search service

Dockerfile

@@ -0,0 +1,15 @@
+FROM debian:stable-slim
+RUN apt-get update
+RUN apt-get -y upgrade
+RUN mkdir /upload /usr/share/man/man1 /usr/share/man/man8
+RUN apt-get -y install --no-install-recommends apt apt-transport-https apt-utils readline-common curl gnupg software-properties-common dirmngr openjdk-8-jdk procps
+RUN echo "deb https://artifacts.elastic.co/packages/6.x/apt stable main" > /etc/apt/sources.list.d/elastic-6.x.list
+RUN apt-key adv --recv-keys D27D666CD88E42B4
+RUN apt-get update
+RUN apt-get -y install --no-install-recommends maven tesseract-ocr tesseract-ocr-spa elasticsearch git
+RUN sed -i "s/#cluster.name: my-application/cluster.name: elasticsearch/" /etc/elasticsearch/elasticsearch.yml
+RUN git clone https://gitlab.com/manalejandro/arjion
+RUN mvn clean install -f /arjion/pom.xml
+RUN echo "/etc/init.d/elasticsearch start && mvn spring-boot:run -f /arjion/pom.xml" > entrypoint.sh
+EXPOSE 8080:8080
+ENTRYPOINT ["bash", "entrypoint.sh"]

README.md

@@ -4,6 +4,11 @@
 
 ### Proof of Concept with [SpringBoot 2.1.0](https://start.spring.io/), [ElasticSearch](https://www.elastic.co/) and [Apache Tika](https://tika.apache.org/)
 
+## Docker image
+
+    $ docker build -t debian:arjion --rm https://gitlab.com/manalejandro/arjion/raw/master/Dockerfile
+    $ docker run -ti -p 8080:8080 debian:arjion
+
 ## License
 
 MIT

pom.xml

@@ -6,7 +6,7 @@
 
 	<groupId>com.manalejandro</groupId>
 	<artifactId>arjion</artifactId>
-	<version>0.1.0-SNAPSHOT</version>
+	<version>0.2.0-SNAPSHOT</version>
 	<packaging>war</packaging>
 
 	<name>arjion</name>
@@ -52,6 +52,31 @@
             <version>1.18</version>
         </dependency>
 
+        <dependency>
+            <groupId>org.apache.tika</groupId>
+            <artifactId>tika-parsers</artifactId>
+            <version>1.18</version>
+        </dependency>
+
+        <!-- https://mvnrepository.com/artifact/com.github.jai-imageio/jai-imageio-jpeg2000 -->
+        <dependency>
+            <groupId>com.github.jai-imageio</groupId>
+            <artifactId>jai-imageio-jpeg2000</artifactId>
+            <version>1.3.0</version>
+        </dependency>
+
+        <dependency>
+            <groupId>com.levigo.jbig2</groupId>
+            <artifactId>levigo-jbig2-imageio</artifactId>
+            <version>2.0</version>
+        </dependency>
+
+        <dependency>
+            <groupId>com.github.jai-imageio</groupId>
+            <artifactId>jai-imageio-core</artifactId>
+            <version>1.4.0</version>
+        </dependency>
+        
 		<dependency>
 			<groupId>org.webjars</groupId>
 			<artifactId>bootstrap</artifactId>

src/main/java/com/manalejandro/arjion/controllers/MainController.java

@@ -1,12 +1,23 @@
 package com.manalejandro.arjion.controllers;
 
+import java.io.File;
 import java.io.IOException;
+import java.net.MalformedURLException;
+import java.net.URLEncoder;
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.nio.file.Paths;
-import java.util.ArrayList;
+import java.text.Normalizer;
+import java.util.HashMap;
+import java.util.Map;
 
+import javax.servlet.http.HttpServletResponse;
+
+import com.fasterxml.jackson.databind.ObjectMapper;
 import com.manalejandro.arjion.model.Archivo;
+import com.manalejandro.arjion.model.Documento;
+import com.manalejandro.arjion.services.MainService;
+import com.manalejandro.arjion.vo.DetailVO;
 import com.manalejandro.arjion.vo.DocumentoVO;
 
 import org.apache.tika.config.TikaConfig;
@@ -16,8 +27,15 @@ import org.apache.tika.language.LanguageIdentifier;
 import org.apache.tika.metadata.Metadata;
 import org.apache.tika.parser.AutoDetectParser;
 import org.apache.tika.parser.ParseContext;
+import org.apache.tika.parser.Parser;
+import org.apache.tika.parser.ocr.TesseractOCRConfig;
+import org.apache.tika.parser.pdf.PDFParserConfig;
 import org.apache.tika.sax.BodyContentHandler;
+import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.beans.factory.annotation.Value;
+import org.springframework.core.io.ByteArrayResource;
+import org.springframework.http.HttpHeaders;
+import org.springframework.http.ResponseEntity;
 import org.springframework.stereotype.Controller;
 import org.springframework.ui.Model;
 import org.springframework.web.bind.annotation.GetMapping;
@@ -31,12 +49,27 @@ import org.xml.sax.SAXException;
 @Controller
 public class MainController {
 
