Initial commit

.github/workflows/build_and_test.yml

@@ -0,0 +1,55 @@
+name: Build and Test
+on:
+  push:
+    branches: [ main ]
+  workflow_dispatch:
+env:
+  MORPH_API_KEY: ${{ secrets.MORPH_API_KEY }}
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+    - uses: actions/setup-python@v5
+      with:
+        python-version: '3.11'
+    - name: Install morphcloud
+      run: |
+        pip install git+https://github.com/morph-labs/morph-python-sdk.git
+    - name: Start instance and get ID
+      run: |
+        echo "INSTANCE_ID=$(morphcloud instance start snapshot_do2mk83p)" >> $GITHUB_ENV
+    - name: Copy files to MorphVM
+      run: |
+        morphcloud instance copy -r . $INSTANCE_ID:.
+    - name: Install dependencies on MorphVM
+      run: |
+        morphcloud instance ssh $INSTANCE_ID -- apt update
+        morphcloud instance ssh $INSTANCE_ID -- apt install -y gcc gdb make tmux python3
+    - name: Build program on MorphVM
+      run: |
+        morphcloud instance ssh $INSTANCE_ID -- make
+    - name: Generate test input
+      run: |
+        morphcloud instance ssh $INSTANCE_ID -- 'python3 test_input.py > test_data.txt'
+    - name: Run test in tmux/gdb
+      id: gdb-run
+      continue-on-error: true
+      run: |
+        morphcloud instance ssh $INSTANCE_ID -- tmux new-session -d 'gdb -ex "handle SIGABRT stop" -ex "run test_data.txt" -ex "bt" ./text_analyzer'
+        sleep 5
+        if morphcloud instance ssh $INSTANCE_ID -- tmux list-sessions; then
+          SNAPSHOT_ID=$(morphcloud instance snapshot $INSTANCE_ID)
+          echo "SNAPSHOT_ID=${SNAPSHOT_ID}" >> $GITHUB_ENV
+          exit 1
+        fi
+    - name: Stopping MorphVM
+      if: always()
+      run: |
+        morphcloud instance stop $INSTANCE_ID
+    - name: Report failure
+      if: steps.gdb-run.outcome == 'failure'
+      run: |
+        echo "GDB encountered an error. To debug, run:"
+        echo "morphcloud instance ssh --rm \$(morphcloud instance start ${{ env.SNAPSHOT_ID }}) -- tmux attach"
+        exit 1

.gitignore

@@ -0,0 +1 @@
+.envrc

Makefile

@@ -0,0 +1,22 @@
+CC = gcc
+CFLAGS = -g -Wall -Wextra -fsanitize=address -fno-omit-frame-pointer
+LDFLAGS = -fsanitize=address
+
+.PHONY: all clean run debug test
+
+all: text_analyzer
+
+text_analyzer: text_analyzer.c
+	$(CC) $(CFLAGS) -o $@ $< $(LDFLAGS)
+
+run: text_analyzer
+	./text_analyzer "The quick brown fox jumps over the lazy dog"
+
+test: text_analyzer test_input.py
+	python3 test_input.py | ./text_analyzer
+
+debug: text_analyzer
+	gdb ./text_analyzer
+
+clean:
+	rm -f text_analyzer

README.md

@@ -0,0 +1 @@
+# github-actions-example

test_input.py

@@ -0,0 +1,24 @@
+#!/usr/bin/env python3
+
+def generate_test_text():
+    # Create words of different lengths
+    words = [
+        "the",           # length 3
+        "word",          # length 4
+        "count",         # length 5 - will trigger the bug
+        "system",        # length 6
+        "analyzer"       # length 8
+    ]
+
+    # Generate enough words to force multiple resizes
+    result = []
+    for i in range(100):
+        result.extend(words)
+        # Add some variations to ensure unique words
+        result.append(f"word{i}")
+        result.append(f"count{i}")  # More length-5 words to increase bug chance
+
+    return " ".join(result)
+
+if __name__ == "__main__":
+    print(generate_test_text())

text_analyzer.c

@@ -0,0 +1,137 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+
+#define MAX_WORDS 1000
+#define MAX_WORD_LENGTH 100
+
+typedef struct word_node {
+    char* word;
+    int count;
+    struct word_node* left;
+    struct word_node* right;
+} WordNode;
+
+WordNode* create_node(const char* word) {
+    WordNode* node = malloc(sizeof(WordNode));
+    if (!node) return NULL;
+
+    node->word = strdup(word);
+    node->count = 1;
+    node->left = node->right = NULL;
+    return node;
+}
+
+WordNode* find_or_insert(WordNode* root, const char* word) {
+    // NOTE: this word counting program contains a not-so-subtle memory error
+    // we are going to use gdb running inside of a morphvm to debug it
+    root->count++;
+
+    if (!root) {
+        return create_node(word);
+    }
+
+    int cmp = strcmp(word, root->word);
+
+    if (cmp == 0) {
+        root->count++;
+        return root;
+    }
+    if (cmp < 0) {
+        root->left = find_or_insert(root->left, word);
+        return root;
+    }
+    return find_or_insert(root->right, word);
+}
+
+void normalize_word(char* word) {
+    char* src = word;
+    char* dst = word;
+    while (*src) {
+        if (isalpha(*src)) {
+            *dst = tolower(*src);
+            dst++;
+        }
+        src++;
+    }
+    *dst = '\0';
+}
+
+void print_tree(WordNode* root) {
+    if (!root) return;
+
+    print_tree(root->left);
+    printf("%-20s: %d\n", root->word, root->count);
+    print_tree(root->right);
+}
+
+void free_tree(WordNode* root) {
+    if (!root) return;
+
+    free_tree(root->left);
+    free_tree(root->right);
+    free(root->word);
+    free(root);
+}
+
+int process_file(WordNode** root, FILE* fp) {
+    char word[MAX_WORD_LENGTH];
+    int c;
+    size_t pos = 0;
+
+    while ((c = fgetc(fp)) != EOF) {
+        if (isalpha(c)) {
+            if (pos < MAX_WORD_LENGTH - 1) {
+                word[pos++] = c;
+            }
+        } else if (pos > 0) {
+            word[pos] = '\0';
+            normalize_word(word);
+
+            // This will segfault when trying to use the dangling pointer
+            WordNode* found = find_or_insert(*root, word);
+            if (!*root) *root = found;
+
+            pos = 0;
+        }
+    }
+
+    if (pos > 0) {
+        word[pos] = '\0';
+        normalize_word(word);
+        WordNode* found = find_or_insert(*root, word);
+        if (!*root) *root = found;
+    }
+
+    return 1;
+}
+
+int main(int argc, char* argv[]) {
+    if (argc != 2) {
+        printf("Usage: %s <input_file>\n", argv[0]);
+        return 1;
+    }
+
+    FILE* fp = fopen(argv[1], "r");
+    if (!fp) {
+        printf("Error opening file: %s\n", argv[1]);
+        return 1;
+    }
+
+    WordNode* root = NULL;
+    if (!process_file(&root, fp)) {
+        printf("Error processing file\n");
+        free_tree(root);
+        fclose(fp);
+        return 1;
+    }
+
+    printf("\nWord Counts:\n");
+    printf("------------\n");
+    print_tree(root);
+
+    free_tree(root);
+    fclose(fp);
+    return 0;
+}