Examples with FilteredRows com.google.refine.browsing.FilteredRows used on opensource projects

Example 6 with FilteredRows

use of com.google.refine.browsing.FilteredRows in project OpenRefine by OpenRefine.

the class ColumnSplitOperation method createHistoryEntry.

@Override
protected HistoryEntry createHistoryEntry(Project project, long historyEntryID) throws Exception {
    Engine engine = createEngine(project);
    Column column = project.columnModel.getColumnByName(_columnName);
    if (column == null) {
        throw new Exception("No column named " + _columnName);
    }
    List<String> columnNames = new ArrayList<String>();
    List<Integer> rowIndices = new ArrayList<Integer>(project.rows.size());
    List<List<Serializable>> tuples = new ArrayList<List<Serializable>>(project.rows.size());
    FilteredRows filteredRows = engine.getAllFilteredRows();
    RowVisitor rowVisitor;
    if ("lengths".equals(_mode)) {
        rowVisitor = new ColumnSplitRowVisitor(column.getCellIndex(), columnNames, rowIndices, tuples) {

            @Override
            protected java.util.List<Serializable> split(String s) {
                List<Serializable> results = new ArrayList<Serializable>(_fieldLengths.length + 1);
                int lastIndex = 0;
                for (int length : _fieldLengths) {
                    int from = lastIndex;
                    int to = Math.min(from + length, s.length());
                    results.add(stringToValue(s.substring(from, to)));
                    lastIndex = to;
                }
                return results;
            }

            ;
        };
    } else if (_regex) {
        Pattern pattern = Pattern.compile(_separator);
        rowVisitor = new ColumnSplitRowVisitor(column.getCellIndex(), columnNames, rowIndices, tuples) {

            Pattern _pattern;

            @Override
            protected java.util.List<Serializable> split(String s) {
                return stringArrayToValueList(_pattern.split(s, _maxColumns));
            }

            ;

            public RowVisitor init(Pattern pattern) {
                _pattern = pattern;
                return this;
            }
        }.init(pattern);
    } else {
        rowVisitor = new ColumnSplitRowVisitor(column.getCellIndex(), columnNames, rowIndices, tuples) {

            @Override
            protected java.util.List<Serializable> split(String s) {
                return stringArrayToValueList(StringUtils.splitByWholeSeparatorPreserveAllTokens(s, _separator, _maxColumns));
            }

            ;
        };
    }
    filteredRows.accept(project, rowVisitor);
    String description = "Split " + rowIndices.size() + " cell(s) in column " + _columnName + " into several columns" + ("separator".equals(_mode) ? " by separator" : " by field lengths");
    Change change = new ColumnSplitChange(_columnName, columnNames, rowIndices, tuples, _removeOriginalColumn);
    return new HistoryEntry(historyEntryID, project, description, this, change);
}

Example 7 with FilteredRows

use of com.google.refine.browsing.FilteredRows in project OpenRefine by OpenRefine.

the class ReconCopyAcrossColumnsOperation method createHistoryEntry.

@Override
protected HistoryEntry createHistoryEntry(final Project project, final long historyEntryID) throws Exception {
    Engine engine = createEngine(project);
    final Column fromColumn = project.columnModel.getColumnByName(_fromColumnName);
    final List<Column> toColumns = new ArrayList<Column>(_toColumnNames.length);
    for (String c : _toColumnNames) {
        Column toColumn = project.columnModel.getColumnByName(c);
        if (toColumn != null) {
            toColumns.add(toColumn);
        }
    }
    final Set<Recon.Judgment> judgments = new HashSet<Recon.Judgment>(_judgments.length);
    for (String j : _judgments) {
        judgments.add(Recon.stringToJudgment(j));
    }
    final List<CellChange> cellChanges = new ArrayList<CellChange>(project.rows.size());
    if (fromColumn != null && toColumns.size() > 0) {
        final Map<Object, Recon> cellValueToRecon = new HashMap<Object, Recon>();
        FilteredRows filteredRows = engine.getAllFilteredRows();
        try {
            filteredRows.accept(project, new RowVisitor() {

