Search Results

Search found 82077 results on 3284 pages for 'better at work'.

Page 194/3284 | < Previous Page | 190 191 192 193 194 195 196 197 198 199 200 201  | Next Page >

  • Does using structure data semantic LocalBusiness schema markup work for local EMD URL's?

    - by ElHaix
    Based on what I have read about Google's recent Panda and Penguin updates, I'm getting the impression that using semantic markup may help improve SEO results. On a EMD (exact match domain) site, that may have been hit, we list location-based products. We are now going to be adding a itemtype="http://schema.org/Product" to each product, with relevant details. However, that product may be available in Los Angeles and also in appear in a Seattle results page. We could add a LocalBusiness item type on each geo page to define the geo location for that page. While the definition states: A particular physical business or branch of an organization. Examples of LocalBusiness include a restaurant, a particular branch of a restaurant chain, a branch of a bank, a medical practice, a club, a bowling alley, etc. We could add use the location property which would simply include the city/state details. I realize that this looks like it is meant for a physical location, however could this be done without seeming black-hat?

    Read the article

  • 2011 - ALMs for your development team and the people they work with.

    - by David V. Corbin
    Welcome to 2011, it is already shaping up to be a very exciting year. The title of the post is not about charitable giving, although that is also a great topic. Application Lifecycle Management and the Systems that support the environment is, and 2011 will be a year where I expect many teams to invest heavily in this area. For those not familiar with ALM, it can be simplified down to "A comprehensive view of all of the iteas, requirements, activities and artifacts that impact an application over the course of its lifecycle, from concept until decommissioning". Obviously, this encompases a large number of different areas even for relatively small and medium sized projects. In recent years, many teams have adapted methodoligies which address individual aspects of this; but the majority of this adoption has resulted in "islands of improvement" rather than the desired comprehensive outcome...Until now! Last year Microsoft released Team Foundation Server 2010 along with Visual Studio 2010 Ultimate Edition, and with these two in combination the situation has drastically changed. At last there is a single environment that is capable of handling all aspects of ALM, and is also capable of dealing with migration and integration with existing systems to make the transition to a single solution much easier. Thse possibilities (and practicalities) are nothing short of amazing, Architecture thru Testing integration? YES. Being able to correlate specific requirement items (and their history) to actual code (and code history)? YES. Identification of which tests will be potentially impacted by a given code change? YES. Resiliant Automated Testing of User Interfaces? YES. Automatic Deployment Management? YES. Integraton Level testing as part of (designated) Builds? YES. I could easily double or triple the above list, but these items should be enough to get you thinking about the "pain points" your team and organization currently face and the fact that there IS a way to relieve the pain. Over the course of the year, I am hoping to bring together some of the "best of breed" information, along with hosting (and participating in) discussions with various experts in the field. There are already a number of groups (including many on LinkedIn) that have an ALM focus, and I encourage everyone out to check them out. I will be posting a list of the ones I find most helpful in the not too distant future. As I said at the beginning, 2011 is shaping up to be a very interesting (and productive) year. Why wait to start investigating and adopting ALM? ps: For those interested in becoming an "Alms Giver" in the charitable sense, I highly recommend checking out GiveCamp. A group of developers, designers and others get together to create a solution for a charity in just under 48 hours. I will be attending the GiveCamp in New York City on Jan 14-16, more information is available at nycgivecamp.org/

    Read the article

  • How does a BSP tree work for Z sorting?

    - by Jenko
    I'm developing a 3D engine in software, and so I must compute Z sorting manually. I'm currently using the painters algorithm to sort triangles and then drawing them back-to-front. This causes artifacts that I'm trying to correct. Would using a dynamic BSP-tree ensure "correct Z sorting" of triangles? Why? Because the bounding volumes of triangles would be similar? Since I would have a single "world" BSP tree, would I have to remove and re-add any moved/scaled/rotated object into the tree? Is it possible to add triangles into a BSP tree without the expensive cutting process? Why do you need to cut triangles on the axis planes anyway? Is it faster to traverse a BSP tree from any angle, than to sort all tris each draw like the painters algorithm?

