Search Results

Search found 1770 results on 71 pages for 'stupid idiot'.

Page 56/71 | < Previous Page | 52 53 54 55 56 57 58 59 60 61 62 63  | Next Page >

  • Mysterious extra hashtable entry

    - by Harm De Weirdt
    Good evening everyone, I'm back :) Let me explain my problem. I have a hashtable in wich I store the products a costumors buys (%orders). It uses the productcode as key and has a reference to an array with the other info as value. At the end of the program, I have to rewrite the inventory to the updated version (i.e. subtract the quantity of the bought items) This is how I do this: sub rewriteInventory{ open(FILE,'>inv.txt'); foreach $key(%inventory){ print FILE "$key\|$inventory{$key}[0]\|$inventory{$key}[1]\|$inventory{$key}[2]\n" } close(FILE); } where $inventory{$key}[x] is 0 - Title, 1 - price, 2 - quantity. The problem here is that when I look at inv.txt afterwards, I see things like this: CD-911|Lady Gaga - The Fame|15.99|21 ARRAY(0x145030c)||| BOOK-1453|The Da Vinci Code - Dan Brown|14.75|12 ARRAY(0x145bee4)||| Where do these "ARRAY(0x145030c)|||" entries come from? Or more important, how do I get rid of them? This is the last part of this school task, I had so much problems programming all this and this stupid little thing comes up now and I'm really fed up with this whole Perl thing. (this aside :p) I hope someone can help me :) Fuji

    Read the article

  • Assigning specific menu administration rights for roles in drupal

    - by Martin Andersson
    Hello folks. I'm trying to give one of my roles the administrative rights to add/remove content in a specific menu (but not all menus). I think I found a module that should enable something like this, http://drupalmodules.com/module/delegate-menu-administration I've followed the instructions, added the role to my user, checked the "administer some menus" value for that role and checked the "Make admin" field for that role and specific menu in Menus. I also gave the role permissions to change page and story content. However, it still wont let the user add any new content it creates in any menu, and I get an error message saying "warning: Invalid argument supplied for foreach() in /home/martin/www/drupal/modules/delegate_menu_admin/delegate_menu_admin.module on line 346." Line 346 looks like this: foreach ($form['menu']['parent']['#options'] as $key => $value) { I did a print_r($form); in the file just before it and there's no such array that I can see: [menu] => Array ( [#access] => 1 [delete] => Array ( [#access] => ) ) When I gave the role "administer menu" permissions, nothing extra was printed at all, leading me to the assumption that the delegate_menu_admin.module file is not used at all while both the "administer menu" and the "administer some menus" (from the delegate-menu-administration module) permissions are set! Is this some incompatibility between the module because of some drupal update? Or am I just too tired and too stupid? :)

    Read the article

  • Too many false positives when using FxCop.

    - by mark
    Dear ladies and sirs. We are using FxCop and it generates too many false positives to our liking. For instance, if a private method is invoked using reflection, then this method is reported as potentially unused - understandable and we suppress this warning explicitly using the SuppressMessage attribute. However, FxCop reports the same warning for the methods invoked from that method, which we already suppressed warnings about. This is stupid and generates too much noise. There are also false reports on member variables used in these methods. Also, there are problems with generic types (I even saw something about it in MS connect). Anyway, I am wondering if anyone knows whether Microsoft is going to upgrade FxCop, because it seems to be stuck in version 1.36 for a long time. BTW, I we do not use StyleCop, because it is way too picky and we just do not have the time to examine all the zillion messages in order to suppress them all. Besides, the StyleCop report seem to augment, rather than replace FxCop. Maybe anyone can suggest a good alternative to FxCop? We are using VS2008 pro. Thanks.

    Read the article

  • [Apache Camel] Generating multiple files based on DB query in a nice way

    - by chantingwolf
    Dear all, I have a following question. I have to generate many files based on sql query. Let's say for example, I have get from database a list of orders made today and genarate file for each order and later store each file on ftp. Ideally I would like to get follewing. Not quite sure how to get it. from(MyBean).to(Ftp) The problem and main question is how to generate multiple messages by custom bean (for example). I am not sure if splitter EIP is ok in this case because in my case I have not just one message to split, but I just have to generate and send many messages. http://camel.apache.org/splitter.html I hope, someone meet this problem before. If the task will be to generate just one file - everything is quite simple - you need just fill Exchange.OutMessage (or something like this). But what about multiple files - I really can't get, how to manage this situation. P.S. Sorry if this question is stupid. I am novice in Camel (working with it just for coupe weeks). It's a great tool. Actually, that's why I want to use in in the best way. Thanks a lot.

