNER JOIN table2 ON [...] INNER JOIN table3 ON [...] LEFT JOIN table3 ON [...] ) SELECT [aggregates_columns of each subset] FROM DataCTE Main LEFT JOIN DataCTE BananasSubset ON [...] AND Product = 'Bananas' AND Quality = 100 LEFT JOIN DataCTE DamagedBananasSubset ON [...] AND Product = 'Bananas' AND Quality < 20 LEFT JOIN DataCTE MangosSubset ON [...] GROUP BY [ I have the feeling that SQL Server gets confused and calls the CTE for each self join, which seems confirmed by looking at the execution plan, although I confess not being an expert at reading those. I would have assumed SQL Server to be smart enough to only perform the data retrieval from the CTE only once, rather than do it several times. I have tried the same approach but rather than using a CTE to get the subset of the data, I used the same select query as in the CTE, but made it output to a temp table instead. The version referring the CTE version takes 40 seconds. The version referring the temp table takes between 1 and 2 seconds. Why isn't SQL Server smart enough to keep the CTE results in memory? I like CTEs, especially in this case as my UDF is a table-valued one, so it allowed me to keep everything in a single statement. To use a temp table, I would need to write a multi-statement table valued UDF, which I find a slightly less elegant solution. Did some of you had this kind of performance issues with CTE, and if so, how did you get them sorted? Thanks, Kharlos

  Read the article

• Setting up a "to-many" relationship value dependency for a transient Core Data attribute

  - by Greg Combs

  I've got a relatively complicated Core Data relationship structure and I'm trying to figure out how to set up value dependencies (or observations) across various to-many relationships. Let me start out with some basic info. I've got a classroom with students, assignments, and grades (students X assignments). For simplicity's sake, we don't really have to focus much on the assignments yet. StudentObj <--->> ScoreObj <<---> AssignmentObj Each ScoreObj has a to-one relation with the StudentObj and the AssignmentObj. ScoreObj has real attributes for the numerical grade, the turnInDate, and notes. AssignmentObj.scores is the set of Score objects for that assignment (N = all students). AssignmentObj has real attributes for name, dueDate, curveFunction, gradeWeight, and maxPoints. StudentObj.scores is the set of Score objects for that student (N = all assignments). StudentObj also has real attributes like name, studentID, email, etc. StudentObj has a transient (calculated, not stored) attribute called gradeTotal. This last item, gradeTotal, is the real pickle. it calculates the student's overall semester grade using the scores (ScoreObj) from all their assignments, their associated assignment gradeWeights, curves, and maxPoints, and various other things. This gradeTotal value is displayed in a table column, along with all the students and their individual assignment grades. Determining the value of gradeTotal is a relatively expensive operation, particularly with a large class, therefore I want to run it only when necessary. For simplicity's sake, I'm not storing that gradeTotal value in the core data model. I don't mind caching it somewhere, but I'm having a bitch of a time determining where and how to best update that cache. I need to run that calculation for each student whenever any value changes that affects their gradeTotal. If this were a simple to-one relationship, I know I could use something like keyPathsForValuesAffectingGradeTotal ... but it's more like a many-to-one-to-many relationship. Does anyone know of an elegant (and KVC correct) solution? I guess I could tear through all those score and assignment objects and tell them to register their students as observers. But this seems like a blunt force approach.

  Read the article

• How to Load Dependent Files on Demand + Check if They're Loaded or Not?

  - by br4inwash3r

  I'm trying to implement an assets/dependency loader that i've found from an old article at 24Ways.org. most of you might be familiar with it. it's from this article by Christian Heilmann: http://24ways.org/2007/keeping-javascript-dependencies-at-bay i've modified the script to load CSS files as well. and it's now quite close to what i want. but i still need to do some checking to see wether an asset have been completely loaded or not. just wondering if you guys have any ideas :) here's what my script currently looked like: var assetLoader = { assets: { products: { js: 'products.js', css: 'products.css', loaded: false }, articles: { js: 'articles.js', css: 'articles.css', loaded: false }, [...] cycle: { js: 'jquery.cycle.min.js', loaded: false }, swfobject: { js: 'jquery.swfobject.min.js', loaded: false } }, add: function(asset) { var comp = assetLoader.assets[asset]; var path = '/path/to/assets/'; if (comp && comp.loaded == false) { if (comp.js) { // load js var js = document.createElement('script'); js.src = path + 'js/' + comp.js; js.type = 'text/javascript'; js.charset = 'utf-8'; // append to document document.getElementsByTagName('body')[0].appendChild(js); } if (comp.css) { // load css var css = document.createElement('link'); css.rel = 'stylesheet'; css.href = path + 'css/' + comp.css; css.type = 'text/css'; css.media = 'screen, projection'; css.charset = 'utf-8'; // append to document document.getElementsByTagName('head')[0].appendChild(css); } } }, check: function(asset) { assetLoader.assets[asset].loaded = true; } } Christian explains this method in his article in great detail. I don't want to confuse you guys anymore with my bad english :P and here's an example of how i run the script: ... // load jquery cycle plugin if (page=='tvc' || page=='products') { if (!assetLoader.assets.cycle.loaded) { assetLoader.add('cycle'); } } // load products page assets if (!assetLoader.assets.products.loaded) { assetLoader.add('products'); } ... this kind of approach is very problematic though. coz assets loads asynchronously, which means some of the code inside products.js that depends on jquery.cycle.js might continue running before jquery.cycle.js is even loaded resulting in errors. while i'm quite aware that scripts can be attached with an onload event, i'm just not really sure how to implement it to my script. anyone care to help me? please... :P

