Search Results

Search found 2628 results on 106 pages for 'lexical analysis'.

Page 5/106 | < Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • Code Analysis Rule Sets in Visual Studio 2010

    - by Anthony Trudeau
    Microsoft Visual Studio 2010 introduces the concept of rule sets when configuring code analysis.  This is a valuable change from Visual Studio 2008 that I didn't even realize I wanted.  Visual Studio 2008 by default selected all rules and then you had to remove rules on an item by item basis. The rule sets fall into logical groups including "Microsoft All Rules", "Microsoft Basic Correctness Rules", "Microsoft Security Rules", et al.  And within the project properties you can select one rule set, multiple rule sets, or you can define your own rule set based upon another. Selecting a single rule set is obviously the easiest option.  The default rule set when you create a new project is the "Microsoft Minimum Recommended Rules".  However, in my opinion the recommended rules are just too permissive.  For that reason you might want to change your rule set to "Microsoft All Rules" until you get around to creating your own rule set; or alternately you can select multiple rule sets which is an option from the rule set combo box.  The Visual Studio documentation has comprehensive help on what is contained within the rule sets. Creating your own rule set is easy if not obvious.  You need to start a rule set from an existing rule set.  To get started select a rule set in the combo box within the Code Analysis tab of the project properties.  I selected the "Microsoft All Rules" for my rule set, but you may find it easier to start with the "Microsoft Minimum Recommended Rules" if your rules are on the more permissive side. Once your rule set is selected click the Open button.  This will display a dialog that is similar in composition to the rules selection from Visual Studio 2008.  Browsing through the tree view you can select or deselect individual rules within their categories; and you can indicate that the rules are flagged as errors instead of the default which is a warning.  A nice touch to the form is that you get a help pane when you select an individual rule.  That helped me considerably when I first configured my rule set. Once you have finished selecting your rules click the Save tool button, specify a location and name, and click the Save button on the Save As dialog.  Once you're back on the Code Analysis tab you'll choose the Browse option within the combo box and open the file you just created.

    Read the article

  • Book Review: Expert Cube Development with Microsoft SQL Server 2008 Analysis Services

    - by Greg Low
    I spent last week on campus in Redmond with the SQL Server Analysis Services Maestro program. It was great to have a chance to focus on SSAS for a week. As part of that, I did quite a bit of reading as I had quite a bit of travelling time. Ironically, I re-read a few books. The first was Marco Russo, Alberto Ferrari and Chris Webb's book Expert Cube Development with Microsoft SQL Server 2008 Analysis Services . I've often told BI classes that I've been teaching that this is a really good book and...(read more)

    Read the article

  • How do you get past the Analysis to Paralysis when working on a new project?

    - by Cape Cod Gunny
    I've been struggling with how to get my project going. I've got an old software package that is in need of desparate rewrite. I haven't compiled the source code since 2004. It still sells, it's stable but does require the “Run this program in compatibility mode for:” on a lot of the newer windows systems. It's also one of those hard coded 640 X 480 screen resolution programs. Yuck! I can't seem to get started with this rewrite. I'm constantly fiddling around with different things. I'll play around with different fluid layouts for a while. Then I start looking around at how the main menu should work/look. I quickly find out that there's this thing called "Cool Bars" and I'll spend hours playing with that. Then I start thinking about stuff like "Oh I need to make sure that the screen sizes are preserved so when the application gets relaunched it remebers how the screens were positioned." Which leads to what happens if they have two monitors? Which leads to what happens if they have a quad screen? Yikes it's got to stop. I have always been a slow starter. I think about stuff long and hard up front. This has always plagued me. Once I get my mind made up then bam... I'm off and running. I'm looking for advice from some other one-person software companies that can help someone like me get off to a quicker start?

    Read the article

  • Code analysis: Global project/assembly suppression

    - by klausbyskov
    I have several CA1704:IdentifiersShouldBeSpelledCorrectly warnings that I want to suppress. Basically they refer to the company name which is deemed to be spelled incorrectly. The company name is part of several namespaces in my project, and in order to suppress all the warnings I need to add a lot of suppressions to the GlobalSuppressions file. Is there any way to suppress all warning in a single line in order to aviod my GlobalSuppressions file to become overly cluttered?

