Update docs for resampling sequentially.

docs/src/resampling/sequential/resampling_indexvalue_sequential.md

@@ -2,13 +2,15 @@
 # Resampling syntax
 
 ## Manually resampling 
+
 Because both the indices and the values of `UncertainIndexValueDataset`s are 
 datasets themselves, you could manually resample them by accessing the `indices` and 
 `values` fields. This gives you full control of the resampling. 
 
 There are some built-in sampling routines you could use instead if you use cases are simple.
 
 ## Built-in resampling methods 
+
 Sequential constraints are always interpreted as belonging to the indices of an 
 [uncertain index-value dataset](../../uncertain_datasets/uncertain_indexvalue_dataset.md). 
 Therefore, when using the built-in function to resample an index-value dataset, you can use 
@@ -23,21 +25,20 @@ The following examples illustrates the syntax. Assume `udata` is an
 - `resample(udata, StrictlyIncreasing())` enforces the sequential constraint only to the 
     indices, applying no constraint(s) on the furnishing distributions of either the 
     indices nor the values of the dataset.
-- `resample(udata, TruncateQuantile(0.1, 0.9), StrictlyIncreasing())` applies the truncating 
+- `resample(udata, StrictlyIncreasing(), TruncateQuantile(0.1, 0.9))` applies the truncating 
     constraint both the indices and the values, then enforces the sequential constraint 
     on the indices. 
-- `resample(udata, TruncateStd(2), TruncateQuantile(0.1, 0.9), StrictlyIncreasing())` 
+- `resample(udata, StrictlyIncreasing(), TruncateStd(2), TruncateQuantile(0.1, 0.9))` 
     applies separate truncating constraints to the indices and to the values, then 
     enforces the sequential constraint on the indices. 
-- `resample(udata, NoConstraint(), TruncateQuantile(0.1, 0.9), StrictlyIncreasing())` does 
+- `resample(udata, StrictlyIncreasing(), NoConstraint(), TruncateQuantile(0.1, 0.9))` does 
     the same as above, but `NoConstraint()` indicates that no constraints are applied to 
     the indices prior to drawing the sequential realization of the indices. 
 
 Of course, like for uncertain value datasets, you can also apply individual constraints to 
 each index and each value in the dataset, by providing a vector of constraints instead 
 of a single constraint.
 
-
 Currently implemented sequential constraints: 
 
 - [StrictlyIncreasing](strictly_increasing.md) 

src/resampling/ordered_resampling/resample_sequential.jl

@@ -11,7 +11,8 @@ const XD = Union{AbstractUncertainDataset, Vector{<:AbstractUncertainValue}}
 
 """
     resample(x, ssc::SequentialSamplingConstraint)
-    resample(x, ssc::SequentialSamplingConstraint, c::Union{SamplingConstraint, Vector{SamplingConstraint}})
+    resample(x, ssc::SequentialSamplingConstraint, c)
+    resample(x::UncertainIndexValueDataset, ssc::SequentialSamplingConstraint, ic, vc)
 
 Sample `x` element-wise such that the samples obey the sequential constraints given by `ssc`.
 Alteratively, apply constrain(s) `c` to `x` *before* sequential sampling is performed. 
@@ -21,7 +22,12 @@ Before the check is performed, the distributions in `x` are truncated element-wi
 to the quantiles provided by `c` to ensure they have finite supports.
 
 If `x` is an uncertain index-value dataset, then the sequential constraint is only applied to 
-the indices.
+the indices. If one set of additional constraints are added to an uncertain index-value dataset,
+then they are applied to both the indices and the values. It is also possible to give separate 
+index constraints `ic` and value constraints `vc`. 
+
+If constraints `c`, or `ic` and `vc`, are given, then `c`/`ic`/`vc` must be either a single constraint,
+or a vector of constraints where constraints are applied element-wise to the distributions in `x`.
 
     resample!(s, x, ssc::SequentialSamplingConstraint, lqs, uqs)