    Read the article

  • Upstart: How does rc job work / order of (contradicting) "start on ..." and "stop on ..." stanzas

    - by Binarus
    Hi, I just can't understand how Upstart's rc job definition in Natty 11.04 works. To illustrate the problem, here is the definition (empty lines and comments are left out): start on runlevel [0123456] stop on runlevel [!$RUNLEVEL] export RUNLEVEL export PREVLEVEL console output env INIT_VERBOSE task exec /etc/init.d/rc $RUNLEVEL Let's suppose we currently are in runlevel 2 and the rc job is stopped (that is exactly the situation after booting my box and logging in via SSH). Now, let's assume that the system switches to runlevel 3, for example due to a command like "telinit 3" given by root. What will happen to the rc job? Obviously, the rc job will be started since it is currently stopped and the event runlevel 3 is matching the start events. But from now on, things are unclear to me: According to the manual $RUNLEVEL evaluates to the new runlevel when the job is started (that means 3 in our example). Therefore, the next stanza "stop on runlevel [!$RUNLEVEL]" translates to "stop on runlevel [!3]"; that means we have a first stanza which will trigger the job, but the second stanza will never stop the job and seems to be useless. Since I know that the Ubuntu / Upstart people won't do useless things, I must be heavily misunderstanding something. I would be grateful for any explanation. While trying to understand this, an additional question came to my mind. If I had contradicting start and stop triggers, for example start on foo stop on foo what would happen? I swear I never will do that, but I am nevertheless very interested in how Upstart handles that on the theoretical level. Thank you very much! Editing the question as a reaction on geekosaur's first answer: I can see the parallelism, but it is not that easy (at least, not to me). Let's assume the job aurrently is still running, and a new runlevel event comes in (of course, the new runlevel is different from the current one). Then, the following should happen: 1) The job is single instance. That means that "start on ..." won't be triggered since the job is currently running; $RUNLEVEL is not touched. 2) "stop on ..." will be triggered since the new runlevel is different from $RUNLEVEL, so the job will be aborted. 3) Now, the job is stopped and waiting. I can't see how it is restarted with the new runlevel. AFAIK, initctl emits events only once, so "start on ..." won't be triggered and the new runlevel won't be entered. I know that I still misunderstanding something, and I am grateful for explanations. Thank you very much!

    Read the article

  • how to make audio and video streaming servers work?

    - by explorex
    I am a php mysql developer ... just an (below) average. and i am interested in the way television and radio are broadcasted over internet live. i want to know how it works and and what are its requirements (which package of which programming language offers the best). i must admit that i am a complete layman but i expect it do by next half month or year or so. And please clarify me Websites are stored in servers. From my desktop, i want to broadcast some video, then i need to connect to webserver(to upstream the video). Is there an application to do that (or do i have to code that or embed in my web application and which programming language would be suitable(does python support that))? and i also need a script to handle the upstreamed video or audio(can i do that with php)?

    Read the article

  • How does rc job work / order of (contradicting) "start on ..." and "stop on ..." stanzas

