Search Results

Search found 1329 results on 54 pages for 'rob'.

Page 3/54 | < Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • Full-text Indexing Books Online

    - by Most Valuable Yak (Rob Volk)
    While preparing for a recent SQL Saturday presentation, I was struck by a crazy idea (shocking, I know): Could someone import the content of SQL Server Books Online into a database and apply full-text indexing to it?  The answer is yes, and it's really quite easy to do. The first step is finding the installed help files.  If you have SQL Server 2012, BOL is installed under the Microsoft Help Library.  You can find the install location by opening SQL Server Books Online and clicking the gear icon for the Help Library Manager.  When the new window pops up click the Settings link, you'll get the following: You'll see the path under Library Location. Once you navigate to that path you'll have to drill down a little further, to C:\ProgramData\Microsoft\HelpLibrary\content\Microsoft\store.  This is where the help file content is kept if you downloaded it for offline use. Depending on which products you've downloaded help for, you may see a few hundred files.  Fortunately they're named well and you can easily find the "SQL_Server_Denali_Books_Online_" files.  We are interested in the .MSHC files only, and can skip the Installation and Developer Reference files. Despite the .MHSC extension, these files are compressed with the standard Zip format, so your favorite archive utility (WinZip, 7Zip, WinRar, etc.) can open them.  When you do, you'll see a few thousand files in the archive.  We are only interested in the .htm files, but there's no harm in extracting all of them to a folder.  7zip provides a command-line utility and the following will extract to a D:\SQLHelp folder previously created: 7z e –oD:\SQLHelp "C:\ProgramData\Microsoft\HelpLibrary\content\Microsoft\store\SQL_Server_Denali_Books_Online_B780_SQL_110_en-us_1.2.mshc" *.htm Well that's great Rob, but how do I put all those files into a full-text index? I'll tell you in a second, but first we have to set up a few things on the database side.  I'll be using a database named Explore (you can certainly change that) and the following setup is a fragment of the script I used in my presentation: USE Explore; GO CREATE SCHEMA help AUTHORIZATION dbo; GO -- Create default fulltext catalog for later FT indexes CREATE FULLTEXT CATALOG FTC AS DEFAULT; GO CREATE TABLE help.files(file_id int not null IDENTITY(1,1) CONSTRAINT PK_help_files PRIMARY KEY, path varchar(256) not null CONSTRAINT UNQ_help_files_path UNIQUE, doc_type varchar(6) DEFAULT('.xml'), content varbinary(max) not null); CREATE FULLTEXT INDEX ON help.files(content TYPE COLUMN doc_type LANGUAGE 1033) KEY INDEX PK_help_files; This will give you a table, default full-text catalog, and full-text index on that table for the content you're going to insert.  I'll be using the command line again for this, it's the easiest method I know: for %a in (D:\SQLHelp\*.htm) do sqlcmd -S. -E -d Explore -Q"set nocount on;insert help.files(path,content) select '%a', cast(c as varbinary(max)) from openrowset(bulk '%a', SINGLE_CLOB) as c(c)" You'll need to copy and run that as one line in a command prompt.  I'll explain what this does while you run it and watch several thousand files get imported: The "for" command allows you to loop over a collection of items.  In this case we want all the .htm files in the D:\SQLHelp folder.  For each file it finds, it will assign the full path and file name to the %a variable.  In the "do" clause, we'll specify another command to be run for each iteration of the loop.  I make a call to "sqlcmd" in order to run a SQL statement.  I pass in the name of the server (-S.), where "." represents the local default instance. I specify -d Explore as the database, and -E for trusted connection.  I then use -Q to run a query that I enclose in double quotes. The query uses OPENROWSET(BULK…SINGLE_CLOB) to open the file as a data source, and to treat it as a single character large object.  In order for full-text indexing to work properly, I have to convert the text content to varbinary. I then INSERT these contents along with the full path of the file into the help.files table created earlier.  This process continues for each file in the folder, creating one new row in the table. And that's it! 5 SQL Statements and 2 command line statements to unzip and import SQL Server Books Online!  In case you're wondering why I didn't use FILESTREAM or FILETABLE, it's simply because I haven't learned them…yet. I may return to this blog after I figure that out and update it with the steps to do so.  I believe that will make it even easier. In the spirit of exploration, I'll leave you to work on some fulltext queries of this content.  I also recommend playing around with the sys.dm_fts_xxxx DMVs (I particularly like sys.dm_fts_index_keywords, it's pretty interesting).  There are additional example queries in the download material for my presentation linked above. Many thanks to Kevin Boles (t) for his advice on (re)checking the content of the help files.  Don't let that .htm extension fool you! The 2012 help files are actually XML, and you'd need to specify '.xml' in your document type column in order to extract the full-text keywords.  (You probably noticed this in the default definition for the doc_type column.)  You can query sys.fulltext_document_types to get a complete list of the types that can be full-text indexed. I also need to thank Hilary Cotter for giving me the original idea. I believe he used MSDN content in a full-text index for an article from waaaaaaaaaaay back, that I can't find now, and had forgotten about until just a few days ago.  He is also co-author of Pro Full-Text Search in SQL Server 2008, which I highly recommend.  He also has some FTS articles on Simple Talk: http://www.simple-talk.com/sql/learn-sql-server/sql-server-full-text-search-language-features/ http://www.simple-talk.com/sql/learn-sql-server/sql-server-full-text-search-language-features,-part-2/

