Remember

gitlab sends notification emails email address used during registration. can change target email address gitlab uses send email notifications to? i've tried setting/changing email address on "profile settings" it's used display; doesn't change gitlab sends notification emails to. ok seems figured out causing issue. as turns out "email" property on "profile settings" does specify gitlab sends notification emails. problem bit more elaborate. there 2 gitlab accounts had same username! both belonged me; 1 regular user , other admin. 2 users had different email addresses; usernames same. this caused kind of anomaly; though action triggered admin user notification email went email address of other user; strange indeed. i changed username of 1 of users make them unique. after that, worked well. hope in future.

i need open sqlite file , read data file show in table. far have button trigger nsopenpanel(), , path. after tried open file have encoding problem. below code in appdelegate.swift @ibaction func importfromkobo(sender: anyobject) { var openpanel = nsopenpanel() let filedb = "kobodbtest.sqlite" openpanel.allowsmultipleselection = false openpanel.canchoosedirectories = true openpanel.cancreatedirectories = true openpanel.canchoosefiles = false openpanel.beginwithcompletionhandler { (result) -> void in if result == nsfilehandlingpanelokbutton { //do //if there's 1 url, surely 'openpanel.url' var deviceurl = openpanel.url! println("path: \(deviceurl)") println(self.resultdev.stringvalue) self.resultdev.stringvalue += "\npath: \(deviceurl)" self.resultinput.stringvalue += "\npath: \(deviceurl)" var fu...

the default spree commerce has "categories" , "brand" taxonomies. have generated extension creates custom testing page taxonomies displayed, in reverse order. practicing purposes. the custom testing page "/testing" seems routing correctly , using view testing.html.erb , can not load taxonomies instance defined in method. gives me undefined method each nil:nilclass . note: i'm working in extension only, not in spree app. spree_custom_extension/controllers/spree/home_controller_extension.rb module spree homecontroller.class_eval def testing @taxonomies = spree::taxonomy.includes(root: :children).reverse_order end end end spree_custom_extension/views/spree/home/testing.html.erb <% @taxonomies.each |taxonomy| %> <%= taxonomy.name %><br> <% end %> spree_custom_extension/config/routes.rb spree::core::engine.routes.draw '/testing' => 'home#testing' end

we developing application on our ios, inside our html page, there "email results" button. when click on button screen scrolls immediately. html: <div class="result-button-wrapper"> <a target="_blank" href="" class="button button-email"><span class="icon"></span>email results</a> </div> css: .button { display:inline-block; padding:12px; position:relative; margin:12px; cursor:pointer; color:#fff !important; font-size:16px; font-weight:bold; text-decoration:none; text-shadow:0px 1px 2px #222222; text-align:center;} .button.button-email { display:block; width:130px; padding-left:28px; margin:10px auto 10px auto;} if removed "button" class like: <a target="_blank" href="" class="button-email"><span class="icon"></span>email results</a> , button works , can click it where issue ...

i have simple spring boot 1.2.2 web application have created logback.xml configuration file: <configuration debug="true" scan="true" scanperiod="33 seconds"> <appender name="console" class="ch.qos.logback.core.consoleappender"> <encoder> <pattern>%d{hh:mm:ss.sss} [%thread] %-5level %logger{36} - %msg%n</pattern> </encoder> </appender> <!-- <appender name="logfile" class="ch.qos.logback.core.fileappender"> <file>myapp.log</file> <append>true</append> <encoder> <pattern>%d{hh:mm:ss.sss %-5level [%thread] %logger{10} [%file:%line] %msg%n</pattern> </encoder> </appender> --> <logger name="com.myapp.logtest.samplecontroller" level="debug"/> <logger name="com.myapp.logtest" level="warn"> </logge...

this question has answer here: where , why have put “template” , “typename” keywords? 5 answers i don't understand why code #include <map> template<typename t, typename u> std::ostream& operator<<(std::ostream& o, const std::map<t,u>& input) { (std::map<typename t,typename u>::iterator it=input.begin(); it!=input.end(); ++it) { o << it->first << " => " << it->second << '\n'; } return o; } returns compilation error: error: wrong number of template arguments (1, should 4) error: provided ‘template<class _key, class _tp, class _compare, class _alloc> class std::map’ can me please?? you should write typename before iterator declaration , use const_iterator: for (typename std::map<t,u>::const_iterator it=input.be...

