Your company here — click to reach over 10,000 unique daily visitors

osmium-export - Man Page

export OSM data


osmium export [Options] OSM-FILE


The OSM data model with its nodes, ways, and relations is very different from the data model usually used for geodata with features having point, linestring, or polygon geometries (or their cousins, the multipoint, multilinestring, or multipolygon geometries).

The export command transforms OSM data into a more usual GIS data model. Nodes will be translated into points and ways into linestrings or polygons (if they are closed ways). Multipolygon and boundary relations will be translated into multipolygons. This transformation is not loss-less, especially information in non-multipolygon, non-boundary relations is lost.

All tags are preserved in this process. Note that most GIS formats (such as Shapefiles, etc.) do not support arbitrary tags. Transformation into other GIS formats will need extra steps mapping tags to a limited list of attributes. This is outside the scope of this command.

The osmium export command has to keep an index of the node locations in memory or in a temporary file on disk while doing its work. There are several different ways it can do that which have different advantages and disadvantages. The default is good enough for most cases, but see the osmium-index-types(5) man page for details.

Objects with invalid geometries are silently omitted from the output. This is the case for ways with less than two nodes or closed ways or relations that can’t be assembled into a valid (multi)polygon. See the options --show-errors/-e and --stop-on-error/-E for how to modify this behaviour.

The input file will be read twice (once for the relations, once for nodes and ways), so this command can not read its input from STDIN.

This command will not work on full history files.

This command will work with negative IDs on OSM objects (for instance on files created with JOSM).


-c,  --config=FILE

Read configuration from specified file.

-C,  --print-default-config

Print the default config to STDOUT. Useful if you want to change it and not write the whole thing manually. If you use this option all other options are ignored.

-e,  --show-errors

Output any geometry errors on STDERR. This includes ways with a single node or areas that can’t be assembled from multipolygon relations. This output is not suitable for automated use, there are other tools that can create very detailed errors reports that are better for that (see https://osmcode.org/osm-area-tools/).

-E,  --stop-on-error

Usually geometry errors (due to missing node locations or broken polygons) are ignored and the features are omitted from the output. If this option is set, any error will immediately stop the program.


Specify the geometry types that should be written out. Usually all created geometries (points, linestrings, and (multi)polygons) are written to the output, but you can restrict the types using this option. TYPES is a comma-separated list of the types (“point”, “linestring”, and “polygon”).

-a,  --attributes=ATTRS

In addition to tags, also export attributes specified in this comma-separated list. By default, none are exported. See the Attributes section below for the known attributes list and an explanation.

-i,  --index-type=TYPE

Set the index type. For details see the osmium-index-types(5) man page.

-I,  --show-index-types

Shows a list of available index types. For details see the osmium-index-types(5) man page. If you use this options all other options are ignored.

-n,  --keep-untagged

If this is set, features without any tags will be in the exported data. By default these features will be omitted from the output. Tags are the OSM tags, not attributes (like id, version, uid, ...) without the tags removed by the exclude_tags or include_tags settings.

-u,  --add-unique-id=TYPE

Add a unique ID to each feature. TYPE can be either counter in which case the first feature will get ID 1, the next ID 2 and so on. The type of object does not matter in this case. Or the TYPE is type_id in which case the ID is a string, the first character is the type of object (`n' for nodes, `w' for linestrings created from ways, and `a' for areas created from ways and/or relations, after that there is a unique ID based on the original OSM object ID(s). If the input file has negative IDs, this can create IDs such as `w-12'. In spaten exports the ID is written into the @fid field. For counter the value will be an integer, for type_id it will be a string.

-x,  --format-option=OPTION(=VALUE)

Set an output format option. The options available depend on the output format. See the Output Format Options section for available options. If the VALUE is not set, the OPTION will be set to “true”. If needed you can specify this option multiple times to set several options. Options set on the command line overwrite options set in the config file.

Common Options

-h,  --help

Show usage help.

-v,  --verbose

Set verbose mode. The program will output information about what it is doing to STDERR.


Show progress bar. Usually a progress bar is only displayed if STDOUT and STDERR are detected to be TTY. With this option a progress bar is always shown. Note that a progress bar will never be shown when reading from STDIN or a pipe.


Do not show progress bar. Usually a progress bar is displayed if STDOUT and STDERR are detected to be a TTY. With this option the progress bar is suppressed. Note that a progress bar will never be shown when reading from STDIN or a pipe.

Input Options

-F,  --input-format=FORMAT

The format of the input file(s). Can be used to set the input format if it can’t be autodetected from the file name(s). This will set the format for all input files, there is no way to set the format for some input files only. See osmium-file-formats(5) or the libosmium manual for details.

Output Options

-f,  --output-format=FORMAT

The format of the output file. Can be used to set the output file format if it can’t be autodetected from the output file name. See the Output Formats section for a list of formats.


Call fsync after writing the output file to force flushing buffers to disk.

-o,  --output=FILE

Name of the output file. Default is `-' (STDOUT).

-O,  --overwrite

Allow an existing output file to be overwritten. Normally osmium will refuse to write over an existing file.