                @Override
                public void start(Project project) {
                // nothing to do
                }

                @Override
                public void end(Project project) {
                // nothing to do
                }

                @Override
                public boolean visit(Project project, int rowIndex, Row row) {
                    Cell cell = row.getCell(fromColumn.getCellIndex());
                    if (cell != null && cell.value != null && cell.recon != null) {
                        if (judgments.contains(cell.recon.judgment)) {
                            cellValueToRecon.put(cell.value, cell.recon);
                        }
                    }
                    return false;
                }
            });
            filteredRows.accept(project, new RowVisitor() {

                @Override
                public void start(Project project) {
                // nothing to do
                }

                @Override
                public void end(Project project) {
                // nothing to do
                }

                @Override
                public boolean visit(Project project, int rowIndex, Row row) {
                    for (Column column : toColumns) {
                        int cellIndex = column.getCellIndex();
                        Cell cell = row.getCell(cellIndex);
                        if (cell != null && cell.value != null) {
                            Recon reconToCopy = cellValueToRecon.get(cell.value);
                            boolean judged = cell.recon != null && cell.recon.judgment != Judgment.None;
                            if (reconToCopy != null && (!judged || _applyToJudgedCells)) {
                                Cell newCell = new Cell(cell.value, reconToCopy);
                                CellChange cellChange = new CellChange(rowIndex, cellIndex, cell, newCell);
                                cellChanges.add(cellChange);
                            }
                        }
                    }
                    return false;
                }
            });
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    String description = "Copy " + cellChanges.size() + " recon judgments from column " + _fromColumnName + " to " + StringUtils.join(_toColumnNames);
    return new HistoryEntry(historyEntryID, project, description, this, new MassChange(cellChanges, false));
}

Example 8 with FilteredRows

use of com.google.refine.browsing.FilteredRows in project OpenRefine by OpenRefine.

the class RowFlagOperation method createHistoryEntry.

@Override
protected HistoryEntry createHistoryEntry(Project project, long historyEntryID) throws Exception {
    Engine engine = createEngine(project);
    List<Change> changes = new ArrayList<Change>(project.rows.size());
    FilteredRows filteredRows = engine.getAllFilteredRows();
    filteredRows.accept(project, createRowVisitor(project, changes));
    return new HistoryEntry(historyEntryID, project, (_flagged ? "Flag" : "Unflag") + " " + changes.size() + " rows", this, new MassChange(changes, false));
}

Example 9 with FilteredRows

use of com.google.refine.browsing.FilteredRows in project OpenRefine by OpenRefine.

the class GetScatterplotCommand method draw.