i using truezip compression. here code looks public string compress() throws ioexception { if (loglocations.isempty()) { throw new illegalstateexception("no logs provided compress"); } removedestinationifexists(desiredarchive); final tfile destinationarchive = new tfile(desiredarchive + "/diagnostics"); (final string loglocation : loglocations) { final tfile log = new tfile(loglocation); if (!log.exists()) { logger.debug("{} not exist, ignoring."); continue; } if (log.isdirectory()) { log.cp_r(destinationarchive); } else { final string newloglocation = new tfile(destinationarchive.getabsolutepath()) + slash + getlognamefrompath(loglocation); log.cp(new tfile(newloglocation)); } } return destinationarchive.getenclarchive().getabsolutepath(); } and test @test public void testbenchmarkwithhprof() throws ioex...

i'm facing bizarre issue when trying scale , crop image using html5 canvas api across browsers. specifically, i'm working trying find center of image , crop square based on width , height using these parameters specified context api context.drawimage(img,sx,sy,swidth,sheight,x,y,width,height); here's example of code i'm using: <!doctype html> <html> <body> <p>image use:</p> <img id="scream" width="220" height="277" src="img_the_scream.jpg" alt="the scream"> <p>canvas:</p> <canvas id="mycanvas" width="150" height="150" style="border:1px solid #d3d3d3;"> browser not support html5 canvas tag. </canvas> <script> window.onload = function() { var c = document.getelementbyid("mycanvas"); var ctx = c.getcontext("2d"); var img = document.getelementbyid("scream"); c...

i need assistance setting solution in visual studio uses umbraco. i'm planning on developing umbraco cms platform in visual studio 2013, using c# , bootstrap. source control done team foundation server. i understand need create empty asp.net mvc application, create c# class, use nuget install umbraco on development server. need else set environment? and bootstrap fit this? first of - umbraco content management system. provides back-office managing content of system. umbraco have project template can download inside visual studio. project contains needed structure , packages in it. here couple of useful links: umbraco template project article called nuget packages , visual studio templates so now, when have umbraco installed , have back-office - need display content somehow end users. here place can use bootstrap apply styling content.

i have following command draw histogram in ordered manner. so code follows: ggplot(upstream, aes(x=type, y=round(..count../sum(..count..) * 100, 2))) + geom_histogram(fill= "red", color = "red") + xlab ("vehicle type") + ylab("percentage of vehicles in category (%)") + ggtitle ("percentage of upstream vehicles type") + stat_bin(geom="text", aes(label=round(..count../sum(..count..) * 100, 2)), vjust=-0.5) the output is: i arrange bars in ordered manner, use reorder() function in aes , gives me following problem: stat_bin requires following missing aesthetics x how can use reorder without getting error? couldn't seem able figure out posted solutions. thanks suggestions in advance. edit 1: fixed looking based on joran's suggestion geom_bar() follows in case needs it: # reorder factor trying plot on x-side (descending manner) upstream$type <- with(upstream, reorder(type, type, functio...

i have 2 time periods of interest , 4 observation points(0 months, 4 months, 12 months, 16 months) subjects. first time period of interest between observation 1 , observation 3. second time period of interest between observation 3 , observation 4. i run hlm account correlation of observations on same subject. have pasted sample data , code , output below. when compare model output actual means similar in case. however, when use actual data set less similar. imply? can tell me if have coded time appropriately? goal compare effect of treatment during time period 1 effect of treatment during time period 2. thank you! library(nlme) #run model , output model=lme(response~time1*treatment+time2*treatment, random=~time1+time2|subject,data=test,control=list(opt="optim")) round(summary(model)$ttable,dig=3) # output value std.error df t-value p-value (intercept) 172.357 2.390 41 72.110 0.000 time1 0.464 0.062 41 7.496 0....

i attempting union 3 different pipe delimited tables 1 can export 1 text file. each pipe delimited table has different structure. here have far. when run union query sorting query 2, query 1, query 3; sure because of output of each query doing final sort on abc order. there way sort order of query? select qry_header.hdr qry_header union select qry_ghd_headers.ghd_header qry_ghd_headers inner join [coverage id] on qry_ghd_headers. [coverage id] = [coverage id].[coverage id] [coverage id].id = 1 union select [qry_pipe_delim].pipe_delim qry_pipe_delim inner join [coverage id] on qry_pipe_delim. [coverage id] = [coverage id].[coverage id] [coverage id].id = 1 output looks this: ghd|ca015015hl|tm|f|20150414 hdr|10|p00000000020123|axqen16e88|rxhub|optumrx|cp000005|20150415|104818|frm|u|20150414|p tdt|a|ca015015hl|00008112312|003|||||this medication increases risk of serious toxicity in patients 65 , older. use safer alternatives.|national guide...