+    private final MainService mainService;
+
     @Value("${arjion.uploadpath}")
     private String uploadpath;
 
+    @Value("${arjion.tesseractpath}")
+    private String tesseractpath;
+
+    @Value("${arjion.tesseractdatapath}")
+    private String tesseractdatapath;
+
+    @Autowired
+    public MainController(MainService mainService) {
+        this.mainService = mainService;
+    }
+
     @RequestMapping(path = "/")
     public String indexPage(final Model model) {
         DocumentoVO documentoVO = new DocumentoVO();
+        documentoVO.setCount(mainService.count());
+        documentoVO.setDocumentos(mainService.findAllDocumento());
         model.addAttribute("documentoVO", documentoVO);
         return "index";
     }
@@ -50,22 +83,83 @@ public class MainController {
     public String uploadPage(final Model model, @RequestParam("archivos") MultipartFile[] archivos)
             throws IOException, TikaException, SAXException {
         DocumentoVO documentoVO = new DocumentoVO();
-        documentoVO.setArchivos(new ArrayList<Archivo>());
-        TikaConfig tikaConfig = TikaConfig.getDefaultConfig();
-        for (int i = 0; i < archivos.length; i++) {
-            byte[] bytes = archivos[i].getBytes();
-            Path path = Paths.get(uploadpath + archivos[i].getOriginalFilename());
-            Files.write(path, bytes);
-            Metadata metadata = new Metadata();
-            AutoDetectParser parser = new AutoDetectParser(tikaConfig);
-            ContentHandler handler = new BodyContentHandler(-1);
-            TikaInputStream stream = TikaInputStream.get(bytes);
-            parser.parse(stream, handler, metadata, new ParseContext());
-            LanguageIdentifier identifier = new LanguageIdentifier(handler.toString());
-            documentoVO.getArchivos().add(new Archivo(archivos[i].getOriginalFilename(), String.valueOf(archivos[i].getSize()),
-                    metadata.toString(), handler.toString(), identifier.getLanguage()));
+        documentoVO.setCount(mainService.count());
+        documentoVO.setDocumentos(mainService.findAllDocumento());
+        if (archivos.length > 0) {
+            // Recupera la configuración de Tika
+            TikaConfig tikaConfig = TikaConfig.getDefaultConfig();
+            // Itera los archivos recibidos
+            for (int i = 0; i < archivos.length; i++) {
+                byte[] bytes = archivos[i].getBytes();
+                // Normaliza el título de los archivos
+                String normalized = Normalizer.normalize(archivos[i].getOriginalFilename(), Normalizer.Form.NFD),
+                        filename = normalized.replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
+                Path path = Paths.get(uploadpath + filename);
+                // Instancias necesarias
+                Metadata metadata = new Metadata();
+                Parser parser = new AutoDetectParser(tikaConfig);
+                PDFParserConfig pdfConfig = new PDFParserConfig();
+                TesseractOCRConfig tesseractConfig = new TesseractOCRConfig();
+                tesseractConfig.setTesseractPath(tesseractpath);
+                tesseractConfig.setTessdataPath(tesseractdatapath);
+                tesseractConfig.setLanguage("spa+eng");
+                pdfConfig.setExtractInlineImages(true);
+                ParseContext parseContext = new ParseContext();
+                parseContext.set(TesseractOCRConfig.class, tesseractConfig);
+                parseContext.set(PDFParserConfig.class, pdfConfig);
+                // Usa -1 para no tener límite de 100000 chars
+                ContentHandler handler = new BodyContentHandler(-1);
+                // Castea los bytes al Stream de Tika
+                TikaInputStream stream = TikaInputStream.get(bytes);
+                // Parsea el contenido
+                parser.parse(stream, handler, metadata, parseContext);
+                // Identifica el idioma del archivo
+                LanguageIdentifier identifier = new LanguageIdentifier(handler.toString());
+                // Almacena en elasticsearch
+                String[] names = metadata.names();
+                Map<String, String> meta = new HashMap<String, String>();
+                for (int j = 0; j < names.length; j++) {
+                    meta.put(names[j], metadata.get(names[j]));
+                }
+                ObjectMapper mapper = new ObjectMapper();
+                if (!mainService.save(new Documento(filename, Long.valueOf(archivos[i].getSize()).intValue(),
+                        mapper.valueToTree(meta), handler.toString(), identifier.getLanguage()))) {
+                    return "exists";
+                } else {
+                    // Guarda el archivo en el directorio configurado en las properties
+                    Files.write(path, bytes);
+                }
+                // Añade los parámetros al VO para mostrar en la vista
+                documentoVO.getArchivos().add(new Archivo(filename, Long.valueOf(archivos[i].getSize()).intValue(), meta,
+                        handler.toString(), identifier.getLanguage()));
+            }
         }
         model.addAttribute("documentoVO", documentoVO);
         return "index";
     }
-}
+
+    @GetMapping(path = "/detail")
+    public String detail(final Model model, @RequestParam(value = "nombre", required = true) String nombre) {
+        DetailVO detailVO = new DetailVO();
+        ObjectMapper mapper = new ObjectMapper();
+        Documento doc = mainService.findOne(nombre);
+        detailVO.setArchivo(new Archivo(doc.getNombre(), doc.getTamano(),
+                mapper.convertValue(doc.getMetadata(), Map.class), doc.getContenido(), doc.getLenguaje()));
+        model.addAttribute("detailVO", detailVO);
+        return "detail";
+    }
+
+    @GetMapping(path = "/download")
+    public ResponseEntity<ByteArrayResource> download(final HttpServletResponse response,
+            @RequestParam(value = "filename", required = true) String filename)
+            throws IOException, MalformedURLException {
+        File file = new File(uploadpath + filename);
+        Path path = Paths.get(file.getAbsolutePath());
+        ByteArrayResource resource = new ByteArrayResource(Files.readAllBytes(path));
+        String type = file.toURL().openConnection().guessContentTypeFromName(filename);
+        HttpHeaders responseHeaders = new HttpHeaders();
+        responseHeaders.add("Content-Disposition", "attachment; filename=" + URLEncoder.encode(filename, "UTF-8"));
+        responseHeaders.add("Content-Type", type);
+        return ResponseEntity.ok().contentLength(file.length()).headers(responseHeaders).body(resource);
+    }
+}