public void draw(OutputStream output, Project project, Engine engine, JSONObject o) throws IOException, JSONException {
    double min_x = 0;
    double min_y = 0;
    double max_x = 0;
    double max_y = 0;
    int columnIndex_x = 0;
    int columnIndex_y = 0;
    Evaluable eval_x = null;
    Evaluable eval_y = null;
    int size = (o.has(ScatterplotFacet.SIZE)) ? o.getInt(ScatterplotFacet.SIZE) : 100;
    double dot = (o.has(ScatterplotFacet.DOT)) ? o.getDouble(ScatterplotFacet.DOT) : 100;
    int dim_x = (o.has(ScatterplotFacet.DIM_X)) ? ScatterplotFacet.getAxisDim(o.getString(ScatterplotFacet.DIM_X)) : ScatterplotFacet.LIN;
    int dim_y = (o.has(ScatterplotFacet.DIM_Y)) ? ScatterplotFacet.getAxisDim(o.getString(ScatterplotFacet.DIM_Y)) : ScatterplotFacet.LIN;
    int rotation = (o.has(ScatterplotFacet.ROTATION)) ? ScatterplotFacet.getRotation(o.getString(ScatterplotFacet.ROTATION)) : ScatterplotFacet.NO_ROTATION;
    String color_str = (o.has(ScatterplotFacet.COLOR)) ? o.getString(ScatterplotFacet.COLOR) : "000000";
    Color color = new Color(Integer.parseInt(color_str, 16));
    String base_color_str = (o.has(ScatterplotFacet.BASE_COLOR)) ? o.getString(ScatterplotFacet.BASE_COLOR) : null;
    Color base_color = base_color_str != null ? new Color(Integer.parseInt(base_color_str, 16)) : null;
    String columnName_x = o.getString(ScatterplotFacet.X_COLUMN_NAME);
    String expression_x = (o.has(ScatterplotFacet.X_EXPRESSION)) ? o.getString(ScatterplotFacet.X_EXPRESSION) : "value";
    if (columnName_x.length() > 0) {
        Column x_column = project.columnModel.getColumnByName(columnName_x);
        if (x_column != null) {
            columnIndex_x = x_column.getCellIndex();
        }
    } else {
        columnIndex_x = -1;
    }
    try {
        eval_x = MetaParser.parse(expression_x);
    } catch (ParsingException e) {
        logger.warn("error parsing expression", e);
    }
    String columnName_y = o.getString(ScatterplotFacet.Y_COLUMN_NAME);
    String expression_y = (o.has(ScatterplotFacet.Y_EXPRESSION)) ? o.getString(ScatterplotFacet.Y_EXPRESSION) : "value";
    if (columnName_y.length() > 0) {
        Column y_column = project.columnModel.getColumnByName(columnName_y);
        if (y_column != null) {
            columnIndex_y = y_column.getCellIndex();
        }
    } else {
        columnIndex_y = -1;
    }
    try {
        eval_y = MetaParser.parse(expression_y);
    } catch (ParsingException e) {
        logger.warn("error parsing expression", e);
    }
    NumericBinIndex index_x = null;
    NumericBinIndex index_y = null;
    String col_x_name = o.getString(ScatterplotFacet.X_COLUMN_NAME);
    Column column_x = project.columnModel.getColumnByName(col_x_name);
    if (column_x != null) {
        columnIndex_x = column_x.getCellIndex();
        index_x = ScatterplotFacet.getBinIndex(project, column_x, eval_x, expression_x);
        min_x = index_x.getMin();
        max_x = index_x.getMax();
    }
    String col_y_name = o.getString(ScatterplotFacet.Y_COLUMN_NAME);
    Column column_y = project.columnModel.getColumnByName(col_y_name);
    if (column_y != null) {
        columnIndex_y = column_y.getCellIndex();
        index_y = ScatterplotFacet.getBinIndex(project, column_y, eval_y, expression_y);
        min_y = index_y.getMin();
        max_y = index_y.getMax();
    }
    if (index_x != null && index_y != null && index_x.isNumeric() && index_y.isNumeric()) {
        ScatterplotDrawingRowVisitor drawer = new ScatterplotDrawingRowVisitor(columnIndex_x, columnIndex_y, min_x, max_x, min_y, max_y, size, dim_x, dim_y, rotation, dot, color);
        if (base_color != null) {
            drawer.setColor(base_color);
            FilteredRows filteredRows = engine.getAllRows();
            filteredRows.accept(project, drawer);
            drawer.setColor(color);
        }
        {
            FilteredRows filteredRows = engine.getAllFilteredRows();
            filteredRows.accept(project, drawer);
        }
        ImageIO.write(drawer.getImage(), "png", output);
    } else {
        ImageIO.write(new BufferedImage(1, 1, BufferedImage.TYPE_4BYTE_ABGR), "png", output);
    }
}

Example 10 with FilteredRows

use of com.google.refine.browsing.FilteredRows in project OpenRefine by OpenRefine.

the class kNNClusterer method computeClusters.

@Override
public void computeClusters(Engine engine) {
    //VPTreeClusteringRowVisitor visitor = new VPTreeClusteringRowVisitor(_distance,_config);
    BlockingClusteringRowVisitor visitor = new BlockingClusteringRowVisitor(_distance, _config);
    FilteredRows filteredRows = engine.getAllFilteredRows();
    filteredRows.accept(_project, visitor);
    _clusters = visitor.getClusters();
}