i have numpy.poly1d polynomial follows: c = np.poly1d([2,-4,-28,62,122,-256,-196,140,392,240,72]) the curve looks when graphed in range -2.5 <= x <= 2.5 : how can find minimum point(s) of curve within given range, without using discrete values used graph curve (by mean using continuous poly1d object)? ok bit different functions @matiasg aim make more copyable code , use vectorized code possible. import numpy np matplotlib.pyplot import * c = np.poly1d([2,-4,-28,62,122,-256,-196,140,392,240,72]) crit = c.deriv().r r_crit = crit[crit.imag==0].real test = c.deriv(2)(r_crit) # compute local minima # excluding range boundaries x_min = r_crit[test>0] y_min = c(x_min) plot( x_min, y_min, 'o' ) xc = np.arange(-2.5, 2.6, 0.02) yc = c(xc) plot( xc, yc) xlim([-2.5,2.5]) show() image 1 : result. note there local minima outside bounds ;)

i'm trying dynamically generate button in angular. on click, button needs call deleterow() function , pass in username. applicable username passed controller , resulting button code appears generated. however, button ends passing undefined deleterow() function. problem how i'm using $compile ? validationapp.controller('userscontroller', function ($scope, $compile, $http, $timeout){ $(document).on("click", ".open-confirm-dialog", function () { var username = $(this).data('id'); var btnhtml = '<button class="btn btn-danger" data-title="delete" ng-click="deleterow(' + username + ')">delete</button>'; var temp = $compile(btnhtml)($scope); angular.element(document.getelementbyid('deletebutton-dynamic')).append(temp); }); $scope.deleterow = function(username){ alert(username); //this shows 'undefined' in pop-up ...

im building exercise game based off of dragon ball z fun. page type in numbers , record progress. live at: http://surrealmayhem.com/dbzgame/record_progress.html now, ive got 99% of page working far. im trying accomplish take value in right-hand "totals" column , have add in long progress bar @ bottom of page. each individual progress bar in totals column works. reason cant figure out progress bar @ bottom wont work. im still beginner suspect syntax. i post entire code since must screwing somewhere. <script> function balanceexp(){ var arr = document.getelementsbyname('balexp'); var tot=0; for(var z=0;z<arr.length;z++){ if(parseint(arr[z].value)) tot += parseint(arr[z].value); } document.getelementbyid('balancetotal').value = tot; } </script> <script> function powerexp(){ var arr = document.getelementsbyname('powexp'); var tot=0; for(var y=0;y<arr.length;y++){ ...

i trying bind visibility property of grid column using boolean visibility converter. grid visible though bool value false. please me fix issue. code works fine datagrid not grid .here code. <window x:class="myapp.firstwindow" xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation" xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml" xmlns:viewmodel="clr-namespace:myapp.viewmodel" xmlns:i="clr-namespace:system.windows.interactivity; assembly=system.windows.interactivity" xmlns:y="clr-namespace:myapp.framework.converter" xmlns:cmd="clr-namespace:galasoft.mvvmlight.command;assembly=galasoft.mvvmlight.extras.wpf4" datacontext="{binding myappviewmodel, source={staticresource locator}}" minheight="530" minwidth="275" height="530" width="275" title="{binding title}" > <window.resources> ...

i want make app searchable in google quick search box. followed examples still doesn't work. official doc isn't clear @ how make case work. now after checking these stackoverflow issues, start wondering if feature still available developer? is global search in android still available developer? how app appear in google now's phone search list? i see few apps still make "phone search" list in quick search box settings. if so, can shed light me? searchable config xml follows <searchable xmlns:android="http://schemas.android.com/apk/res/android" android:label="@string/app_name" android:hint="@string/search_hint" android:searchsuggestauthority="@string/search_authority" android:searchsuggestintentaction="android.intent.action.view" android:searchsuggestintentdata="content://some_search_authority" android:includeinglobalsearch="true"> </searchable> ...