  Read the article

• JSF 2.0: Preserving component state across multiple views

  - by tlind

  The web application I am developing using MyFaces 2.0.3 / PrimeFaces 2.2RC2 is divided into a content and a navigation area. In the navigation area, which is included into multiple pages using templating (i.e. <ui:define>), there are some widgets (e.g. a navigation tree, collapsible panels etc.) of which I want to preserve the component state across views. For example, let's say I am on the home page. When I navigate to a product details page by clicking on a product in the navigation tree, my Java code triggers a redirect using navigationHandler.handleNavigation(context, null, "/detailspage.jsf?faces-redirect=true") Another way of getting to that details page would be by directly clicking on a product teaser that is shown on the home page. The corresponding <h:link> would lead us to the details page. In both cases, the expansion state of my navigation tree (a PrimeFaces tree component) and my collapsible panels is lost. I understand this is because the redirect / h:link results in the creation of a new view. What is the best way of dealing with this? I am already using MyFaces Orchestra in my project along with its conversation scope, but I am not sure if this is of any help here (since I'd have to bind the expansion/collapsed state of the widgets to a backing bean... but as far as I know, this is not possible). Is there a way of telling JSF which component states to propagate to the next view, assuming that the same component exists in that view? I guess I could need a pointer into the right direction here. Thanks! Update 1: I just tried binding the panels and the tree to a session-scoped bean, but this seems to have no effect. Also, I guess I would have to bind all child components (if any) manually, so this doesn't seem like the way to go. Update 2: Binding UI components to non-request scoped beans is not a good idea (see link I posted in a comment below). If there is no easier approach, I might have to proceed as follows: When a panel is collapsed or the tree is expanded, save the current state in a session-scoped backing bean (!= the UI component itself) The components' states are stored in a map. The map key is the component's (hopefully) unique, relative ID. I cannot use the whole absolute component path here, since the IDs of the parent naming containers might change if the view changes, assuming these IDs are generated programmatically. As soon as a new view gets constructed, retrieve the components' states from the map and apply them to the components. For example, in case of the panels, I can set the collapsed attribute to a value retrieved from my session-scoped backing bean.

  Read the article

• Web development - relative URLs without duplicating files

  - by eshriek

  I have a site with index.php in the root folder, images in /img , and overview.php in /content . I have a sidebar.php file that is included in both index.php and overview.php . How should I refer to /img/image.gif if I include a link in each file? The location of image.gif changes relative to the location of the file that references it. Using /img/image.gif in sidebar.php will work in index.php, but it fails for the file located at /content/overview.php. The only solution that I can see is to either include a seperate sidebar.php in each subdirectory, or include an /img directory in every sub-directory. The best suggestion that I can find is to use the <base html tag as suggested here: Change relative link paths for included content in PHP However, in the same link, SamGoody suggests that the <base tag "is no longer properly supported in Internet Explorer, since version 7." I'd like some insight on the matter before committing to a course of action. Thanks. EDIT: I am using the wrong approach below with "../" Example- root/index.php: ... <link rel="stylesheet" type="text/css" href="style.css" /> <title>title</title> </head> <body> <?php include('include/header.php'); ?> <?php include('include/menu.php'); ?> ... root/include/header.php: ... <div id="header"> <span class="fl"><img src="img/dun1.png"/></span><span class="fr"><img src="img/dun2.png"/></span> ... root/content/overview.php: ... <link rel="stylesheet" type="text/css" href="../style.css" media="screen" /> <title>Overview</title> </head> <body> <?php include('../include/header.php'); ?> <?php include('../include/menu.php'); ?> ...

  Read the article

• WndProc(ref Message m), Prevent minimize Games, Send key strokes.