    Read the article

  • Can I perform some processing on the POST data before ASP.NET MVC UpdateModel happens?

    - by Domenic
    I would like to strip out non-numeric elements from the POST data before using UpdateModel to update the copy in the database. Is there a way to do this? // TODO: it appears I don't even use the parameter given at all, and all the magic // happens via UpdateModel and the "controller's current value provider"? [HttpPost] public ActionResult Index([Bind(Include="X1, X2")] Team model) // TODO: stupid magic strings { if (this.ModelState.IsValid) { TeamContainer context = new TeamContainer(); Team thisTeam = context.Teams.Single(t => t.TeamId == this.CurrentTeamId); // TODO HERE: apply StripWhitespace() to the data before using UpdateModel. // The data is currently somewhere in the "current value provider"? this.UpdateModel(thisTeam); context.SaveChanges(); this.RedirectToAction(c => c.Index()); } else { this.ModelState.AddModelError("", "Please enter two valid Xs."); } // If we got this far, something failed; redisplay the form. return this.View(model); } Sorry for the terseness, up all night working on this; hopefully my question is clear enough? Also sorry since this is kind of a newbie question that I might be able to get with a few hours of documentation-trawling, but I'm time-pressured... bleh.

    Read the article

  • StringTokenizer split at "<br/>"

    - by AnAmuser
    Maybe I am stupid but I don't understand why the behaviour of StringTokenizer here: import static org.apache.commons.lang.StringEscapeUtils.escapeHtml; String object = (String) value; String escaped = escapeHtml(object); StringTokenizer tokenizer = new StringTokenizer(escaped, escapeHtml("<br/>")); If fx. value is Hej<br/>$user.get(0).name Har vundet<br/><table border='1'><tr><th>Name</th><th>Played</th><th>Brewed</th></tr>#foreach( $u in $user )<tr><td>$u.name</td> <td>$u.played</td> <td>$u.brewed</td></tr>#end</table><br/> Then the result is Hej $use . e (0).name Ha vunde a e o de ='1' h Name h h P ayed h h B ewed h #fo each( $u in $use ) d $u.name d d $u.p ayed d d $u. ewed d #end a e It makes no sense to me. How can I make it behave as I expect to.

    Read the article

  • making arrays from tab-delimited text file column

    - by absolutenewbie
    I was wondering if anyone could help a desperate newbie with perl with the following question. I've been trying all day but with my perl book at work, I can't seem to anything relevant in google...or maybe am genuinely stupid with this. I have a file that looks something like the following: Bob April Bob April Bob March Mary August Robin December Robin April The output file I'm after is: Bob April April March Mary August Robin December April So that it lists each month in the order that it appears for each person. I tried making it into a hash but of course it wouldn't let me have duplicates so I thought I would like to have arrays for each name (in this example, Bob, Mary and Robin). I'm afraid to upload the code I've been trying to tweak because I know it'll be horribly wrong. I think I need to define(?) the array. Would this be correct? Any help would be greatly appreciated and I promise I will be studying more about perl in the meantime. Thank you for your time, patience and help. #!/usr/bin/perl -w while (<>) { chomp; if (defined $old_name) { $name=$1; $month=$2; if ($name eq $old_name) { $array{$month}++; } else { print "$old_name"; foreach (@array) { push (@array, $month); print "\t@array"; } print "\n"; @array=(); $array{$month}++; } } else { $name=$1; $month=$2; $array{month}++; } $old_name=$name; } print "$old_name"; foreach (@array) { push (@array, $month); print "\t@array"; } print "\n";

    Read the article

  • Trouble using latex in Matplotlib / Scipy etc.