    Read the article

  • What static analysis tool do you prefer?

    - by glutz78
    Ideally, I'm looking for something to integrate into visual studio 2005, but if i could run it on the command line on windows or on linux for gcc, that would be okay also. I'm looking for something in the $1000 range so I understand I wont get the best tools available. But I also prefer something better than cppcheck, which is free. Any ideas? Thanks.

    Read the article

  • Perfmon File Analysis Tools

    - by Daniel Pollard
    I have a bunch of perfmon files that have captured information over a period of time. Whats the best tool to crunch this information? Idealy I'd like to be able to see avg stats per hour for the object counters that have been monitored.

    Read the article

  • Trend analysis using iterative value increments

    - by Dave Jarvis
    We have configured iReport to generate the following graph: The real data points are in blue, the trend line is green. The problems include: Too many data points for the trend line Trend line does not follow a Bezier curve (spline) The source of the problem is with the incrementer class. The incrementer is provided with the data points iteratively. There does not appear to be a way to get the set of data. The code that calculates the trend line looks as follows: import java.math.BigDecimal; import net.sf.jasperreports.engine.fill.*; /** * Used by an iReport variable to increment its average. */ public class MovingAverageIncrementer implements JRIncrementer { private BigDecimal average; private int incr = 0; /** * Instantiated by the MovingAverageIncrementerFactory class. */ public MovingAverageIncrementer() { } /** * Returns the newly incremented value, which is calculated by averaging * the previous value from the previous call to this method. * * @param jrFillVariable Unused. * @param object New data point to average. * @param abstractValueProvider Unused. * @return The newly incremented value. */ public Object increment( JRFillVariable jrFillVariable, Object object, AbstractValueProvider abstractValueProvider ) { BigDecimal value = new BigDecimal( ( ( Number )object ).doubleValue() ); // Average every 10 data points // if( incr % 10 == 0 ) { setAverage( ( value.add( getAverage() ).doubleValue() / 2.0 ) ); } incr++; return getAverage(); } /** * Changes the value that is the moving average. * @param average The new moving average value. */ private void setAverage( BigDecimal average ) { this.average = average; } /** * Returns the current moving average average. * @return Value used for plotting on a report. */ protected BigDecimal getAverage() { if( this.average == null ) { this.average = new BigDecimal( 0 ); } return this.average; } /** Helper method. */ private void setAverage( double d ) { setAverage( new BigDecimal( d ) ); } } How would you create a smoother and more accurate representation of the trend line?

    Read the article

  • ASP .NET Code analysis tool to check cross site scripting

    - by Prashant
    I am aware of a tool which MS has provided which tells you about coss site scripting attack etc. The tool is http://www.microsoft.com/downloads/details.aspx?FamilyId=0178e2ef-9da8-445e-9348-c93f24cc9f9d&displaylang=en But are there tools which you have used for ASP .NET applications which do similar to this and which one is widely used in ASP .Net applications ?

    Read the article

  • Alternatives to CAT.NET for website security analysis

    - by Gavin Miller
    I'm looking for an alternative tool to CAT.NET for performing static security scans on .NET code. Currently the CAT.NET tooling/development is at a somewhat fragile stage and doesn't offer the reliability that I'm looking for. Are there any alternative static code analyzers that you use for detecting security issues?

    Read the article

  • Fowler Analysis Patterns lately?

    - by Berryl
    As much as I've always loved this one is how much I always wished there were more meaty examples of how to apply some of the concepts available. Is anyone aware of anything out there worth looking at that attempts to that? Cheers, Berryl

    Read the article

  • Power Analysis in [R] for Two-Way Anova

    - by Thomas
    I am trying to calculate the necessary sample size for a 2x2 factorial design. I have two questions. 1) I am using the package pwr and the one way anova function to calculate the necessary sample size using the following code pwr.anova.test(k = , n = , f = , sig.level = , power = ) However, I would like to look at two way anova, since this is more efficient at estimating group means than one way anova. There is no two-way anova function that I could find. Is there a package or routine in [R] to do this? 2) Moreover, am I safe in assuming that since I am using a one-way anova power calculations, that the sample size will be more conservative (i.e. larger)?