  - by Stanomatic

  Overview: I am going to create a touch application that interfaces with games and other apps. This concept is similar to the app found on touch-buddy.com but I will be using C# and WPF instead of how the application is written in Perl. I have a few challenges I would like to evaluate. The touch-buddy app uses two approaches while interacting with games; 1. Client mode (Same machine runs both game and touch-buddy). 2. Server / Client mode where a separate box sends commands to the game machine. The reason I believe for this method was to circumvent the issue with games minimizing. In Client only mode I am faced with the issue where I touch a screen OTHER than the main screen where the game is viewed and then the game minimizes. Not all games have this behavior but I would like to conquer the games that do minimize and prevent it. Is it possible to keep a game front and center Focused and prevent minimizing utilizing C# WndProc(ref Message m)? I have been experimenting with WndProc(ref Message m) where I created a win form and when I press minimize on my own Win form and it will close an instance of notepad. This proves to me that I can capture a message, prevent that message from bubbling up and then send a message to another application. I then tried to click on notepad with my touch screen and keep my win form application in focus and not minimize. At this point I am unsuccessful. I need more time understanding message codes. Is this the right approach? Can it be done? Should I look at other libraries such as Windows Automation? Key input is my other concern. What is the best way to send key strokes to other apps/games. Should I tap into DirectX, use some kind of send key, Automation Framework? Can any of these handle the multiple key strokes that some simulation games require? I appreciate any links and or insight you may have. If you have gone down this path for any reason I would love to hear your comments. Stan

  Read the article

• Persistent (purely functional) Red-Black trees on disk performance

  - by Waneck

  I'm studying the best data structures to implement a simple open-source object temporal database, and currently I'm very fond of using Persistent Red-Black trees to do it. My main reasons for using persistent data structures is first of all to minimize the use of locks, so the database can be as parallel as possible. Also it will be easier to implement ACID transactions and even being able to abstract the database to work in parallel on a cluster of some kind. The great thing of this approach is that it makes possible implementing temporal databases almost for free. And this is something quite nice to have, specially for web and for data analysis (e.g. trends). All of this is very cool, but I'm a little suspicious about the overall performance of using a persistent data structure on disk. Even though there are some very fast disks available today, and all writes can be done asynchronously, so a response is always immediate, I don't want to build all application under a false premise, only to realize it isn't really a good way to do it. Here's my line of thought: - Since all writes are done asynchronously, and using a persistent data structure will enable not to invalidate the previous - and currently valid - structure, the write time isn't really a bottleneck. - There are some literature on structures like this that are exactly for disk usage. But it seems to me that these techniques will add more read overhead to achieve faster writes. But I think that exactly the opposite is preferable. Also many of these techniques really do end up with a multi-versioned trees, but they aren't strictly immutable, which is something very crucial to justify the persistent overhead. - I know there still will have to be some kind of locking when appending values to the database, and I also know there should be a good garbage collecting logic if not all versions are to be maintained (otherwise the file size will surely rise dramatically). Also a delta compression system could be thought about. - Of all search trees structures, I really think Red-Blacks are the most close to what I need, since they offer the least number of rotations. But there are some possible pitfalls along the way: - Asynchronous writes -could- affect applications that need the data in real time. But I don't think that is the case with web applications, most of the time. Also when real-time data is needed, another solutions could be devised, like a check-in/check-out system of specific data that will need to be worked on a more real-time manner. - Also they could lead to some commit conflicts, though I fail to think of a good example of when it could happen. Also commit conflicts can occur in normal RDBMS, if two threads are working with the same data, right? - The overhead of having an immutable interface like this will grow exponentially and everything is doomed to fail soon, so this all is a bad idea. Any thoughts? Thanks! edit: There seems to be a misunderstanding of what a persistent data structure is: http://en.wikipedia.org/wiki/Persistent_data_structure

  Read the article

• Unsure how to design JavaScript / jQuery functionality which uses XML to create HTML objects

  - by Jack Roscoe

  Hi, I'm using JavScript and jQuery to read an XML document and subsequently use the information from the XML to create HTML objects. The main 'C' nodes in the XML document all have a type attribute, and depending on the type I want to run a function which will create a new html object using the other attributes assigned to that particular 'C' node node. Currently, I have a for loop which extracts each 'C' node from the XML and also it's attributes (e.g. width, height, x, y). Also inside the for loop, I have an if statement which checks the 'type' attribute of the current 'C' node being processed, and depending on the type it will run a different function which will then create a new HTML object with the attributes which have been drawn from the XML. The problem is that there may be more than one 'C' node of the same type, so for example when I'm creating the function that will run when a 'C' node of 'type=1' is detected, I cannot use the 'var p = document.createElement('p')' because if a 'C' node of the same type comes up later in the loop it will clash and override that element with that variable that has just been created. I'm not really sure how to approach this? Here is my entire script. If you need me to elaborate on any parts please ask, I'm sure it's not written in the nicest possible way: var arrayIds = new Array(); $(document).ready(function(){ $.ajax({ type: "GET", url: "question.xml", dataType: "xml", success: function(xml) { $(xml).find("C").each(function(){ arrayIds.push($(this).attr('ID')); }); var svgTag = document.createElement('SVG'); // Create question type objects function ctyp3(x,y,width,height,baC) { alert('test'); var r = document.createElement('rect'); r.x = x; r.y = y; r.width = width; r.height = height; r.fillcolor = baC; svgTag.appendChild(r); } // Extract question data from XML var questions = []; for (j=0; j<arrayIds.length; j++) { $(xml).find("C[ID='" + arrayIds[j] + "']").each(function(){ // pass values questions[j] = { typ: $(this).attr('typ'), width: $(this).find("I").attr('wid'), height: $(this).find("I").attr('hei'), x: $(this).find("I").attr('x'), y: $(this).find("I").attr('x'), baC: $(this).find("I").attr('baC'), boC: $(this).find("I").attr('boC'), boW: $(this).find("I").attr('boW') } alert($(this).attr('typ')); if ($(this).attr('typ') == '3') { ctyp3(x,y,width,height,baC); // alert('pass'); } else { // Add here // alert('fail'); } }); } } }); });

