Incorporated latest changes

Project.toml

@@ -4,6 +4,7 @@ authors = ["Alex Wietek <[email protected]> and contributors"]
 version = "1.0.0-DEV"
 
 [deps]
+DataStructures = "864edb3b-99cc-5e75-8d2d-829cb0a9cfe8"
 HDF5 = "f67ccb44-e63f-5c2f-98bd-6dc0ccc4ba2f"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
 

src/Dumper.jl

@@ -2,7 +2,9 @@ module Dumper
 using HDF5
 using Printf
 
-export DumpFile, write_data!, dump!, read_data, FileCollection, filenames, read_collection_h5
+import Base.keys, Base.read
+export DumpFile, write_data!, dump!, read_data
+export FileCollection, files
 
 include("create_extensible.jl")
 include("append_extensible.jl")

src/file_collection.jl

@@ -1,9 +1,19 @@
+import DataStructures: OrderedDict
+
 struct FileCollection
     directory::AbstractString
     regex::Regex
 end
 
-function filenames(collection::FileCollection)
+"""
+    files
+
+Returns which files are matched by the file collection
+
+# Arguments
+- `collection::FileCollection`: the file collection to be matched
+"""
+function files(collection::FileCollection)
     files = readdir(collection.directory)
     matched_files = String[]
     for file in files
@@ -15,9 +25,56 @@ function filenames(collection::FileCollection)
     return map(fl -> joinpath(collection.directory, fl), matched_files)
 end
 
-function read_collection_h5(collection::FileCollection, tag::AbstractString)
+"""
+    keys
+
+Returns which keys are defined in the hdf5 files
+
+# Arguments
+- `collection::FileCollection`: the file collection for which to get the keys
+"""
+function keys(collection::FileCollection)
+    files = readdir(collection.directory)
+    data = OrderedDict()
+
+    for file in files
+        m = match(collection.regex, file)
+        if !isnothing(m)
+            if length(m) == 1
+                param = m[Base.keys(m)[1]]
+            else
+                param = ntuple(i -> m[Base.keys(m)[i]], length(m))
+            end
+
+            # read hdf5 data
+            file = joinpath(collection.directory, file)
+            try
+                f = h5open(file)
+                keys = Base.keys(f)
+                data[param] = keys
+            catch e
+                showerror(stdout, e)
+                println()
+                println("cannot read from file", file)
+            end
+        end
+    end
+    return data
+end
+
+
+"""
+    read
+
+reads the data from a file collection given a key
+
+# Arguments
+- `collection::FileCollection`: the file collection to be matched
+- `key::String`: key for which the data is supposed to be read
+"""
+function Base.read(collection::FileCollection, key::String)
     files = readdir(collection.directory)
-    data = Dict()
+    data = OrderedDict()
     for file in files
         m = match(collection.regex, file)
         if !isnothing(m)
@@ -30,13 +87,15 @@ function read_collection_h5(collection::FileCollection, tag::AbstractString)
             # read hdf5 data
             file = joinpath(collection.directory, file)
             try
-                dset = h5read(file, tag)
+                dset = h5read(file, key)
                 ndims = length(size(dset))
                 data[param] = permutedims(dset, ndims:-1:1)
-            catch
+            catch e
+                showerror(stdout, e)
+                println()
                 println("cannot read from file", file)
             end
         end
     end
-    return sort(collect(data), by = x->x[1])
+    return data
 end