i have following problem rotating ppm image right first 2 lines in result image black (or color rainbow) here's code sets buffer image (the variables g_width , g_height set function) struct pixel *image = malloc(sizeof(struct pixel) * g_width * g_height); here's function pointer passed it void rotate90(struct pixel *img) { int i, j, size, th; size = sizeof(struct pixel) * g_width * g_height; struct pixel *buffer = malloc(size); if (buffer == null) { fprintf(stderr, "unable allocate memory\n"); exit(exit_failure); } (i = 0; < g_height; i++) { (j=0; j < g_width; j++) { buffer[(g_height*j)+(g_height-i)] = img[(g_width*i) + j]; } } //copy buffer image pointer memcpy(img, buffer, size); //free buffer , swap width , height around free(buffer); th = g_height; g_height = g_width; g_width = th; } if print image buffer comes out fine, if rotate it comes ou...

"subquery returned more 1 value. not permitted when subquery follows =, !=, <, <= , >, >= or when subquery used expression." i got wired error message while debugging following code: when not exists(select 1 defs with(nolock) defaultname = @default_currency , defaultvalue= @currency) , not exists( select 1 supp with(nolock) pint = @id , currency= @currency) "not valid." i couldn't find wrong code regarding error message. moreover, error message disappear when changed 'and' 'or' . when not exists(select 1 defs with(nolock) defaultname = @default_currency , defaultvalue= @currency) or not exists( select 1 supp with(nolock) pint = @id ...

in c header file have: long test_api callandsave( ___out_ char param1[31], ___out_ char param2[5], ___out_ char param3[21], ___out_ char* pointerparam ); i trying set , pass parameters in python , pass them callandsave : import ctypes ctypes import * lib = cdll.loadlibrary('./testlib.so') param1 = (ctypes.c_char*31)() param2 = (ctypes.c_char*5)() param3 = (ctypes.c_char*21)() pointerparam = ctypes.c_char_p() lib.efttrx(param1,param2,param3,pointerparam) but getting " output parameter null " error pointerparam . how should define " ___out_ char* pointerparam " in python correctly? ___out_ char* pointerparam the ___out_ in specification means pointer have data written it. must allocate buffer write to. ctypes.c_char_p() doesn't make buffer - it's null pointer. look ctypes.create_string_buffer . must figure out how big of buffer callandsave expects.

i have modified version of github project https://github.com/lkwdwrd/git-deploy [i modified reason json wasnt being parsed on server (mediatemple grid server) & added custom logging] script autodeploys github repos using git hooks. the script working fine while stopped working. i've isolated problem git fetch command. exact code i'm using: exec( 'git fetch ' . $this->_remote . ' ' . $this->_branch, $fetch_output, $fetch_return_var ); the $fetch_output array blank , $fetch_return_var 255. believe 255 means -1, error git command. however able execute command when ssh'd server , running same user exact same string being build 1st argument of exec() function. so i'm lost. is there effective difference between running commands manually via ssh , using exec()? is there way can actual git fetch error being returned exec()? could kind of server configuration has been changed host , should contact them for? thanks in advance ...

apparently microsoft has added share intent support onenote android app: http://www.xda-developers.com/massive-facelift-to-microsoft-onenote-for-android-brings-share-intent-and-multi-window-support/ how can use share intent on android launch onenote app? you can not directly launch onenote through share intent. behaviour occurs is: content sent onenote app page created title matching extra_subject , body sent in extra_text . low priority notification indicating page has been created fires. target set open onenote view page activity. i suppose if asked notifications permission intercept notification , open onenote @ point. if want open onenote, why not use like: string onpackagename = "com.microsoft.office.onenote"; intent launchintent = getpackagemanager(). getlaunchintentforpackage(onpackagename); startactivity(launchintent);

hi guys looked @ topic jsf chrome save password on login form similar problem. solution doesn't work in case because don't use primefaces. form looks following: <h:form id="loginform" class="form-signin " > <h1 class="form-signin-heading "> <span class="glyphicon glyphicon-user"/> login form</h1> <h:inputtext id="username" value="#{loginhandler.inputusername}" styleclass="form-control" pt:placeholder="username" required="true" requiredmessage="please type in username" /> <h:message style="color: red" for="username"/> <br></br> <h:inputsecret id="password" value="#{loginhandler.inputpassword}" class="form-control" pt:placeholder="password" required="true" pt:type="password" ...

