Initial public release (#15)

* Initial public release

Author: PyDataBlog

Project.toml

@@ -7,14 +7,17 @@ version = "0.1.0"
 CircularArrays = "7a955b69-7140-5f4e-a0ed-f168c5e2e749"
 DataStructures = "864edb3b-99cc-5e75-8d2d-829cb0a9cfe8"
 OffsetArrays = "6fe1bfb0-de20-5000-8ca7-80f57d26f881"
-ProgressMeter = "92933f4c-e287-5a05-a399-4b506db050ca"
 
 [compat]
+CircularArrays = "1"
+DataStructures = "0.18"
+OffsetArrays = "1"
 julia = "1"
 
 [extras]
 Faker = "0efc519c-db33-5916-ab87-703215c3906f"
 Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+Suppressor = "fd094767-a336-5f1f-9728-57cf17d0bbfb"
 
 [targets]
-test = ["Test", "Faker"]
+test = ["Test", "Faker", "Suppressor"]

README.md

@@ -17,13 +17,16 @@ This package is be particulary useful for natural language processing tasks whic
 - [X] Support for unicodes
 - [ ] Custom user defined feature generation methods
 - [ ] Mecab-based tokenizer support
+- [X] Support for building databases directly from text files
+- [ ] Support for persistent databases
 
 ## Suported String Similarity Measures
 
 - [X] Dice coefficient
 - [X] Jaccard coefficient
 - [X] Cosine coefficient
 - [X] Overlap coefficient
+- [X] Exact match
 
 ## Installation
 

docs/src/index.md

@@ -16,7 +16,7 @@ This package is be particulary useful for natural language processing tasks whic
 - [X] Support for unicodes
 - [ ] Custom user defined feature generation methods
 - [ ] Mecab-based tokenizer support
-- [ ] Support for building databases directly from text files
+- [X] Support for building databases directly from text files
 - [ ] Support for persistent databases
 
 ## Suported String Similarity Measures
@@ -64,6 +64,8 @@ push!(db, "fooo");
 
 # Convinient approach is to use an array of strings for multiple entries: `append!(db, ["foo", "bar", "fooo"]);`
 
+# OR: Build database from text files: `append!(db, "YOUR_FILE_NAME.txt");
+
 # Retrieve the closest match(es)
 res = search(Dice(), db, "foo"; α=0.8, ranked=true)
 # 2-element Vector{Tuple{String, Float64}}:
@@ -72,7 +74,7 @@ res = search(Dice(), db, "foo"; α=0.8, ranked=true)
 
