A clojure library designed to make conversions between EDN and XML a little easier.
This repository follows the guidelines and standards of the Wall Brew Open Source Policy.
A deployed copy of the most recent version of clj-xml can be found on clojars. To use it, add the following as a dependency in your project.clj file:
The next time you build your application, Leiningen or deps.edn should pull it automatically. Alternatively, you may clone or fork the repository to work with it directly.
clj-xml runs its test suite in a matrix, replacing the clojure version with multiple previous versions of the language. If your project is not running the most recent version of clojure, please consult the most recent compatibility testing results in the CI/CD pipeline.
This library is built on top of clojure.data.xml
and extracts the element structure they've implemented.
Based on where your XML is and how it's stored, you'll want to use one of the three following functions to extract it.
- When you have already parsed an XML document into
clojure.data.xml
elements:xml->edn
- When you have a string containing an XML document:
xml-str->edn
- When you have a
java.io.InputStream
orjava.io.Reader
(generally used for HTTP/File System):xml-source->edn
Each of these functions accepts an option map as an optional second argument, supporting the following keys:
Option | Default Value | Description |
---|---|---|
:preserve-keys? |
false |
Maintain the exact keyword structure provided by clojure.xml/parse |
:preserve-attrs? |
false |
Maintain embedded XML attributes |
:remove-empty-attrs? |
false |
Remove any empty attribute maps |
:stringify-values? |
false |
Coerce non-nil, non-string, non-collection values to strings |
:remove-newlines? |
false |
Remove any newline characters in xml-str . Only applicable for xml-str->edn |
:force-seq? |
false |
Coerce all child XML nodes into an array of maps. |
:force-seq-for-paths |
[] |
A sequence of key-path sequences that will be selectively coerced into sequences. Read more about Key Pathing below. |
xml-str->edn
and xml-source->edn
also support the parsing options from clojure.data.xml
and Java's XMLInputFactory
class.
Additional documentation from Oracle is available.
This library does not override the default behavior of XMLInputFactory
.
Option | Default Value | Description |
---|---|---|
:include-node? |
#{:element :characters} |
A subset of #{:element :characters :comment}, representing the XML nodes to keep |
:location-info |
true |
Wether or not location metadata should be generated and attached to results |
:allocator |
Object created Just-In-Time | An instance of an XMLInputFactory/ALLOCATOR to allocate events |
:coalescing |
true |
Convert CDATA nodes to text, and append any adjacent text nodes |
:namespace-aware |
false |
Wether or not XML 1.0 namespacing support is enabled |
:replacing-entity-references |
false |
Wether or not internal entity text should be replaced with its XML entity form |
:supporting-external-entities |
false |
Wether or not externally hosted entities will be resolved at parsing time and evaluated |
:validating |
false |
Wether or not attached DTDs will be resolved and used to validate the XML document |
:reporter |
Object created Just-In-Time | An instance of a XMLInputFactory/REPORTER to use in place of default |
:resolver |
Object created Just-In-Time | An instance of a XMLInputFactory/RESOLVER to use in place of defaults |
:support-dtd |
false |
Wether or not attached DTDs will be read by the parser |
:skip-whitespace |
false |
Wether or not any whitespace only elements will be preserved as nodes |
Let's see how it works:
(require [clj-xml.core :as xml])
(def xml-example
{:tag :TEST_DOCUMENT
:attrs {:XMLNS "https://www.fake.not/real"}
:content
[{:tag :HEAD
:attrs nil
:content
[{:tag :META_DATA :attrs {:TYPE "title"} :content ["Some Fake Data!"]}
{:tag :META_DATA :attrs {:TYPE "tag"} :content ["Example Content"]}]}
{:tag :FILE
:attrs
{:POSTER "JANE DOE <j.doe@fake-email.not-real>"
:DATE "2020/04/12"
:SUBJECT "TEST DATA"}
:content
[{:tag :GROUPS
:attrs nil
:content
[{:tag :GROUP :attrs nil :content ["test-data-club"]}]}
{:tag :SEGMENTS
:attrs nil
:content
[{:tag :SEGMENT
:attrs {:BITS "00111010" :NUMBER "58"}
:content ["more data"]}
{:tag :SEGMENT
:attrs {:BYTES "10100010" :NUMBER "-94"}
:content ["more fake data"]}]}]}]})
