I’d like to share a basic interview question I’ve used in the past. I’ve used this in a number of different guises over the years, both at Sanger and at ONT but the (very small!) core remains the same. It still seems to be able to trip up a lot of people who sell themselves as senior developers on their CVs and demand \u00c2\u00a335k+ salaries.<\/p>\n
You have a list of characters.<\/p>\n
The time taken for the interviewee to scratch their head determines whether they’re a Perl programmer, or at least think like one – this is an idomatic question in Perl. It’s a fairly standard solution to anyone who uses hashes, maps or associative arrays in any language. It’s certainly a lot harder without them.<\/p>\n
The answer I would expect to see would run something like this:<\/p>\n
#########\r\n# pass in an array ref of characters, e.g.\r\n# remove_dupes([qw(a e r o i g n o s e w f e r g e r i g e o n k)]);\r\n#\r\nsub remove_dupes {\r\n my $chars_in = shift;\r\n my $seen = {};\r\n my $chars_out = [];\r\n\r\n for my $char (@{$chars_in}) {\r\n if(!$seen->{$char}++) {\r\n push @{$chars_out}, $char;\r\n }\r\n }\r\n\r\n return $chars_out;\r\n}<\/code><\/pre>\nOr for the more adventurous, using a string rather than an array:<\/p>\n
#########\r\n# pass in a string of characters, e.g.\r\n# remove_dupes(q[uyavubnopwemgnisudhjopwenfbuihrpgbwogpnskbjugisjb]);\r\n#\r\nsub remove_dupes {\r\n my $str = shift;\r\n my $seen = {};\r\n $str =~ s\/(.)\/( !$seen->{$1}++ ) ? $1 : q[]\/smegx;\r\n return $str;\r\n}<\/code><\/pre>\nThe natural progression from Q1 then follows. It should be immediately obvious to the interviewee if they answered Q1 inappropriately.<\/p>\n
\n- List duplicates<\/li>\n<\/ol>\n
#########\r\n# pass in an array ref of characters, e.g.\r\n# list_dupes([qw(a e r o i g n o s e w f e r g e r i g e o n k)]);\r\n#\r\nsub list_dupes {\r\n my $chars_in = shift;\r\n my $seen = {};\r\n my $chars_out = [];<\/del>\r\n\r\n for my $char (@{$chars_in}) {\r\n $seen->{$char}++;\r\n }\r\n\r\n return [ grep { $seen->{$_} > 1 } keys %{$seen} ];\r\n}<\/code><\/pre>\nand with a string<\/p>\n
#########\r\n# pass in a string of characters, e.g.\r\n# list_dupes(q[uyavubnopwemgnisudhjopwenfbuihrpgbwogpnskbjugisjb]);\r\n#\r\nsub list_dupes {\r\n my $str = shift;\r\n my $seen = {};\r\n $str =~ s\/(.)\/( $seen->{$1}++ > 1) ? $1 : q[]\/smegx;\r\n return $str;\r\n}<\/code><\/pre>\nThe standard follow-up is then “Given more time, what would you do to improve this?”. Well? What would you<\/em> do? I know what I would do before I even started – WRITE SOME TESTS<\/strong>!<\/p>\nIt’s pretty safe to assume that any communicative, personable candidate who starts off writing a test on the board will probably be head and shoulders above any other.<\/p>\n
If I’m interviewing you tomorrow and you’re reading this now, it’s also safe to mention it. Interest in the subject and a working knowledge of the intertubes generally comes in handy for a web developer. I’m hiring you as an independent thinker!<\/p>\n","protected":false},"excerpt":{"rendered":"
I’d like to share a basic interview question I’ve used in the past. I’ve used this in a number of different guises over the years, both at Sanger and at ONT but the (very small!) core remains the same. It still seems to be able to trip up a lot of people who sell themselves … Continue reading “An Interview Question”<\/span><\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_uf_show_specific_survey":0,"_uf_disable_surveys":false,"footnotes":""},"categories":[11],"tags":[13,227,21,229,228],"class_list":["post-140","post","type-post","status-publish","format-standard","hentry","category-programming","tag-code","tag-interviews","tag-perl","tag-simple","tag-tests"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t