This is an automated email from the ASF dual-hosted git repository.
nick pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/tika.git
The following commit(s) were added to refs/heads/master by this push:
new c05c524 Serialise the details of multiple parsers
c05c524 is described below
commit c05c5245bdea38ddb9ef24e99ddcdc7bf7803d67
Author: Nick Burch <n...@gagravarr.org>
AuthorDate: Sun Apr 8 13:56:38 2018 +0100
Serialise the details of multiple parsers
---
.../apache/tika/config/TikaConfigSerializer.java | 14 ++++++++++
.../parser/multiple/AbstractMultipleParser.java | 4 +++
.../multiple/PickBestTextEncodingParser.java | 30 ++++++++++++++++------
3 files changed, 40 insertions(+), 8 deletions(-)
diff --git
a/tika-core/src/main/java/org/apache/tika/config/TikaConfigSerializer.java
b/tika-core/src/main/java/org/apache/tika/config/TikaConfigSerializer.java
index c67b03b..dda1675 100644
--- a/tika-core/src/main/java/org/apache/tika/config/TikaConfigSerializer.java
+++ b/tika-core/src/main/java/org/apache/tika/config/TikaConfigSerializer.java
@@ -45,6 +45,7 @@ import org.apache.tika.parser.DefaultParser;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.parser.Parser;
import org.apache.tika.parser.ParserDecorator;
+import org.apache.tika.parser.multiple.AbstractMultipleParser;
import org.apache.tika.utils.XMLReaderUtils;
import org.w3c.dom.Document;
import org.w3c.dom.Element;
@@ -254,6 +255,9 @@ public class TikaConfigSerializer {
(mode == Mode.STATIC || mode == Mode.STATIC_FULL)) {
outputParser = false;
}
+ } else if (parser instanceof AbstractMultipleParser) {
+ // Always output the parsers that go into the multiple
+ children = ((AbstractMultipleParser)parser).getAllParsers();
}
if (outputParser) {
@@ -290,6 +294,16 @@ public class TikaConfigSerializer {
parserElement.setAttribute("class", className);
rootElement.appendChild(parserElement);
+ // TODO Output configurable parameters in a genric way, see TIKA-1508
+ if (parser instanceof AbstractMultipleParser) {
+ Element paramsElement = doc.createElement("params");
+ Element paramElement = doc.createElement("param");
+ paramElement.setAttribute("name", "metadataPolicy");
+ paramElement.setAttribute("value",
((AbstractMultipleParser)parser).getMetadataPolicy().toString());
+ paramsElement.appendChild(paramElement);
+ parserElement.appendChild(paramsElement);
+ }
+
for (MediaType type : addedTypes) {
Element mimeElement = doc.createElement("mime");
mimeElement.appendChild(doc.createTextNode(type.toString()));
diff --git
a/tika-core/src/main/java/org/apache/tika/parser/multiple/AbstractMultipleParser.java
b/tika-core/src/main/java/org/apache/tika/parser/multiple/AbstractMultipleParser.java
index d687e41..1a58f89 100644
---
a/tika-core/src/main/java/org/apache/tika/parser/multiple/AbstractMultipleParser.java
+++
b/tika-core/src/main/java/org/apache/tika/parser/multiple/AbstractMultipleParser.java
@@ -25,6 +25,7 @@ import java.io.InputStream;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
+import java.util.Collections;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
@@ -173,6 +174,9 @@ public abstract class AbstractMultipleParser extends
AbstractParser {
public MetadataPolicy getMetadataPolicy() {
return policy;
}
+ public List<Parser> getAllParsers() {
+ return Collections.unmodifiableList(new ArrayList<>(parsers));
+ }
/**
* Used to allow implementations to prepare or change things
diff --git
a/tika-core/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
b/tika-parsers/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
similarity index 87%
rename from
tika-core/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
rename to
tika-parsers/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
index f043a5a..b1a0caa 100644
---
a/tika-core/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
+++
b/tika-parsers/src/main/java/org/apache/tika/parser/multiple/PickBestTextEncodingParser.java
@@ -31,9 +31,9 @@ import org.apache.tika.detect.NonDetectingEncodingDetector;
import org.apache.tika.exception.TikaException;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.mime.MediaTypeRegistry;
-import org.apache.tika.parser.EmptyParser;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.parser.Parser;
+import org.apache.tika.parser.txt.TXTParser;
import org.apache.tika.sax.BodyContentHandler;
import org.apache.tika.sax.ContentHandlerFactory;
import org.xml.sax.ContentHandler;
@@ -49,8 +49,8 @@ import org.xml.sax.SAXException;
* This is not recommended for actual production use... It is mostly to
* prove that the {@link AbstractMultipleParser} environment is
* sufficient to support this use-case
- *
- * TODO Move this to the parsers package so it can get {@link TXTParser}
+ *
+ * TODO Implement proper "Junk" detection
*
* @deprecated Currently not suitable for real use, more a demo / prototype!
*/
@@ -66,7 +66,6 @@ public class PickBestTextEncodingParser extends
AbstractMultipleParser {
private String[] charsetsToTry;
public PickBestTextEncodingParser(MediaTypeRegistry registry, String[]
charsets) {
- // TODO Actually give 1 more TXTParser than we have charsets
super(registry, MetadataPolicy.DISCARD_ALL, makeParsers(charsets));
this.charsetsToTry = charsets;
}
@@ -74,8 +73,7 @@ public class PickBestTextEncodingParser extends
AbstractMultipleParser {
// One more TXTParser than we have charsets, for the real thing
List<Parser> parsers = new ArrayList<>(charsets.length+1);
for (int i=0; i<charsets.length+1; i++) {
- // TODO Actually get the right parser, TXTParser
- parsers.set(i, new EmptyParser());
+ parsers.set(i, new TXTParser());
}
return parsers;
}
@@ -104,9 +102,25 @@ public class PickBestTextEncodingParser extends
AbstractMultipleParser {
charsetTester.charsetText.put(charset, handler.toString());
// If this was the last real charset, see which one is best
+ // TODO Do this in a more generic, less english-only way!
if (! charsetTester.moreToTest()) {
- // TODO Properly work out the best!
- charsetTester.pickedCharset = charsetsToTry[0];
+ int numEnglish = 0;
+ String bestcharset = null;
+ for (String pcharset : charsetTester.charsetText.keySet()) {
+ String text = charsetTester.charsetText.get(pcharset);
+ int cEnglish = 0;
+ for (char c : text.toCharArray()) {
+ if ((c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') ||
+ (c >= '0' && c <= '9')) {
+ cEnglish++;
+ }
+ }
+ if (cEnglish > numEnglish) {
+ numEnglish = cEnglish;
+ bestcharset = pcharset;
+ }
+ }
+ charsetTester.pickedCharset = bestcharset;
}
}
--
To stop receiving notification emails like this one, please contact
n...@apache.org.