    - by ajhall
    I'm having some issues with my first attempts at using matplotlib and scipy to make some scatter plots of my data (too many variables, trying to see many things at once). Here's some code of mine that is working fairly well... import numpy from scipy import * import pylab from matplotlib import * import h5py FileID = h5py.File('3DiPVDplot1.mat','r') # (to view the contents of: list(FileID) ) group = FileID['/'] CurrentsArray = group['Currents'].value IvIIIarray = group['IvIII'].value PFarray = group['PF'].value growthTarray = group['growthT'].value fig = pylab.figure() ax = fig.add_subplot(111) cax = ax.scatter(IvIIIarray, growthTarray, PFarray, CurrentsArray, alpha=0.75) cbar = fig.colorbar(cax) ax.set_xlabel('Cu / III') ax.set_ylabel('Growth T') ax.grid(True) pylab.show() I tried to change the code to include latex fonts and interpreting, none of it seems to work for me, however. Here's an example attempt that didn't work: import numpy from scipy import * import pylab from matplotlib import * import h5py rc('text', usetex=True) rc('font', family='serif') FileID = h5py.File('3DiPVDplot1.mat','r') # (to view the contents of: list(FileID) ) group = FileID['/'] CurrentsArray = group['Currents'].value IvIIIarray = group['IvIII'].value PFarray = group['PF'].value growthTarray = group['growthT'].value fig = pylab.figure() ax = fig.add_subplot(111) cax = ax.scatter(IvIIIarray, growthTarray, PFarray, CurrentsArray, alpha=0.75) cbar = fig.colorbar(cax) ax.set_xlabel(r'Cu / III') ax.set_ylabel(r'Growth T') ax.grid(True) pylab.show() I'm using fink installed python26 with corresponding packages for scipy matplotlib etc. I've been using iPython and manual work instead of scripts in python. Since I'm completely new to python and scipy, I'm sure I'm making some stupid simple mistakes. Please enlighten me! I greatly appreciate the help!

    Read the article

  • Autocomplete Error Question - Ruby on Rails

    - by bgadoci
    I have built a very simple blog application using Ruby on Rails. New to both Ruby and Rails so excuse the stupid questions. I currently have two tables that relate to this question. I have a Post table and a Tag table. Basically I set it up such that Post has_many :tags and Tag belongs_to :post. I am using AJAX to process and display the tags in the show view of the post. I installed the auto_complete plugin and I am getting an error when I enter the letters in the text_field_with_auto_complete for tag creation. My suspicion is this is because the form is a remote_form_for or something I am doing wrong in the routes.rb. Here is the error and code: Error Processing PostsController#show (for 127.0.0.1 at 2010-04-13 23:25:46) [GET] Parameters: {"tag"=>{"tag_name"=>"f"}, "id"=>"auto_complete_for_tag_tag_name"} Post Load (0.1ms) SELECT * FROM "posts" WHERE ("posts"."id" = 0) ActiveRecord::RecordNotFound (Couldn't find Post with ID=auto_complete_for_tag_tag_name): app/controllers/posts_controller.rb:22:in `show' Rendered rescues/_trace (26.0ms) Rendered rescues/_request_and_response (0.2ms) Rendering rescues/layout (not_found) remote_form_for located in /views/posts/show.html.erb <% remote_form_for [@post, Tag.new] do |f| %> <p> <%= f.label :tag_name, "Tag" %><br/> <%= text_field_with_auto_complete :tag, :tag_name, {}, {:method => :get} %> </p> <p><%= f.submit "Add Comment" %></p> <% end %> tags_controller.rb (I'll spare you all the actions but added the following here) auto_complete_for :tag, :tag_name routes.rb map.resources :posts, :has_many => :comments map.resources :posts, :has_many => :tags map.resources :tags, :collection => {:auto_complete_for_tag_tag_name => :get }

    Read the article

  • How to disable vertical bounce/scroll on iPhone in a mobile web application

    - by Kasper Skov
    As the title says, i need to disable vertical bounce on iphone on my mobile web form application. Ive tried alot of different things, but most of them disables my form or horizontal scroll and bounce as well. Any ideas? Im using jquery.mobile btw :) Update: I actually managed to get the code from the first answer working somewhat: function stopScrolling( touchEvent ) { touchEvent.preventDefault(); } document.addEventListener( 'touchstart' , stopScrolling , false ); document.addEventListener( 'touchmove' , stopScrolling , false ); The reason why I couldnt get it to work in the first place, was that there actually was some margin on my body (stupid me). But. As the layout is fluid and im using jquery.mobile and have <meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1"> in the header (I think) it doesnt work properly. The page is zoomed out (view from like a desktop browser) and zooming is disabled. Without the code, the page scales perfectly right from an 50" tv to the smallest nokia on the planet. Am I doing something wrong? Im starting to think the problem is caused by the body/content somehow being over 100% of the viewport. No idea how though.