src/main/java/com/manalejandro/arjion/model/Archivo.java

@@ -1,14 +1,16 @@
 package com.manalejandro.arjion.model;
 
+import java.util.Map;
+
 public class Archivo {
 
     private String nombre;
-    private String tamano;
-    private String metadata;
+    private Integer tamano;
+    private Map metadata;
     private String contenido;
     private String lenguaje;
 
-    public Archivo(String nombre, String tamano, String metadata, String contenido, String lenguaje) {
+    public Archivo(String nombre, Integer tamano, Map metadata, String contenido, String lenguaje) {
         this.nombre = nombre;
         this.tamano = tamano;
         this.metadata = metadata;
@@ -26,14 +28,14 @@ public class Archivo {
     /**
      * @return the tamano
      */
-    public String getTamano() {
+    public Integer getTamano() {
         return tamano;
     }
 
     /**
      * @return the metadata
      */
-    public String getMetadata() {
+    public Map getMetadata() {
         return metadata;
     }
 
@@ -54,14 +56,14 @@ public class Archivo {
     /**
      * @param tamano the tamano to set
      */
-    public void setTamano(String tamano) {
+    public void setTamano(Integer tamano) {
         this.tamano = tamano;
     }
 
     /**
      * @param metadata the metadata to set
      */
-    public void setMetadata(String metadata) {
+    public void setMetadata(Map metadata) {
         this.metadata = metadata;
     }
 

src/main/java/com/manalejandro/arjion/model/Consulta.java

@@ -0,0 +1,52 @@
+package com.manalejandro.arjion.model;
+
+import java.util.ArrayList;
+import java.util.List;
+
+public class Consulta {
+    private List<Documento> documentos = new ArrayList<Documento>();
+    private String suggest;
+    private List<String> autocomplete = new ArrayList<String>();
+
+    /**
+     * @return the documentos
+     */
+    public List<Documento> getDocumentos() {
+        return documentos;
+    }
+
+    /**
+     * @return the suggest
+     */
+    public String getSuggest() {
+        return suggest;
+    }
+
+    /**
+     * @return the autocomplete
+     */
+    public List<String> getAutocomplete() {
+        return autocomplete;
+    }
+
+    /**
+     * @param documentos the documentos to set
+     */
+    public void setDocumentos(List<Documento> documentos) {
+        this.documentos = documentos;
+    }
+
+    /**
+     * @param suggest the suggest to set
+     */
+    public void setSuggest(String suggest) {
+        this.suggest = suggest;
+    }
+
+    /**
+     * @param autocomplete the autocomplete to set
+     */
+    public void setAutocomplete(List<String> autocomplete) {
+        this.autocomplete = autocomplete;
+    }
+}

src/main/java/com/manalejandro/arjion/model/Documento.java

@@ -1,32 +1,108 @@
 package com.manalejandro.arjion.model;
 