    - by Binarus
    Hi, I just can't understand how Upstart's rc job definition in Natty 11.04 works. To illustrate the problem, here is the definition (empty lines and comments are left out): start on runlevel [0123456] stop on runlevel [!$RUNLEVEL] export RUNLEVEL export PREVLEVEL console output env INIT_VERBOSE task exec /etc/init.d/rc $RUNLEVEL Let's suppose we currently are in runlevel 2 and the rc job is stopped (that is exactly the situation after booting my box and logging in via SSH). Now, let's assume that the system switches to runlevel 3, for example due to a command like "telinit 3" given by root. What will happen to the rc job? Obviously, the rc job will be started since it is currently stopped and the event runlevel 3 is matching the start events. But from now on, things are unclear to me: According to the manual $RUNLEVEL evaluates to the new runlevel when the job is started (that means 3 in our example). Therefore, the next stanza "stop on runlevel [!$RUNLEVEL]" translates to "stop on runlevel [!3]"; that means we have a first stanza which will trigger the job, but the second stanza will never stop the job and seems to be useless. Since I know that the Ubuntu / Upstart people won't do useless things, I must be heavily misunderstanding something. I would be grateful for any explanation. While trying to understand this, an additional question came to my mind. If I had contradicting start and stop triggers, for example start on foo stop on foo what would happen? I swear I never will do that, but I am nevertheless very interested in how Upstart handles that on the theoretical level. Thank you very much! Editing the question as a reaction on geekosaur's first answer: I can see the parallelism, but it is not that easy (at least, not to me). Let's assume the job aurrently is still running, and a new runlevel event comes in (of course, the new runlevel is different from the current one). Then, the following should happen: 1) The job is single instance. That means that "start on ..." won't be triggered since the job is currently running; $RUNLEVEL is not touched. 2) "stop on ..." will be triggered since the new runlevel is different from $RUNLEVEL, so the job will be aborted. 3) Now, the job is stopped and waiting. I can't see how it is restarted with the new runlevel. AFAIK, initctl emits events only once, so "start on ..." won't be triggered and the new runlevel won't be entered. I know that I still misunderstanding something, and I am grateful for explanations. Thank you very much!

    Read the article

  • Will Ubuntu 12.04 work on an Acer Aspire One D257-1682?

    - by Avid
    Is it possible to install Ubuntu 12.04 on an Acer Aspire One D257-1682? I have a disc that I used to run Ubuntu 12.04 on my desktop without incident. When trying to run it on my Aspire One, using an external DVD drive via USB, I get a ton of errors about missing files while it tries to load Ubuntu. I found a list of which versions of Ubuntu to run depending on which Aspire One model you have, but the D257 was not listed at all. I want to have Ubuntu on my Aspire One for coding Java and C++ since I don't have to sit at my desk for hours. Please help! -Avid

    Read the article

  • How do Expires headers and cache manifest rules work together?

    - by Robert K
    I find the W3C's official Offline Web Applications specification to be rather vague about how the cache manifest interacts with headers such as ETag, Expires, or Pragma on cached assets. I know that the manifest should be checked with each request so that the browser knows when to check the other assets for updates. But because the specification doesn't define how the cache manifest interacts with normal cache instructions, I can't predict precisely how the browser will react. Will assets with a future expiration date be refreshed (no matter the cache headers) when the cache manifest is updated? Or, will those assets obey the normal caching rules? Which caching mechanism, HTTP cache versus cache manifest, will take precedence, and when?

    Read the article

  • Better way to get int from FragmentActivity in a Fragment?

