Search Results

Search found 9008 results on 361 pages for 'undocumented feature'.

Page 300/361 | < Previous Page | 296 297 298 299 300 301 302 303 304 305 306 307  | Next Page >

• Best way to test a Delphi application

  - by Osama ALASSIRY

  I have a Delphi application that has many dependencies, and it would be difficult to refactor it to use DUnit (it's huge), so I was thinking about using something like AutomatedQA's TestComplete to do the testing from the front-end UI. My main problem is that a bugfix or new feature sometimes breaks old code that was previously tested (manually), and used to work. I have setup the application to use command-line switches to open-up a specific form that could be tested, and I can create a set of values and clicks needed to be done. But I have a few questions before I do anything drastic... (and before purchasing anything) Is it worth it? Would this be a good way to test? The result of the test should in my database (Oracle), is there an easy way in testcomplete to check these values (multiple fields in multiple tables)? I would need to setup a test database to do all the automated testing, would there be an easy way to automate re-setting the test db? Other than drop user cascade, create user,..., impdp. Is there a way in testcomplete to specify command-line parameters for an exe? Does anybody have any similar experiences.

  Read the article

• Save Dialog on top of another custom dialog is behaving strangely!

  - by Raghu

  I have a save as image feature for charts in my application. The chart control is a custom user control with custom logic in them. It also has some scaling based on size, zoom etc. However, while saving them as an image I would like to give the user the option to set the size of the image (eg: 800x600 px @ 300 DPI). To do this I have created a Form with textboxes/checkboxes etc for various settings for image. One of these TextBoxes is for the file name. The file name textbox is readonly and is accompanied with a browse button which shows a SaveFileDialog when clicked. The user clicks "Save As Image" in the main form's menu. I show the ImageExportDialog using the code below: using(ImageExportDialog dlg = new ImageExportDialog()) { if(dlg.ShowDialog() == DialogResult.OK) { //get the settings selected by the user and generate the image } } In the ImageExportDialog, the user clicks on the browse button and the SaveFileDialog is shown as follows: using(SaveFileDialog dlg = new SaveFileDialog()) { if(dlg.ShowDialog() == DialogResult.OK) { txtFileName.Text = dlg.FileName; } } Now the problem is, when the user clicks on "Save" button in the SaveFileDialog, as expected the txtFileName.Text is set, but the parent custom dialog also seems to return from the ShowDialog method and the DialogResult is the same as the one for SaveFileDialog! The control then goes on to the "get the settings selected by the user and generate the image" part of the code above. Not really sure what I am doing wrong here!

  Read the article

• Reducing moire when downsampling halftone comic images.

  - by drawnonward

  How can I reduce moire effects when downsampling halftone comic book images during live zoom on an iPhone or iPad? I am writing a comic book viewer. It would be nice to provide higher resolution images and allow the user to zoom in while reading the comic book. However, my client is averse to moire effects and will not allow this feature if there are noticeable moire artifacts while zooming, which of course there are. Modifying the images to be less susceptible to moire would only work if the modifications were not perceptible. Blur was specifically prohibited, as is anything that removes the beloved halftone dots. The images are black and white halftone and line art. The originals are 600 dpi but what we ship with the application will be half that at best, so probably 2500 pixels or less tall. So what are my options? If I write a custom downsampling algorithm would it be fast enough for real time on these devices? Are there other tricks I can do? Would it work to just avoid the size ratios that have the most visual moire effects? As you zoom in an out, there are definitely peaks where the moire effects are worst. Is there a way to calculate what those points are and just zoom to a nearby scale that is not as bad? Any suggestions are welcome. I have very little experience with image and signal processing, but am enjoying the opportunity to learn. I know nothing of wavelets and acutance and other jargon, so please be verbose.

  Read the article

• what is the point of heterogenous arrays?