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.databind.JsonNode;
+
 import org.springframework.data.annotation.Id;
 import org.springframework.data.elasticsearch.annotations.Document;
 import org.springframework.data.elasticsearch.annotations.Mapping;
 import org.springframework.data.elasticsearch.annotations.Setting;
 
-import com.fasterxml.jackson.annotation.JsonCreator;
-import com.fasterxml.jackson.annotation.JsonProperty;
-
 @Document(indexName = "#{@indexName}", type = "#{@documentType}")
 @Setting(settingPath = "/elasticsearch/settings.json")
 @Mapping(mappingPath = "/elasticsearch/mapping.json")
 public class Documento {
-	@Id
-	public Integer id;
+    @Id
+    public String nombre;
+    public Integer tamano;
+    public JsonNode metadata;
+    public String contenido;
+    public String lenguaje;
 
-	@JsonCreator
-	public Documento(@JsonProperty("id") Integer id) {
-		super();
-		this.id = id;
-	}
+    @JsonCreator
+    public Documento(@JsonProperty("nombre") String nombre, @JsonProperty("tamano") Integer tamano,
+            @JsonProperty("metadata") JsonNode metadata, @JsonProperty("contenido") String contenido,
+            @JsonProperty("lenguaje") String lenguaje) {
+        this.nombre = nombre;
+        this.tamano = tamano;
+        this.metadata = metadata;
+        this.contenido = contenido;
+        this.lenguaje = lenguaje;
+    }
 
-	@JsonProperty("id")
-	public Integer getId() {
-		return id;
-	}
+    /**
+     * @return the nombre
+     */
+    @JsonProperty("nombre")
+    public String getNombre() {
+        return nombre;
+    }
 
-	public void setId(Integer id) {
-		this.id = id;
-	}
+    /**
+     * @param nombre the nombre to set
+     */
+    public void setNombre(String nombre) {
+        this.nombre = nombre;
+    }
+
+    /**
+     * @return the tamano
+     */
+    @JsonProperty("tamano")
+    public Integer getTamano() {
+        return tamano;
+    }
+
+    /**
+     * @param tamano the tamano to set
+     */
+    public void setTamano(Integer tamano) {
+        this.tamano = tamano;
+    }
+
+    /**
+     * @return the metadata
+     */
+    @JsonProperty("metadata")
+    public JsonNode getMetadata() {
+        return metadata;
+    }
+
+    /**
+     * @param metadata the metadata to set
+     */
+    public void setMetadata(JsonNode metadata) {
+        this.metadata = metadata;
+    }
+
+    /**
+     * @return the contenido
+     */
+    @JsonProperty("contenido")
+    public String getContenido() {
+        return contenido;
+    }
+
+    /**
+     * @param contenido the contenido to set
+     */
+    public void setContenido(String contenido) {
+        this.contenido = contenido;
+    }
+
+    /**
+     * @return the lenguaje
+     */
+    @JsonProperty("lenguaje")
+    public String getLenguaje() {
+        return lenguaje;
+    }
+
+    /**
+     * @param lenguaje the lenguaje to set
+     */
+    public void setLenguaje(String lenguaje) {
+        this.lenguaje = lenguaje;
+    }
 }

src/main/java/com/manalejandro/arjion/services/MainService.java

@@ -1,8 +1,23 @@
 package com.manalejandro.arjion.services;
 
-import org.springframework.stereotype.Service;
+import java.util.List;
+
+import com.manalejandro.arjion.model.Consulta;
+import com.manalejandro.arjion.model.Documento;
+
+import org.springframework.data.domain.Pageable;
 
-@Service
 public interface MainService {
 
+    public boolean save(Documento doc);
+
+    public long count();
+
+    public List<Documento> findAllDocumento();
+
+    public Documento findOne(String nombre);
+
+    public Integer maxTamano();
+
+    public Consulta search(String busqueda, String[] tipo, Integer tamano, Pageable pageable);
 }

src/main/java/com/manalejandro/arjion/services/MainServiceImpl.java

@@ -1,5 +1,127 @@
 package com.manalejandro.arjion.services;
 