    - by Gimberg
    Hi im trying to get an int from my FragmentActivity and i have a way to do this but the code get very cluttered and long and i did this to shorten the problem and i don't get any errors while in the editor but when i run the app it eminently crashes. Any suggestions? An example of the code that doesn't work MainActivity mainActivity = ((MainActivity)getActivity()); public View onCreateView(LayoutInflater inflater, ViewGroup container, Bundle savedInstanceState) { View view = inflater.inflate(R.layout.upgrades_fragment, container, false); TextView AirFreshenersCost = (TextView) view.findViewById(R.id.AirFreshenersCost ); if(mainActivity.amountAirFresheners == 5){ AirFreshenersCost.setText("5"); } return view; } LogCat 10-27 22:16:37.333: E/AndroidRuntime(5593): FATAL EXCEPTION: main 10-27 22:16:37.333: E/AndroidRuntime(5593): java.lang.NullPointerException 10-27 22:16:37.333: E/AndroidRuntime(5593): at com.free.dennisg.clickingbad.fragments.UpgradesFragment.onCreateView(UpgradesFragment.java:40) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.Fragment.performCreateView(Fragment.java:1478) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.FragmentManagerImpl.moveToState(FragmentManager.java:927) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.FragmentManagerImpl.moveToState(FragmentManager.java:1104) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.BackStackRecord.run(BackStackRecord.java:682) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.FragmentManagerImpl.execPendingActions(FragmentManager.java:1460) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.FragmentManagerImpl.executePendingTransactions(FragmentManager.java:472) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.app.FragmentPagerAdapter.finishUpdate(FragmentPagerAdapter.java:141) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.view.ViewPager.populate(ViewPager.java:1068) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.view.ViewPager.populate(ViewPager.java:914) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.support.v4.view.ViewPager$3.run(ViewPager.java:244) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.view.Choreographer$CallbackRecord.run(Choreographer.java:749) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.view.Choreographer.doCallbacks(Choreographer.java:562) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.view.Choreographer.doFrame(Choreographer.java:531) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.view.Choreographer$FrameDisplayEventReceiver.run(Choreographer.java:735) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.os.Handler.handleCallback(Handler.java:730) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.os.Handler.dispatchMessage(Handler.java:92) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.os.Looper.loop(Looper.java:137) 10-27 22:16:37.333: E/AndroidRuntime(5593): at android.app.ActivityThread.main(ActivityThread.java:5289) 10-27 22:16:37.333: E/AndroidRuntime(5593): at java.lang.reflect.Method.invokeNative(Native Method) 10-27 22:16:37.333: E/AndroidRuntime(5593): at java.lang.reflect.Method.invoke(Method.java:525) 10-27 22:16:37.333: E/AndroidRuntime(5593): at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:739) 10-27 22:16:37.333: E/AndroidRuntime(5593): at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:555) 10-27 22:16:37.333: E/AndroidRuntime(5593): at dalvik.system.NativeStart.main(Native Method) An example of the code that work public View onCreateView(LayoutInflater inflater, ViewGroup container, Bundle savedInstanceState) { View view = inflater.inflate(R.layout.upgrades_fragment, container, false); TextView AirFreshenersCost = (TextView) view.findViewById(R.id.AirFreshenersCost ); if(((MainActivity)getActivity()).amountAirFresheners == 5){ AirFreshenersCost.setText("5"); } return view; }

    Read the article

  • How do client-server cooperation based games like Diablo 3 work?

    - by edgar
    Diablo 3 cooperates with Blizzard servers even during single player games. In fact, Blizzard has had problems with the games "melting their servers." I would like to ask: How do the client and the server communicate? What details does the client leave to the server, and vice versa? What details are redundant - both the client and the server know - and how often do they disagree? The previous paragraph contains the important questions, but I have a few more that I must explain my motivation towards. I am interested in the programming of botting. Ethical botting - I don't plan on actually abusing the automation to run 24/7. I just find it to be a great programming challenge to glean information from a game, and then make decisions from that information. I am stuck in the starting gate. The unofficial questions from this post would be: How can I make a bot (language, tools, libraries)? Can I get information through the communication between client and server, rather than the brute force pixel detection easily used in more static games? There probably is a trust issue, and to that all I can say is that I promise not to abuse the answers. But please feel free to answer any of the questions you feel comfortable with. Thank you!

    Read the article

  • Getting the Dell XPS 16 Synaptics touchpad to work after hibernation in Ubuntu.

    - by Aron Rotteveel
    I am currently experimenting with using Ubuntu as a my main OS instead of Windows 7. So far, pretty much everything is working fine, except for an issue that I am having with my touchpad. I have a Dell XPS 16 (1640) with a Synaptics touchpad. It works out of the box, but it seems that it stops working after returning from hibernation mode. This problem has also been addressed in an earlier bug, but judging from this bugreport, it should be fixed by now. I am running the 2.6.35-23-generic-pae kernel. Any ideas?

    Read the article

  • Is there a better (i.e vectorised) way to put part of a column name into a row of a data frame in R

    - by PaulHurleyuk
    I have a data frame in R that has come about from running some stats on the result fo a melt/cast operation. I want to add a row into this dataframe containing a Nominal value. That Nominal Value is present in the names for each column df<-as.data.frame(cbind(x=c(1,2,3,4,5),`Var A_100`=c(5,4,3,2,1),`Var B_5`=c(9,8,7,6,5))) > df x Var A_100 Var B_5 1 1 5 9 2 2 4 8 3 3 3 7 4 4 2 6 5 5 1 5 So, I want to create a new row, that contains '100' in the column Var A_100 and '5' in Var B_5. Currently this is what I'm doing but I'm sure there must be a better, vectorised way to do this. temp_nom<-NULL for (l in 1:length(names(df))){ temp_nom[l]<-strsplit(names(df),"_")[[l]][2] } temp_nom [1] NA "100" "5" df[6,]<-temp_nom > df x Var A_100 Var B_5 1 1 5 9 2 2 4 8 3 3 3 7 4 4 2 6 5 5 1 5 6 <NA> 100 5 rm(temp_nom) Typically I'd have 16-24 columns. Any ideas ?