    Read the article

  • NSPredicate as a constraint solver?

    - by Felixyz
    I'm working on a project which includes some slightly more complex dynamic layout of interface elements than what I'm used to. I always feel stupid writing complex code that checks if so-and-so is close to such-and-such and in that case move it x% in some direction, etc. That's just not how programming should be done. Programming should be as declarative as possible! Precisely because what I'm going to do is fairly simple, I thought it would be a good opportunity to try something new, and I thought of using NSPredicate as a simple constraints solver. I've only used NSPredicate for very simple tasks so far, but I know that it capable of much more. Are there any ideas, experiences, examples, warnings, insights that could be useful here? I'll give a very simple example so there will be something concrete to answer. How could I use NSPredicate to solve the following constraints: viewB.xmid = (viewB.leftEdge + viewB.width) / 2 viewB.xmid = max(300, viewA.rightEdge + 20 + viewB.width/2) ("viewB should be horizontally centered on coordinate 300, unless its left edge gets within 20 pixels of viewB's right edge, in which case viewA's left edge should stay fixed at 20 pixels to the right of viewB's right edge and viewA's horizontal center get pushed to the right.") viewA.rightEdge and viewB.width can vary, and those are the 'input variables'. EDIT: Any solution would probably have to use the NSExpression method -(id)expressionValueWithObject:(id)object context:(NSMutableDictionary *)context. This answer is relevant.

    Read the article

  • C# 4.0: casting dynamic to static

    - by Kevin Won
    This is an offshoot question that's related to another I asked here. I'm splitting it off because it's really a sub-question: I'm having difficulties casting an object of type dynamic to another (known) static type. I have an ironPython script that is doing this: import clr clr.AddReference("System") from System import * def GetBclUri(): return Uri("http://google.com") note that it's simply newing up a BCL System.Uri type and returning it. So I know the static type of the returned object. now over in C# land, I'm newing up the script hosting stuff and calling this getter to return the Uri object: dynamic uri = scriptEngine.GetBclUri(); System.Uri u = uri as System.Uri; // casts the dynamic to static fine Works no problem. I now can use the strongly typed Uri object as if it was originally instantiated statically. however.... Now I want to define my own C# class that will be newed up in dynamic-land just like I did with the Uri. My simple C# class: namespace Entity { public class TestPy // stupid simple test class of my own { public string DoSomething(string something) { return something; } } } Now in Python, new up an object of this type and return it: sys.path.append(r'C:..path here...') clr.AddReferenceToFile("entity.dll") import Entity.TestPy def GetTest(): return Entity.TestPy(); // the C# class then in C# call the getter: dynamic test = scriptEngine.GetTest(); Entity.TestPy t = test as Entity.TestPy; // t==null!!! here, the cast does not work. Note that the 'test' object (dynamic) is valid--I can call the DoSomething()--it just won't cast to the known static type string s = test.DoSomething("asdf"); // dynamic object works fine so I'm perplexed. the BCL type System.Uri will cast from a dynamic type to the correct static one, but my own type won't. There's obviously something I'm not getting about this...