  - by aharon

  I know that more-dynamic-than-Java languages, like Python and Ruby, often allow you to place objects of mixed types in arrays, like so: ["hello", 120, ["world"]] What I don't understand is why you would ever use a feature like this. If I want to store heterogenous data in Java, I'll usually create an object for it. For example, say a User has int ID and String name. While I see that in Python/Ruby/PHP you could do something like this: [["John Smith", 000], ["Smith John", 001], ...] this seems a bit less safe/OO than creating a class User with attributes ID and name and then having your array: [<User: name="John Smith", id=000>, <User: name="Smith John", id=001>, ...] where those <User ...> things represent User objects. Is there reason to use the former over the latter in languages that support it? Or is there some bigger reason to use heterogenous arrays? N.B. I am not talking about arrays that include different objects that all implement the same interface or inherit from the same parent, e.g.: class Square extends Shape class Triangle extends Shape [new Square(), new Triangle()] because that is, to the programmer at least, still a homogenous array as you'll be doing the same thing with each shape (e.g., calling the draw() method), only the methods commonly defined between the two.

  Read the article

• Java SSH2 libraries in depth: Trilead/Ganymed/Orion [/other?]

  - by Bernd Haug

  I have been searching for a pure Java SSH library to use for a project. The single most important needed feature is that it has to be able to work with command-line git, but remote-controlling command-line tools is also important. A pretty common choice, e.g. used in the IntelliJ IDEA git integration (which works very well), seems to be Trilead SSH2. Looking at their website, it's not being maintained any more. Trilead seems to have been a fork of Ganymed SSH2, which was a ETH Zurich project that didn't see releases for a while, but had a recent release by its new owner, Christian Plattner. There is another actively maintained fork from that code base, Orion SSH, that saw an even more recent release, but which seems to get mentioned online much less than the other 2 forks. Has anybody here worked with any of (or, if possible, both) of Ganymed and Orion and could kindly describe the development experience with either/both? Accuracy of documentation [existence of documentation?], stability, buggyness... - all of these would be highly interesting to me. Performance is not so important for my current project. If there is another pure-Java SSH implementation that should be used instead, please feel free to mention it, but please don't just mention a name...describe your judgment from actual experience. Sorry if this question may seem a bit "do my homework"-y, but I've really searched for reviews. Everything out there seems to be either a listing of implementations or short "use this! it's great!" snippets.

  Read the article

• Detecting branch reintegration or merge in pre-commit script

  - by Shawn Chin

  Within a pre-commit script, is it possible (and if so, how) to identify commits stemming from an svn merge? svnlook changed ... shows files that have changed, but does not differentiate between merges and manual edits. Ideally, I would also like to differentiate between a standard merge and a merge --reintegrate. Background: I'm exploring the possibility of using pre-commit hooks to enforce SVN usage policies for our project. One of the policies state that some directories (such as /trunk) should not be modified directly, and changed only through the reintegration of feature branches. The pre-commit script would therefore reject all changes made to these directories apart from branch reintegrations. Any ideas? Update: I've explored the svnlook command, and the closest I've got is to detect and parse changes to the svn:mergeinfo property of the directory. This approach has some drawback: svnlook can flag up a change in properties, but not which property was changed. (a diff with the proplist of the previous revision is required) By inspecting changes in svn:mergeinfo, it is possible to detect that svn merge was run. However, there is no way to determine if the commits are purely a result of the merge. Changes manually made after the merge will go undetected. (related post: Diff transaction tree against another path/revision)

  Read the article

• Undefined method `add' on a cucumber step that usually works.

  - by Josiah Kiehl

  I have a path defined: when /the admin home\s?page/ "/admin/" I have scenario that is passing: Scenario: Let admins see the admin homepage Given "pojo" is logged in And "pojo" is an "admin" And I am on the admin home page Then I should see "Hi there." And I have a scenario that is failing: Scenario: Review flagged photo Given "pojo" is logged in And "pojo" is an "admin" ...bunch of steps that create stuff in the database... And I am on the admin home page Then ... the rest of the steps The step that fails in the second one is "And I am on the admin home page" which passes just fine in the first scenario. Here's the error I get: And I am on the admin home page # features/step_definitions/web_steps.rb:18 undefined method `add' for {}:Hash (NoMethodError) ./app/controllers/admin_controller.rb:13:in `index' ./app/controllers/admin_controller.rb:11:in `each' ./app/controllers/admin_controller.rb:11:in `index' /usr/lib/ruby/1.8/benchmark.rb:308:in `realtime' ./features/step_definitions/web_steps.rb:19:in `/^(?:|I )am on (.+)$/' features/admin.feature:52:in `And I am on the admin home page' This is very odd... why would it be fine in the first case, and not in the second where the only difference are a bunch of steps that create records in the db? [edit] Here's the add stuff to database step: Given /^there is a "([^\"]*)" with the following:$/ do |model, table| model.constantize.create!(table.rows_hash) end

  Read the article

• Can you recommend a full-text search engine?