    Read the article

  • DMX Analysis Services question

    - by user282382
    Hi, I am have two mining models, both are time series. One is [Company_Inputs] and the other is [Booking_Projections]. What I want to do is use EXTEND_MODEL_CASES to join the results of [Company_Inputs] as the extended cases. So basically something like: Select Flattened PredictTimeSeries([Bookings], 1, 6, EXTEND_MODEL_CASES) FROM [Booking_Projections] Natural Prediction Join (Select Flattened PredictTimeSeries([Metric1], 1, 6) From [Company_Inputs]) AS T This code of course doesn't work, but the idea is to use the predictions made from [Company_Inputs] as cases for predicting future values of [Booking_Projections] If anyone has an idea of how I can accomplish this I would appreciate it very much.

    Read the article

  • Visual Studio 2008 profiler analysis - missing time

    - by Scott Vercuski
    I ran the Visual Studio 2008 profiler against my ASP.NET application and came up with the following result set. CURRENT FUNCTION TIME (msec) ---------------------------------------------------|-------------- Data.GetItem(params) | 10,158.12 ---------------------------------------------------|-------------- Functions that were called by Data.GetItem(params) TIME (msec) ---------------------------------------------------|-------------- Model.GetSubItem(params) | 0.83 Model.GetSubItem2(params) | 0.77 Model.GetSubItem3(params) | 0.76 etc. The issue I'm facing is that the sum of the Functions called by Data.GetItem(params) do not sum up to the 10,158.12 msec total. This would lead me to believe that the bulk of the time is actually spent executing the code within that method. My question is ... does Visual Studio provide a way to analyze the method itself so I can see which sections of code are taking the longest? if it does not are there any recommended tools to do this? or should I start writing my own timing scripts? Thank you

    Read the article

  • Accelerometer data analysis

    - by jrrt
    Hello, I would like to know if there are some libraries/algorithms/techniques (python, if at all possible) that help to extract features from accelerometer data (extracted from and android phone, btw), like periodicity of movements, energy of acceleration and the like. Has anyone done this kind of task before? Thank you very much in advance :)

    Read the article

  • Investment advice data dump analysis

    - by portoalet
    For my year-end pet project, I'd like to analyze investment advices and their correlation to the stock market performance. The problem is, where do I get the dump of investment advice data (free) ? something like stackoverflow.com data dump will be nice. Or maybe it's easier to do distributed crawling and crawl the public finance webpages for investment advices? Investment advice is buy/sell advice for stocks/forex, issued by institution/investment advisor.

    Read the article

  • Code Coverage Analysis for Embedded C++ projects

    - by Steve Hawkins
    I have recently started working on a very large C++ project that, after completing 90% of the implementation, has determined that they need to demonstrate 100% branch coverage during testing. The project is hosted on an embedded platform (Green Hills Integrity). I'm looking for suggestions and experiences from others on StackOverflow that have used code coverage products in similar environments. I'm interested in both positive and negative comments regarding these types of tools.

    Read the article

  • Blob ID matching over multiple frames in C++ (image analysis)

    - by pollux
    Dear reader, I'm working on a blob matching and tracking library in C++. Currently I'm using openCV to detect blobs and try to match blobs in a new frame by checking the position, velocity and size of the blob. This works quite okay and I'm receiving a high blob match rate (95% or higher). Sometimes blobs fall out of the image or new blobs appear. Now I need to give matched blobs the same ID as they had before. I'm wondering if there are typical or commonly used techniques for doing this. Or even some keywords I can use to google on. Thanks

    Read the article

  • SQL Analysis Services - Dimension attributes with a "many" cardinality

    - by MonkeyBrother
    I am creating a cube with the following tables: Customer CustomerID, Name Customer Rep CustomerID, RepID Rep RepID, Name The important thing here is that there is a many to many relationship between Reps and Customers. I want to be able to ask the question "How much sales for customers working with rep 'A'?" In the data source view i set up the relationships between both customerid columns and both repid columns. I set up the rep attribute in the dimension builder and when I try to build the cube I get this error: Errors in the high-level relationship engine. the 'Rep' table that is required for a join cannot be reached based on the relationships in the data source view.

    Read the article

  • difference between a lexical, morphological and semantic mistakes?