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.List;
+
+import com.fasterxml.jackson.databind.ObjectMapper;
+import com.manalejandro.arjion.model.Consulta;
+import com.manalejandro.arjion.model.Documento;
+import com.manalejandro.arjion.repositories.MainRepository;
+
+import org.elasticsearch.action.search.SearchResponse;
+import org.elasticsearch.client.Client;
+import org.elasticsearch.index.query.BoolQueryBuilder;
+import org.elasticsearch.index.query.QueryBuilders;
+import org.elasticsearch.search.aggregations.AggregationBuilder;
+import org.elasticsearch.search.aggregations.AggregationBuilders;
+import org.elasticsearch.search.suggest.Suggest.Suggestion.Entry;
+import org.elasticsearch.search.suggest.Suggest.Suggestion.Entry.Option;
+import org.elasticsearch.search.suggest.SuggestBuilder;
+import org.elasticsearch.search.suggest.SuggestBuilders;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.beans.factory.annotation.Value;
+import org.springframework.context.ApplicationContext;
+import org.springframework.data.domain.Pageable;
+import org.springframework.data.domain.Sort;
+import org.springframework.stereotype.Service;
+
+@Service
 public class MainServiceImpl implements MainService {
 
+    private final ApplicationContext appContext;
+    private final MainRepository mainRepository;
+
+    @Value("#{@indexName}")
+    private String index;
+    @Value("#{@documentType}")
+    private String document;
+
+    @Autowired
+    public MainServiceImpl(MainRepository mainRepository, ApplicationContext appContext) {
+        this.mainRepository = mainRepository;
+        this.appContext = appContext;
+    }
+
+    @Override
+    public boolean save(Documento doc) {
+        if (!mainRepository.existsById(doc.nombre)) {
+            if (mainRepository.save(doc) != null)
+                return true;
+            else
+                return false;
+        } else
+            return false;
+    }
+
+    @Override
+    public long count() {
+        return mainRepository.count();
+    }
+
+    @Override
+    public List<Documento> findAllDocumento() {
+        List<Documento> docs = new ArrayList<Documento>();
+        mainRepository.findAll().forEach(doc -> {
+            docs.add((Documento) doc);
+        });
+        return docs;
+    }
+
+    @Override
+    public Documento findOne(String nombre) {
+        return mainRepository.findById(nombre).get();
+    }
+
+    @Override
+    public Integer maxTamano() {
+        return mainRepository.findAll(new Sort(Sort.Direction.DESC, "tamano")).iterator().next().getTamano();
+    }
+
+    @Override
+    public Consulta search(String busqueda, String[] tipo, Integer tamano, Pageable pageable) {
+        Client client = (Client) appContext.getBean("client");
+        BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
+        if (busqueda != null && !"null".equals(busqueda) && !busqueda.isEmpty()) {
+            boolQueryBuilder.must(QueryBuilders.matchQuery("nombre", busqueda));
+            boolQueryBuilder.should(QueryBuilders.matchQuery("contenido", busqueda));
+        }
+        if (tipo != null && tipo.length > 0)
+            boolQueryBuilder.filter(QueryBuilders.termsQuery("tipo", tipo));
+        if (tamano != null && tamano >= 0)
+            boolQueryBuilder.must(QueryBuilders.rangeQuery("tamano").to(tamano).includeUpper(true));
+        AggregationBuilder aggregation = AggregationBuilders.terms("by_xarchivo").field("x_archivo").size(10000);
+        SuggestBuilder suggest = new SuggestBuilder()
+                .addSuggestion("suggest", SuggestBuilders.completionSuggestion("nombre").text(busqueda).size(10))
+                .addSuggestion("phrase", SuggestBuilders.phraseSuggestion("nombre").text(busqueda).size(1)
+                        .realWordErrorLikelihood((float) 0.95).maxErrors((float) 0.5).gramSize(2));
+        System.out.println(boolQueryBuilder);
+        SearchResponse response = client.prepareSearch(index).setQuery(boolQueryBuilder).addAggregation(aggregation)
+                .suggest(suggest).setSize(pageable.getPageSize()).setFrom(pageable.getPageNumber()).execute()
+                .actionGet();
+        Consulta consulta = new Consulta();
+        consulta.setSuggest(response.getSuggest().getSuggestion("phrase").getEntries().get(0).getOptions().size() > 0
+                ? response.getSuggest().getSuggestion("phrase").getEntries().get(0).getOptions().get(0).getText()
+                        .string()
+                : "");
+        for (Entry<? extends Option> entry : response.getSuggest().getSuggestion("suggest").getEntries()) {
+            entry.getOptions().forEach(option -> {
+                String suggestText = option.getText().string().trim(),
+                        autocompleteClean = busqueda.replaceAll("[^\\p{Alnum}\\p{IsAlphabetic} ]", "");
+                for (String item : autocompleteClean.split(" ")) {
+                    if (item.length() > 0) {
+                        consulta.getAutocomplete().add(
+                                suggestText.replaceAll("(?i)((?!<)" + item + "(?![^<>]*>))", "<strong>$1</strong>"));
+                    }
+                }
+            });
+        }
+        ObjectMapper mapper = new ObjectMapper();
+        try {
+			consulta.setDocumentos(mapper.readValue(response.getHits().getHits().toString(), List.class));
+		} catch (IOException e) {
+			e.printStackTrace();
+		}
+        return consulta;
+    }
 }

