Upgrade to Pro — share decks privately, control downloads, hide ads and more …

GPT + TDD: Create Doclet Parser

GPT + TDD: Create Doclet Parser

Julio Faerman

March 20, 2023
Tweet

More Decks by Julio Faerman

Other Decks in Education

Transcript

  1. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 1/29 Help

    me write a test class in Java. I will give you the requirements and you should write me the test class when i say "i'm done" Sure, please provide the requirements for the class that you want to test. First, name the test class DocletParserTest and the class under test DocletParser Got it. Please continue with the requirements for the DocletParser class and any
  2. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 2/29 specific

    methods or scenarios you want me to cover in the test class. Let the parser accept a String or a Text file Understood. Please provide the methods available in the DocletParser class for parsing a String and a Text file, and any additional requirements or scenarios you want me to cover in the test class.
  3. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 3/29 In

    the content, search for code comments. accept comments in either bash, java or xml style Great, now I understand that the DocletParser class should have methods to parse content and find comments in the bash, Java, or XML style. I'll assume the following methods for the DocletParser class: 1. parseString(String content) : This method will accept a String and parse it for comments. 2. parseFile(File file) : This method will accept a File object and parse it for comments. Here is the test class DocletParserTest : java Copy code import org.junit.jupiter.api.BeforeEach; import org.junit.jupiter.api.Test; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import static org.junit.jupiter.api.Assertions.assertEquals; public class DocletParserTest {
  4. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 4/29 private

    DocletParser parser; @BeforeEach public void setUp() { parser = new DocletParser(); } @Test public void testParseStringJava() { String content = "/* This is a Java comment */\n" + "public class HelloWorld {}"; assertEquals("/* This is a Java comment */", parser.parseString } @Test public void testParseStringBash() { String content = "# This is a Bash comment\n" + "echo \"Hello, World!\""; assertEquals("# This is a Bash comment", parser.parseString(con } @Test public void testParseStringXML() { String content = "<!-- This is an XML comment -->\n" + "<root></root>"; assertEquals("<!-- This is an XML comment -->", parser.parse
  5. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 5/29 let

    the parser return only the content of the comment, not the content delimiters Understood. I've updated the test cases in the DocletParserTest class to reflect the requirement of returning only the content of the comments, without the comment delimiters: java Copy code import org.junit.jupiter.api.BeforeEach; import org.junit.jupiter.api.Test; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import static org.junit.jupiter.api.Assertions.assertEquals;
  6. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 6/29 public

    class DocletParserTest { private DocletParser parser; @BeforeEach public void setUp() { parser = new DocletParser(); } @Test public void testParseStringJava() { String content = "/* This is a Java comment */\n" + "public class HelloWorld {}"; assertEquals("This is a Java comment", parser.parseString( } @Test public void testParseStringBash() { String content = "# This is a Bash comment\n" + "echo \"Hello, World!\""; assertEquals("This is a Bash comment", parser.parseString( } @Test public void testParseStringXML() { String content = "<!-- This is an XML comment -->\n" + "<root></root>"; assertEquals("This is an XML comment", parser.parseString( }
  7. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 7/29 //

    Add test cases for the parseFile method with different file } Please note that you'll need to implement the parseString and parseFile add a method where there is no comment at all I added a new test case to the DocletParserTest class to test the scenario where there is no comment in the content. The expected result should be an empty string or null depending on your implementation: java Copy code
  8. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 8/29 import

    org.junit.jupiter.api.BeforeEach; import org.junit.jupiter.api.Test; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import static org.junit.jupiter.api.Assertions.assertEquals; public class DocletParserTest { private DocletParser parser; @BeforeEach public void setUp() { parser = new DocletParser(); } @Test public void testParseStringJava() { String content = "/* This is a Java comment */\n" + "public class HelloWorld {}"; assertEquals("This is a Java comment", parser.parseString( } @Test public void testParseStringBash() { String content = "# This is a Bash comment\n" + "echo \"Hello, World!\""; assertEquals("This is a Bash comment", parser.parseString(
  9. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 9/29 }

    @Test public void testParseStringXML() { String content = "<!-- This is an XML comment -->\n" + "<root></root>"; assertEquals("This is an XML comment", parser.parseString( } @ consider a doclet to be a instruction inside comments in the format "$key: $value" where $key must be a single word and $value is the remainder of the comment. the key may be preceded by spaces. create a new record to represent a doclet. make the Parser return List<Doclet> containing the doclet, if found, or empty
  10. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 10/29 list

    if not found. for example in the string "# assert: something else" the key would be "assert" and the value "something else" Based on your requirements, I have created a Doclet record to represent the doclet and updated the DocletParserTest class to include tests for parsing doclets inside comments. The DocletParser class is now expected to return a list of Doclet objects, or an empty list if no doclet is found. Here's the updated test class: java Copy code import org.junit.jupiter.api.BeforeEach; import org.junit.jupiter.api.Test; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import java.util.List; import static org.junit.jupiter.api.Assertions.assertEquals; public class DocletParserTest { private DocletParser parser; @BeforeEach public void setUp() {
  11. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 11/29 parser