    Read the article

  • Merge replication stopping without errors in SQL 2008 R2

    - by Rob Farley
    A non-SQL MVP friend of mine, who also happens to be a client, asked me for some help again last week. I was planning on writing this up even before Rob Volk (@sql_r) listed his T-SQL Tuesday topic for this month. Earlier in the year, I (well, LobsterPot Solutions, although I’d been the person mostly involved) had helped out with a merge replication problem. The Merge Agent on the subscriber was just stopping every time, shortly after it started. With no errors anywhere – not in the Windows Event Log, the SQL Agent logs, not anywhere. We’d managed to get the system working again, but didn’t have a good reason about what had happened, and last week, the problem occurred again. I asked him about writing up the experience in a blog post, largely because of the red herrings that we encountered. It was an interesting experience for me, also because I didn’t end up touching my computer the whole time – just tapping on my phone via Twitter and Live Msgr. You see, the thing with replication is that a useful troubleshooting option is to reinitialise the thing. We’d done that last time, and it had started to work again – eventually. I say eventually, because the link being used between the sites is relatively slow, and it took a long while for the initialisation to finish. Meanwhile, we’d been doing some investigation into what the problem could be, and were suitably pleased when the problem disappeared. So I got a message saying that a replication problem had occurred again. Reinitialising wasn’t going to be an option this time either. In this scenario, the subscriber having the problem happened to be in a different domain to the publisher. The other subscribers (within the domain) were fine, just this one in a different domain had the problem. Part of the problem seemed to be a log file that wasn’t being backed up properly. They’d been trying to back up to a backup device that had a corruption, and the log file was growing. Turned out, this wasn’t related to the problem, but of course, any time you’re troubleshooting and you see something untoward, you wonder. Having got past that problem, my next thought was that perhaps there was a problem with the account being used. But the other subscribers were using the same account, without any problems. The client pointed out that that it was almost exactly six months since the last failure (later shown to be a complete red herring). It sounded like something might’ve expired. Checking through certificates and trusts showed no sign of anything, and besides, there wasn’t a problem running a command-prompt window using the account in question, from the subscriber box. ...except that when he ran the sqlcmd –E –S servername command I recommended, it failed with a Named Pipes error. I’ve seen problems with firewalls rejecting connections via Named Pipes but letting TCP/IP through, so I got him to look into SQL Configuration Manager to see what kind of connection was being preferred... Everything seemed fine. And strangely, he could connect via Management Studio. Turned out, he had a typo in the servername of the sqlcmd command. That particular red herring must’ve been reflected in his cheeks as he told me. During the time, I also pinged a friend of mine to find out who I should ask, and Ted Kruger (@onpnt) ‘s name came up. Ted (and thanks again, Ted – really) reconfirmed some of my thoughts around the idea of an account expiring, and also suggesting bumping up the logging to level 4 (2 is Verbose, 4 is undocumented ridiculousness). I’d just told the client to push the logging up to level 2, but the log file wasn’t appearing. Checking permissions showed that the user did have permission on the folder, but still no file was appearing. Then it was noticed that the user had been switched earlier as part of the troubleshooting, and switching it back to the real user caused the log file to appear. Still no errors. A lot more information being pushed out, but still no errors. Ted suggested making sure the FQDNs were okay from both ends, in case the servers were unable to talk to each other. DNS problems can lead to hassles which can stop replication from working. No luck there either – it was all working fine. Another server started to report a problem as well. These two boxes were both SQL 2008 R2 (SP1), while the others, still working, were SQL 2005. Around this time, the client tried an idea that I’d shown him a few years ago – using a Profiler trace to see what was being called on the servers. It turned out that the last call being made on the publisher was sp_MSenumschemachange. A quick interwebs search on that showed a problem that exists in SQL Server 2008 R2, when stored procedures have more than 4000 characters. Running that stored procedure (with the same parameters) manually on SQL 2005 listed three stored procedures, the first of which did indeed have more than 4000 characters. Still no error though, and the problem as listed at http://support.microsoft.com/kb/2539378 describes an error that should occur in the Event log. However, this problem is the type of thing that is fixed by a reinitialisation (because it doesn’t need to send the procedure change across as a transaction). And a look in the change history of the long stored procs (you all keep them, right?), showed that the problem from six months earlier could well have been down to this too. Applying SP2 (with sufficient paranoia about backups and how to get back out again if necessary) fixed the problem. The stored proc changes went through immediately after the service pack was applied, and it’s been running happily since. The funny thing is that I didn’t solve the problem. He had put the Profiler trace on the server, and had done the search that found a forum post pointing at this particular problem. I’d asked Ted too, and although he’d given some useful information, nothing that he’d come up with had actually been the solution either. Sometimes, asking for help is the most useful thing you can do. Often though, you don’t end up getting the help from the person you asked – the sounding board is actually what you need. @rob_farley