src/main/java/com/manalejandro/arjion/vo/DetailVO.java

@@ -0,0 +1,22 @@
+package com.manalejandro.arjion.vo;
+
+import com.manalejandro.arjion.model.Archivo;
+
+public class DetailVO {
+
+    private Archivo archivo;
+
+    /**
+     * @return the archivo
+     */
+    public Archivo getArchivo() {
+        return archivo;
+    }
+
+    /**
+     * @param archivo the archivo to set
+     */
+    public void setArchivo(Archivo archivo) {
+        this.archivo = archivo;
+    }
+}

src/main/java/com/manalejandro/arjion/vo/DocumentoVO.java

@@ -1,11 +1,16 @@
 package com.manalejandro.arjion.vo;
 
+import java.util.ArrayList;
 import java.util.List;
 
 import com.manalejandro.arjion.model.Archivo;
+import com.manalejandro.arjion.model.Documento;
 
 public class DocumentoVO {
-    private List<Archivo> archivos;
+
+    private List<Archivo> archivos = new ArrayList<Archivo>();
+    private long count;
+    private List<Documento> documentos = new ArrayList<Documento>();
 
     /**
      * @return the archivos
@@ -20,4 +25,26 @@ public class DocumentoVO {
     public void setArchivos(List<Archivo> archivos) {
         this.archivos = archivos;
     }
+
+    public long getCount() {
+        return count;
+    }
+
+    public void setCount(long count) {
+        this.count = count;
+    }
+
+    /**
+     * @return the documentos
+     */
+    public List<Documento> getDocumentos() {
+        return documentos;
+    }
+
+    /**
+     * @param documentos the documentos to set
+     */
+    public void setDocumentos(List<Documento> documentos) {
+        this.documentos = documentos;
+    }
 }

src/main/resources/application.properties

@@ -7,10 +7,12 @@ elasticsearch.nodename=arjion
 arjion.indexName=documentos
 arjion.documentType=documento
 arjion.uploadpath=/upload/
+arjion.tesseractpath=/usr/bin
+arjion.tesseractdatapath=/usr/share/tesseract-ocr
 spring.main.allow-bean-definition-overriding=true
 spring.thymeleaf.enabled=true
 spring.thymeleaf.prefix=classpath:/templates/
 spring.thymeleaf.suffix=.html
 spring.thymeleaf.cache=false
-spring.servlet.multipart.max-file-size=10MB
-spring.servlet.multipart.max-request-size=20MB
+spring.servlet.multipart.max-file-size=20MB
+spring.servlet.multipart.max-request-size=100MB

src/main/resources/elasticsearch/mapping.json

@@ -1,16 +1,38 @@
 {
-	"documento": {
-		"properties": {
-			"@timestamp": {
-				"type": "date",
-				"format": "strict_date_optional_time||epoch_millis"
-			},
-			"@version": {
-				"type": "keyword"
-			},
-			"id": {
-				"type": "long"
-			}
-		}
-	}
+    "documento": {
+        "dynamic_templates": [
+            {
+                "metadata_as_keywords": {
+                    "path_match": "metadata.*",
+                    "mapping": {
+                        "type": "keyword"
+                    }
+                }
+            }
+        ],
+        "properties": {
+            "@timestamp": {
+                "type": "date",
+                "format": "strict_date_optional_time||epoch_millis"
+            },
+            "@version": {
+                "type": "keyword"
+            },
+            "nombre": {
+                "type": "text"
+            },
+            "tamano": {
+                "type": "long"
+            },
+            "metadata": {
+                "type": "object"
+            },
+            "contenido": {
+                "type": "text"
+            },
+            "lenguaje": {
+                "type": "keyword"
+            }
+        }
+    }
 }

