def "can normalize line endings in files (eol = '#description')"() {
        def unnormalized = file('unnormalized.txt') << content.textWithLineEndings(eol)
        def normalized = file('normalized.txt') << content.textWithLineEndings('\n')

        expect:
        hasher.hashContent(unnormalized).get() == hasher.hashContent(normalized).get()

        where:
        eol     | description
        '\r'    | 'CR'

	op_AU      uint32 = 0x7E00 // FORMAT_RX1        ADD UNNORMALIZED (short HFP)
	op_AUR     uint32 = 0x3E00 // FORMAT_RR         ADD UNNORMALIZED (short HFP)
	op_AW      uint32 = 0x6E00 // FORMAT_RX1        ADD UNNORMALIZED (long HFP)
	op_AWR     uint32 = 0x2E00 // FORMAT_RR         ADD UNNORMALIZED (long HFP)
	op_AXBR    uint32 = 0xB34A // FORMAT_RRE        ADD (extended BFP)

  let summary = "Draws samples from a categorical distribution.";

  let description = [{
The generated values will have a categorical distribution based on the `logits`
or unnormalized log-probabilities provided for all classes.
  }];

  let arguments = (ins
    TFL_FpTensor:$logits,
    TFL_I32Tensor:$num_samples,
    DefaultValuedOptionalAttr<I64Attr, "0">:$seed,

  let summary = "Draws samples from a multinomial distribution.";

  let arguments = (ins
    Arg<TF_IntOrFpTensor, [{2-D Tensor with shape `[batch_size, num_classes]`.  Each slice `[i, :]`
represents the unnormalized log probabilities for all classes.}]>:$logits,
    Arg<TF_Int32Tensor, [{0-D.  Number of independent samples to draw for each row slice.}]>:$num_samples,

    DefaultValuedOptionalAttr<I64Attr, "0">:$seed,