    Read the article

  • nhibernate - mapping with contraints

    - by Tobias Müller
    Hello everybody, I am having a Problem with my nhibernate-mapping and I can't find a solution by searching on stackoverflow/google/documentation. The database I am using has (amongst others) two tables. One is unit with the following fields: id enduring_id starts ends damage_enduring_id [...] The other one is damage, which has the following fields: id enduring_id starts ends [...] The units are assigned to a damage and one damage can have zero, one or more units working on it. Every time a unit moves to annother damage, the dataset is copied. The field "ends" of the old record and "starts" of the new record are set to the current time stamp, enduring_id stays the same. So if I want to know which units were working on a damage at a certain time, I do the following select: select * from unit join damage on damage.enduring_id = unit.damage_enduring_id where unit.starts <= 'time' and unit.ends = 'time' (This is not an actualy query from the database, I made it up to make clear what I mean. The the real database is a little more complex) Now I want to map it that way, so I can load all the damages which are valid at one time (starts <= wanted time <= ends) and that each of them has a Bag with all the attached units at that time (again starts <= wanted time <= ends). Is this possible within the mapping? Sorry if this is a stupid question, but I am pretty new to nhibernate and I have no clue how to do it. Thanks a lot for reading my post! Bye, Tobias

    Read the article

  • Most unintuitive behaviour in the .Net framework?

    - by BlueRaja
    Intended behavior is often another phrase for bug-which-we-knew-about-when-we-wrote-it, but-we-wrote-it-anyways. Because it was "intended" (or perhaps it is now too late or too difficult), many of these extremely-unintuitive bugs never get fixed. For instance, consider the following code (C#): TextInfo textInfo = Thread.CurrentThread.CurrentCulture.TextInfo; textInfo.ToTitleCase("hello world!"); //Returns "Hello World!" textInfo.ToTitleCase("hElLo WoRld!"); //Returns "Hello World!" textInfo.ToTitleCase("Hello World!"); //Returns "Hello World!" What would you expect textInfo.ToTitleCase("HELLO WORLD!"); to return? In fact, it returns "HELLO WORLD!". This was well-documented "intended behavior," but, in my eyes, is extremely unintuitive, and therefore a bug. What is some other unintuitive behavior like this in this in the .Net framework? Bonus points if you can provide a fix that does not break backwards-compatibility. Remember! Always keep these two simple rules in mind when designing an API (or anything else): Make the common case the default, and Keep It Simple, Stupid!

    Read the article

  • TCP: Address already in use exception - possible causes for client port? NO PORT EXHAUSTION

    - by TomTom
    Hello, stupid problem. I get those from a client connecting to a server. Sadly, the setup is complicated making debugging complex - and we run out of options. The environment: *Client/Server system, both running on the same machine. The client is actually a service doing some database manipulation at specific times. * The cnonection comes from C# going through OleDb to an EasySoft JDBC driver to a custom written JDBC server that then hosts logic in C++. Yeah, compelx - but the third party supplier decided to expose the extension mechanisms for their server through a JDBC interface. Not a lot can be done here ;) The Symptom: At (ir)regular intervals we get a "Address already in use: connect" told from the JDBC driver. They seem to come from one particular service we run. Now, I did read all the stuff about port exhaustion. This is why we have a little tool running now that counts ports and their states every minute. Last time this happened, we had an astonishing 370 ports in use, with the count rising to about 900 AFTER the error. We aleady patched the registry (it is a windows machine) to allow more than the 5000 client ports standard, but even then, we are far far from that limit to start with. Which is why I am asking here. Ayneone an ide what ELSE could cause this? It is a Windows 2003 Server machine, 64 bit. The only other thing I can see that may cause it (but this functionality is supposedly disabled) is Symantec Endpoint Protection that is installed on the server - and being capable of actinc as a firewall, it could possibly intercept network traffic. I dont want to open a can of worms by pointing to Symantec prematurely (if pointing to Symantec can ever be seen as such). So, anyone an idea what else may be the cause? Thanks

    Read the article

  • passive view and display logic

    - by genesys
    Hi! In MVC and MVP and similar patterns there's often the approach of the "passive view" which is as stupid as possible. This should facilitate unit testing and create a clearer separation of view and model. I know that those patterns come in very different flavours and especially the understanding of MVP seems to differ from article to article. Therefore my question is not "how do i implement this pattern correctly". I want to improve view and model separation and go for better testability of the application. Therefore i'd like to go for a passive view. But my question is, where would you put logic that is clearly only view related? like a textviewer should scroll the text when the scrollbar is moved. would you put the logic for this into the Presenter? Let's say the textviewer has some extended functionality. like setting markings on textpassages. The logic for this makes clearly sense to be put into the Presenter. However, if it is mixed with all the 'direct' logic of the view (like scrolling the text) the Presenter could become very big, which is also not really a nice design. So my question is where to put display related logic of a passive view and what functionallity to mix in the Presenter. Thanks!