    Read the article

  • My wireless/WiFi connection does not work. What can I do?

    - by Wild Man
    Your situation You have successfully installed Ubuntu. You have just downloaded and booted Ubuntu live media. The latest LTS (see also HWE) or latest non-LTS release are preferred. See the list of Ubuntu releases that are currently supported.) You upgraded your Ubuntu installation to the latest release that the software updater offered you. WiFi worked before, but not now on the new release. You migrated your existing Ubuntu installation to new hardware. Your problem The wireless of your laptop or dekstop is not working. You tried switching the wireless switch off and on and you tried rebooting several times, but you don't see any WiFi access points. You can see your wireless access point, but you cannot establish a connection. You want to analyze the problem, but you don't know where to start or what information you can provide. Related questions I have a hardware detection problem, what logs do I need to look into?

    Read the article

  • What does this snippet work? It uses session cookie [closed]

    - by Emerald214
    My website is a gadget application. It uses CakePHP framework. It has the below snippet in AppController. I searched the cookie variable in the project but can't find any other occurences. I also don't understand what it works. Why does it always check this in every request. if ( isset( $_COOKIE[session_name()] ) ) { if ( !isset( $this->sns_id ) ) { // ????????? $this->cakeError( 'session' ); } } else { // ?????????????????? if ( !isset( $_REQUEST['post_pf_params'] ) ) { $this->cakeError( 'cookie' ); } }

    Read the article

  • Is it common to purchase an insurance policy for contract development work?

    - by Matthew Patrick Cashatt
    I am not sure if this is the best place for the question, but I am not sure where else to ask. Background I am a contract developer and have just been asked to provide a general liability policy for my next gig. In 6-7 years this has never been asked of me. Question Is this common? If so, can anyone recommend a good underwriter that focuses on what we do as contract software developers? I realize that Google could help me find underwriters but it won't give me unbiased public opinion about which companies actually understand what we do and factor that into the price of the policy. Thanks, Matt

    Read the article

  • Why doesn't the "scroll bar" on my laptop touchpad work?

    - by Antonio Guzman Belmont
    EDIT, from Google translate: Well buy my laptop with windows 7 and the touchpad has some arrows up and down to q deslises finger on the right side as if the q scroll the scroll bar handles, recently installed Ubuntu 11.10 and there is also served by the touchpad scroll so to speak but then asked me and I did update (late hours) and longer term attempt q dezlizar my finger down the right side of the touchpad and now low, anyone can tell me I can do about q to q and windows function as and ubuntu 11.10, I think now I have ubuntu 4.12 EDIT.2 (minimal tidy-up) I bought my laptop with Windows 7 installed. The touchpad has up/down arrows on the right side, as if this would scroll. When I installed Ubuntu 11.10, I was able to use this touchpad scroll function. I recently did an update (at some late hours) and i can no longer use right side of the touchpad to scroll. Can anyone tell me how to get the same function as I had in Windows and Ubuntu 11.10 ? I think I now have Ubuntu 12.04 Original question: Bueno compre mi lap con windows 7 y en el touchpad tiene unas flechitas arriba y abajo para q deslises el dedo del lado derecho como si fuese el scroll q maneja la barra de desplazamiento, recientemente instale ubuntu 11.10 y ahi servia tambien el scroll del touchpad por asi decirlo pero luego me pidio actualizar y lo hice(tardo horas) y ya q termino intento dezlizar mi dedo para bajar por la parte derecha del touchpad y ahora no baja, alguien me puede decir q puedo hacer al respecto para q funcione como e windows y en ubuntu 11.10, ahora tengo ubuntu 12.04 creo

