-
Notifications
You must be signed in to change notification settings - Fork 6
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
3 changed files
with
221 additions
and
24 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
193 changes: 193 additions & 0 deletions
193
g419-liner2-cli/src/main/java/g419/liner2/cli/action/ActionAnonimization.java
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,193 @@ | ||
| package g419.liner2.cli.action; | ||
|
|
||
| import com.google.common.collect.Lists; | ||
| import com.google.common.collect.Maps; | ||
| import com.google.common.collect.Sets; | ||
| import g419.corpus.io.reader.AbstractDocumentReader; | ||
| import g419.corpus.io.reader.ReaderFactory; | ||
| import g419.corpus.io.writer.AbstractDocumentWriter; | ||
| import g419.corpus.io.writer.WriterFactory; | ||
| import g419.corpus.structure.Document; | ||
| import g419.corpus.structure.Sentence; | ||
| import g419.corpus.structure.Token; | ||
| import g419.lib.cli.Action; | ||
| import g419.lib.cli.CommonOptions; | ||
| import io.vavr.collection.Stream; | ||
| import org.apache.commons.cli.CommandLine; | ||
| import org.apache.commons.cli.Option; | ||
| import org.apache.commons.csv.CSVFormat; | ||
| import org.apache.commons.csv.CSVParser; | ||
|
|
||
| import java.io.FileReader; | ||
| import java.io.IOException; | ||
| import java.nio.file.Path; | ||
| import java.nio.file.Paths; | ||
| import java.util.Collection; | ||
| import java.util.List; | ||
| import java.util.Map; | ||
| import java.util.Set; | ||
| import java.util.stream.Collectors; | ||
|
|
||
| public class ActionAnonimization extends Action { | ||
|
|
||
| class PhraseType { | ||
| private String phrase; | ||
| private String type; | ||
|
|
||
| public PhraseType(final String phrase, final String type) { | ||
| this.phrase = phrase; | ||
| this.type = type; | ||
| } | ||
|
|
||
| public String getPhrase() { | ||
| return phrase; | ||
| } | ||
|
|
||
| public void setPhrase(final String phrase) { | ||
| this.phrase = phrase; | ||
| } | ||
|
|
||
| public String getType() { | ||
| return type; | ||
| } | ||
|
|
||
| public void setType(final String type) { | ||
| this.type = type; | ||
| } | ||
| } | ||
|
|
||
| private final String OPTION_DICTIONARY_FILE = "d"; | ||
|
|
||
| private String inputFile = null; | ||
| private String inputFormat = null; | ||
| private String outputFolder = null; | ||
| private String outputFormat = null; | ||
| private String dictFile = null; | ||
|
|
||
| public ActionAnonimization() { | ||
| super("anonim"); | ||
| setDescription("anonimize according to given dictionary"); | ||
| options.addOption(CommonOptions.getInputFileFormatOption()); | ||
| options.addOption(CommonOptions.getInputFileNameOption()); | ||
| options.addOption(CommonOptions.getOutputFileNameOption()); | ||
| options.addOption(CommonOptions.getOutputFileFormatOption()); | ||
| options.addOption(getDictionaryFile()); | ||
| } | ||
|
|
||
| @Override | ||
| public void parseOptions(final CommandLine line) throws Exception { | ||
| inputFile = line.getOptionValue(CommonOptions.OPTION_INPUT_FILE); | ||
| inputFormat = line.getOptionValue(CommonOptions.OPTION_INPUT_FORMAT, "ccl"); | ||
| outputFolder = line.getOptionValue(CommonOptions.OPTION_OUTPUT_FILE); | ||
| outputFormat = line.getOptionValue(CommonOptions.OPTION_OUTPUT_FORMAT, "ccl"); | ||
| dictFile = line.getOptionValue(OPTION_DICTIONARY_FILE); | ||
| } | ||
|
|
||
| public Option getDictionaryFile() { | ||
| return Option.builder(OPTION_DICTIONARY_FILE).longOpt("dict") | ||
| .hasArg().argName("path").desc("dictionary of phrases per document").build(); | ||
| } | ||
|
|
||
|
|
||
| @Override | ||
| public void run() throws Exception { | ||
| final Map<String, List<PhraseType>> phrases = loadPhrases(Paths.get(dictFile)); | ||
| getLogger().info("Number of documents with phrases to anonymize: {}", phrases.size()); | ||