src/main/resources/static/css/main.css

@@ -0,0 +1,4 @@
+hr {
+    width: 100%;
+    text-align: center;
+}

src/main/resources/templates/detail.html

@@ -0,0 +1,46 @@
+<!DOCTYPE html>
+<html>
+
+<head>
+    <meta charset="UTF-8">
+    <title>Arjion</title>
+    <link rel="stylesheet" th:href="@{/webjars/bootstrap/3.3.7-1/css/bootstrap.min.css}">
+    <link rel="stylesheet" th:href="@{/css/main.css}">
+    <script th:src="@{/webjars/jquery/3.1.1-1/jquery.min.js}"></script>
+    <script th:src="@{/webjars/bootstrap/3.3.7-1/js/bootstrap.min.js}"></script>
+    <script th:src="@{/js/main.js}"></script>
+</head>
+
+<body>
+    <header class="text-center">
+        <a th:href="@{/}">
+            <h1 class="text-primary">Arjion</h1>
+        </a>
+        <h3 class="text-warning">[[${detailVO.archivo.nombre}]]</h3>
+    </header>
+    <section class="col-md-12">
+        <hr>
+    </section>
+    <section>
+        <span class="col-md-1 text-muted">Tamaño</span>
+        <span class="col-md-11 text-muted">[[${detailVO.archivo.tamano}]] bytes</span>
+        <span class="col-md-1 text-muted">Lenguaje</span>
+        <span class="col-md-11 text-muted">[[${detailVO.archivo.lenguaje}]]</span>
+        <span class="col-md-1 text-success">Metadatos</span>
+        <span class="col-md-11 text-success"><ul><li th:each="meta : ${detailVO.archivo.metadata}"><span th:text="${meta.key}"></span>: <span th:text="${meta.value}"></span></li></ul></span>
+        <span class="col-md-1 text-warning">Contenido</span>
+        <pre class="col-md-11 text-warning">[[${detailVO.archivo.contenido}]]</pre>
+        </div>
+    </section>
+    <section class="col-md-12">
+        <hr>
+    </section>
+    <footer class="col-md-12 text-center">
+        <span class="col-md-12">
+            <a class="btn btn-primary" th:href="@{/}">Volver</a>
+        </span>
+        <span>2018</span>
+    </footer>
+</body>
+
+</html>

src/main/resources/templates/exists.html

@@ -0,0 +1,18 @@
+<!DOCTYPE html>
+<html>
+
+<head>
+    <meta charset="UTF-8">
+    <title>Error</title>
+    <link rel="stylesheet" th:href="@{/webjars/bootstrap/3.3.7-1/css/bootstrap.min.css}">
+    <script th:src="@{/webjars/jquery/3.1.1-1/jquery.min.js}"></script>
+    <script th:src="@{/webjars/bootstrap/3.3.7-1/js/bootstrap.min.js}"></script>
+</head>
+
+<body class="text-center text-danger">
+    <h2>Error</h2>
+    <h4>El archivo ya existe o hubo un error</h4>
+    <button class="btn btn-primary" th:onclick="'window.location.pathname=\'' + @{/} + '\''">Volver</button>
+</body>
+
+</html>

src/main/resources/templates/index.html

@@ -2,57 +2,83 @@
 <html>
 
 <head>
-	<meta charset="UTF-8">
-	<title>Arjion</title>
-	<link rel="stylesheet" th:href="@{/webjars/bootstrap/3.3.7-1/css/bootstrap.min.css}">
-	<link rel="stylesheet" th:href="@{/css/main.css}">
-	<script th:src="@{/webjars/jquery/3.1.1-1/jquery.min.js}"></script>
-	<script th:src="@{/webjars/bootstrap/3.3.7-1/js/bootstrap.min.js}"></script>
-	<script th:src="@{/js/main.js}"></script>
+    <meta charset="UTF-8">
+    <title>Arjion</title>
+    <link rel="stylesheet" th:href="@{/webjars/bootstrap/3.3.7-1/css/bootstrap.min.css}">
+    <link rel="stylesheet" th:href="@{/css/main.css}">
+    <script th:src="@{/webjars/jquery/3.1.1-1/jquery.min.js}"></script>
+    <script th:src="@{/webjars/bootstrap/3.3.7-1/js/bootstrap.min.js}"></script>
+    <script th:src="@{/js/main.js}"></script>
 </head>
 