  Read the article

• Is Abstract Factory Pattern implemented correctly for given scenario.... ???

  - by Amit

  First thing... I am novice to pattern world, so correct me if wrong anywhere Scenario: There are multiple companies providing multiple products of diff size so there are 3 entities i.e. Companies, Their Product and size of product I have implement Abstract Pattern on this i.e. so that I will create instance of IProductFactory interface to get desired product... Is below implementation of Abstract Factory Pattern correct ??? If not then please correct the approach + Also tell me if any other pattern can be used for such scenario Thanks in advance... public enum Companies { Samsung = 0, LG = 1, Philips = 2, Sony = 3 } public enum Product { PlasmaTv = 0, DVD = 1 } public enum ProductSize { FortyTwoInch, FiftyFiveInch } interface IProductFactory { IPhilips GetPhilipsProduct(); ISony GetSonyProduct(); } interface ISony { string CreateProducts(Product product, ProductSize size); } interface IPhilips { string CreateProducts(Product product, ProductSize size); } class ProductFactory : IProductFactory { public IPhilips GetPhilipsProduct() { return new Philips(); } public ISony GetSonyProduct() { return new Sony(); } } class Philips : IPhilips { #region IPhilips Members public string CreateProducts(Product product, ProductSize size) {// I have ingnore size for now.... string output = string.Empty; if (product == Product.PlasmaTv) { output = "Plasma TV Created !!!"; } else if (product == Product.DVD) { output = "DVD Created !!!"; } return output; } #endregion } class Sony : ISony {// I have ingnore size for now.... #region ISony Members public string CreateProducts(Product product, ProductSize size) { string output = string.Empty; if (product == Product.PlasmaTv) { output = "Plasma TV Created !!!"; } else if (product == Product.DVD) { output = "DVD Created !!!"; } return output; } #endregion } IProductFactory prodFactory = new ProductFactory(); IPhilips philipsObj = prodFactory.GetPhilipsProduct(); MessageBox.Show(philipsObj.CreateProducts(Product.DVD, ProductSize.FortyTwoInch)); or //ISony sonyObj = prodFactory.GetSonyProduct(); //MessageBox.Show(sonyObj.CreateProducts(Product.DVD, ProductSize.FortyTwoInch));

  Read the article

• Python-How to call up a function from another module?

  - by user2691540

  I am making a project where I need several modules which are imported into one main module to make a pizza ordering service. Upon finishing this (to a working standard) I decided to re-code it so that the whole order is completed with one tkinter window, now instead of using input(), I use tkinter.Entry() etc. etc. To do this I had to use functions for each step so it is broken up nicely, e.g. the code asks if the user wants pickup or delivery, the user clicks a button which sets some variables and one of the buttons is then configured to say "continue" and the command leads to the next step of the pizza ordering process e.g. getting the name. The problem I have is that when I get past the last function of the first module the configured button has a command to go to a function in the second module, but it says that the command is not defined??? I have tried my way around this but cannot import the configured button variable into the next module, and anything else I tried gave no result, it simply doesn't go to the next module after the first module is done. I have made the main tkinter window in the main module and have it that it will mainloop after importing the other modules so shouldn't the function I want to call upon be defined? How can I get from one function to the next if the latter is in a seperate module? Is this possible or do I have to rethink my approach and if so how? Ok then, have made some more code to show what my problem is, this isn't what I am actually using but it's a lot shorter and has the same issue: this is the main module: import tkinter mainwindow = tkinter.Tk() # here i set the window to a certain size etc. import mod1 import mod2 mainwindow.mainloop() this is mod1: import tkinter def button1(): label.destroy() button1.destroy() button2.config(text = "continue", command = func2) def button2(): label.destroy() button1.destroy() button2.config(text = "continue", command = func2) label = tkinter.Label(text = "example label") button1 = tkinter.Button(text = "button1", command = button1) button2 = tkinter.Button(text = "button2", command = button2) label.pack() button1.pack() button2.pack() this is mod2: def func2(): button2.destroy() print ("haha it works...") I still get the problem that func2 is not defined? Thanks in advance

  Read the article

• how to refactor user-permission system?