i working on web scraping project , follower data public twitter account. find request url can't seem find on twitter. wondering how can list of followers. using python 3 this page describes how it. you'll need install tweepy , use following code: import tweepy import time #insert twitter keys here consumer_key ='xxxxxxx' consumer_secret='xxxxxxx' access_token='xxxxxxx' access_secret='xxxxxxx' twitter_handle='handle' auth = tweepy.auth.oauthhandler(consumer_key, consumer_secret) auth.set_access_token(access_token, access_secret) api = tweepy.api(auth) list= open('twitter_followers.txt','w') if(api.verify_credentials): print 'we logged in' user = tweepy.cursor(api.followers, screen_name=twitter_handle).items() while true: try: u = next(user) list.write(u.screen_name +' \n') except: time.sleep(15*60) print 'we got timeout ... sleeping 1...

i'm trying add instances following enummap: class activeg<d extends gdata, t extends gtemplate> { enummap<gstatetype, gstate<activeg<d,t>> map; .. .. } class tgdata extends gdata {...} class tgtemplate extends gtemplate {....} class activetg extends activeg<tgdata, tgtemplate> { // fails compilation error super.states.put(gstatetype.waiting, new tgwaitingstate<activetg>(this)); ... ... } the error get: the method put(gstatetype, gstate<activeg<tgdata,tgtemplate>>) in type enummap<gstatetype,gstate<activeg<tgdata,tgtemplate>>> not applicable arguments (gstatetype, tgwaitingstate<activetg>) can try me figure out what's missing make work?? thanks! in brief, generics in java not covariance. e.g. if have class parent {...} class child extends parent {...} foo<child> not foo<parent> go code: your enum map in activetg expecting gstate<...

trying understand using joins hql. def result=food.findall('from mealitems m, nutrition n m.food=n.food); the result returns, i'm not understanding how/what returns. can access nutrition object on each result? for(row in result){ var ntr=row.nutrition; println ntr.calories; } suggest should read on hibernate. it transforms result set list of object arrays. utimately if able map result corresponding classes so. have @ these links. should how hibernate works hql query if transformer not explicit declared http://levelup.lishman.com/hibernate/hql/joins.php

i trying apply ordered bayer dithering algo on image in openframeworks , not able result. wikipedia link : http://en.wikipedia.org/wiki/ordered_dithering the algo :-- foreach y foreach x oldpixel := pixel[x][y] + (pixel[x][y] * threshold_map_4x4[x mod 4][y mod 4]) newpixel := find_closest_palette_color(oldpixel) pixel[x][y] := newpixel my code :-- void testapp::setup(){ int factor = 0; float array[4][4] = {{1,9,3,11},{13,5,15,7},{4,12,2,10},{16,8,14,6}}; for(int = 0 ; i<4;i++){ for(int j = 0 ; j<4;j++){ array[i][j] = array[i][j]/17; } } img1.loadimage("hello.jpg"); img1.resize(999,999); img1.update(); img2 = img1; pix = img1.getpixels(); pix2 = img2.getpixels(); for(int k = 0 ; k < 999 ; k++){ for(int j = 0 ; j < 999*3 ; j+=3){ factor = array[k%4][j%4]; pix2...

i have image in public/images folder.i want show image in webpage , trying this {{ html::image('public/images/20140208-img_2538.jpg') }} but cant see image on webpage.instead shows broken image icon.what can can suggest? the html::image() expects path relative public . try this: {{ html::image('images/20140208-img_2538.jpg') }}

this question has answer here: using cpan proxy failing after o conf init /proxy/ 3 answers after solving few issues, i've succeeded cpan on work windows 7 machine dwimperl via perl -mcpan -e shell , o conf init /proxy/ ftp: http: http://192.168.1.1:8080 no: username: chris@work password: superdedupersecretpassword cpan works. i'm able install/build. it seems no matter do, cpanm doesn't want past proxy: c:\>cpanm www::mechanize ! finding www::mechanize on cpanmetadb failed. 501 protocol scheme 'wwwproxy.work.com' not supported c:\> cpan www::mechanize ... ... ... c:\programs\dwimperl\c\bin\dmake.exe install -- ok what's different between 2 , cause cpan succeed while cpanm fails 501? 10 minutes after posted this, had brief conversation coworker. mentioned environment variables. verified had indeed added windows environ...

