Dev

.github/workflows/lint.yaml

@@ -29,3 +29,8 @@ jobs:
           go install honnef.co/go/tools/cmd/staticcheck@latest
           staticcheck ./...
           go vet ./...
+
+      - name: revive
+        run: |
+          go install github.com/mgechev/revive@latest
+          revive -config .revive.toml ./...

.testcoverage.yml

@@ -24,27 +24,3 @@ threshold:
   # (optional; default 0) 
   # The minimum total coverage project should have
   total: 0
-
-# Holds regexp rules which will override thresholds for matched files or packages 
-# using their paths.
-#
-# First rule from this list that matches file or package is going to apply 
-# new threshold to it. If project has multiple rules that match same path, 
-# override rules should be listed in order from specific to more general rules.
-override:
-  # Increase coverage threshold to 100% for `foo` package 
-  # (default is 80, as configured above in this example)
-  - threshold: 100
-    path: ^pkg/lib/foo$
-
-# Holds regexp rules which will exclude matched files or packages 
-# from coverage statistics
-exclude:
-  # Exclude files or packages matching their paths
-  paths:
-    - \.pb\.go$    # excludes all protobuf generated files
-    - ^pkg/bar     # exclude package `pkg/bar`
-
-# NOTES:
-# - symbol `/` in all path regexps will be replaced by current OS file path separator
-#   to properly work on Windows

decoder.go

@@ -1,6 +1,9 @@
 package seltabl
 
-import "io"
+import (
+	"fmt"
+	"io"
+)
 
 // Decoder is a struct for decoding a reader into a slice of structs.
 //
@@ -72,7 +75,12 @@ func NewDecoder[T any](r io.ReadCloser) *Decoder[T] {
 // This allows for decoding a reader into a slice of structs.
 //
 // Similar to the json.Decoder for brevity.
-func (d *Decoder[T]) Decode(value *T) ([]T, error) {
+func (d *Decoder[T]) Decode() ([]T, error) {
 	defer d.reader.Close()
-	return NewFromReader[T](d.reader)
+	var result []T
+	result, err := NewFromReader[T](d.reader)
+	if err != nil {
+		return nil, fmt.Errorf("failed to decode: %w", err)
+	}
+	return result, nil
 }

decoder_test.go

@@ -1 +1,139 @@
 package seltabl
+
+import (
+	"io"
+	"strings"
+	"testing"
+)
+
+// DecodeExStruct is a test struct
+type DecodeExStruct struct {
+	A string `json:"a" seltabl:"a" hSel:"tr:nth-child(1) td:nth-child(1)" dSel:"tr td:nth-child(1)" cSel:"$text"`
+	B string `json:"b" seltabl:"b" hSel:"tr:nth-child(1) td:nth-child(2)" dSel:"tr td:nth-child(2)" cSel:"$text"`
+}
+
+// TestDecoder_Decode tests the Decoder.Decode function
+func TestDecoder_Decode(t *testing.T) {
+	testCases := []struct {
+		name     string
+		input    string
+		expected []DecodeExStruct
+		hasError bool
+	}{
+		{
+			name: "Valid input",
+			input: `
+                <table>
+                    <tr>
+                        <td>a</td>
+                        <td>b</td>
+                    </tr>
+                    <tr>
+                        <td>1</td>
+                        <td>2</td>
+                    </tr>
+                    <tr>
+                        <td>3</td>
+                        <td>4</td>
+                    </tr>
+                </table>
+            `,
+			expected: []DecodeExStruct{
+				{A: "1", B: "2"},
+				{A: "3", B: "4"},
+			},
+			hasError: false,
+		},
+		{
+			name: "Invalid input",
+			input: `
+                <table>
+                    <tr>
+                        <td>a</td>
+                        <td>b</td>
+                    </tr>
+                    <tr>
+                        <td>1</td>
+                    </tr>
+                </table>
+            `,
+			expected: nil,
+			hasError: true,
+		},
+		{
+			name: "Invalid input with invalid html",
+			input: `
+                <table>
+                    <tr>
+                        <td>a</td>
+                        <td>b</td>
+                    </tr>
+                    <tr>
+                        <td>1</td>
+                    </tr>
+                </table>
+            `,
+			expected: nil,
+			hasError: true,
+		},
+		{
+			name: "Invalid input with invalid json",
+			input: `
+                <table>
+                    <tr>
+                        <td>a</td>
+                        <td>b</td>
+                    </tr>
+                    <tr>
+                        <td>1</td>
+                    </tr>
+                </table>
+            `,
+			expected: nil,
+			hasError: true,
+		},
+		{
+			name: "Invalid input with invalid json",
+			input: `
+                <table>
+                    <tr>
+                        <td>a</td>
+                        <td>b</td>
+                        <td>1</td>
+                    </tr>
+                </table>
+            `,
+			expected: nil,
+			hasError: true,
+		},
+	}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			r := io.NopCloser(strings.NewReader(tc.input))
+			decoder := NewDecoder[DecodeExStruct](r)
+			result, err := decoder.Decode()
+
+			if tc.hasError {
+				if err == nil {
+					t.Errorf("Expected an error, but got none")
+				}
+				return
+			}
+
+			if err != nil {
+				t.Errorf("Unexpected error: %v", err)
+			}
+
+			if len(result) != len(tc.expected) {
+				t.Errorf("Expected %d results, but got %d", len(tc.expected), len(result))
+			}
+
+			for i, expected := range tc.expected {
+				if result[i].A != expected.A || result[i].B != expected.B {
+					t.Errorf("Expected %+v, but got %+v", expected, result[i])
+				}
+			}
+		})
+	}
+}

examples/huggingface-leader-board/main.go

@@ -0,0 +1,22 @@
+// Package main shows how to use the seltabl package to scrape a table from a given url.
+// The table used in this example is from the huggingface llm leader board.
+package main
+
+import (
+	"fmt"
+	"os"
+)
+
+// main scrapes from: https://huggingface.co/spaces/HuggingFaceH4/LLM-Leaderboard
+func main() {
+	if err := run(); err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+}
+
+// run runs the example
+func run() error {
+	fmt.Println("Hello, World from llm leader board!")
+	return nil
+}

examples/huggingface-leader-board/main_test.go

@@ -0,0 +1,13 @@
+package main
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+// TestRun tests the run function
+func TestRun(t *testing.T) {
+	err := run()
+	assert.Nil(t, err)
+}

examples/ncaa/doc.go

@@ -0,0 +1,3 @@
+// Package main is the an example of how to use the seltabl package.
+// for the seltabl package
+package main

examples/penguins-wikipedia/main.go

@@ -0,0 +1,24 @@
+// Package main is the an example of how to use the seltabl package.
+// for the seltabl package to scrape a html table from a given url.
+// The table used in this example is from the wikipedia page for
+// penguins.
+package main
+
+import (
+	"fmt"
+	"os"
+)
+
+// main scrapes from: https://en.wikipedia.org/wiki/List_of_penguins
+func main() {
+	if err := run(); err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+}
+
+// run runs the example
+func run() error {
+	fmt.Println("Hello, World from list of penguins!")
+	return nil
+}

examples/penguins-wikipedia/main_test.go

@@ -0,0 +1,13 @@
+package main
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+// TestRun tests the run function
+func TestRun(t *testing.T) {
+	err := run()
+	assert.Nil(t, err)
+}