    = new DocletParser(); } @Test public void testParseStringJava() { String content = "/* $assert: This is a Java doclet */\n" "public class HelloWorld {}"; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Java doclet", doc Use CDI, adding @ApplicationScoped and @Inject instead of @BeforeEach
  12. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 12/29 I've

    updated the DocletParserTest the key does not need to be preceded by $ I've updated the DocletParserTest class to consider the key without the $ symbol. Here's the revised test class using CDI with @ApplicationScoped and @Inject annotations: java Copy code import org.junit.jupiter.api.Test; import javax.inject.Inject; import javax.enterprise.context.ApplicationScoped; import java.io.File; import java.io.IOException; import java.nio.file.Files;
  13. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 13/29 import

    java.nio.file.Path; import java.util.List; import static org.junit.jupiter.api.Assertions.assertEquals; @ApplicationScoped public class DocletParserTest { @Inject private DocletParser parser; @Test public void testParseStringJava() { String content = "/* assert: This is a Java doclet */\n" "public class HelloWorld {}"; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Java doclet", doclets.get(0).valu } @Test public void testParseStringBash() { String content = "# assert: This is a Bash doclet\n" + "echo \"Hello, World!\""; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Bash doclet", doclets.get(0).valu
  14. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 14/29 }

    @Test public void testParseStringXML() { String content = "<!-- assert: This is an XML doclet -->\ "<root></root>"; List<Doclet> doclets = parser.parseString(content); first write me the test declaration, then each method separately Here's the test class declaration with CDI annotations and the injected DocletParser :
  15. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 15/29 java

    Copy code import org.junit.jupiter.api.Test; import javax.inject.Inject; import javax.enterprise.context.ApplicationScoped; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import java.util.List; import static org.junit.jupiter.api.Assertions.assertEquals; @ApplicationScoped public class DocletParserTest { @Inject private DocletParser parser; } Now, I'll provide each test method separately. 1. Test for Java-style comments: java Copy code
  16. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 16/29 @Test

    public void testParseStringJava() { String content = "/* assert: This is a Java doclet */\n" + "public class HelloWorld {}"; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Java doclet", doclets.get(0).value()) } 2. Test for Bash-style comments: java Copy code @Test public void testParseStringBash() { String content = "# assert: This is a Bash doclet\n" + "echo \"Hello, World!\""; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This
  17. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 18/29 show

    me the Doclet record Here's the Doclet record that represents a doclet with a key and a value: java Copy code public record Doclet(String key, String value) { } This record will store the key and value of a doclet found in the comments. The DocletParser class is expected to return a list of Doclet objects when parsing the content.
  18. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 19/29 now

    testParseStringBash please Here's the test method for Bash-style comments: java Copy code @Test public void testParseStringBash() { String content = "# assert: This is a Bash doclet\n" + "echo \"Hello, World!\""; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Bash doclet", doclets.get(0).value()) }
  19. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 20/29 This

    test method checks whether the DocletParser correctly parses a Bash-style comment containing a doclet and returns a list with a single Doclet object having the expected key and value. now the testParseStringXml and the testParseStringWithoutDoclets Here's the test method for XML-style comments: java Copy code @Test public void testParseStringXML() { String content = "<!-- assert: This is an XML doclet -->\n" + "<root></root>"; List<Doclet> doclets = parser.parseString(content); assertEquals(1, doclets.size());
  20. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 21/29 assertEquals("assert",

    doclets.get(0).key()); assertEquals("This is an XML doclet", doclets.get(0).value()) } And here's the test method for a scenario without any doclets: java Copy code @Test public void testParseStringWithoutDoclets() { String content = "/* This is a regular Java comment */\n" + "public class HelloWorld {}"; List<Doclet> doclets = parser.parseString(content); assertEquals(0, doclets.size()); } These test methods check whether the DocletParser correctly parses an XML-style comment containing a doclet and a string without any doclets, returning the expected Doclet objects or an empty list, respectively.
  21. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 22/29 now

    please implement the parser class that would pass this test: package doclet; import javax.inject.Inject; import javax.enterprise.context.ApplicationScoped; import java.io.File java.nio.file.Files; import java.nio.file.Path; import java.util.List; import static org.junit.jupiter.ap @ApplicationScoped public class DocletParserTest { @Inject private DocletParser parser; @Te String content = "/* assert: This is a Java doclet */\n" + "public class HelloWorld {}"; List<Docle parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.ge Java doclet", doclets.get(0).value()); } @Test public void testParseStringBash() { String conten + "echo \"Hello, World!\""; List<Doclet> doclets = parser.parseString(content); assertEquals(1 assertEquals("assert", doclets.get(0).key()); assertEquals("This is a Bash doclet", doclets.get testParseStringXML() { String content = "<!-- assert: This is an XML doclet -->\n" + "<root></ parser.parseString(content); assertEquals(1, doclets.size()); assertEquals("assert", doclets.ge XML doclet", doclets.get(0).value()); } @Test public void testParseStringWithoutDoclets() { Str Java comment */\n" + "public class HelloWorld {}"; List<Doclet> doclets = parser.parseString(c doclets.size()); } } Here's the implementation of the DocletParser class that should pass the provided te
  22. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 23/29 java