  - by John

  Sorry for lengthy question. I can't tell if this should be a programming question or a project management question. Any advice will help. I inherited a reasonably large web project (1 year old) from a solo freelancer who architected it then abandoned it. The project was a mess, but I cleaned up what I could, and now the system is more maintainable. I need suggestions on how to extend the user-permission system. As it is now, the database has a t_user table with the column t_user.membership_type. Currently, there are 4 membership types with the following properties: 3 of the membership types are almost functionally the same, except for the different monthly fees each must pay 1 of the membership type is a "fake-user" type which has limited access ( different business logic also applies) With regards to the fake-user type, if you look in the system's business logic files, you will see a lot of hard-coded IF statements that do something like if (fake-user) { // do something } else { // a paid member of type 1,2 or 3 // proceed normally } My client asked me to add 3 more membership types to the system, each of them with unique features to be implemented this month, and substantive "to-be-determined" features next month. My first reaction is that I need to refactor the user-permission system. But it concerns me that I don't have enough information on the "to-be-determined" membership type features for next month. Refactoring the user-permission system will take a substantive amount of time. I don't want to refactor something and throw it out the following month. I get substantive feature requests on a monthly basis that come out of the blue. There is no project road map. I've asked my client to provide me with a roadmap of what they intend to do with the new membership types, but their answer is along the lines of "We just want to do [feature here] this month. We'll think of something new next month." So questions that come to mind are: 1) Is it dangerous for me to refactor the user permission system not knowing what membership type features exist beyond a month from now? 2) Should I refactor the user permission system regardless? Or just continue adding IF statements as needed in all my controller files? Or can you recommend a different approach to user permission systems? Maybe role-based ? 3) Should this project have a road map? For a 1 year old project like mine, how far into the future should this roadmap project? 4) Any general advice on the best way to add 3 new membership types?

  Read the article

• Converting C source to C++

  - by Barry Kelly

  How would you go about converting a reasonably large (300K), fairly mature C codebase to C++? The kind of C I have in mind is split into files roughly corresponding to modules (i.e. less granular than a typical OO class-based decomposition), using internal linkage in lieu private functions and data, and external linkage for public functions and data. Global variables are used extensively for communication between the modules. There is a very extensive integration test suite available, but no unit (i.e. module) level tests. I have in mind a general strategy: Compile everything in C++'s C subset and get that working. Convert modules into huge classes, so that all the cross-references are scoped by a class name, but leaving all functions and data as static members, and get that working. Convert huge classes into instances with appropriate constructors and initialized cross-references; replace static member accesses with indirect accesses as appropriate; and get that working. Now, approach the project as an ill-factored OO application, and write unit tests where dependencies are tractable, and decompose into separate classes where they are not; the goal here would be to move from one working program to another at each transformation. Obviously, this would be quite a bit of work. Are there any case studies / war stories out there on this kind of translation? Alternative strategies? Other useful advice? Note 1: the program is a compiler, and probably millions of other programs rely on its behaviour not changing, so wholesale rewriting is pretty much not an option. Note 2: the source is nearly 20 years old, and has perhaps 30% code churn (lines modified + added / previous total lines) per year. It is heavily maintained and extended, in other words. Thus, one of the goals would be to increase mantainability. [For the sake of the question, assume that translation into C++ is mandatory, and that leaving it in C is not an option. The point of adding this condition is to weed out the "leave it in C" answers.]

  Read the article

• Hibernate Communications Link Failure in Restlet-Hibernate Based Java application powered by MySQL

  - by Vatsala

  Let me describe my question - I have a Java application - Hibernate as the DB interfacing layer over MySQL. I get the communications link failure error in my application. The occurence of this error is a very specific case. I get this error , When I leave mysql server unattended for more than approximately 6 hours (i.e. when there are no queries issued to MySQL for more than approximately 6 hours). I am pasting a top 'exception' level description below, and adding a pastebin link for a detailed stacktrace description. javax.persistence.PersistenceException: org.hibernate.exception.JDBCConnectionException: Cannot open connection - Caused by: org.hibernate.exception.JDBCConnectionException: Cannot open connection - Caused by: com.mysql.jdbc.exceptions.jdbc4.CommunicationsException: Communications link failure - The last packet successfully received from the server was 1,274,868,181,212 milliseconds ago. The last packet sent successfully to the server was 0 milliseconds ago. - Caused by: com.mysql.jdbc.exceptions.jdbc4.CommunicationsException: Communications link failure - The last packet successfully received from the server was 1,274,868,181,212 milliseconds ago. The last packet sent successfully to the server was 0 milliseconds ago. - Caused by: java.net.ConnectException: Connection refused: connect the link to the pastebin for further investigation - http://pastebin.com/4KujAmgD What I understand from these exception statements is that MySQL is refusing to take in any connections after a period of idle/nil activity. I have been reading up a bit about this via google search, and came to know that one of the possible ways to overcome this is to set values for c3p0 properties as c3p0 comes bundled with Hibernate. Specifically, I read from here http://www.mchange.com/projects/c3p0/index.html that setting two properties idleConnectionTestPeriod and preferredTestQuery will solve this for me. But these values dont seem to have had an effect. Is this the correct approach to fixing this? If not, what is the right way to get over this? The following are related Communications Link Failure questions at stackoverflow.com, but I've not found a satisfactory answer in their answers. http://stackoverflow.com/questions/2121829/java-db-communications-link-failure http://stackoverflow.com/questions/298988/how-to-handle-communication-link-failure Note 1 - i dont get this error when I am using my application continuosly. Note 2 - I use JPA with Hibernate and hence my hibernate.dialect,etc hibernate properties reside within the persistence.xml in the META-INF folder (does that prevent the c3p0 properties from working?)