    Read the article

  • Deploying ASP.NET MVC to IIS6: pages are just blank

    - by BryanGrimes
    I have an MVC app that is actually on a couple other servers but I didn't do the deploy. For this deploy I have added the wildcard to aspnet_isapi.dll which has gotten rid of the 404 error. But the pages are not pulling up, rather everything is just blank. I can't seem to find any IIS configuration differences. The Global asax.cs file does have routing defined, but as I've seen on a working server, that file isn't just hanging out in the root or anything so obvious. What could I be missing here? All of the servers are running IIS6 and I have compared the setups and they look the same to me at this point. Thanks... Bryan EDIT for the comments thus far: I've looked in the event logs with no luck, and scoured various IIS logs per David Wang: blogs.msdn.com. Below is the Global.asax.cs file... public class MvcApplication : System.Web.HttpApplication { public static void RegisterRoutes(RouteCollection routes) { routes.IgnoreRoute("{resource}.axd/{*pathInfo}"); routes.IgnoreRoute("error.axd"); // for Elmah // For deployment to IIS6 routes.Add(new Route ( "{controller}.mvc/{action}/{id}", new RouteValueDictionary(new { action = "Index", id = (string)null }), new MvcRouteHandler() )); routes.MapRoute( "WeeklyTimeSave", "Time/Save", new { controller = "Time", action = "Save" } ); routes.MapRoute( "WeeklyTimeAdd", "Time/Add", new { controller = "Time", action = "Add" } ); routes.MapRoute( "WeeklyTimeEdit", "Time/Edit/{id}", new { controller = "Time", action = "Edit", id = "" } ); routes.MapRoute( "FromSalesforce", "Home/{id}", new { controller = "Home", action = "Index", id = "" }); routes.MapRoute( "Default2", "{controller}/{id}", new { controller = "Home", action = "Index", id = "" } ); routes.MapRoute( "Default", "{controller}/{action}/{id}", new { controller = "Home", action = "Index", id = "" } ); } protected void Application_Start() { RegisterRoutes(RouteTable.Routes); } } Maybe this is as stupid as the asax file not being somewhere it needs to be, but heck if I know at this point.

    Read the article

  • Am I Writing Assembly Or NASM?

    - by cam
    I'm fed up with this. I've been trying to just get a grip on assembly for awhile, but I feel like I'm coding towards my compiler rather than a language. I've been using this tutorial, and so far it's giving me hell. I'm using NASM, which may be the problem, but I figured it was the most popular one. I'm simply trying to learn the most general form of assembly, so I decided to learn x86. I keep running into stupid errors, like not being able to increment a variable. Here's the latest one: not being able to use div. mov bx, 0; mov cx, 0; jmp start; start: inc cx; mov ax, cx; div 3; <-- invalid combination of opcode and operand cmp ah,0; jz totalvalue; mov ax, cx; div 5; <-- invalid combination of opcode and operand cmp ah, 0; jz totalvalue; cmp cx, 1000; jz end; totalvalue: add bx,cx; jmp start; jmp end; end: mov ah,4ch; mov al,00; int 21h; Should I change compilers? It seems like division should be standard. Do I need to read two tutorials (one on NASM, and one on x86?). Any specific help on this problem?

    Read the article

  • why unsigned int 0xFFFFFFFF is equal to int -1?

    - by conejoroy
    perhaps it's a very stupid question but I'm having a hard time figuring this out =) in C or C++ it is said that the maximum number a size_t (an unsigned int data type) can hold is the same as casting -1 to that data type. for example see http://stackoverflow.com/questions/1420982/invalid-value-for-sizet Why?? I'm confused.. I mean, (talking about 32 bit ints) AFAIK the most significant bit holds the sign in a signed data type (that is, bit 0x80000000 to form a negative number). then, 1 is 0x00000001.. 0x7FFFFFFFF is the greatest positive number a int data type can hold. then, AFAIK the binary representation of -1 int should be 0x80000001 (perhaps I'm wrong). why/how this binary value is converted to anything completely different (0xFFFFFFFF) when casting ints to unsigned?? or.. how is it possible to form a binary -1 out of 0xFFFFFFFF? I have no doubt that in C: ((unsigned int)-1) == 0xFFFFFFFF or ((int)0xFFFFFFFF) == -1 is equally true than 1 + 1 == 2, I'm just wondering why. thanks!

