Initial template based rendering

README.md

@@ -465,3 +465,20 @@ There are three basic strategies supported:
 * *OPTIMISTIC* With this convention, no quoting of strings is done.  This should not normally be used since it is very easy to get unparseable data.
 
 The default convention is DOUBLE_QUOTE.
+
+Template based Data Generation
+=====================
+
+This approach uses Freemarker templating engine to render custom templates. The data variables in the template are fed from a specified schema.
+
+## Command-line options:
+
+`-format TEMPLATE` to have the output format coming from the template
+
+`-template file` link to a Freemarker template
+
+`-schema file` to specify the schema (see above)
+
+## Template notation
+
+To print the value of a variable in the template, use ${name.asText()} placeholder.

pom.xml

@@ -76,6 +76,11 @@
             <artifactId>jackson-databind</artifactId>
             <version>2.1.1</version>
         </dependency>
+        <dependency>
+            <groupId>org.freemarker</groupId>
+            <artifactId>freemarker</artifactId>
+            <version>2.3.21</version>
+        </dependency>
         <dependency>
             <groupId>com.google.guava</groupId>
             <artifactId>guava</artifactId>

src/main/java/com/mapr/synth/Synth.java

@@ -24,17 +24,18 @@
 import com.google.common.base.Preconditions;
 import com.google.common.collect.Lists;
 import com.mapr.synth.samplers.SchemaSampler;
+import freemarker.template.Configuration;
+import freemarker.template.Template;
+import freemarker.template.TemplateException;
+import freemarker.template.TemplateExceptionHandler;
 import org.kohsuke.args4j.CmdLineException;
 import org.kohsuke.args4j.CmdLineParser;
 import org.kohsuke.args4j.Option;
 import org.kohsuke.args4j.OptionDef;
 import org.kohsuke.args4j.spi.IntOptionHandler;
 import org.kohsuke.args4j.spi.Setter;
 
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.io.PrintStream;
+import java.io.*;
 import java.lang.management.ManagementFactory;
 import java.lang.management.ThreadMXBean;
 import java.nio.file.Files;
@@ -64,14 +65,18 @@ public static void main(String[] args) throws IOException, CmdLineException, Int
                     "[ -count <number>G|M|K ] " +
                     "-schema schema-file " +
                     "[-quote DOUBLE_QUOTE|BACK_SLASH|OPTIMISTIC] " +
-                    "[-format JSON|TSV|CSV ] " +
+                    "[-format JSON|TSV|CSV|TEMPLATE ] " +
                     "[-threads n] " +
                     "[-output output-directory-name] ");
             throw e;
         }
+
         Preconditions.checkArgument(opts.threads > 0 && opts.threads <= 2000,
                 "Must have at least one thread and no more than 2000");
 
+        Preconditions.checkArgument(opts.format == Format.TEMPLATE && opts.template != null,
+                "Please specify a template file");
+
         if (opts.threads > 1) {
             Preconditions.checkArgument(!"-".equals(opts.output),
                     "If more than on thread is used, you have to use -output to set the output directory");
@@ -226,9 +231,30 @@ public static void header(Format format, List<String> names, PrintStream out) {
         }
 
         public static int generateFile(Options opts, SchemaSampler s, PrintStream out, int count) {
-            for (int i = 0; i < count; i++) {
-                format(opts.format, opts.quote, s.getFieldNames(), s.sample(), out);
+            if (opts.format == Format.TEMPLATE) {
+                Configuration cfg = new Configuration(Configuration.VERSION_2_3_21);
+                cfg.setDefaultEncoding("UTF-8");
+                cfg.setTemplateExceptionHandler(TemplateExceptionHandler.RETHROW_HANDLER);
+
+                Template template = null;
+                try {
+                    cfg.setDirectoryForTemplateLoading(opts.template.getParentFile());
+                    template = cfg.getTemplate(opts.template.getName());
+                } catch (IOException e) {
+                    e.printStackTrace();
+                }
+
+                PrintWriter writer = new PrintWriter(out);
+                for (int i = 0; i < count; i++) {
+                    formatTemplate(opts.format, opts.quote, s.getFieldNames(), s.sample(), template, writer);
+                }
+
+            } else {
+                for (int i = 0; i < count; i++) {
+                    format(opts.format, opts.quote, s.getFieldNames(), s.sample(), out);
+                }
             }
+
             return count;
         }
 
@@ -270,10 +296,18 @@ public ThreadReport() {
         }
     }
 
-
     static Joiner withCommas = Joiner.on(",");
     static Joiner withTabs = Joiner.on("\t");
 
+    private static void formatTemplate(Format format, Quote quoteConvention, List<String> names, JsonNode fields, Template temp, PrintWriter writer) {
+        try {
+            temp.process(fields, writer);
+        } catch (TemplateException e) {
+            e.printStackTrace();
+        } catch (IOException e) {
+            e.printStackTrace();
+        }
+    }
 
     private static void format(Format format, Quote quoteConvention, List<String> names, JsonNode fields, PrintStream out) {
         switch (format) {
@@ -309,7 +343,7 @@ private static void printDelimited(Quote quoteConvention, List<String> names, Js
     }
 
     public static enum Format {
-        JSON, TSV, CSV
+        JSON, TSV, CSV, TEMPLATE
     }
 
     public static enum Quote {
@@ -326,9 +360,12 @@ private static class Options {
         @Option(name = "-count", handler = SizeParser.class)
         int count = 1000;
 
-        @Option(name = "-schema")
+        @Option(name = "-schema", required = false)
         File schema;
 
+        @Option(name = "-template", required = false)
+        File template;
+
         @Option(name = "-format")
         Format format = Format.CSV;