Unit tests and string comparisons

Hi,

I'm testing the equality of two rather large strings. They look the
same to me, but doing

assert_equal "big ass long string", "big ass long string"

says that they're different. And they look very similar to me. I
know that they're different though, as doing "big ass long string".sum
gives a different checksum than the other one.

Any ideas on how I can easily view the differences between two long strings?

Use unit_diff from ZenTest.

···

On Jan 11, 2006, at 4:57 PM, Joe Van Dyk wrote:

I'm testing the equality of two rather large strings. They look the
same to me, but doing

assert_equal "big ass long string", "big ass long string"

says that they're different. And they look very similar to me. I
know that they're different though, as doing "big ass long string".sum
gives a different checksum than the other one.

Any ideas on how I can easily view the differences between two long strings?

--
Eric Hodel - drbrain@segment7.net - http://segment7.net
This implementation is HODEL-HASH-9600 compliant

http://trackmap.robotcoop.com

stupid - but works:

   harp:~ > cat a.rb
   a = "foobar"
   b = "foobaz"

   def sdiff a, b
     a = a.split //
     b = b.split //
     a.zip(b).each_with_index{|cc,idx| printf "%d: %s\n", idx, cc.inspect unless cc.uniq.size == 1}
   end

   sdiff a, b

   harp:~ > ruby a.rb
   5: ["r", "z"]

or just dump them to two files using 'od -c' and then use 'diff -u'.

regards.

-a

···

On Thu, 12 Jan 2006, Joe Van Dyk wrote:

Hi,

I'm testing the equality of two rather large strings. They look the
same to me, but doing

assert_equal "big ass long string", "big ass long string"

says that they're different. And they look very similar to me. I
know that they're different though, as doing "big ass long string".sum
gives a different checksum than the other one.

Any ideas on how I can easily view the differences between two long strings?

--
strong and healthy, who thinks of sickness until it strikes like lightning?
preoccupied with the world, who thinks of death, until it arrives like
thunder? -- milarepa

Just for fun... here's another quick visual method:

s1 = 'abcdefghijk1mnOpqrstuvwxyz'
s2 = 'abcdefghijklmn0pqrstuvwxyz'

puts s1, s2
# XOR the binary of each character, will be 0 if the same
s1.size.times {|i| print s1[i]^s2[i].to_i==0?'-':'#'}

Results:
abcdefghijk1mnOpqrstuvwxyz
abcdefghijklmn0pqrstuvwxyz

···

-----------#--#-----------

(above looks better in a fixed font display)

-Joby

Thanks, I ended up doing something similar. I thought I recalled
reading a post here (or maybe on the Rails mailing list) about some
guy who was comparing large hashes and was having difficulties seeing
the differences between the two in the unit test output. This is sort
of a related problem, I guess.

···

On 1/12/06, ara.t.howard@noaa.gov <ara.t.howard@noaa.gov> wrote:

On Thu, 12 Jan 2006, Joe Van Dyk wrote:

> Hi,
>
> I'm testing the equality of two rather large strings. They look the
> same to me, but doing
>
> assert_equal "big ass long string", "big ass long string"
>
> says that they're different. And they look very similar to me. I
> know that they're different though, as doing "big ass long string".sum
> gives a different checksum than the other one.
>
> Any ideas on how I can easily view the differences between two long strings?

stupid - but works:

   harp:~ > cat a.rb
   a = "foobar"
   b = "foobaz"

   def sdiff a, b
     a = a.split //
     b = b.split //
     a.zip(b).each_with_index{|cc,idx| printf "%d: %s\n", idx, cc.inspect unless cc.uniq.size == 1}
   end

   sdiff a, b

   harp:~ > ruby a.rb
   5: ["r", "z"]

or just dump them to two files using 'od -c' and then use 'diff -u'.

uh, unit_diff, from ZenTest.

http://rubyforge.org/frs/?group_id=419&release_id=3133

No, really.

···

On Jan 11, 2006, at 5:14 PM, Joe Van Dyk wrote:

On 1/12/06, ara.t.howard@noaa.gov <ara.t.howard@noaa.gov> wrote:

On Thu, 12 Jan 2006, Joe Van Dyk wrote:

Hi,

I'm testing the equality of two rather large strings. They look the
same to me, but doing

assert_equal "big ass long string", "big ass long string"

says that they're different. And they look very similar to me. I
know that they're different though, as doing "big ass long string".sum
gives a different checksum than the other one.

Any ideas on how I can easily view the differences between two long strings?

stupid - but works:

   harp:~ > cat a.rb
   a = "foobar"
   b = "foobaz"

   def sdiff a, b
     a = a.split //
     b = b.split //
     a.zip(b).each_with_index{|cc,idx| printf "%d: %s\n", idx, cc.inspect unless cc.uniq.size == 1}
   end

   sdiff a, b

   harp:~ > ruby a.rb
   5: ["r", "z"]

or just dump them to two files using 'od -c' and then use 'diff -u'.

Thanks, I ended up doing something similar. I thought I recalled
reading a post here (or maybe on the Rails mailing list) about some
guy who was comparing large hashes and was having difficulties seeing
the differences between the two in the unit test output. This is sort
of a related problem, I guess.

--
Eric Hodel - drbrain@segment7.net - http://segment7.net
This implementation is HODEL-HASH-9600 compliant

http://trackmap.robotcoop.com

I have to agree. This does what you all are building code to do. :slight_smile:

James Edward Gray II

···

On Jan 11, 2006, at 11:17 PM, Eric Hodel wrote:

uh, unit_diff, from ZenTest.

http://rubyforge.org/frs/?group_id=419&release_id=3133

No, really.