    - by AnH
    Hi there, I just want to make sure that I understood the differences between a lexical, morphological and semantic mistakes correctly. If am not mistaken semantic mistakes deal with problems concerning meanings, for example writing a sentence that is correct grammar wise but doesn't make any sense is a semantic mistake, morphological mistakes deals more or less on how the word should look like, for example childrenS is a morphological mistake, so that leaves lexical mistakes, what are those exactly?? can someone sum up the differences between those 3 mistakes please so that I may know for sure that I got them down correctly? Thank you in advance

    Read the article

  • Analysis Services with excel as front end - is it possible to get the nicer UI that powerpivot provi

    - by AJM
    I have been looking into PowerPivot and concluded that for "self service BI" and ahoc buidling of cubes it has its uses. In particular I like the enhanced UI that you get from using PowerPivot rather than just using a PivotTable hooked up to an analysis services datasource. However it seems that hooking up PowerPivot to an existing analysis services cube is not a solution for "organisational BI". It is not always desireable to suck millions of rows into excel at once and the interface between PowerPivot and analysis services is very poor in my book. Hence the question is can an existing analysis services solution get the enhanced ui features that power pivot brings, withoout using powerpivot as the design tool? If powerpivot is aimed ad self service/personal BI then it seems bizare that the UI for this is better than for bigger/more costly analysis services solutions.

    Read the article

  • Specify a custom dictionary for FxCop and Visual Studio source analysis

    - by Marko Apfel
    Renaming the default custom dictionary from CustomDictionary.xml to an other name – for instance FxCop.CustomDictionary.xml needs some additional changes to work in involved applications. Visual Studio Team System code analysis For Visual Studio Team System code analysis this file should be added as a link to all projects and setted to be the Build Action CodeAnalysisDirectory. Build target In a build target the command line tool FxCopCmd should be called with the /dictionary parameter: <Target Name="FxCop"> <Exec Command="&quot;$(ProjectDir)..\..\build\FxCop\FxCopCmd.exe&quot; /file:&quot;$(TargetPath)&quot; /project:&quot;$(ProjectDir)..\EsriDE.SfgPraxair.FxCop&quot; /directory:&quot;$(ProjectDir)..\..\lib\Esri.ArcGIS&quot; /directory:&quot;$(ProjectDir)..\..\lib\Microsoft&quot; /dictionary:&quot;$(ProjectDir)..\FxCop.CustomDictionary.xml&quot; /out:&quot;$(OutDir)..\$(ProjectName).FxCopReport.xml&quot; /console /forceoutput /ignoregeneratedcode"> </Exec> <Message Text="FxCop finished." /> </Target> FxCop-GUI (standalone application) In FxCop-GUI is no option to specify an own file name – but you could add a hint in the FxCop project file. Open your this file and look for the line: <CustomDictionaries SearchFxCopDir="True" SearchUserProfile="True" SearchProjectDir="True" /> Then change it to: <CustomDictionaries SearchFxCopDir="True" SearchUserProfile="True" SearchProjectDir="True"> <CustomDictionary Path="FxCop.CustomDictionary.xml"/> </CustomDictionaries> Ready :-)

    Read the article

  • Great Example of a Simple Cost-Benefit Analysis

    - by BuckWoody
    I saw a post the other day that you should definitely go check out. It’s a cost/benefit decision, and although the author gives it a quick treatment and doesn’t take all points in the decision into account, you should focus on the process he follows. It’s a quick and simple example of the kind of thought process we should have as data professionals when we pick a server, a process, or application and even platform software. The key is to include more than just the price of a piece of software or hardware. You need to think about the “other” costs in the decision, and then make the right one. Sometimes the cheapest option is the cheapest, and other times, well, it isn’t. I’ve seen this played out not only in the decision to go with a certain selection, but in the options or editions it comes in. You have to put all of the decision points in the analysis to come up with the right answer, and you have to be able to explain your logic to your team and your company. This is the way you become a data professional, not just a DBA. You can check out the post here – it deals with Azure, but the point is the process, not Azure itself: http://blogs.msdn.com/eugeniop/archive/2010/03/19/windows-azure-guidance-a-simplistic-economic-analysis-of-a-expense-migration.aspx Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

< Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >