Please refer to the ffmpeg documentation. You will find an option -frames. That option can be use to specify for a given input stream (in the following the stream 0:0 is the 1st input file, first video stream) the number of frame to record. That option can be combined with other options to start somewhere in the input file (time offset, etc ....) ffmpeg -i intput.ts -frames:0:0 100 -vcodec copy test.ts that command demux and remux only the first 100 frame of the video (no re-encoding). as said you can combine it with a jump. Using ' ‘-ss offset (input)’ ' you can specify a "Frame Accurate" position ie. frame 14 after 1min10seconds = 0:1:10:14. that option should be use before the input like below. ffmpeg -ss 00:00:10.0 -i intput.ts -frames:0:0 100 -vcodec copy test.ts ffmpeg dis

I just wrote a blog post that covers encoding_h264_for_ios that shows examples of h.264 encoding that do and do not work on iOS hardware. The specific command line that I use is this: ffmpeg -y -i INPUT.mov -c:v libx264 -pix_fmt yuv420p -preset:v slow -profile:v baseline -crf 23 OUTPUT.m4v If I were you, I would try to encode the videos on the command line and test them on iOS hardware without the web interface. Then, go step by step with specific command line options until you found the one that was breaking the iOS playback.

You probably do not have a keyframe at the 3 second mark. Because non-keyframes encode differences from other frames, they require all of the data starting with the previous keyframe. With the mp4 container it is possible to cut at a non-keyframe without re-encoding using an edit list. In other words, if the closest keyframe before 3s is at 0s then it will copy the video starting at 0s and use an edit list to tell the player to start playing 3 seconds in. If you are using the latest ffmpeg from git master it will do this using an edit list when invoked using the command that you provided. If this is not working for you then you are probably either using an older version of ffmpeg, or your player does not support edit lists. Some players will ignore the edit list and always play all o

You can use concat to append all the videos one by one after converting them to a single format. You can also use the below command to convert differently formatted video to one format: ./ffmpeg -i 1.mp4 -acodec libvo_aacenc -vcodec libx264 -s 1920x1080 -r 60 -strict experimental 1.mp4 Convert everything to mp4 and then follow the instructions given in the link above. This will enable you to join all the videos in a single file.

Do you have root on the device? Mount '/data' and then enter your same 'ffmpeg' command in the shell and see whether the error is the same. Try using the shell to test out different command expressions. Try 'ffmpeg' alone and with just one input file. See whether those commands produce expected output. My wild guess would be that there is an issue with calling 'ffmpeg.main()' that relates to the details of your build.

I understand that this is a crash? If so, the answer is pretty simple: there's a bug in codec, from the name of the DLL I think that's ffmpeg H264 codec. You should try to update it, or just avoid it.

You've tackled this problem in a way that makes it very hard to debug. I'd recommend switching things around so you can more easily check each step. (Using informative variable names also helps!) The code is unlikely to work as is, but it should be much easier to work iteratively, checking that each step has succeeded before continuing to the next. paths <- dir("C:/Data/", pattern = "\.csv$") # Read in CSV files as data frames raw <- lapply(paths, read.csv, str) # Extract needed rows filter_metrics <- c("MaxTemp", "AvgTemp", "MinTemp") filter_years <- 1970:2000 filtered <- lapply(raw, subset, !is.na(Value) & Year %in% filter_years & Element %in% filter_metrics) # Drop any empty data frames rows <- vapply(filtered, nrow, integer(1)) filtered <- filtered[r

You're using the correct Method but have you passed any code in the "// do some tasks here when playback completes " AREA. If not then the callback wont make any sense. mMediaPlayer.setOnCompletionListener(new OnCompletionListener() { public void onCompletion(MediaPlayer mp) { Toast.makeText(getApplicationContext(),"Playback Completes", Toast.LENGTH_SHORT).show(); } }); This method will pop a Toast on completion of playback.

Not 100% sure about all your query but sounds like you might need to UNION the two sub-queries into one, then you can sum in the SELECT clause... SELECT sum(T) AS d, Y.ApplicantId, Y.Applicantname FROM ( SELECT tblexperience.ApplicantId, SUM(tblExpeRange.score)as T FROM tblexperience LEFT OUTER JOIN tblExpeRange ON tblexperience.ExpRange=tblExpeRange.expeRange GROUP BY tblexperience.ApplicantId UNION SELECT tblAcademicInfo.ApplicantId, SUM(tblGradPoint.score) AS T FROM tblAcademicInfo LEFT OUTER JOIN tblGradPoint ON tblAcademicInfo.cgpa=tblGradPoint.[cgpa/division] GROUP BY tblAcademicInfo.ApplicantId

It seems that your password field isn't actually a varbinary, despite looking like one. Run these for an example: SELECT CAST('0x3E6AFF88BEB29D9B234B9E5A5AD329D8D3F33200EC4EE02749C56AD58D040976' AS varbinary(max)) SELECT CAST(CAST('0x3E6AFF88BEB29D9B234B9E5A5AD329D8D3F33200EC4EE02749C56AD58D040976' AS varbinary(max))AS VARCHAR(MAX)) SELECT CAST('dog' AS VARBINARY(MAX)) SELECT CAST(CAST('dog' AS VARBINARY(max))AS VARCHAR(MAX)) You see that your result is the equivalent of your initial string, just in varbinary format.

Your options include: 1). Ensuring autovacuum is enabled and set aggressively. 2). Recreating the table as I mentioned in an earlier comment (create-table-as-select + truncate + reload the original table). 3). Running CLUSTER on the table if you can afford to be locked out of that table (exclusive lock). 4). VACUUM FULL, though CLUSTER is more efficient and recommended. 5). Running a plain VACUUM ANALYZE a few times and leaving the table as-is, to eventually fill the space back up as new data comes in. 6). Dump and reload the table via pg_dump 7). pg_repack (though I haven't used it in production)

From Parthik Gosar's link in this comment with slight modification to let you enter each value independently or all at once as an object /* accepts parameters * h Object = {h:x, s:y, v:z} * OR * h, s, v */ function HSVtoRGB(h, s, v) { var r, g, b, i, f, p, q, t; if (arguments.length === 1) { s = h.s, v = h.v, h = h.h; } i = Math.floor(h * 6); f = h * 6 - i; p = v * (1 - s); q = v * (1 - f * s); t = v * (1 - (1 - f) * s); switch (i % 6) { case 0: r = v, g = t, b = p; break; case 1: r = q, g = v, b = p; break; case 2: r = p, g = v, b = t; break; case 3: r = p, g = q, b = v; break; case 4: r = t, g = p, b = v; break; case 5: r = v, g = p, b = q; break; } return { r: Math.round(r *

I'm going to make an assumption that you're using Apache. It should be possible to parse the Apache logs to extract connections from unique IP addresses. Hopefully the URL requested by the client will contain some sort of path indicating that it was directed from Facebook. Link: Get unique visitors from apache log file

Assuming you havent yet, try playing with the motiongraphs sample application (in xcode). It'll give you a better understanding of how you might use a combination of "devicemotion" output such as roll, pitch, yaw; as well as gyroscope readings. Relying on accelerometer readings alone is going to be tricky for your application idea (but maybe not impossible.) Accelerometer gives you a measure of G-force, so while there is a detectable difference between "flat" and "not so flat," there may be too much additional noise from other factors such as hand movement or variance between legacy device types to figure out a reliable algorithm.

It's easy to see that some inputs would produce outputs that cannot be represented by the return type int64_t. For example, fn(INT64_MAX, 2, 1, 1). However, the following approach will should allow you to return a correct answer for any combination of inputs that does in fact fit within the range of int64_t. int64_t fn(int64_t a, int32_t b, int32_t c, int32_t d) { /* find the integer and remainder portions of a/d */ int64_t leftI = a / d; int64_t leftR = a % d; /* multiply the integer portion of the result by b and c */ int64_t resultI = leftI * b * c; /* multiply the remainder portion by b */ int64_t resultR = leftR * b; resultI = resultI + (resultR / d) * c; /* multiply the remainder portion by c */ resultR = (resultR % d) * c; return resu

So you're depending on sleep being accurate. It isn't. It isn't designed to be. It will sleep for at least that time, but may sleep for more. The OS will not do a task switch back to that thread as soon as the sleep is up, but as soon as it becomes the top priority thread in its scheduling queue. You need a far more reliable way of waiting- something based on an RTC. An alarm based on the RTC alarm type might do it. I'm not sure if it actually is an RTC interrupt that will send you an immediate alarm, or if its just the relative clock it uses.

I think you mean a replicaset with 3 nodes instead of "3 replica set". If you are still experiencing the same issue. Here is my opinion: Since you are running your server in linode.com. Your server is actually a virtual machine and you are sharing resources with others. The periodic slow down may be due to others running having disk load periodically. Since you have looked into so many different possibilities already, this may be an option for you even it takes a bit of effort. This is definitely caused by a job ran by mongodb or your system. Please try to look for any job that runs regularly. For example, try to remove the 3600 seconds delay on one of your secondary. Even that is not 2 hour and 10 minutes, but that may be a trigger of it. I can't post my suggestions in the comment si

From Apple docs Discussion When a view’s bounds change, that view automatically resizes its subviews according to each subview’s autoresizing mask. You specify the value of this mask by combining the constants described in UIViewAutoresizing using the C bitwise OR operator. Combining these constants lets you specify which dimensions of the view should grow or shrink relative to the superview. The default value of this property is UIViewAutoresizingNone, which indicates that the view should not be resized at all. When more than one option along the same axis is set, the default behavior is to distribute the size difference proportionally among the flexible portions. The larger the flexible portion, relative to the other flexible portions, the more it is lik

The simplest way to avoid problems with floating point precision is to cast your number to a string and then only print what's after the point. EDIT I'm not sure how to do it with arduino's version of c++ so feel free to edit my answer to add an example everyone!

I hope this will help you get toward your goal. I made a regex, that matches what the browser is outputting (the rgb is just the hex as an rgb). It may need to be updated since I didn't test it in all browsers, but I think it converts it to rgb so if there is an alpha, that can be handled easily. Here's the fiddle: http://jsfiddle.net/HQ7NF/2/ And here's the regex I am using: var reBoxShadow = /(?:rgb((d+), ?(d+), ?(d+))([^,]*))+/g; while ( style = reBoxShadow.exex(element.style.boxShadow) ) { /* this is where stuff happens */ } I then split the color and the args into separate variables, so you can decide what to do with them. I hope that can at least get you on the right track.

You can use force from Control.DeepSeq to fully evaluate a data structure (and thus demand and measure its computation). One problem is that forcing a large data structure takes some time itself! This is because a deepseq (used by force) will walk down your algebraic data type tree, visiting every node (but not doing anything with it). When you perform only a cheap operation to each node, such as map (*2) mylist, and try to measure how long it takes, this overhead can suddenly become significant, messing up your measurements. import Control.DeepSeq import Control.Exception (evaluate) import Data.Time (diffUTCTime, getCurrentTime) -- | Measures how long a computation takes, printing both the time and the -- overhead of `force` to stdout. So it forces *twice*. benchmarkForce :: NFData

Are the "incorrect values" within a few dozen meters of the original location? If so, what you might be seeing is just the inaccuracy of GPS. GPS isn't 100% accurate. Even if your device isn't moving, the location where the device's GPS thinks it is can change. In fact, your calculated location might not be based on true GPS at all, but on Wi-Fi or on nearby cell towers (A-GPS), both of which are much less accurate. See this StackOverflow thread for more information. For debugging purposes, I suggest displaying the HorizontalAccuracy somewhere on your UI to see if the GPS drift is in line with expectations.

Once the process terminates, even though the process entry still exists in the kernel, all of the resources for the process are released by the kernel. So, I would not expect to be able to read the memory size of the process from /proc. In addition, there's no "maximum memory used" statistic for the process. The only ways I can think to determine the maximum amount of memory for a process are the following: Periodically sample the process state while it's executing and grab the current memory usage, Use strace to capture all of the system calls made by the process and process this to determine just how much memory the process used, or Use a shared library to intercept the memory-allocation system calls as they are made by the process, capture the details, and find a way to report I

We have recently encountered a similar problem as yours. We found out that file_get_contents() eventually calls getaddrinfo(), which implements some sorting algorithms from RFC3484. The piece of snippet from php source code PHPAPI int php_network_getaddresses(const char *host, int socktype, struct sockaddr ***sal, zend_string **error_string) { // skip... if ((n = getaddrinfo(host, NULL, &hints, &res))) { if (error_string) { *error_string = strpprintf(0, "php_network_getaddresses: getaddrinfo failed: %s", PHP_GAI_STRERROR(n)); php_error_docref(NULL, E_WARNING, "%s", ZSTR_VAL(*error_string)); } else { php_error_docref(NULL, E_WARNING, "php_network_getaddresses: getaddrinfo failed: %s", PHP_GAI_STRERROR(n)); }

You should have added the error parameter when copying the file and checked what it contained... Before you copy the file, ensure the destination directory exists using createDirectoryAtPath:withIntermediateDirectories:attributes:error:.

The problem is that the number of colors is so small that the threshold to be colored is not being met by most of pairs which have only "1" as an entry. to tlook at the distribution of counts try: table(test_matrix) test_matrix 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 9434 2016 1 1 1 1 1 1 1 1 1 1 1 1 1 1 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 1

The only workaround that occurs to me would involve applying -webkit-transform: scale(x), example: div[style*="10."] {-webkit-transform-origin: 0 50%;} div[style*="10.1"] { -webkit-transform: scale(1.01);} div[style*="10.2"] { -webkit-transform: scale(1.02);} div[style*="10.3"] { -webkit-transform: scale(1.03);} div[style*="10.4"] { -webkit-transform: scale(1.04);} div[style*="10.5"] { -webkit-transform: scale(1.05);} div[style*="10.6"] { -webkit-transform: scale(0.96);} div[style*="10.7"] { -webkit-transform: scale(0.97);} div[style*="10.8"] { -webkit-transform: scale(0.98);} div[style*="10.9"] { -webkit-transform: scale(0.99);} The problem, I guess, is that since you are auto-fitting text you are probably getting element.offsetWidth which will give you the width before the scaling. T

Change delay = ((60/HR)/2)-(0.5*pulse_width); into delay = 30/HR; tripuls does not change anything to the time input t1, so the pulse width should not be subtracted from the time vector. You can see this is correct by setting pulse_width = 60e-4; % (try) pulse_width = 60e-10; % (try again) You should see that your results slowly go more and more towards the correct HR (provided your external equipment is capable of handling such short pulses).

Simply add the style white-space: normal; to the p. You have set white-space: nowrap; on the parent table element, and the p inherited that. white-space: normal undoes that style.

Personaly I would use Screen pixel density (dpi) for each layout. That is: ldpi: Low-density screens; approximately 120dpi. mdpi: Medium-density (on traditional HVGA) screens; approximately 160dpi. hdpi: High-density screens; approximately 240dpi. xhdpi: Extra high-density screens; approximately 320dpi. Added in API Level 8 See http://developer.android.com/guide/topics/resources/providing-resources.html#AlternativeResources

You should run supervisorctl with -c as well. From the documentation (my emphasis): The Supervisor configuration file is conventionally named supervisord.conf. It is used by both supervisord and supervisorctl. If either application is started without the -c option (the option which is used to tell the application the configuration filename explicitly), the application will look for a file named supervisord.conf within the following locations, in the specified order. It will use the first file it finds. $CWD/supervisord.conf $CWD/etc/supervisord.conf /etc/supervisord.conf

Your table is close before including the <input> and the <script> tags. You have a rogue </td> and </tr>. Move </table> after the last </tr>.

You forgot to position the header - add position: relative; to the <h2> http://jsfiddle.net/Ms7Qr/1/ (z-index doesn't apply to a statically positioned elements) From the Visual Formatting Model: Other stacking contexts are generated by any positioned element (including relatively positioned elements) having a computed value of 'z-index' other than 'auto'.

Try adding wmode, it seems to have two parameters. &wmode=Opaque &wmode=transparent <iframe title="YouTube video player" width="480" height="390" src="http://www.youtube.com/embed/lzQgAR_J1PI?wmode=transparent" frameborder="0" wmode="Opaque"> For more info check YouTube Video Embedded via iframe Ignoring z-index?

Multipart data is handled by PHP only on POST requests. With PUT you have direct access to bytestream and you need to handle it for yourself. If something changes, this bug report / feature request will be closed

Just modify your call to boot() like this: data.boot <- boot(data, boot.huber, 1999, strata=data$Subject) ?boot provides this description of the strata= argument, which does exactly what you are asking for: strata: An integer vector or factor specifying the strata for multi-sample problems. This may be specified for any simulation, but is ignored when ‘sim = "parametric"’. When ‘strata’ is supplied for a nonparametric bootstrap, the simulations are done within the specified strata. Additional note: To confirm that it's working as you'd like, you can call debugonce(boot), run the call above, and step through the debugger until the object i (whose rows contain the indices used to resample rows of data to create each bootstr

My suggestion would be to set the delay to 0, and place the subsequent UIView animation blocks in the completion block of the previous animateWithDuration statements: [UIView animateWithDuration:0.4 delay:0.0 options:UIViewAnimationOptionCurveEaseInOut animations:^(void) { [_IntroLabel1 setAlpha:1]; } completion:^(BOOL finished){ [UIView animateWithDuration:0.4 delay:2.6 options:UIViewAnimationOptionCurveEaseInOut animations:^(void) { [_IntroLabel2 setAlpha:1]; [_IntroLabel1 setAlpha:0]; [_IntroLabel1 setCenter:CGPointMake(0, _IntroLabe

body { max-width: 300px; background-image: url("data:image/png;base64,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"); backgroun

you must use an xml resource for that and place your src into the xml <?xml version="1.0" encoding="utf-8"?> <nine-patch xmlns:android="http://schemas.android.com/apk/res/android" android:src="@drawable/bg" > </nine-patch> and your bg should be like bg.9.png

This tends to stump people a lot - apparently the roadmap for MVC 5 includes more intuitive DropDownList helpers. When you do: @Html.DropDownListFor(x => x.TimeOption ...) ... the HTML helper will do everything in its power not to use the value you set to Selected = true in your SelectList. It will: Try to get it from ModelState - which it won't get on the GET request Try to get it from ViewData - which it will get, because you have a TimeOption property on your model. Only if both of those attempts fail, will it succumb, and use what you asked it to use. So, in your case, it will use the current value of TimeOption on your model as the selected value. Which should be fine, because that's what you asked for in your SelectList anyway ((int)Model.TimeOption), right? Wrong. Becau