 # Describe a working database collection
 desc = describe_collection(db)
-# (total_collection = 3, avg_num_ngrams = 4.5, total_ngrams = 13)
+# (total_collection = 3, avg_size_ngrams = 4.5, total_ngrams = 13)
 ```
 
 ## TODO: Benchmarks

extras/benchmark_sim.jl

@@ -0,0 +1,48 @@
+using SimString
+using Faker
+using BenchmarkTools
+using DataStructures
+
+################################# Benchmark Bulk addition #####################
+db = DictDB(CharacterNGrams(3, " "));
+Faker.seed(2020)
+@time fake_names = [string(Faker.first_name(), " ", Faker.last_name()) for i in 1:100_000];
+
+
+f(d, x) = append!(d, x)
+@time f(db, fake_names)
+
+
+
+################################ Simple Addition ###############################
+
+db = DictDB(CharacterNGrams(2, " "));
+push!(db, "foo");
+push!(db, "bar");
+push!(db, "fooo");
+
+f(x, c, s, a, r) = search(x, c, s; α=a, ranked=r)
+test = "foo";
+col = db;
+sim = Cosine();
+a = 0.8;
+r = true;
+
+f(Cosine(),  db, "foo", 0.8, true)
+
+@btime f($sim,  $col, $test, $a, $r)
+@btime search(Cosine(), db, "foo"; α=0.8, ranked=true)
+
+
+
+db2 = DictDB(CharacterNGrams(3, " "));
+append!(db2, ["foo", "bar", "fooo", "foor"]) # also works via multiple dispatch on a vector
+
+results = search(Cosine(), db, "foo"; α=0.8, ranked=true)  # yet to be implemented
+
+bs = ["foo", "bar", "foo", "foo", "bar"]
+SimString.extract_features(CharacterNGrams(3, " "), "prepress")
+SimString.extract_features(WordNGrams(2, " ", " "), "You are a really really really cool dude.")
+
+db = DictDB(WordNGrams(2, " ", " "))
+push!(db, "You are a really really really cool dude.")

src/SimString.jl

@@ -2,7 +2,6 @@ module SimString
 
 import Base: push!, append!
 using DataStructures: DefaultOrderedDict, DefaultDict
-using ProgressMeter
 using CircularArrays
 using OffsetArrays
 

src/dictdb.jl

@@ -87,6 +87,7 @@ Basic summary stats for the DB
 db = DictDB(CharacterNGrams(2, " "));
 append!(db, ["foo", "bar", "fooo"]);
 describe_collection(db)
+(total_collection = 3, avg_size_ngrams = 4.5, total_ngrams = 13)
 
 # Returns
 * NamedTuples: Summary stats for the DB
@@ -98,7 +99,7 @@ function describe_collection(db::DictDB)
 # Total number of strings in collection
 ∑ = length(db.string_collection)
 
-# Average number of ngram features
+# Average size of ngram features
 n = [x for x in keys(db.string_size_map)]
 μ = sum(n) / length(n)
 
@@ -108,7 +109,19 @@ for i in values(db.string_feature_map)
     total_ngrams += length(i)
 end
 
-return (total_collection = ∑, avg_num_ngrams = μ, total_ngrams = total_ngrams)
+return (total_collection = ∑, avg_size_ngrams = μ, total_ngrams = total_ngrams)
+end
+
+
+"""
+Pretty print summary stats for the DB
+"""
+function Base.show(io::IO, x::DictDB)
+    metrics = describe_collection(x)
+    println(io, "DictDB($(x.feature_extractor))")
+    println(io, "Total collection: ", metrics.total_collection)
+    println(io, "Average number of ngram features: ", metrics.avg_size_ngrams)
+    println(io, "Total number of ngram features: ", metrics.total_ngrams)
 end
 
 

src/features.jl

@@ -99,8 +99,25 @@ end
 
 
 """
+    push!(db::AbstractSimStringDB, str::AbstractString)
+
 Add a new item to a new or existing collection of strings using
 the custom AbstractSimStringDB type.
+
+# Arguments:
+* `db`: AbstractSimStringDB - The collection of strings to add to
+* `str`: AbstractString - The string to add to the collection
+
+# Example:
+```julia
+db = DictDB(CharacterNGrams(2, " "));
+push!(db, "foo")
+push!(db, "bar")
+push!(db, "fooo")
+````
+
+# Returns:
+* `db`: AbstractSimStringDB - The collection of strings with the new string added
 """
 function push!(db::AbstractSimStringDB, str::AbstractString)
     # Extract features based on the specified feature extractor
@@ -125,11 +142,54 @@ end
 
 
 """
+    append!(db::AbstractSimStringDB, str::Vector)
+
 Add bulk items to a new or existing collection of strings using
 the custom AbstractSimStringDB type.
+
+# Arguments:
+* db: AbstractSimStringDB - The database to add the strings to
+* str: Vector of AbstractString - Vector/Array of strings to add to the database
+
+# Example:
+```julia
+db = DictDB(CharacterNGrams(2, " "));
+append!(db, ["foo", "foo", "fooo"]);
+```
+
+# Returns:
+* db: AbstractSimStringDB - The database with the new strings added
 """
 function append!(db::AbstractSimStringDB, str::Vector)
     @inbounds @simd for i in str
         push!(db, i)
     end
+end
+
+
+"""
+    append!(db::AbstractSimStringDB, file::AbstractString)
+
+Add bulk items to a new or existing collection of strings using
+from a file using the custom AbstractSimStringDB type.
+
+# Arguments:
+* `db``: AbstractSimStringDB - The database to add the items to
+* `file`: AbstractString - Path to the file to read from
+
+# Example:
+```julia
+db = DictDB(CharacterNGrams(2, " "));
+append!(db, "./data/test.txt")
+```
+
+# Returns:
+* `db`: AbstractSimStringDB - The database with the items added
+"""
+function append!(db::AbstractSimStringDB, file::AbstractString)
+    open(file) do f
+        for line in eachline(f)
+            push!(db, line)
+        end
+    end
 end

src/search.jl

@@ -74,7 +74,7 @@ function overlap_join(db_collection::AbstractSimStringDB, features, τ, candidat
     results = String[]
 
     for (candidate, match_count) in candidate_match_counts
-        for i in (query_feature_length - τ + 1) : query_feature_length # TODO: Verify
+        for i in (query_feature_length - τ + 1) : query_feature_length
             if candidate in lookup_feature_set_by_size_feature(db_collection, candidate_size, features[i])
                 match_count += 1
             end
@@ -103,16 +103,16 @@ function search!(measure::AbstractSimilarityMeasure, db_collection::DictDB, quer
     features = extract_features(db_collection.feature_extractor, query)
 
     # Metadata from the generated features (length, min & max sizes)
-    length_of_features = length(features)
-    min_feature_size = minimum_feature_size(measure, length_of_features, α)
-    max_feature_size = maximum_feature_size(measure, db_collection, length_of_features, α)
+    # length_of_features = length(features)
+    # min_feature_size = minimum_feature_size(measure, length_of_features, α)
+    # max_feature_size = maximum_feature_size(measure, db_collection, length_of_features, α)
 
     results = String[]
 
     # Generate and return results from the potential candidate size pool
-    @inbounds for candidate_size in min_feature_size:max_feature_size
+    @inbounds for candidate_size in minimum_feature_size(measure, length(features), α) : maximum_feature_size(measure, db_collection, length(features), α)
         # Minimum overlap
-        τ = minimum_overlap(measure, length_of_features, candidate_size, α)
+        τ = minimum_overlap(measure, length(features), candidate_size, α)
 
         # Generate approximate candidates from the overlap join
         append!(results, overlap_join(db_collection, features, τ, candidate_size))

test/dummy_sents.txt

@@ -0,0 +1,2 @@
+You are a really really really cool dude.
+Sometimes you are not really really cool tho

test/dummy_words.txt

@@ -0,0 +1,3 @@
+foo
+bar
+fooo