 <body>
-	<header class="text-center">
-		<a th:href="@{/}"><h1 class="text-primary">Arjion</h1></a>
-	</header>
-	<section class="text-center col-md-4 col-md-offset-4">
-		<form class="form-horizontal form-label-left" method="post" enctype="multipart/form-data" novalidate="novalidate" th:action="@{/upload}"
-		 th:object="${documentForm}">
-			<div class="input-group">
-				<span class="input-group-addon">
-					<span class="glyphicon glyphicon-file"></span>
-				</span>
-				<label class="input-group-addon custom-file">
-					<input type="file" name="archivos" class="custom-file-input" multiple="multiple" />
-					<span class="custom-file-control"></span>
-				</label>
-				<span class="input-group-addon">
-					<button class="btn btn-primary" type="submit">
-						<span class="glyphicon glyphicon-level-up"></span>
-					</button>
-				</span>
-			</div>
-		</form>
-	</section>
-	<hr class="col-md-12">
-	<section>
-		<div th:each="doc : ${documentoVO.archivos}" class="col-md-12">
-			<span class="col-md-1 text-primary lead">Nombre</span>
-			<span class="col-md-11 text-primary lead">[[${doc.nombre}]]</span>
-			<span class="col-md-1 text-muted">Tamaño</span>
-			<span class="col-md-11 text-muted">[[${doc.tamano}]] bytes</span>
-			<span class="col-md-1 text-muted">Lenguaje</span>
-			<span class="col-md-11 text-muted">[[${doc.lenguaje}]]</span>
-			<span class="col-md-1 text-success">Metadatos</span>
-			<span class="col-md-11 text-success">[[${doc.metadata}]]</span>
-			<span class="col-md-1 text-warning">Contenido</span>
-			<pre class="col-md-11 text-warning">[[${doc.contenido}]]</pre>
-			<hr class="col-md-12">
-		</div>
-	</section>
-	<footer class="col-md-12 text-center">
-		<span>2018</span>
-	</footer>
+    <header class="text-center">
+        <a th:href="@{/}">
+            <h1 class="text-primary">Arjion</h1>
+        </a>
+        <h3 class="text-warning">[[${documentoVO.count}]] archivos</h3>
+    </header>
+    <section class="text-center col-md-4 col-md-offset-4">
+        <form class="form-horizontal form-label-left" method="post" enctype="multipart/form-data" novalidate="novalidate" th:action="@{/upload}"
+            th:object="${documentForm}">
+            <div class="input-group">
+                <span class="input-group-addon">
+                    <span class="glyphicon glyphicon-file"></span>
+                </span>
+                <label class="input-group-addon custom-file">
+                    <input type="file" name="archivos" class="custom-file-input" multiple="multiple" />
+                    <span class="custom-file-control"></span>
+                </label>
+                <span class="input-group-addon">
+                    <button class="btn btn-primary" type="submit">
+                        <span class="glyphicon glyphicon-level-up"></span>
+                    </button>
+                </span>
+            </div>
+        </form>
+    </section>
+    <section class="col-md-12">
+        <hr>
+    </section>
+    <section class="col-md-12">
+        <div th:each="arc : ${documentoVO.archivos}" class="col-md-12">
+            <span class="col-md-1 text-primary lead">Nombre</span>
+            <span class="col-md-11 text-primary lead">[[${arc.nombre}]]</span>
+            <span class="col-md-1 text-muted">Tamaño</span>
+            <span class="col-md-11 text-muted">[[${arc.tamano}]] bytes</span>
+            <span class="col-md-1 text-muted">Lenguaje</span>
+            <span class="col-md-11 text-muted">[[${arc.lenguaje}]]</span>
+            <span class="col-md-1 text-success">Metadatos</span>
+            <span class="col-md-11 text-success"><ul><li th:each="meta : ${arc.metadata}"><span th:text="${meta.key}"></span>: <span th:text="${meta.value}"></span></li></ul></span>
+            <span class="col-md-1 text-warning">Contenido</span>
+            <pre class="col-md-11 text-warning">[[${arc.contenido}]]</pre>
+            <span class="col-md-12">
+                <hr>
+            </span>
+        </div>
+    </section>
+    <section class="col-md-12">
+        <hr>
+    </section>
+    <section class="col-md-12 text-center">
+        <div th:each="doc : ${documentoVO.documentos}">
+            <div>
+                <span class="col-md-12 text-primary">
+                    <a th:href="@{'/detail'(nombre=${doc.nombre})}" class="lead">[[${doc.nombre}]]</a> -
+                    <a th:href="@{'/download'(filename=${doc.nombre})}">download</a>
+                    <br>[[${doc.tamano}]] bytes
+                    <br>[[${doc.lenguaje}]]</span>
+                <span class="col-md-12 text-success">[[${#strings.abbreviate(doc.metadata,200)}]]</span>
+                <span class="col-md-12 text-warning">[[${#strings.abbreviate(doc.contenido,200)}]]</span>
+            </div>
+            <span class="col-md-12">
+                <hr>
+            </span>
+        </div>
+    </section>
+    <footer class="col-md-12 text-center">
+        <span>2018</span>
+    </footer>
 </body>
 
 </html>