    Read the article

  • How to use regex to extract nested patterns

    - by Rob Romanek
    Hi I'm struggling with some regex I've got a string like this: a:b||c:{d:e||f:g}||h:i basically name value pairings. I want to be able to parse out the pairings so I get: a:b c:{d:e||f:g} h:i then I can further parse the pairings contained in { } if required It is the nesting that is making me scratch my head. Any regex experts out there that can give me a hand? thanks, Rob

    Read the article

  • BizTalk FTP - File Rename

    - by Rob Bowman
    Hi I need to send files using the BizTalk 2006 (non wcf) FTP adapter. After successful transmission of each file it needs to be renamed from an "A" prefix to a "U" prefix. I don't want to issue a command using wildcards because I can't be sure of other files in the destination folder. Any ideas? Thanks Rob.

    Read the article

  • PHP object access using string

    - by Rob Waring
    Hi all, I have an object in PHP with some very odd property names. I just need to know how to access a property when it's name is "//www.w3.org/1999/02/22-rdf-syntax-ns#type". I found something that suggested $object->{'//www.w3.org/1999/02/22-rdf-syntax-ns#type'}; but that doesn't seem to work. Thanks in advance Rob

    Read the article

  • Android: RTL support - digits embedded in a right to left sentence (Hebrew)

    - by Rob
    Hi There, My application displays Hebrew text which comes from a Web Service. When a sentence contains digits (in the middle of it), the digits appear in a mirror view: 29 appears as 92, 21:45 appears as 54:12 and 2,000 appears as 000,2. Also, when a sentence starts with digits or English characters, they get thrown to the end of the sentence messing it all up... Does anyone have an idea how can this be solved? Is RTL support in Android still immature? Thanks, Rob

    Read the article

  • Has there been a Firefox update recently?

    - by Rob Nicholson
    Has there been an update to Firefox recently? Users are reporting they are unable to use my web application when using Firefox (can't enter text into Infragistics grid). I've not changed anything recently and it works when I try with Firefox. So more of a general "Has there been an significant upgrade recently?" kind of question. Thanks, Rob.

    Read the article

  • Android: how to convert a Drawable to Bitmap

    - by Rob
    How can I convert a Drawable to a Bitmap? I would like to set a certain drawable as the device's wallpaper, but all wallpaper functions accept Bitmaps only (I cannot use WallpaperManager - I'm pre 2.1). Also, my drawables are downloaded from the web and do not reside in R.drawable. Thanks, Rob

    Read the article

  • Get Doxygen to show XCode compile errors if documentation is missing

    - by Rob
    I have been successful with this in the past but now I can't seem to find the setting that tells XCode or Doxygen (or both) to display compile errors if documentation is missing. I have looked at the Apple documentation that tells how to create docsets and that works fine but XCode does not throw compiling errors on missing comments in the source code. Anyone know how to get this turned on? Thanks, Rob

    Read the article

  • Possible to split one JAX-WS service across multiple source files?

