Nine

i have website crashing internet explorer on users computers. i'm lucky enought hat few people h\ve contacted me tell me otherwise wouldnt know anuthing it. it doesnt happen on everyones pc i'm bit lost causing ie crash. website uses jquery 1.4.2. best way ot trapping error can find out what's going on? thanks al first, find computer can reproduce problem. compare browser versions , windows versions. microsoft has script debugger pretty once running, in case, more primitive simple , effective method put alert() source code until point reached things crash.

could explain me following compiler issue error: type of conditional expression cannot determined because there no implicit conversion between 'string' , 'int' // works string text = string.format( "the id {0}", _obj.id.tostring()); // works, without implicit conversion <<< string text = string.format( "the id {0}", _obj.id); // works string text = string.format( "the id {0}", (_obj == null) ? "unknown" : _obj.id.tostring()); // no way <<< string text = string.format( "the id {0}", (_obj == null) ? "unknown" : _obj.id); in last example, there no implicit conversion, well. eric lippert's cast operators not obey distributive law

when using ggplot or subset, r cmd check command on package generates following notes: * checking r code possible problems ... note foo: no visible binding global variable ‘bar’ for mock function such foo <- function(dataframe) { subset(dataframe,bar>10) } that r cmd check section useful when forgetting usage of global variables in package. therfore i'd have clean. is there possibility let r know of special nature of ggplot/subset? that's reason note not warning or error - there's no automated way detect use of non-standard evaluation in function.

i have multi module maven project , i'd able run operation (antrun) after submodules have finished execution. in projects build rpms each sub-module , in deploy phase copy rpms (by ssh) rpm repository. rpm repo requires me run createdb command after adding new rpms can index them. have antrun runs scp , copies rpms repo. runs each submodule produces rpm , hooked maven deploy phase. can after each scp run createrepo command that'd wasteful, have more 10 submodules , each createrepo takes minute i'll wasting precious build time. what i'd after submodules have finished, run createrepo. once. my first attempt hook antrun calls createrepo command parent pom's (the super-pom) deploy phase. did , problem parent's deploy phase runs before submodules' deploy phase. wanted run after can't attach createrepo antrun other later phase since deploy phase last in lifecycle. so question is: there way run "cleanup" antrun or plugin in gene...

i want have standalone network (no internet access) testing. there way spoof apns servers test notifications , mdm. there way create own apns server. sure. apns protocol defined in apple's documentation here: the binary interface , notification formats . basically, it's ssl server speaks simple binary protocol. have implemented mock push server , feedback server our internal testing.

i've been having problems database repeatedly having crashed tables. luckily it's easy enough fix repairing crashed table, it's hardly practice have watch table crash (or have client tell me) , fix it. crashes tend occur after change has been made database client has access via cms. i noticed recent time table crashed referenced number - found 57 of 89; noticed in cardinality primary key. putting 2 , 2 googled cardinality , found optimising table in way related , thought optimising table regularly, after update, prevent crashes. true or have managed 73 rather 4? i can send mysql functions database when client makes changes via php, perspective great. any other table crashes appreciated. is not true optimize table prevent table crashing in fact, u should not run optimize table (table locked) although designed compact free data try mysqlcheck -c or mysqlcheck -c

i have mfc gui app has multiple frames (sort of outlook'ish behavior main window, , message/appointment windows being created in new frames, or skype multi frame syncronization), , need postmessage malloc'ed data through window hierarchy. say, string, _wcsdup it, postmessage(wm_newstring, ...), and, control somewhere deep down hierarchy processes it, or if there no subscribers, message get's cleaned. what looking now, messages posted application thread, thread finds active frame or best fit frame, passes message it, frame passes message it's view, view passes message subview , on, if there no view process message, get's free'd. the problem these chaining commands pretty tiring write, have duplicate message forwarding code in each cwnd class have. @ same time, resource cleanup pretty unpleasant, because if there no window catch message, has call free. therefore posting message main message pump without handling, hoping catch it, not valid approach. post...