kanji placement in radkfile

[z <z-one@***********>]

Hi,

I'm compiling a file similar to radkfile, which lists radicals and the 
kanji containing them, while taking into account the placement where the 
radical is found. An example from it:

$46 山 3 やま
出山密両満端島微徴催掘屈綱鋼缶堀懲拙仙岳薇剛瑞癌黴咄辿窟凱喘鎧岡搗嶮艷嗤 
揣摧柮眄繃閊倔剴徽疝豐黜丗圸岔岱嵒嶝巒惴攜朏杣棡榿槝樒汕湍澂猯皚 磑糶覬

$46 山 3 やまかんむり
岸崩岩炭嵐崇嵌崖嶺嵩巌嶽祟崗岌豈峯嶄巍巓嶌崔妛屶岑峇峩崋崑崙崟嵜嵬嵳嶐巖蚩嶷

$46 山 3 やまへん
乢峰岐峡峠岬崎屹峙峻巉嶋岨嶼嵯峨岫岶岷岻岼岾峅峪峭峺峽崕崘崚崛崢嵋嵎嵶嶂 
嶇嶢嶬
(kanji number, radical, radical name depending on position)

There are still mistakes but this is just a work-in-progress.

I've noticed, that in kradfile and radkfile many kanji are under both 夂 
and 攵. For example 数 is in the list of 夂, and 条 is listed for 攵 as 
well. Is this a mistake or is there a valid reason for this? I'm only 
trying to place kanji by their visual appearance. Historical/scientific 
accuracy is not my concern. Are the mentioned kanji examples appear in 
modern text with different forms that makes it reasonable to list them 
for both radicals?