Config File

The config file is in JSON format. The top-level is an object which contains the following optional names:

The area_tags and linear_tags can have the following values:


All tags match. (An empty list [] can also be used to mean the same, but this use is deprecated because it can be confusing.)


No tags match.


The array contains one or more expressions as described in the FILTER EXPRESSION section.


If the area_tags or linear_tags is set to null or not set at all, the inverse of the other setting is used. So if you do not set the linear_tags but have some expressions in area_tags, areas will be created for all objects matching those expressions and linestrings for everything else. This can be simpler, because you only have to keep one list, but in cases where an object can be interpreted as both an area and a linestring, only one interpretation will be used.

The exclude_tags and include_tags options are mutually exclusive. If you want to just exclude some tags but leave most tags untouched, use the exclude_tags setting. If you only want a defined list of tags, use include_tags.

When no config file is specified, the following settings are used:

    "attributes": {
        "type":      false,
        "id":        false,
        "version":   false,
        "changeset": false,
        "timestamp": false,
        "uid":       false,
        "user":      false,
        "way_nodes": false
    "format_options": {
    "linear_tags":  true,
    "area_tags":    true,
    "exclude_tags": [],
    "include_tags": []

Filter Expressions

A filter expression specifies a tag or tags that should be matched in the data.

Some examples:


Matches all objects with the key “amenity”.


Matches all objects with the key “highway” and value “primary”.


Matches all objects with the key “highway” and a value other than “primary”.


Matches all objects with key “type” and value “multipolygon” or “boundary”.


Matches any object with a “name” or “name:de” tag with the value “Kastanienallee” or “Kastanienstrasse”.


Matches all objects with any key starting with “addr:”


Matches all objects with a name that contains the word “Paris”.

If there is no equal sign (“=”) in the expression only keys are matched and values can be anything. If there is an equal sign (“=”) in the expression, the key is to the left and the value to the right. An exclamation sign (“!”) before the equal sign means: A tag with that key, but not the value(s) to the right of the equal sign. A leading or trailing asterisk (“*”) can be used for substring or prefix matching, respectively. Commas (“,”) can be used to separate several keys or values.

All filter expressions are case-sensitive. There is no way to escape the special characters such as “=”, “*” and “,”. You can not mix comma-expressions and “*”-expressions.


All OSM objects (nodes, ways, and relations) have attributes, areas inherit their attributes from the ways and/or relations they were created from. The attributes known to osmium export are:

For areas, the type will be way or relation if the area was created from a closed way or a multipolygon or boundary relation, respectively. The id for areas is the id of the closed way or the multipolygon or boundary relation.

By default the attributes will not be in the export, because they are not necessary for most uses of OSM data. If you are interested in some (or all) attributes, add an attributes object to the config file. Add a member for each attribute you are interested in, the value can be either false (do not output this attribute), true (output this attribute with the attribute name prefixed by the @ sign) or any string, in which case the string will be used as the attribute name.

Another option is to specify attributes list in a comma-separated string for the --attributes/-a command-line option. This way you cannot control column names, but also you won’t have to create a config file.

Depending on your choice of values for the attributes objects, attributes can have the same name as tag keys. If this is the case, the conflicting tag is silently dropped. So if there is a tag “@id=foo” and you have set id to true in the attributes object, the tag will not show up in the output.

Note that the id is not necessarily unique. Even the combination type and id is not unique, because a way may end up in the output file as LineString and as (Multi)Polygon. See the --add-unique-id/-u option for a unique ID.

Area Handling

Multipolygon relations will be assembled into multipolygon geometries forming areas. Some closed ways will also form areas. Here are the detailed rules:

Non-closed way

A non-closed way (with the last node location not the same as the first node location) is always (regardless of any tags) a linestring, not an area.


A relation tagged type=multipolygon or type=boundary is always (regardless of any tags) assembled into an area.

Closed way

For a closed way (with the last node location the same as the first node location) the tags are checked: If the way has an area=yes tag, an area is created. If the way has an area=no tag, a linestring is created. An area tag with a value other than yes or no is ignored. The configuration settings area_tags and linear_tags can be used to augment the area check. If any of the tags matches the area_tags, an area is created. If any of the tags matches the linear_tags, a linestring is created. If both match, an area and a linestring is created. This is important because some objects have tags that make them both, an area and a linestring.

Output Formats

The following output formats are supported:

Output Format Options


osmium export exits with exit code


if everything went alright,


if there was an error processing the data, or


if there was a problem with the command line arguments.

Memory Usage

osmium export will usually keep all node locations and all objects needed for assembling the areas in memory. For larger data files, this can need several tens of GBytes of memory. See the osmium-index-types(5) man page for details.


Export into GeoJSON format:

osmium export data.osm.pbf -o data.geojson

Use a config file and export into GeoJSON Text Sequence format:

osmium export data.osm.pbf -o data.geojsonseq -c export-config.json

See Also


If you have any questions or want to report a bug, please go to https://osmcode.org/contact.html


Jochen Topf <jochen@topf.org>.

Referenced By

osmium(1), osmium-index-types(5).