  Read the article

• Sanitizing DB inputs with XSLT

  - by azathoth

  Hello I've been looking for a method to strip my XML content of apostrophes (') like: <name> Jim O'Connor</name> since my DBMS is complaining of receiving those. By looking at the example described here, that is supposed to replace ' with '', I constructed the following script: <xsl:stylesheet version="1.0" xmlns:xsl="http://www.w3.org/1999/XSL/Transform"> <xsl:output omit-xml-declaration="yes" indent="yes" /> <xsl:template match="node()|@*"> <xsl:copy> <xsl:apply-templates select="node()|@*" /> </xsl:copy> </xsl:template> <xsl:template name="sqlApostrophe"> <xsl:param name="string" /> <xsl:variable name="apostrophe">'</xsl:variable> <xsl:choose> <xsl:when test="contains($string,$apostrophe)"> <xsl:value-of select="concat(substring-before($string,$apostrophe), $apostrophe,$apostrophe)" disable-output-escaping="yes" /> <xsl:call-template name="sqlApostrophe"> <xsl:with-param name="string" select="substring-after($string,$apostrophe)" /> </xsl:call-template> </xsl:when> <xsl:otherwise> <xsl:value-of select="$string" disable-output-escaping="yes" /> </xsl:otherwise> </xsl:choose> </xsl:template> <xsl:template match="node()|@*"> <xsl:apply-templates name="sqlApostrophe"/> </xsl:template> </xsl:stylesheet> However, the processor isn't accepting it. What am I missing here? Is there a better way to get rid of the apostrophes? Perhaps another approach for sanitizing DB inputs by using XSLT? Thanks for your help

  Read the article

• Unlimited SMS API/Gateway (Sending and Receiving)

  - by Naif

  I am creating a chat application which requires that users be able to send and receive sms messages through a web interface. It would be somewhat similar to the text messaging service available in yahoo mail or in aol instant messenger. The situation is this: Given the high quantity of messages that would be sent and received, paying on a per message basis is not economically feasible. How is it that sites such as yahoo, aim, and twitter can send and receive unlimited sms messages? Essentially, I am looking for a way to send and receive unlimited sms from my computer. Below is a list of some approaches I've come up with but have run into problems with as well. If just one of the approaches can be utilized effectively, then I am fine. As a note on the nature of my application: I will only be sending messages to users that explicitly sign up for the service and permit the receiving of messages. They can unsubscribe at any time. This is to prevent spam. I am aware of software such as Kannel which allows one to connect to a providers smsc gateway. However, this adds the risk of not being approved by the provider which would be unacceptable. Is there any way to significantly mitigate this risk? Utilizing a gateway provider eliminates this risk, but adds the issue of per message pricing. I am also aware of email to sms. However, I have done some testing with that and it appears that this method results in many messages being undelivered or delivered VERY late. If it weren't for that, this approach would have been ideal. Is there any way to negotiate with carriers to remove me from their filters (considering the nature of my service as stated before)? I could use a gsm modem, but even with an "unilimited" plan on a sim card, there are still limits (around 100,000 messages or so). Furthermore, from my understanding, gsm modems are capable of only sending out around a dozen messages per minute. I need to be able to send out as much as several hundred messages per second. During the first 2 months, around 10 messages per second would suffice. There are ways to send out ads with messages to cover the per message costs. However, this is a deal-breaker since it has a high chance of tarnishing the quality of service. Furthermore, I know it is possible to do it without such ads since yahoo, aim, and twitter do not send ads with their messages.