|
|
||
| try (final AbstractDocumentReader reader = ReaderFactory.get().getStreamReader(inputFile, inputFormat); | ||
| final AbstractDocumentWriter writer = WriterFactory.get().getStreamWriter(outputFolder, inputFormat)) { | ||
| reader.forEachRemaining(document -> { | ||
| final String docId = document.getName(); | ||
| final List<PhraseType> documentPhrases = phrases.computeIfAbsent(docId, d -> Lists.newArrayList()); | ||
| anonymizePhrases(document, documentPhrases); | ||
| writer.writeDocument(document); | ||
| }); | ||
| } | ||
| } | ||
|
|
||
| private void anonymizePhrases(final Document d, final List<PhraseType> documentPhrases) { | ||
| getLogger().info("[{}] Number of phrases: {}", d.getName(), documentPhrases.size()); | ||
| final Set<String> replaced = Sets.newHashSet(); | ||
| documentPhrases.forEach(phrase -> { | ||
| if (replaced.contains(phrase.getPhrase())) { | ||
| getLogger().error("[{}] PHRASE '{}' already replaced", d.getName(), phrase.getPhrase()); | ||
| } else { | ||
| final int replaceCount = anonymizePhrase(d, phrase); | ||
| if (replaceCount > 0) { | ||
| getLogger().info("[{}] PHRASE '{}' found {} times and replaced with '@{}'", | ||
| d.getName(), phrase.getPhrase(), replaceCount, phrase.getType()); | ||
| replaced.add(phrase.getPhrase()); | ||
| } else { | ||
| getLogger().error("[{}] PHRASE '{}' not found", d.getName(), phrase.getPhrase()); | ||
| } | ||
| } | ||
| }); | ||
| } | ||
|
|
||
| private int anonymizePhrase(final Document d, final PhraseType phrase) { | ||
| final List<List<Token>> matches = matchPhrase(d, phrase.getPhrase()); | ||
| matches.forEach(m -> anonymizePhrase(m, "@" + phrase.getType())); | ||
| return matches.size(); | ||
| } | ||
|
|
||
| private void anonymizePhrase(final List<Token> tokens, final String label) { | ||
| tokens.forEach(t -> { | ||
| t.setOrth(label); | ||
| t.getTags().forEach(tag -> tag.setBase(label)); | ||
| }); | ||
| } | ||
|
|
||
| private List<List<Token>> matchPhrase(final Document d, final String phrase) { | ||
| return d.getSentences() | ||
| .stream() | ||
| .map(s -> matchPhrase(s, phrase)) | ||
| .flatMap(Collection::stream) | ||
| .collect(Collectors.toList()); | ||
| } | ||
|
|
||
|
|
||
| private List<List<Token>> matchPhrase(final Sentence s, final String phrase) { | ||
| final List<Token> tokens = s.getTokens(); | ||
| final List<List<Token>> matches = Lists.newArrayList(); | ||
| int n = 0; | ||
| while (n < tokens.size()) { | ||
| final List<Token> match = matchPhraseAtPos(tokens, phrase, n); | ||
| if (match != null) { | ||
| matches.add(match); | ||
| n += match.size(); | ||
| } | ||
| n++; | ||
| } | ||
| return matches; | ||
| } | ||
|
|
||
| private List<Token> matchPhraseAtPos(final List<Token> tokens, final String phrase, final int pos) { | ||
| String concat = tokens.get(pos).getOrth(); | ||
| int n = pos; | ||
| while (!concat.equals(phrase) && n + 1 < tokens.size() && concat.length() < phrase.length()) { | ||
| if (!tokens.get(n).getNoSpaceAfter()) { | ||
| concat += " "; | ||
| } | ||
| concat += tokens.get(++n).getOrth(); | ||
| } | ||
| if (concat.equals(phrase)) { | ||
| return Stream.rangeClosed(n, pos) | ||
| .map(tokens::get) | ||
| .collect(Collectors.toList()); | ||
| } else { | ||
| return null; | ||
| } | ||
| } | ||
|
|
||
| public Map<String, List<PhraseType>> loadPhrases(final Path path) throws IOException { | ||
| final CSVParser csv = new CSVParser(new FileReader(path.toFile()), CSVFormat.TDF); | ||
| final Map<String, List<PhraseType>> phrases = Maps.newHashMap(); | ||
| csv.forEach(record -> { | ||
| final String docId = record.get(0); | ||
| final PhraseType phraseType = new PhraseType(record.get(4), record.get(3)); | ||
| phrases.computeIfAbsent(docId, d -> Lists.newArrayList()).add(phraseType); | ||
| }); | ||
| return phrases; | ||
| } | ||
|
|
||
| } |