    Copy code package doclet; import javax.enterprise.context.ApplicationScoped; import java.io.File; import java.io.IOException; import java.nio.file.Files; import java.util.ArrayList; import java.util.List; import java.util.regex.Matcher; import java.util.regex.Pattern; @ApplicationScoped public class DocletParser { private static final Pattern JAVA_COMMENT_PATTERN = Pattern.compil private static final Pattern BASH_COMMENT_PATTERN = Pattern.compil private static final Pattern XML_COMMENT_PATTERN = Pattern.compile private static final Pattern DOCLET_PATTERN = Pattern.compile("^\\ public List<Doclet> parseString(String content) { List<Doclet> doclets = new ArrayList<>(); Matcher javaMatcher = JAVA_COMMENT_PATTERN.matcher(content); Matcher bashMatcher = BASH_COMMENT_PATTERN.matcher(content); Matcher xmlMatcher = XML_COMMENT_PATTERN.matcher(content);
  23. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 24/29 extractDoclets(javaMatcher,

    doclets use var declarations where possible Here's the updated implementation of the DocletParser class using var declarations java Copy code package doclet; import javax.enterprise.context.ApplicationScoped; import java.io.File; import java.io.IOException; import java.nio.file.Files;
  24. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 25/29 import

    java.util.ArrayList; import java.util.List; import java.util.regex.Matcher; import java.util.regex.Pattern; @ApplicationScoped public class DocletParser { private static final Pattern JAVA_COMMENT_PATTERN = Pattern.compil private static final Pattern BASH_COMMENT_PATTERN = Pattern.compil private static final Pattern XML_COMMENT_PATTERN = Pattern.compile private static final Pattern DOCLET_PATTERN = Pattern.compile("^\\ public List<Doclet> parseString(String content) { var doclets = new ArrayList<Doclet>(); var javaMatcher = JAVA_COMMENT_PATTERN.matcher(content); var bashMatcher = BASH_COMMENT_PATTERN.matcher(content); var xmlMatcher = XML_COMMENT_PATTERN.matcher(content); extractDoclets(javaMatcher, doclets); extractDoclets(bashMatcher, doclets); extractDoclets(xmlMatcher, doclets); return doclets; } private void extractDoclets(Matcher matcher, List<Doclet> doclets)
  25. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 26/29 while

    (matcher.find()) { var comment = matcher.group(1); var docletMatcher = DOCLET_PATTERN.matcher(comment); while (docletMatcher.find()) { var key = docletMatcher.group(1); var value = docletMatcher.group(2); doc write only the extractDoclets method Here's the extractDoclets method in isolation: java Copy code
  26. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 27/29 private

    void extractDoclets(Matcher matcher, List<Doclet> doclets while (matcher.find()) { var comment = matcher.group(1); var docletMatcher = DOCLET_PATTERN.matcher(comment); while (docletMatcher.find()) { var key = docletMatcher.group(1); var value = docletMatcher.group(2); doclets.add(new Doclet(key, value)); } } } This method takes a Matcher object that matches comments and a List<Doclet> object to store the extracted doclets. It iterates through the matched comments, finds the doclets within the comments, and adds them to the provided list.
  27. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 28/29 thanks!

    how can i share this conversation? You're welcome! To share this conversation, you can simply copy the URL from your web browser's address bar and share it with whoever you'd like to share it with. This will allow others to view the entire conversation between us. print me that url I'm sorry, as an AI language model, I don't have the capability to access your web browser's address bar or any external services. However, you can simply copy the URL from the address bar at the top of your web browser to share this conversation.
  28. 20/03/2023, 12:03 Export ChatGPT Conversation As Images https://gptsave.xyz/images?id=9d1bf121-e14a-48af-ba1b-08e65f660055 29/29 Unable

    to load conversation 4d802020-7cd2-4e01-9d9b-3199807fc3f1 I apologize for the inconvenience. As an AI language model, I don't have the capability to store conversations or access past conversations. Once the conversation is completed, it cannot be accessed again. If you have any other questions or concerns, feel free to ask! Buy me a coffee © 2023 GPTSave, all rights reserved