    Read the article

  • Is it immoral to write crappy code even if readability and correctness is not a requirement?

    - by mafutrct
    There are cases when crappy (i.e. unreadable and buggy) code is not much of a problem. For instance, imagine you need to generate a big text file that mostly follows a simple pattern with a few very complex exceptions. What do you do? You quickly write a simple algorithm and insert the exceptional bits in the output manually to save 4 hours. The code is unreadable, and the output is flawed, but it's still the correct way since it is way faster. But let's get this straight: I hate bad code. I've had to read and work with code that caused my stomach to hurt. I care a lot about good code. And actually, I caught myself thinking that it is immoral to write bad code even though the dirty approach is sometimes superior. I was surprised by myself and found my idea to be very irrational. Did you ever experience this? Should I just get rid of this stupid idea and use the most efficient approach to coding?

    Read the article

  • Translating CURL to FLEX HTTPRequests

    - by Joshua
    I am trying to convert from some CURL code to FLEX/ActionScript. Since I am 100% ignorant about CURL and 50% ignorant about Flex and 90% ignorant on HTTP in general... I'm having some significant difficulty. The following CURL code is from http://code.google.com/p/ga-api-http-samples/source/browse/trunk/src/v2/accountFeed.sh I have every reason to believe that it's working correctly. USER_EMAIL="[email protected]" #Insert your Google Account email here USER_PASS="secretpass" #Insert your password here googleAuth="$(curl https://www.google.com/accounts/ClientLogin -s \ -d Email=$USER_EMAIL \ -d Passwd=$USER_PASS \ -d accountType=GOOGLE \ -d source=curl-accountFeed-v2 \ -d service=analytics \ | awk /Auth=.*/)" feedUri="https://www.google.com/analytics/feeds/accounts/default\ ?prettyprint=true" curl $feedUri --silent \ --header "Authorization: GoogleLogin $googleAuth" \ --header "GData-Version: 2" The following is my abortive attempt to translate the above CURL to AS3 var request:URLRequest=new URLRequest("https://www.google.com/analytics/feeds/accounts/default"); request.method=URLRequestMethod.POST; var GoogleAuth:String="$(curl https://www.google.com/accounts/ClientLogin -s " + "-d [email protected] " + "-d Passwd=secretpass " + "-d accountType=GOOGLE " + "-d source=curl-accountFeed-v2" + "-d service=analytics " + "| awk /Auth=.*/)"; request.requestHeaders.push(new URLRequestHeader("Authorization", "GoogleLogin " + GoogleAuth)); request.requestHeaders.push(new URLRequestHeader("GData-Version", "2")); var loader:URLLoader=new URLLoader(); loader.dataFormat=URLLoaderDataFormat.BINARY; loader.addEventListener(Event.COMPLETE, GACompleteHandler); loader.addEventListener(IOErrorEvent.IO_ERROR, GAErrorHandler); loader.addEventListener(SecurityErrorEvent.SECURITY_ERROR, GAErrorHandler); loader.load(request); This probably provides you all with a good laugh, and that's okay, but if you can find any pity on me, please let me know what I'm missing. I readily admit functional ineptitude, therefore letting me know how stupid I am is optional.

    Read the article

  • "multiply frog enemy" timer and array AS3

    - by VideoDnd
    How can I use the counter value to multiply the frogs in the array? My counter goes from 0-100. I want to prove that I can increment the enemies using a counter. EXPLAINED BETTER I have 10 frogs in an array. I want to use a timer to add 10 more frogs on each iteration of the TimerEvent.TIMER firing. //currentCount var timer:Timer = new Timer(1000, 50); timer.addEventListener(TimerEvent.TIMER, countdown); timer.start(); function countdown(event:TimerEvent) { // myText.text = String(0 + timer.currentCount); } //Creates 10 enemies "I want enemies to multiply 0-100" var enemyArray:Array = new Array(); for (var i:int = 0; i < 10; i++) { var noname:FrogClass = new FrogClass(); noname.x = i*10; //this will just assign some different x and y value depending on i. noname.y = i*11; enemyArray.push(noname); //put the enemy into the array addChild(noname); //puts it on the stage } SYMBOL PROPERTIES NAME "noname" CLASS "FrogClass" WHY I need specific examples using strings and arrays, because I'm stuck in a learning curve. Stupid examples are more fun!

