drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From paul-rogers <...@git.apache.org>
Subject [GitHub] drill pull request #574: DRILL-4726: Dynamic UDFs support
Date Tue, 06 Sep 2016 23:38:21 GMT
Github user paul-rogers commented on a diff in the pull request:

    https://github.com/apache/drill/pull/574#discussion_r77738257
  
    --- Diff: exec/java-exec/src/main/java/org/apache/drill/exec/expr/fn/DrillFunctionRegistry.java
---
    @@ -64,62 +76,134 @@
           .put("CONVERT_FROM", Pair.of(2, 2))
           .put("FLATTEN", Pair.of(1, 1)).build();
     
    +  /** Registers all functions present in Drill classpath on start-up. All functions will
be marked as built-in.*/
       public DrillFunctionRegistry(ScanResult classpathScan) {
    +    validate(BUILT_IN, classpathScan);
    +    register(BUILT_IN, classpathScan, this.getClass().getClassLoader());
    +    if (logger.isTraceEnabled()) {
    +      StringBuilder allFunctions = new StringBuilder();
    +      for (DrillFuncHolder method: registryHolder.getAllFunctionsWithHolders().values())
{
    +        allFunctions.append(method.toString()).append("\n");
    +      }
    +      logger.trace("Registered functions: [\n{}]", allFunctions);
    +    }
    +  }
    +
    +  /**
    +   * Validates all functions, present in jars.
    +   * Will throw {@link FunctionValidationException} if:
    +   * 1. Jar with the same name has been already registered.
    +   * 2. Conflicting function with the similar signature is found.
    +   * 3. Aggregating function is not deterministic.
    +   *
    +   * @return list of validated functions
    +   */
    +  public List<Func> validate(String jarName, ScanResult classpathScan) {
    +    List<Func> functions = Lists.newArrayList();
         FunctionConverter converter = new FunctionConverter();
         List<AnnotatedClassDescriptor> providerClasses = classpathScan.getAnnotatedClasses();
     
    -    // Hash map to prevent registering functions with exactly matching signatures
    -    // key: Function Name + Input's Major Type
    -    // value: Class name where function is implemented
    -    //
    -    final Map<String, String> functionSignatureMap = new HashMap<>();
    +    if (registryHolder.containsJar(jarName)) {
    +      throw new FunctionValidationException(String.format("Jar %s is already registered",
jarName));
    +    }
    +
    +    final ListMultimap<String, String> allFuncWithSignatures = registryHolder.getAllFunctionsWithSignatures();
    +
         for (AnnotatedClassDescriptor func : providerClasses) {
    -      DrillFuncHolder holder = converter.getHolder(func);
    +      DrillFuncHolder holder = converter.getHolder(func, ClassLoader.getSystemClassLoader());
           if (holder != null) {
    -        // register handle for each name the function can be referred to
    -        String[] names = holder.getRegisteredNames();
     
    -        // Create the string for input types
             String functionInput = "";
    +        List<MajorType> types = Lists.newArrayList();
             for (DrillFuncHolder.ValueReference ref : holder.parameters) {
               functionInput += ref.getType().toString();
    +          types.add(ref.getType());
             }
    +
    +        String[] names = holder.getRegisteredNames();
             for (String name : names) {
               String functionName = name.toLowerCase();
    -          registeredFunctions.put(functionName, holder);
               String functionSignature = functionName + functionInput;
    -          String existingImplementation;
    -          if ((existingImplementation = functionSignatureMap.get(functionSignature))
!= null) {
    -            throw new AssertionError(
    -                String.format(
    -                    "Conflicting functions with similar signature found. Func Name: %s,
Class name: %s " +
    -                " Class name: %s", functionName, func.getClassName(), existingImplementation));
    -          } else if (holder.isAggregating() && !holder.isDeterministic() ) {
    -            logger.warn("Aggregate functions must be deterministic, did not register
function {}", func.getClassName());
    +
    +          if (allFuncWithSignatures.get(functionName).contains(functionSignature)) {
    +            throw new FunctionValidationException(
    +                String.format("Conflicting function with similar signature found. " +
    +                        "Function name: %s, class name: %s, input parameters : %s",
    +                    functionName, func.getClassName(), functionInput));
    +          } else if (holder.isAggregating() && !holder.isDeterministic()) {
    +            throw new FunctionValidationException(
    +                String.format("Aggregate functions must be deterministic: %s", func.getClassName()));
               } else {
    -            functionSignatureMap.put(functionSignature, func.getClassName());
    +            functions.add(Func.newBuilder().setName(functionName).addAllMajorType(types).build());
    +            allFuncWithSignatures.put(functionName, functionSignature);
               }
             }
           } else {
             logger.warn("Unable to initialize function for class {}", func.getClassName());
           }
         }
    -    if (logger.isTraceEnabled()) {
    -      StringBuilder allFunctions = new StringBuilder();
    -      for (DrillFuncHolder method: registeredFunctions.values()) {
    -        allFunctions.append(method.toString()).append("\n");
    +    return functions;
    +  }
    +
    +  /**
    +   * Registers all functions present in jar.
    +   * If jar name is already registered, all jar related functions are overridden.
    +   */
    +  public void register(String jarName, ScanResult classpathScan, ClassLoader classloader)
{
    +    FunctionConverter converter = new FunctionConverter();
    +    List<AnnotatedClassDescriptor> providerClasses = classpathScan.getAnnotatedClasses();
    +    Map<String, Pair<String, DrillFuncHolder>> functions = Maps.newHashMap();
    +    for (AnnotatedClassDescriptor func : providerClasses) {
    +      DrillFuncHolder holder = converter.getHolder(func, classloader);
    +      if (holder != null) {
    +        String functionInput = "";
    +        for (DrillFuncHolder.ValueReference ref : holder.parameters) {
    +          functionInput += ref.getType().toString();
    --- End diff --
    
    Looks like the functionInput is of the form MAPTINYINTUINT1. This will be hard to debug,
but may also introduce ambiguity if we ever have types such as DATE, TIME, DATETIME as "DATETIME"
could be (DATE, TIME) or "(DATETIME)" Simple solution: insert a delimiter such as a comma:
"DATE,TIME" for the two-argument first case, "DATETIME" for the one-argument second case.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message