  Read the article

• dynamic module creation

  - by intuited

  I'd like to dynamically create a module from a dictionary, and I'm wondering if adding an element to sys.modules is really the best way to do this. EG context = { a: 1, b: 2 } import types test_context_module = types.ModuleType('TestContext', 'Module created to provide a context for tests') test_context_module.__dict__.update(context) import sys sys.modules['TestContext'] = test_context_module My immediate goal in this regard is to be able to provide a context for timing test execution: import timeit timeit.Timer('a + b', 'from TestContext import *') It seems that there are other ways to do this, since the Timer constructor takes objects as well as strings. I'm still interested in learning how to do this though, since a) it has other potential applications; and b) I'm not sure exactly how to use objects with the Timer constructor; doing so may prove to be less appropriate than this approach in some circumstances. EDITS/REVELATIONS/PHOOEYS/EUREKAE: I've realized that the example code relating to running timing tests won't actually work, because import * only works at the module level, and the context in which that statement is executed is that of a function in the testit module. In other words, the globals dictionary used when executing that code is that of main, since that's where I was when I wrote the code in the interactive shell. So that rationale for figuring this out is a bit botched, but it's still a valid question. I've discovered that the code run in the first set of examples has the undesirable effect that the namespace in which the newly created module's code executes is that of the module in which it was declared, not its own module. This is like way weird, and could lead to all sorts of unexpected rattlesnakeic sketchiness. So I'm pretty sure that this is not how this sort of thing is meant to be done, if it is in fact something that the Guido doth shine upon. The similar-but-subtly-different case of dynamically loading a module from a file that is not in python's include path is quite easily accomplished using imp.load_source('NewModuleName', 'path/to/module/module_to_load.py'). This does load the module into sys.modules. However this doesn't really answer my question, because really, what if you're running python on an embedded platform with no filesystem? I'm battling a considerable case of information overload at the moment, so I could be mistaken, but there doesn't seem to be anything in the imp module that's capable of this. But the question, essentially, at this point is how to set the global (ie module) context for an object. Maybe I should ask that more specifically? And at a larger scope, how to get Python to do this while shoehorning objects into a given module?

  Read the article

• Multi-tier applications using L2S, WCF and Base Class

  - by Gena Verdel

  Hi all. One day I decided to build this nice multi-tier application using L2S and WCF. The simplified model is : DataBase-L2S-Wrapper(DTO)-Client Application. The communication between Client and Database is achieved by using Data Transfer Objects which contain entity objects as their properties. abstract public class BaseObject { public virtual IccSystem.iccObjectTypes ObjectICC_Type { get { return IccSystem.iccObjectTypes.unknownType; } } [global::System.Data.Linq.Mapping.ColumnAttribute(Storage = "_ID", AutoSync = AutoSync.OnInsert, DbType = "BigInt NOT NULL IDENTITY", IsPrimaryKey = true, IsDbGenerated = true)] [global::System.Runtime.Serialization.DataMemberAttribute(Order = 1)] public virtual long ID { //get; //set; get { return _ID; } set { _ID = value; } } } [DataContract] public class BaseObjectWrapper<T> where T : BaseObject { #region Fields private T _DBObject; #endregion #region Properties [DataMember] public T Entity { get { return _DBObject; } set { _DBObject = value; } } #endregion } Pretty simple, isn't it?. Here's the catch. Each one of the mapped classes contains ID property itself so I decided to override it like this [global::System.Data.Linq.Mapping.TableAttribute(Name="dbo.Divisions")] [global::System.Runtime.Serialization.DataContractAttribute()] public partial class Division : INotifyPropertyChanging, INotifyPropertyChanged { [global::System.Data.Linq.Mapping.ColumnAttribute(Storage="_ID", AutoSync=AutoSync.OnInsert, DbType="BigInt NOT NULL IDENTITY", IsPrimaryKey=true, IsDbGenerated=true)] [global::System.Runtime.Serialization.DataMemberAttribute(Order=1)] public override long ID { get { return this._ID; } set { if ((this._ID != value)) { this.OnIDChanging(value); this.SendPropertyChanging(); this._ID = value; this.SendPropertyChanged("ID"); this.OnIDChanged(); } } } } Wrapper for division is pretty straightforward as well: public class DivisionWrapper : BaseObjectWrapper<Division> { } It worked pretty well as long as I kept ID values at mapped class and its BaseObject class the same(that's not very good approach, I know, but still) but then this happened: private CentralDC _dc; public bool UpdateDivision(ref DivisionWrapper division) { DivisionWrapper tempWrapper = division; if (division.Entity == null) { return false; } try { Table<Division> table = _dc.Divisions; var q = table.Where(o => o.ID == tempWrapper.Entity.ID); if (q.Count() == 0) { division.Entity._errorMessage = "Unable to locate entity with id " + division.Entity.ID.ToString(); return false; } var realEntity = q.First(); realEntity = division.Entity; _dc.SubmitChanges(); return true; } catch (Exception ex) { division.Entity._errorMessage = ex.Message; return false; } } When trying to enumerate over the in-memory query the following exception occurred: Class member BaseObject.ID is unmapped. Although I'm stating the type and overriding the ID property L2S fails to work. Any suggestions?