    - by Rob S.
    Hi everyone, Is it possible to split a web service in to multiple classes and still provide a single path to the web service? I know this isn't possible because of the duplicate url-pattern values. It sort of illustrates where we're wanting to go :) <endpoint name="OneBigService" implementation="SmallImpl1" url-pattern="/OneBigService"/> <endpoint name="OneBigService" implementation="SmallImpl2" url-pattern="/OneBigService"/> Basically, how do avoid having one monolithic @WebService class? Thanks! Rob

    Read the article

  • Is it poor practice to build an SQL query using WHERE 1=1 AND ...

    - by Rob
    I'm writing a script that builds an SQL query by concatenating the string and adding conditions to the WHERE clause as needed. Would it be better practice to use WHERE 1=1 so that the first condition is satisfied and the script can just concatenate an AND x = 'y' to the query, or should I write the extra code to check if a clause has been added and if not, add the AND ? Confusing question, I know. Let me know if I need to be more clear. Rob

    Read the article

  • tryin' to enumerate all available Wifi network with J2SE

    - by Rob
    Hi, I'd like to find out a way to enumerate all available wifi networks using Java 6.0 and any third-party API. Unfortunately, I'm not able to find a solution to this problem, all I got is a library that I can't use because I don't have any example. I'll be very pleased if someone could help me. The target platform is Win XP/7. Edit: the library I found is named jwlanscan Rob

    Read the article

  • MSQL upgrade on Ubuntu - any heads ups?

    - by Rob Sedge
    I am needing to upgrade MYSQL on Ubuntu, it is a production server and naturally cautious. My many googles look to be essentially saying that I need to : 1) Backup my current mysql database and tables/data 2) Uninstall current mysql 3) Install new MYSQL 5+ 4) Restore Databases/ tables and data 5) Hope and Pray I got it right ?? Something doesn't seem right, sounds like a lot of down time and risk Am I missing something / or any simple solutions? Upgrading from mSQL 4 to 5 on Ubuntu 10 Many Thanks, Rob

    Read the article

  • Displaying fancy routes in cloudmade based service

    - by Rob
    I look for a way to display a route in a fancy way using the Cloudmade service. Currently, I can see computed routes like on this tutorial http://developers.cloudmade.com/projects/web-maps-lite/examples/routing, but I look for a fancier way to do it -- without A and B tags, and with colors, etc. Is this possible ? Thanks for your help Rob

    Read the article

  • Jquery slider going the wrong way

    - by Rob
    Im using a jquery slider and having an issue with what direction it scrolls on click. It seems to screw up after clicking the right arrow then left, but will scroll left if clicked first, weird. Is this a bug or am I missing something here? PS- Im using Chrome (Mac) but seems to behave the same with Firefox and Safari See example http://www.warface.co.uk/clients/warface.co.uk/blog/ Please click the red central arrow to reveal the slider. Many thanks Rob

    Read the article

  • Run Explorer in SYSTEM account on Windows Vista or 7 using Sysinternal’s psexec tool?

    - by Rob
    Has anyone been successful at launching an instance of Windows Explorer in the SYSTEM account on Windows Vista or 7? It is possible to do this on XP, but I haven't been able to get it to completely work in Vista or 7. Trying to launch Explorer as SYSTEM into session 1 (my user session) results in Explorer exiting immediately and returning an error code of 1. I can launch Explorer as SYSTEM into session 0 with the following command: psexec -i 0 -s explorer That will create an instance of explorer running as SYSTEM with a taskbar and start menu on the hidden session 0 desktop, but won't let you open a file browser window. If you switch to the hidden session 0 desktop and try to open an Explorer window from there to browse files, the following error message appears: "The server process could not be started because the configured identity is incorrect. Check the username and password." I have set the following registry key to 1 for my user account and the SYSTEM account: \Software\Microsoft\Windows\CurrentVersion\Explorer\Advanced\SeparateProcess There has got to be a way to make this work? If it is not possible, can anyone explain why? -Rob