  - by Jen

  Can you recommend a full-text search engine? (Preferably open source) I have a database of many (though relatively short) HTML documents. I want users to be able to search this database by entering one or more search words in my C++ desktop application. Hence, I’m looking for a fast full-text search solution to integrate with my app. Ideally, it should: Skip common words, such as the, of, and, etc. Support stemming, i.e. search for run also finds documents containing runner, running and ran. Be able to update its index in the background as new documents are added to the database. Be able to provide search word suggestions (like Google Suggest) Have a well-documented API To illustrate, assume the database has just two documents: Document 1: This is a test of text search. Document 2: Testing is fun. The following words should be in the index: fun, search, test, testing, text. If the user types t in the search box, I want the application to be able to suggest test, testing and text (Ideally, the application should be able to query the search engine for the 10 most common search words starting with t). A search for testing should return both documents. Other points: I don't need multi-user support I don't need support for complex queries The database resides on the user's computer, so the indexing should be performed locally. Can you suggest a C or C++ based solution? (I’ve briefly reviewed CLucene and Xapian, but I’m not sure if either will address my needs, especially querying the search word indexes for the suggest feature).

  Read the article

• Enforce strong type checking in C (type strictness for typedefs)

  - by quinmars

  Is there a way to enforce explicit cast for typedefs of the same type? I've to deal with utf8 and sometimes I get confused with the indices for the character count and the byte count. So it be nice to have some typedefs: typedef unsigned int char_idx_t; typedef unsigned int byte_idx_t; With the addition that you need an explicit cast between them: char_idx_t a = 0; byte_idx_t b; b = a; // compile warning b = (byte_idx_t) a; // ok I know that such a feature doesn't exist in C, but maybe you know a trick or a compiler extension (preferable gcc) that does that. EDIT: I still don't really like the Hungarian notation in general, I couldn't used it for this problem because of project coding conventions, but I used it now in another similar case, where also the types are the same and the meanings are very similar. And I have to admit: it helps. I never would go and declare every integer with a starting "i", but as in Joel's example for overlapping types, it can be life saving.

  Read the article

• The best way to implement drawing features like Keynote

  - by Shamseddine

  Hi all, I'm trying to make a little iPad tool's for drawing simple geometrical objects (rect, rounded rect, ellipse, star, ...). My goal is to make something very close to Keynote (drawing feature), i.e. let the user add a rect (for instance), resizing it and moving it. I want too the user can select many objects and move them together. I've thought about at least 3 differents ways to do that : Extends UIView for each object type, a class for Rect, another for Ellipse, ... With custom drawing method. Then add this view as subview of the global view. Extends CALayer for each object type, a class for Rect, another for Ellipse, ... With custom drawing method. Then add this layer as sublayer of the global view layer's. Extends NSObject for each object type, a class for Rect, another for Ellipse, ... With just a drawing method which will get as argument a CGContext and a Rect and draw directly the form in it. Those methods will be called by the drawing method of the global view. I'm aware that the two first ways come with functions to detect touch on each object, to add easily shadows,... but I'm afraid that they are a little too heavy ? That's why I thought about the last way, which it seems to be straight forward. Which way will be the more efficient ??? Or maybe I didn't thought another way ? Any help will be appreciated ;-) Thanks.

  Read the article

• Java many to many association map

  - by Raphael Jolivet

  Hi, I have to classes, ClassA and ClassB and a "many to many" AssociationClass. I want to use a structure to hold the associations between A and B such as I can know, for each instance of A or B, which are their counterparts. I thought of using a Hashmap, with pair keys: Hasmap<Pair<ClassA, ClassB>, AssociationClass> associations; This way, I can add and remove an association between two instances of ClassA and ClassB, and I can query a relation for two given instances. However, I miss the feature of getting all associations defined for a given instance of ClassA or ClassB. I could do it by brute force and loop over all keys of the map to search for associations between a given instance, but this is inefficient and not elegant. Do you know of any data structure / free library that enables this ? I don't want to reinvent the wheel. Thanks in advance for your help, Raphael NB: This is not a "database" question. These objects are pure POJO used for live computation, I don't need persistence stuff.