  Read the article

• functional dependencies vs type families

  - by mhwombat

  I'm developing a framework for running experiments with artificial life, and I'm trying to use type families instead of functional dependencies. Type families seems to be the preferred approach among Haskellers, but I've run into a situation where functional dependencies seem like a better fit. Am I missing a trick? Here's the design using type families. (This code compiles OK.) {-# LANGUAGE TypeFamilies, FlexibleContexts #-} import Control.Monad.State (StateT) class Agent a where agentId :: a -> String liveALittle :: Universe u => a -> StateT u IO a -- plus other functions class Universe u where type MyAgent u :: * withAgent :: (MyAgent u -> StateT u IO (MyAgent u)) -> String -> StateT u IO () -- plus other functions data SimpleUniverse = SimpleUniverse { mainDir :: FilePath -- plus other fields } defaultWithAgent :: (MyAgent u -> StateT u IO (MyAgent u)) -> String -> StateT u IO () defaultWithAgent = undefined -- stub -- plus default implementations for other functions -- -- In order to use my framework, the user will need to create a typeclass -- that implements the Agent class... -- data Bug = Bug String deriving (Show, Eq) instance Agent Bug where agentId (Bug s) = s liveALittle bug = return bug -- stub -- -- .. and they'll also need to make SimpleUniverse an instance of Universe -- for their agent type. -- instance Universe SimpleUniverse where type MyAgent SimpleUniverse = Bug withAgent = defaultWithAgent -- boilerplate -- plus similar boilerplate for other functions Is there a way to avoid forcing my users to write those last two lines of boilerplate? Compare with the version using fundeps, below, which seems to make things simpler for my users. (The use of UndecideableInstances may be a red flag.) (This code also compiles OK.) {-# LANGUAGE MultiParamTypeClasses, FunctionalDependencies, FlexibleInstances, UndecidableInstances #-} import Control.Monad.State (StateT) class Agent a where agentId :: a -> String liveALittle :: Universe u a => a -> StateT u IO a -- plus other functions class Universe u a | u -> a where withAgent :: Agent a => (a -> StateT u IO a) -> String -> StateT u IO () -- plus other functions data SimpleUniverse = SimpleUniverse { mainDir :: FilePath -- plus other fields } instance Universe SimpleUniverse a where withAgent = undefined -- stub -- plus implementations for other functions -- -- In order to use my framework, the user will need to create a typeclass -- that implements the Agent class... -- data Bug = Bug String deriving (Show, Eq) instance Agent Bug where agentId (Bug s) = s liveALittle bug = return bug -- stub -- -- And now my users only have to write stuff like... -- u :: SimpleUniverse u = SimpleUniverse "mydir"

  Read the article

• What (tf) are the secrets behind PDF memory allocation (CGPDFDocumentRef)

  - by Kai

  For a PDF reader I want to prepare a document by taking 'screenshots' of each page and save them to disc. First approach is CGPDFDocumentRef document = CGPDFDocumentCreateWithURL((CFURLRef) someURL); for (int i = 1; i<=pageCount; i++) { NSAutoreleasePool *pool = [[NSAutoreleasePool alloc]init]; CGPDFPageRef page = CGPDFDocumentGetPage(document, i); ...//getting + manipulating graphics context etc. ... CGContextDrawPDFPage(context, page); ... UIImage *resultingImage = UIGraphicsGetImageFromCurrentImageContext(); ...//saving the image to disc [pool drain]; } CGPDFDocumentRelease(document); This results in a lot of memory which seems not to be released after the first run of the loop (preparing the 1st document), but no more unreleased memory in additional runs: MEMORY BEFORE: 6 MB MEMORY DURING 1ST DOC: 40 MB MEMORY AFTER 1ST DOC: 25 MB MEMORY DURING 2ND DOC: 40 MB MEMORY AFTER 2ND DOC: 25 MB .... Changing the code to for (int i = 1; i<=pageCount; i++) { CGPDFDocumentRef document = CGPDFDocumentCreateWithURL((CFURLRef) someURL); NSAutoreleasePool *pool = [[NSAutoreleasePool alloc]init]; CGPDFPageRef page = CGPDFDocumentGetPage(document, i); ...//getting + manipulating graphics context etc. ... CGContextDrawPDFPage(context, page); ... UIImage *resultingImage = UIGraphicsGetImageFromCurrentImageContext(); ...//saving the image to disc CGPDFDocumentRelease(document); [pool drain]; } changes the memory usage to MEMORY BEFORE: 6 MB MEMORY DURING 1ST DOC: 9 MB MEMORY AFTER 1ST DOC: 7 MB MEMORY DURING 2ND DOC: 9 MB MEMORY AFTER 2ND DOC: 7 MB .... but is obviously a step backwards in performance. When I start reading a PDF (later in time, different thread) in the first case no more memory is allocated (staying at 25 MB), while in the second case memory goes up to 20 MB (from 7). In both cases, when I remove the CGContextDrawPDFPage(context, page); line memory is (nearly) constant at 6 MB during and after all preparations of documents. Can anybody explain whats going on there?

  Read the article

< Previous Page | 341 342 343 344 345 346 347 348 349 350 351 352  | Next Page >