html: <h2>hello world</h2> css h2{ padding: 0px; margin: 0px; font-size: 100px; background-color: #f00; } http://jsfiddle.net/mekaron/w50mz2n3/ margin , padding set 0. why doesnt text fit border? using javascript option here, have no idea start. ideas? you want make h2 inline element. add display: inline; css. jsfiddle

this question has answer here: how round number n decimal places in java 28 answers so calculator gives me answer 30.122345353 or that. question how can round 30.122 ? here part of code. need round answer comes (ans + "") this code if (v.getid() == r.id.badd) if (num2 == 0) flag = true; else ans = 2 * num1; tv2.settext(ans + ""); easy way rounding double int: int result = ((int)(yourdouble+0.5)); or: public double roundmyvalue(string value){ decimalformat df=new decimalformat("0.000"); return (double)df.parse(df.format(value)) ; } or easy: public double roundmyvalue(string value){ double current=double.parsedouble(value); int temp = current*1000; return (double)temp/1000; }

i have created code error says "javax.imageio.iioexception: can't read input file!". please resolve. <%@ page language="java" contenttype="text/html" import="java.awt.alphacomposite,java.awt.color,java.awt.font, java.awt.graphics,java.awt.graphics2d,java.awt.image,java.awt.image.bufferedimage,java.io.file,java.io.fileoutputstream, javax.imageio.imageio,javax.swing.imageicon,com.sun.image.codec.jpeg.jpegcodec,com.sun.image.codec.jpeg.jpegencodeparam, com.sun.image.codec.jpeg.jpegimageencoder" %> <% try { font font; file _file = new file("images/1.jpg"); image src = imageio.read(_file); int width = src.getwidth(null); int height = src.getheight(null); system.out.println("x = "+width+" , y = "+height); bufferedimage image = new bufferedimage(width, height, bufferedimage.type_int_rgb); graphics g = image.creategraphics(); g.drawimage(src, 0, 0, width, height, null); ...

i'm using jenkins continuous integration. right have job command in jenkins in command line arguments build step: this command: "%workspace%\oevizion\itvizion.oevizion.web\itvizion.oevizion.web.csproj" /p:deployonbuild=true /p:publishprofile="it vizion - web deploy package given domain.pubxml" it works fine, is, web deploy package (.zip) created in specified folder defined in .pubxml file. however i'd generate .zip web deploy package each of .pubxml files have (right 3) specific itvizion.oevizion.web.csproj in single shot\command. with i'd have multiple .zip packages different settings ready deployed different servers every time commit pushed repository , project builds successfully. is possible? how should approach this? after posting question saw way go this: add multiple build steps . 1 each .pubxml file. way build process run 3 times 3 publish profiles , you'll end 3 web deploy .zip packages @ end of job execut...

i have bucket name 0001 when use following n1ql statement "5000" syntax error: cbq> select * 0001; { "requestid": "f2b70856-f80c-4c89-ab37-740e82d119b5", "errors": [ { "code": 5000, "msg": "syntax error" } ], "status": "fatal", "metrics": { "elapsedtime": "349.733us", "executiontime": "204.442us", "resultcount": 0, "resultsize": 0, "errorcount": 1 } } i think takes 0001 number , not bucket name, there easy way rename it? in case can use ticks in n1ql escape bucket name: cbq> select * `0001`; { "requestid": "f48527e6-6035-47e7-a34f-90efe9f90d4f", "signature": { "*": "*" }, "results": [ { ...

given depth-prepass renderer, have minimum depth values given fragment can possibly contain- thus, makes no sense consider fragments farther denoted. now, have shader writes gl_fragdepth , guaranteed write value greater or equal depth value of polygonal face. how can still execute depth-pretest (ie, if fragment depth farther buffer value, discard without shader execution), allows me write value different (greater) interpolated face depth if passes pre-test? starting opengl 4.2 (glsl 4.20), functionality you're looking available layout qualifier on gl_fragdepth . allows specify intent on how going modify depth output in fragment shader. example, following specifies going change depth value make in greater: layout (depth_greater) out float gl_fragdepth; this allow depth test still used when depth modified in fragment shader. if not follow "contract" establish qualifier, undefined behavior. example, qualifier above, if make depth smaller, fragments otherwi...