  Read the article

• Is MVVM killing silverlight development?

  - by DeanMc

  This is a question I have had rattling around in my head for some time. I had a chat with a guy the other night who told me he would not be using the navigational framework because he could not figure out how it works with MVVM. As much as I tried to explain that patterns should be taken with a pinch of salt he would not listen. My point is this, patterns are great when they solve some problem. Sometimes only part of the pattern solves a particular problem while the other parts of it cause different problems. The goal of any developer is to build a solid application using a combination of patterns know how and foresight. I feel MVVM is becoming the one pattern to rule them all. As it is not directly supported by .Net some fancy business is needed to make it work. I feel that people are missing the point of the pattern, which is loosely coupled, testable code and instead jumping through hoops and missing out on great experiences trying to follow MVVM to the letter. MVVM is great but I wish it came with a warning or disclaimer for newbies as my fear is people will shy away from silverlight development for fear of being smacked with the mvvm stick. EDIT: Can I just add as an edit, I use and agree with MVVM as a pattern I know when it is and isn't feasible in my projects. My issue is with the encompassing nature it is taking, as if it HAS to be used as part of development. It is being used as an integral feature and not a pattern, which it is.

  Read the article

• How to convert source code to a xml based representation of the ast?

  - by autobiographer

  i wanna get a xml representation of the ast of java and c code. 3 months ago, i asked this question yet but the solutions weren't comfortable for me srcml seems to be a good solution for this problem but it does not support line numbers and columns but i need that feature. about elsa: cite: "There is ongoing effort to export the Elsa AST as an XML document; we expect to be able to advertise this in the next public release." dms... didn't understand that. especially for java, there is javaml which supports line numbers. but the sourceforge page doesn't list any files. question: there's software available which supports conversion of ast into xml which supports line numbers (and columns) [especially for java and c/c++]? is there an alternative to javaml and srcml? ps: i don't wanne have parser generators. i hope to find a tool which can be used on the console typing: ./my-xml-generator Test.java [or something like that]... or a java implementation would be great too.

  Read the article

• Microsoft products such as Visual Studio 2010 does not require to enter serial number

  - by MainMa

  Hi, I am member of WebsiteSpark and was member of DreamSpark. Both programs enable to download software and provide serial keys to use. Some software like Windows Server has an ISO file to download and a serial number displayed on the website which I must enter during installation. Some other software does not have any serial key. For example, when I downloaded Visual Studio 2010, there was just a link to an ISO file. During installation, there was no such a field as serial number (whereas Visual Studio 2008 had this field at the beginning of installation process). There is the same thing with SQL Server 2008 and Microsoft Expression Studio 3. Even when I've downloaded the public trial RTM version of Windows Seven Enterprise, there were no serial number to enter. I don't think that such expensive products as SQL Server 2008 Enterprise are delivered without serials and online validation, so I suppose that the serial is embedded into the product itself, either in installation binaries or in a separate config file, so is already in the ISO I download so I do not have to enter it. So my question is, how it is done technically? Is each 2 GBs ISO generated on-demand on the server to embed a serial each time this ISO is requested? I suppose that if it is done, it has a huge impact on servers performance (no caching, no streaming...), so what may be the techniques used behind? I want to implement the same feature in a product I intend to ship (to simplify installation by avoiding to ask to enter serial number), but I really don't see how to do it with low impact on server performance.

  Read the article

• How to use Data aware controls "correctly"?

  - by lyborko

  Hi, I would like to ask experienced users, if you prefer to use data aware controls to add, insert, delete and edit data in DB or you favor to do it manualy. I developed some DB applications, in which for the sake of "user friendly policy" I run into complicated web of table events (afterinsert, afteredit, after... and beforeedit, beforeinsert, before...). After that it was a quite nasty work to debug the application. Aware of this risk (later by another application) I tried to avoid this problem, so I paid increased attention to write code well, readable and comprehensive. It seemed everything all right from the beginning, but as I needed to handle some preprocessing stuff before sending and loading data etc, I run into the same problems again, "slowly and inevitably". Sometime I could not use dataaware controls anyway, and what seemed to be a "cool" feature of DAControl at the beginning it turned to an obstacle on the end. I "had to" write special routine for non-dataaware controls, in order to behave as dataaware. Then I asked myself, why on earth should I use dataaware controls? Is it better to found application architecture on non-dataaware controls? It requires more time to write bug-proof code, of course, but does it worth of it? I do not know... I happened to me several times, like jinxed : paradise on the beginning hell on the end... I do not know, if I use wrong method to write DB program, if there is some standard common practice how to proceed. Or if it is common problem to everybody? Thanx for advices and your experiences

  Read the article

• Multiple Concurrent Postbacks when using UpdatePanels

  - by d4nt

  Here's an example app that I built to demonstrate my problem. A single aspx page with the following on it: <form id="form1" runat="server"> <asp:ScriptManager runat="server" /> <asp:Button runat="server" ID="btnGo" Text="Go" OnClick="btnGo_Click" /> <asp:UpdatePanel runat="server"> <ContentTemplate> <asp:TextBox runat="server" ID="txtVal1" /> </ContentTemplate> </asp:UpdatePanel> </form> Then, in code behind, we have the following: protected void btnGo_Click(object sender, EventArgs e) { Thread.Sleep(5000); Debug.WriteLine(string.Format("{0}: {1}", DateTime.Now.ToString("HH:MM:ss.fffffff"), txtVal1.Text)); txtVal1.Text = ""; } If you run this and click on the "Go" button multiple times you will see multiple debug statements on the "Output" window showing that multiple requests have been processed. This appears to contradict the documented behaviour of update panels (i.e. If you make a request while one is processing, the first requests gets terminated and the current one is processed). Anyway, the point is I want to fix it. The obvious option would be to use Javascript to disable the button after the first press, but that strikes me as hard to maintain, we potentially have the same issue on a lot of screens it could be easily broken if someone renames a button. Do you have any suggestions? Perhaps there is something I could do in BeginRequest in Global.asax to detect a duplicate request? Is there some setting or feature on the UpdatePanel to stop it doing this, or maybe something in the AjaxControlToolkit that will prevent it?

  Read the article

• SVN checkout or export for production environment?

  - by Eran Galperin

  In a project I am working on, we have an ongoing discussion amongst the dev team - should the production environment be deployed as a checkout from the SVN repository or as an export? The development environment is obviously a checkout, since it is constantly updated. For the production, I'm personally for checking out the main trunk, since it makes future updates easier (just run svn update). However some of the devs are against it, as svn creates files with the group/owner and permissions of the svn process (this is on a linux OS, so those things matter), and also having the .svn directories on the production seem to them to be somewhat dirty. Also, if it is a checkout - how do you push individual features to the production without including in-development code? do you use tags or branch out for each feature? any alternatives? EDIT: I might not have been clear - one of the requirement is to be able to constantly be able to push fixes to the production environment. We want to avoid a complete build (which takes much longer than a simple update) just for pushing critical fixes.

  Read the article

• Why does my .NET 4 application know .NET 4 is not installed

  - by Tergiver

  I developed an application that targeted .NET 4 the other day and XCOPY-installed it to a Windows XP machine. I had told the owner of the machine that they would need to install .NET Framework 4 to run my app and he told me he did (not a reliable source). When I ran the application I was presented with a message box that said this app requires .NET Framework 4, would I like to install it? Clicking the Yes button took me to the Microsoft web site and a few clicks later .NET 4 was installed, and the application successfully launched. Now I normally don't develop applications that target the latest version of .NET, I always target the lowest version I can (what features do I really need?). So this was my first .NET 4 app (and I only targeted 4 because it used a library that did). In the past, XCOPY-installing .NET applications to a machine that didn't have the correct version of .NET installed resulted in the application simply crashing on startup with no useful information presented to the user. Was it built into my app because I targeted .NET X? Was it something already installed on the target machine? I love the feature, I just want to know precisely how to leverage it in the future.

  Read the article

• Internet explorer 8 opens file in browser instead of the client

  - by Rogier

  Our company is working with a great Business Intelligence tool CorVu 4.2 to analyse the operational and strategic data. Since several years we are successfully working with Sharepoint 2007 to collaborate and share information with colleagues. Most of my colleagues are working with Internet Explorer 7, but step by step Internet Explorer 8 is implemented in the company. We share a lot of CorVu files thought Sharepoint, but since we are using Internet Explorer 8, we have a problem that is new for us. If we click on a CorVu file in Internet Explorer 8 (not necessarily in Sharepoint) a pop-up shows how to open the file, if we save the file, there is no problem. But if we open the file, the file is shown in the browser and not in the CorVu client! See the screenshot below: link (I removed some unnecessary information) So far my colleagues accept this 'feature' in Internet Explorer 8. But I we open and closes more CorVu files, multiple errors (more than 10) show up starting with: (unable to place more hyperlinks) By pressing Enter the errors disappear, but it's not professional! I contacted the creators of CorVu, but they don't have a solution for in their client. There may be a solution in Internet Explorer 8? The extensions of a CorVu file can be a .sqy, .tab or .qrp. But is it possible to force the files to open in the standard client instead of the browser?

  Read the article

• Switching textstorage of NSTextViews back and forth

  - by Jakob Dam Jensen

  I'm trying to make a feature in a product which gives the user the ability to split a textview into two. The way this is done is by removing the textview from it's superview, making a NSSplitView and adding the textview as well as a new NSTextView instance to this splitview. Lastly I make these two textviews share the same textstorage in order to make them share the same content. It works great. But the problem is when I want to make one of the two textviews change textstorage. The replaceTextStorage method in NSLayoutManager causes both NSTextView to change textStorage. The API documentation states: replaceTextStorage: All NSLayoutManager objects sharing the original NSTextStorage object then share the new one. This method makes all the adjustments necessary to keep these relationships intact, unlike setTextStorage:. So it makes sense that it would do this. But the question is how do I make it possible to have two (or more) textviews first share the same storage and after that having them using their own? I've tried replacing the layoutManager and even making new instances of NSTextViews but no luck... Any suggestions?

  Read the article

• Are there solutions for streamlining the update of legacy code in multiple places?

  - by ccomet

  I'm working in some old code which was originally designed for handling two different kinds of files. I was recently tasked with adding a new kind of file to this code. Most of my problems were solved by filling out an extensive XML file with a new entry that handled everything from what lists were named to how the file is written in plural lower case. But this ended up being insufficient, as there were maybe 50 different places in 24 different code files where I had to update hardcoded switch-statements that only branched for the original two file types. Unfortunately there is no consistency in this; there are methods which operate half from the XML file, and half off of hardcode. Some of the files which look like they would operate off of the XML file don't, and some that I would expect that I'd need to update the hardcode don't need it. So the only way to find the majority of these is to run through testing the whole system when only part of it is operational, finding that one step to fix (when I'm lucky that error logging actually tells me what is going on), and then running the whole thing again. This wastes time testing the parts of the code which are already confirmed to work, time better spent testing the new parts I have to add on top of it all. It's a hassle and a half, and to my luck I can expect that I will have to add yet another new kind of file in the near future. Are there any solutions out there which can aid in this kind of endeavour? Something which I can input some parameters of current features, document what points in a whole code project actually need to be updated, and run something nice the next time I need to add a new feature to the code. It needn't even be fully automated, something that'll help me navigate straight to the specific points in everything and maybe even record what kind of parameters need to be loaded. Doubt it matters specifically, but the code is comprised of ASP.NET pages, some ASP.NET controls, hundreds of C# code files, and a handful of additional XML files. It's all currently in a couple big Visual Studio 2008 projects.

  Read the article

• Strategy for animating a lot of "LED's" - thread?, UIView animations? NSOperation? (iPhone)

  - by RickiG

  Hi I have to do some different views containing 72 LED lights. I built an LED Class so I can loop through the LED's and set them to different colors (Green, Red, Orange, Blue None etc.). The LED then loads the appropriate .png. This works fine, I loop over the LED's and set them. Now I know that at some time they will need to not just turn on/off change color, but will have to turn on with a small delay. Like an equalizer. I have a 5-10 views containing the 72 LED's and I would like to achieve the above with the minimum amount of memory/CPU strain. for(LED *l in self.ledArray) { [l display:Green]; } I simply loop as shown above and inside the LED is a switch case that does the correct logic. If this were actual LED's and a microController I would use sleep(100) or similar in the loop, but I would really like to avoid stuff like that for obvious reasons. I was thinking that doing a performOnThread withDelay would really be consuming, so would UIView animation changing the alpha and NSOperation would also be a lot of lifting for a small feature. Is there a both efficient and clever way to go around this? Thanks for any inspiration given:)

  Read the article

• PostgreSQL, Foreign Keys, Insert speed & Django

  - by Miles

  A few days ago, I ran into an unexpected performance problem with a pretty standard Django setup. For an upcoming feature, we have to regenerate a table hourly, containing about 100k rows of data, 9M on the disk, 10M indexes according to pgAdmin. The problem is that inserting them by whatever method literally takes ages, up to 3 minutes of 100% disk busy time. That's not something you want on a production site. It doesn't matter if the inserts were in a transaction, issued via plain insert, multi-row insert, COPY FROM or even INSERT INTO t1 SELECT * FROM t2. After noticing this isn't Django's fault, I followed a trial and error route, and hey, the problem disappeared after dropping all foreign keys! Instead of 3 minutes, the INSERT INTO SELECT FROM took less than a second to execute, which isn't too surprising for a table <= 20M on the disk. What is weird is that PostgreSQL manages to slow down inserts by 180x just by using 3 foreign keys. Oh, disk activity was pure writing, as everything is cached in RAM; only writes go to the disks. It looks like PostgreSQL is working very hard to touch every row in the referred tables, as 3MB/sec * 180s is way more data than the 20MB this new table takes on disk. No WAL for the 180s case, I was testing in psql directly, in Django, add ~50% overhead for WAL logging. Tried @commit_on_success, same slowness, I had even implemented multi row insert and COPY FROM with psycopg2. That's another weird thing, how can 10M worth of inserts generate 10x 16M log segments? Table layout: id serial primary, a bunch of int32, 3 foreign keys to small table, 198 rows, 16k on disk large table, 1.2M rows, 59 data + 89 index MB on disk large table, 2.2M rows, 198 + 210MB So, am I doomed to either drop the foreign keys manually or use the table in a very un-Django way by defining saving bla_id x3 and skip using models.ForeignKey? I'd love to hear about some magical antidote / pg setting to fix this.

  Read the article

• Conventions for the behavior of double or triple "click to select text" features?

  - by John Sullivan

  Almost any mature program that involves text implements "double click to select the word" and, in some cases, "triple click to select additional stuff like an entire line" as a feature. I find these features useful but they are often inconsistent between programs. Example - some programs' double clicks do not select the ending space after a word, but most do. Some recognize the - character as the end of a word, others do not. SO likes to select the entire paragraph as I write this post when I triple click it, VS web developer 2005 has no triple click support, and ultra-edit 32 will select one line upon triple clicking. We could come up with innumerable inconsistencies about how double and triple click pattern matching is implemented across programs. I am concerned about how to implement this behavior in my program if nobody else has achieved a convention about how the pattern matching should work. My question is, does a convention (conventions? maybe an MS or Linux convention?) exist that dictates how these features are supposed to behave to the end user? What, if any, are they?

  Read the article

• Sell me Distributed revision control

  - by ring bearer

  I know 1000s of similar topics floating around. I read at lest 5 threads here in SO But why am I still not convinced about DVCS? I have only following questions (note that I am selfishly worried only about Java projects) What is the advantage or value of committing locally? What? really? All modern IDEs allows you to keep track of your changes? and if required you can restore a particular change. Also, they have a feature to label your changes/versions at IDE level!? what if I crash my hard drive? where did my local repository go? (so how is it cool compared to checking in to a central repo?) Working offline or in an air plane. What is the big deal?In order for me to build a release with my changes, I must eventually connect to the central repository. Till then it does not matter how I track my changes locally. Ok Linus Torvalds gives his life to Git and hates everything else. Is that enough to blindly sing praises? Linus lives in a different world compared to offshore developers in my mid-sized project? Pitch me!

  Read the article

< Previous Page | 296 297 298 299 300 301 302 303 304 305 306 307  | Next Page >