    Read the article

  • Executable Resumes

    - by Liam McLennan
    Over the past twelve months I have been thinking a lot about executable specifications. Long considered the holy grail of agile software development, executable specifications means expressing a program’s functionality in a way that is both readable by the customer and computer verifiable in an automatic, repeatable way. With the current generation of BDD and ATDD tools executable specifications seem finally within the reach of a significant percentage of the development community. Lately, and partly as a result of my craftsmanship tour, I have decided that soon I am going to have to get a job (gasp!). As Dave Hoover describes in Apprenticeship Patters, “you … have mentors and kindred spirits that you meet with periodically, [but] when it comes to developing software, you work alone.” The time may have come where the only way for me to feel satisfied and enriched by my work is to seek out a work environment where I can work with people smarter and more knowledgeable than myself. Having been on both sides of the interview desk many times I know how difficult and unreliable the process can be. Therefore, I am proposing the idea of executable resumes. As a journeyman programmer looking for a fruitful work environment I plan to write an application that demonstrates my understanding of the state of the art. Potential employers can download, view and execute my executable resume and judge wether my aesthetic sensibility matches their own. The concept of the executable resume is based upon the following assertion: A line of code answers a thousand interview questions Asking people about their experiences and skills is not a direct way of assessing their value to your organisation. Often it simple assesses their ability to mislead an interviewer. An executable resume demonstrates: The highest quality code that the person is able to produce. That the person is sufficiently motivated to produce something of value in their own time. That the person loves their craft. The idea of publishing a program to demonstrate a developer’s skills comes from Rob Conery, who suggested that each developer should build their own blog engine since it is the public representation of their level of mastery. Rob said: Luke had to build his own lightsaber – geeks should have to build their own blogs. And that should be their resume. In honour of Rob’s inspiration I plan to build a blog engine as my executable resume. While it is true that the world does not need another blog engine it is as good a project as any, it is a well understood domain, and I have not found an existing blog engine that I like. Executable resumes fit well with the software craftsmanship metaphor. It is not difficult to imagine that under the guild system master craftsmen may have accepted journeymen based on the quality of the work they had produced in the past. We now understand that when it comes to the functionality of an application that code is the final arbiter. Why not apply the same rule to hiring?

    Read the article

  • Tab Sweep: Dynamic JSF Forms, GlassFish on VPS, Upgrading to 3.1.2, Automated Deployment Script, ...

    - by arungupta
    Recent Tips and News on Java, Java EE 6, GlassFish & more : • Dynamic forms, JSF world was long waiting for (Oleg Varaksin) • Creating a Deployment Pipeline with Jenkins, Nexus, Ant and Glassfish (Rob Terp) • Installing Java EE 6 SDK with Glassfish included on a VPS without GUI (jvm host) • GlassFish multimode Command for Batch Processing (javahowto) • Servlet Configuration in Servlet 3.0 api (Nikos Lianeris) • Creating a Simple Java Message Service (JMS) Producer with NetBeans and GlassFish (Oracle Learning Library) • GlassFish 3.1 to JBoss AS 7.1.1 EJB Invocation (java howto) • Tests In Java Ee For Zero-error Applications (Dylan Rodriguez) • Upgrading GlassFish 3.1.1 to 3.1.2 on Oracle Linux 6.2 64-bit (Matthias Hoys) • Migrating an Automated Deployment Script from Glassfish v2 to Glassfish v3 (Rob Terp) • Installer updates, Glassfish, Confluence and more…! (Rimu Hosting)

    Read the article

  • X.509 Certificate validation with Java and Bouncycastle

    - by Rob
    Hi, through the bouncycastle wiki page I was able to understand how to create a X.509 root certificate and a certification request, but I do not quite understand how to proceed concept- and programming wise after that. Lets assume party A does a cert request and gets his client certificate from the CA. How can some party B validate A's certificate? What kind of certificate does A need? A root certificate? A 'normal' client certificate? And how does the validation work on programming level, if we assume that A has successfully send his certificate in DER or PEM format to B? Any help is much appreciated. Best Regards, Rob

    Read the article

  • Passing variable to Google Charts URL

    - by Rob A
    Hi All, This is probably something really simple, however I am quite new to PHP, and havent done any HTML in years. I need to get a PHP variable filled with an array of figures into Google Charts. My code for this so far is: <img src="http://chart.apis.google.com/chart? &chs=340x175 &chd=t:<?=$filedetail[1]?> &cht=lc &chtt=Test "> However, Google reports an error, as it stops at the ?=$filedetail[1] for some reason. It doesnt seem that reading the variable is the problem, more that the API simply cant read past the start of the PHP tags. Thanks, Rob A.

    Read the article

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