smaato ... automatic test is failing

[lionux2]

Dear Ben;

Recently, we had 1.5 issues running through the Smaato automatic test. 1 (+ another potential issue?).

A: the file related to the Smaato textad template was wrong: see the none matching <clickText> with </imgUrl>.

static String  TEXTAD_TEMPLATE = "" +
			"<ad>" +
			"<textAd>" +
			"<clickUrl>__CLICKURL__</clickUrl>" +
			"<clickTextl>__TEXT__</imgUrl>" +
			"<toolTip>__TOOLTIP__</toolTip>" +
			"<additionalText>__ADDITIONALTEXT__</additionalText>" +
			"<beacons>" +
			"<beacon>__PIXELURL__</beacon>" +
			"</beacons>" +
			"</textAd>" +
			"</ad>";

Obviously, it should be </clickText> and not </imgUrl>. I can provide the code I used to pass the test.

B) Another issue is the rounding of the Win Cost during that run. This is a simple rounding, although I had the impression that the last win cost was not accounted for. On the Web admin interface, it was correct, and via the accountant (spark) is was just a bit off, and in case of rounding ... you could fail the test.

Hope this helps, and happy to send more details on the textad template issue if this is not clear. For the rounding, this was annoying, although this is just a test, and in real life, such a small rounding is not a real issue at all.

Thanks,

P.

[benmfaul]

Lionux: Here's what the textad template looks like now: Problem 1: static String TEXTAD_TEMPLATE = "" + "<ad>" + "<textAd>" + "<clickUrl>__CLICKURL__</clickUrl>" + "<clickTextl>__TEXT__</imgUrl>" + "<toolTip>__TOOLTIP__</toolTip>" + "<additionalText>__ADDITIONALTEXT__</additionalText>" + "<beacons>" + "<beacon>__PIXELURL__</beacon>" + "</beacons>" + "</textAd>" + "</ad>"; Can you point out what the error is? I can't tell exactly what you are referring to. Problem 2: Spark is counting using doubles. Probably should be changed to using BigDecimal instead of Double. I have an implementation of AtomicBigDecimal too we could use. What do you think, is it worth changing? Comment: Smaato needs to get their act together, they also support regular RTB responses now too, the XML junk should be phased out, but I know their tester program still uses it. Thanks,

…

-Ben

On Fri, Apr 7, 2017 at 1:47 PM, Lionux ***@***.***> wrote: Dear Ben; Recently, we had 1.5 issues running through the Smaato automatic test. 1 (+ another potential issue?). A: the file related to the Smaato textad template was wrong: see the none matching with . static String TEXTAD_TEMPLATE = "" + "" + "" + "*CLICKURL*" + "*TEXT*" + "*TOOLTIP*" + "*ADDITIONALTEXT*" + "" + "*PIXELURL*" + "" + "" + ""; Obviously, it should be and not . I can provide the code I used to pass the test. B) Another issue is the rounding of the Win Cost during that run. This is a simple rounding, although I had the impression that the last win cost was not accounted for. On the Web admin interface, it was correct, and via the accountant (spark) is was just a bit off, and in case of rounding ... you could fail the test. Hope this helps, and happy to send more details on the textad template issue if this is not clear. For the rounding, this was annoying, although this is just a test, and in real life, such a small rounding is not a real issue at all. Thanks, P. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#15>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACoB4tUhYnLQrM7Xb7n3HtSm2Em6qEaoks5rtqDqgaJpZM4M3X33> .

[lionux2]

Hi Ben; Apology for the delay: the issue was with __TEXT__ (which only few are using, but whatever)... <clickText> do not match with </imgUrl>. I changed it into: static String TEXTAD_TEMPLATE = "" + "<ad xmlns:xsi=\\\"http://www.w3.org/2001/XMLSchema-instance\\\" x si:noNamespaceSchemaLocation=\\\"smaato_ad_v0.9.xsd\\\" modelVersion=\\\"0.9\\\">" + "<textAd>" + "<clickUrl>__CLICKURL__</clickUrl>" + "<clickText>__IMAGEURL__</clickText>" + "<toolTip>__TOOLTIP__</toolTip>" + "<additionalText>__ADDITIONALTEXT__</additionalText>" + "<beacons>" + "<beacon>__PIXELURL__</beacon>" + "</beacons>" + "</textAd>" + "</ad>"; Smaato complained about the <ad> not having xmlns definition, and the <clickTextl> just made it worse (notice the l typo, and the none matching </imgUrl>. For the problem 2, well, this might indeed be the trouble. It would be good indeed to have it corrected (I wish I could do this ;(. Thanks.

…

On 4/7/2017 11:59 PM, benmfaul wrote: Lionux: Here's what the textad template looks like now: Problem 1: static String TEXTAD_TEMPLATE = "" + "<ad>" + "<textAd>" + "<clickUrl>__CLICKURL__</clickUrl>" + "<clickTextl>__TEXT__</imgUrl>" + "<toolTip>__TOOLTIP__</toolTip>" + "<additionalText>__ADDITIONALTEXT__</additionalText>" + "<beacons>" + "<beacon>__PIXELURL__</beacon>" + "</beacons>" + "</textAd>" + "</ad>"; Can you point out what the error is? I can't tell exactly what you are referring to. Problem 2: Spark is counting using doubles. Probably should be changed to using BigDecimal instead of Double. I have an implementation of AtomicBigDecimal too we could use. What do you think, is it worth changing? Comment: Smaato needs to get their act together, they also support regular RTB responses now too, the XML junk should be phased out, but I know their tester program still uses it. Thanks, -Ben On Fri, Apr 7, 2017 at 1:47 PM, Lionux ***@***.***> wrote: > Dear Ben; > > Recently, we had 1.5 issues running through the Smaato automatic test. 1 > (+ another potential issue?). > > A: the file related to the Smaato textad template was wrong: see the none > matching with . > > static String TEXTAD_TEMPLATE = "" + > "" + > "" + > "*CLICKURL*" + > "*TEXT*" + > "*TOOLTIP*" + > "*ADDITIONALTEXT*" + > "" + > "*PIXELURL*" + > "" + > "" + > ""; > > Obviously, it should be and not . I can provide the code I used to pass > the test. > > B) Another issue is the rounding of the Win Cost during that run. This is > a simple rounding, although I had the impression that the last win cost was > not accounted for. On the Web admin interface, it was correct, and via the > accountant (spark) is was just a bit off, and in case of rounding ... you > could fail the test. > > Hope this helps, and happy to send more details on the textad template > issue if this is not clear. For the rounding, this was annoying, although > this is just a test, and in real life, such a small rounding is not a real > issue at all. > > Thanks, > > P. > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#15>, or mute the thread > <https://github.com/notifications/unsubscribe-auth/ACoB4tUhYnLQrM7Xb7n3HtSm2Em6qEaoks5rtqDqgaJpZM4M3X33> > . > — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#15 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFW3vvTEou6pOaub85_n446LrQUSSv5mks5rtrHFgaJpZM4M3X33>.