    Read the article

  • Web-app currency input/manipulation/calculation with javascript .. there has got to be a better (fra

    - by dreftymac
    BACKGROUND: I am of the "user-input-lockdown" school of thought. Whenever possible, I try to mistrust and sanitize user input, both client side and server side; and I try to take multiple opportunities to restrict possible inputs to a known subset of possibilities, usually this means providing a lot of checkboxes and select lists. (This is from the usability side of things, I know security-wise that malicious users can easily bypass fixed user input GUI controls). PROBLEM: Anyway, the problem always arises with non-fixed input of currency. Whenever I have to accept a freely-specified dollar amount as user input, I always have to confront these problems/annoyances and it is always painful: 1) Make sure to give the user two input boxes for each currency_datapoint, one for the whole_dollar_part and another for the fractional_pennies_part 2) Whenever the user changes a currency_datapoint, provide keystroke-by-keystroke GUI feedback to let them know whether the currency_datapoint is well-formed, with context-appropriate validation rules (e.g., no negatives?, nonzero only?, numeric only!, no non-numeric punctuation! no symbols!) 3) For display purposes, every user-provided currency_datapoint should be translated to human-readable currency formatting (dollar sign, period, commas provided by the app, where appropriate) 4) For calculation purposes, every user-provided currency_datapoint has to be converted to integer (all pennies, to avoid floating point errors) and summed into a grand total with zero or more subtotals. 5) Every user-provided currency_datapoint should be displayed or displayable in a nice "tabular" format, which auto-updates as the user enters each currency_datapoint, including a baloon that warns when one or more currency_datapoints is not well-formed. I seem to be re-inventing this wheel every time I have to work with currency in Javascript on the client side (server side is a bit more flexible since most programming languages have higher-level currency formatting logic). QUESTION: Has anyone out there solved the problem of dealing with the above issues, client side, in a way that is server-side-technology-stack agnostic, (preferrably plain javascript or jquery)? This is getting old, there has to be a better way.

    Read the article

  • How to program something with the expectation that it will work the first time?

    - by Peter Turner
    I had a friend in college who programmed something that worked the first time, that was pretty amazing. But as for me, I just fire up the debugger as soon as I finally get whatever I'm working on to compile - saves me time (kidding of course, I sometimes hold out a little bit of hope or use a lot of premeditated debug strings). What's the best way to approach the Dijkstrain ideal for our programs? -or- Is this just some sort of pie-in-the-sky old fools quest for greatness applicable only to finite tasks that no one should hope for in our professional lives because programming is just too complex?

    Read the article

  • How can I get my meshes to work with Bullet Physics?

    - by Molmasepic
    The problem is that I'm trying to use my meshes with Bullet Physics for the collision part of my game. When I attempted doing this method with my GLM(model loading library by nate robins) model, I get a segmentation fault in the debug, so I figured that it doesnt like the coordinate variables of the model. If i use blender to export my model as a collision file, what type of file should I use? I have heard of a .bullet exporter, but i dont know hot to integrate this python script into my Blender 2.5 program.

    Read the article

  • Python's asyncore to periodically send data using a variable timeout. Is there a better way?