    Read the article

  • [Python] name 'OptionGroup' is not defined

    - by Cawas
    Ok, so I made this rookie mistake below, but in my defense I was led to it thanks to how the help about this subject is on python docs, which states how to use optparse. It is actually an error under the gigantic tutorial section. In contrast and to my offense, I may be one of the very few stupid people who can't read very well and pay close attention on what I do. But since this took me so long to discover, I wanted to "document" it here: Python 2.6.1 (r261:67515, Feb 11 2010, 00:51:29) >>> from optparse import OptionParser >>> outputGroup = OptionGroup(parser, 'Output handling') Traceback (most recent call last): File "<stdin>", line 1, in <module> NameError: name 'OptionGroup' is not defined This is strictly done with examples found on the docs, and you can't find anything about it anywhere, be it that long long docs page, google or stackoverflow. Plus, reading optparse.py shows OptionGroup is there, so that adds to the confusion. I bet it will take less than 1 minute for someone to spot my error. For that I'll only add proper tags and / or modify the title later on. :)

    Read the article

  • Is there a good way of automatically generating javascript client code from server side python

    - by tat.wright
    I basically want to be able to: Write a few functions in python (with the minimum amount of extra meta data) Turn these functions into a web service (with the minimum of effort / boiler plate) Automatically generate some javascript functions / objects for rpc (this should prevent me from doing as many stupid things as possible like mistyping method names, forgetting the names of methods, passing the wrong number of arguments) Example python: def hello_world(): return "Hello world" javascript: ... <!-- This file is automatically generated (either dynamically or statically) --> <script src="http://myurl.com/webservice/client_side_javascript"> </script> ... <script> $('#button').click(function () { hello_world(function (data){ $('#label').text(data))) } </script> A bit of research has shown me some approaches that come close to this: Automatic generation of json-rpc services from functions with a little boiler plate code in python and then using jquery and json to do the calls (still easy to make mistakes with method names - still need to be aware of urls when calling, very irritating to write these calls yourself in the firebug shell) Using a library like soaplib to generate wsdl from python (by adding copious type information). And then somehow convert this into javascript (not sure if there is even a library to do this) But are there any approaches closer to what I want?

    Read the article

  • Stop Excel from changing cell contents, ever

    - by Enable Manual-Correct
    I work with card numbers, like credit card and ID numbers. We do not do any calculations with card numbers, obviously. They are "text." I format them as text, I type them like text. I know how that works. Excel doesn't care. 16 digit card numbers get their last digit turned into a zero, changed into scientific notation, stupid stuff that I did not tell Excel to do. I need to do things like Find/Remove spaces from cells in files downloaded from our currently imperfect web-system. The system sends me files with 16 digit numbers, cells formatted as text, but due to bugs there are spaces at the end. I do Find/Remove all spaces and all my card numbers are transformed into scientific notation and the last digit turned into a 0. THEY ARE TEXT, they are formatted as text, I yelled into the screen that they are text, why does Excel refuse to acknowledge that they are text? (I would rather find a way to stop Excel's action than find a way to tell our programmers to put an apostrophe in every cell) How do I make it so that Excel just STOPS doing anything that I didn't tell it to do? Or at least stop it from doing anything to numbers it doesn't like. Maybe I can write a macro for whenever it discovers "Uhoh I should change that number to something different!" I'll make it format that cell to text a thousand times instead. Give me an error when I try calculating with a number larger than 15 digits, make my computer explode violently, that's fine. Just stop changing the numbers. Is it possible? I have many thousands of numbers that need changing in many different scenarios. I just want to stop it from trying to help. I can't understand why that would be difficult. I have 2007, but answers for other versions would be great as well. Thank you!

    Read the article

< Previous Page | 52 53 54 55 56 57 58 59 60 61 62 63  | Next Page >