    - by Nick Sonneveld
    I wanted to write a server that a client could connect to and receive periodic updates without having to poll. The problem I have experienced with asyncore is that if you do not return true when dispatcher.writable() is called, you have to wait until after the asyncore.loop has timed out (default is 30s). The two ways I have tried to work around this is 1) reduce timeout to a low value or 2) query connections for when they will next update and generate an adequate timeout value. However if you refer to 'Select Law' in 'man 2 select_tut', it states, "You should always try to use select() without a timeout." Is there a better way to do this? Twisted maybe? I wanted to try and avoid extra threads. I'll include the variable timeout example here: #!/usr/bin/python import time import socket import asyncore # in seconds UPDATE_PERIOD = 4.0 class Channel(asyncore.dispatcher): def __init__(self, sock, sck_map): asyncore.dispatcher.__init__(self, sock=sock, map=sck_map) self.last_update = 0.0 # should update immediately self.send_buf = '' self.recv_buf = '' def writable(self): return len(self.send_buf) > 0 def handle_write(self): nbytes = self.send(self.send_buf) self.send_buf = self.send_buf[nbytes:] def handle_read(self): print 'read' print 'recv:', self.recv(4096) def handle_close(self): print 'close' self.close() # added for variable timeout def update(self): if time.time() >= self.next_update(): self.send_buf += 'hello %f\n'%(time.time()) self.last_update = time.time() def next_update(self): return self.last_update + UPDATE_PERIOD class Server(asyncore.dispatcher): def __init__(self, port, sck_map): asyncore.dispatcher.__init__(self, map=sck_map) self.port = port self.sck_map = sck_map self.create_socket(socket.AF_INET, socket.SOCK_STREAM) self.bind( ("", port)) self.listen(16) print "listening on port", self.port def handle_accept(self): (conn, addr) = self.accept() Channel(sock=conn, sck_map=self.sck_map) # added for variable timeout def update(self): pass def next_update(self): return None sck_map = {} server = Server(9090, sck_map) while True: next_update = time.time() + 30.0 for c in sck_map.values(): c.update() # <-- fill write buffers n = c.next_update() #print 'n:',n if n is not None: next_update = min(next_update, n) _timeout = max(0.1, next_update - time.time()) asyncore.loop(timeout=_timeout, count=1, map=sck_map)

    Read the article

  • Why does nothing work after I randomly changed some file permissions?

    - by Josh B
    Ok so last night i was trying to set permissions to some folders in my File System, since apparently im not an admin on my own computer. And now everything got messed up today and i don't know what to do.. I lost my internet, the icon is not showing in the taskbar anymore. I lost my sound, there is no sound devices listed when i go into the sound menu. I can not log into root anymore, it gives me "sudo: must be setuid root" I can not plug anything in anymore, it will not recognize flash drives or external hard drives. It gives me a Internal Error message everytime i log in It doesn't let me log into the Grub screen anymore on boot up. What did i do? I have a lot of files on here i wish to put on a flash drive but it won't recognize it.

    Read the article

  • Can we put percentage on amount of work of a certain role in project's lifecycle?

    - by deviDave
    The title may be confusing, but I will elaborate it here. I am trying to figure our how much time and effort each person spend during some project. I divided roles into: - junior developer (works mainly on UI and some light things) - senior developer (develops complex logic, database structures, etc.) - lead developer (leads the team, usually most experienced person) - negotiator/resolver (a person who directly talk to a client trying to either negotiate terms and timeframe or to clarify vagueness presented by a team leader) My AIM is to calculate percentage of role's involvement based on quality, not time (obviously a junior will spend most time in project, but with the least quality). In the end I would get a table which may look like this: Total: 100% ---------------- Junior: 10% Senior: 50% Lead: 30% Negotiator: 10% Can this be achieved? Has anyone found any source which may help me?

    Read the article

  • Is ZeroMQ a good choice to make a Python app and a C# managed assembly work together?

    - by Alex Bausk
    I have a task that involves talking to a .NET-based API (namely AutoCAD) to retrieve data, send commands, and react to events. I want to separate the API operations and the proper program logic (largely already implemented in Python) by using natural tools for both: a C# DLL for the former and a Python app for the latter. To connect these two pieces, I began exchanging JSON in ZeroMQ messages. I'm at early development stages but having recently discovered that ZeroMQ does not guarantee message delivery/order, I have reservations about whether this is a feasible way to go. Right now my app is a very basic REQ/REP pair and I plan to handle reacting to events and executing different commands by adding some sort of 'recipient-function' field to my message format. The reason that I want to use ZMQ is that I might be able to scale the software into a larger, multi-user, distributed solution sometime. I am a lay programmer so I would ask for your advice about this architecture. Should I just go ahead with it and plan to deal with message reliability/ordering when problems appear? Should I consider developing some kind of a REST wrapper around ZMQ?

    Read the article

< Previous Page | 190 191 192 193 194 195 196 197 198 199 200 201  | Next Page >