(xml/xml->edn xml-example)
;; => {:test-document
;; {:head [{:meta-data "Some Fake Data!"}
;; {:meta-data "Example Content"}]
;; :file {:groups [{:group "test-data-club"}]
;; :segments [{:segment "more data"}
;; {:segment "more fake data"}]}}}
;; Parse a string instead
(def xml-test-string
"<?xml version=\"1.0\" encoding=\"UTF-8\"?><TEST_DOCUMENT XMLNS=\"https://www.fake.not/real\"><HEAD><META_DATA TYPE=\"title\">Some Fake Data!</META_DATA><META_DATA TYPE=\"tag\">Example Content</META_DATA></HEAD><FILE POSTER=\"JANE DOE <j.doe@fake-email.not-real>\" DATE=\"2020/04/12\" SUBJECT=\"TEST DATA\"><GROUPS><GROUP>test-data-club</GROUP></GROUPS><SEGMENTS><SEGMENT BITS=\"00111010\" NUMBER=\"58\">more data</SEGMENT><SEGMENT BYTES=\"10100010\" NUMBER=\"-94\">more fake data</SEGMENT></SEGMENTS></FILE></TEST_DOCUMENT>")
(xml/xml-str->edn xml-test-string)
;; => {:test-document
;; {:head [{:meta-data "Some Fake Data!"}
;; {:meta-data "Example Content"}]
;; :file {:groups [{:group "test-data-club"}]
;; :segments [{:segment "more data"}
;; {:segment "more fake data"}]}}}
;; Preserve the XML_CASE
(xml/xml->edn xml-example {:preserve-keys? true})
;; => {:TEST_DOCUMENT
;; {:HEAD [{:META-DATA "Some Fake Data!"}
;; {:META-DATA "Example Content"}]
;; :FILE {:GROUPS [{:GROUP "test-data-club"}]
;; :SEGMENTS [{:SEGMENT "more data"}
;; {:SEGMENT "more fake data"}]}}}
;; Preserve the XML attributes
(xml/xml->edn xml-example {:preserve-attrs? true})
;; => {:test-document
;; {:head [{:meta-data "Some Fake Data!" :meta-data-attrs {:type "title"}}
;; {:meta-data "Example Content" :meta-data-attrs {:type "tag"}}]
;; :file {:groups [{:group "test-data-club"}]
;; :segments [{:segment "more data" :segment-attrs {:bits "00111010" :number "58"}}
;; {:segment "more fake data" :segment-attrs {:bytes "10100010" :number "-94"}}]}
;; :file-attrs {:poster "JANE DOE <j.doe@fake-email.not-real>"
;; :date "2020/04/12"
;; :subject "TEST DATA"}}
;; :test-document-attrs {:xmlns "https://www.fake.not/real"}}
When you want to ensure selective paths in the returned XML are coerced to sequences, you may pick one of two options:
force-seq?
- to coerce all child XML nodes into a collection of nodes.force-seq-for-paths
- to coerce selective child XML nodes into collections of nodes.
In the case of force-seq-for-paths
, you will supply a sequence of key paths, each of which direct to children a la assoc-in
.
This key path may contain three types of paths, which may be used together.
- The namespace qualified keywords:
:clj-xml.core/first
,:clj-xml.core/last
, and:clj-xml.core/every
- These will modify a sequence's first, last, or every child, respectively
- The provided alias symbols for the above keywords:
first-child
,last-child
, andevery-child
- These will modify a sequence's first, last, or every child, respectively
- Bare keywords
- These will modify the matching given keyword in a map a la
update
- These will modify the matching given keyword in a map a la
If the key path is incongruent with the current data structure, (e.g. every-child
and a hash-map), an exception will be thrown.
(require [clj-xml.core :as xml])
(xml/xml->edn' xml-example {:force-seq-for-paths [[:test-document :file :segments xml/every-child]
[:test-document]]})
;; => {:test-document
;; [{:head [{:meta-data "Some Fake Data!"}
;; {:meta-data "Example Content"}]
;; :file {:groups [{:group "test-data-club"}]
;; :segments [[{:segment "more data"}]
;; [{:segment "more fake data"}
(xml/xml->edn' xml-example {:force-seq-for-paths [[xml/every-child]]})
;; => java.lang.IllegalArgumentException: The key clj-xml.core/every is incompatible with class clojure.lang.PersistentArrayMap
To convert EDN into XML, you'll want to use one of the following functions based on the target location.
- When you want
clojure.data.xml
elements:edn->xml
- When you want an unindented string containing an XML document:
edn->xml-str
- When you have a
java.io.OutputputStream
orjava.io.Writer
(generally used for HTTP/File System):edn->xml-stream
Each of these functions accepts an option map as an optional final argument, supporting the following keys:
Option | Default Value | Description |
---|---|---|
:to-xml-case? |
false |
Wether or not the keys representing XML tags will be coerced to XML_CASE |
:from-xml-case? |
false |
Wether or not the source EDN is in XML_CASE |
:stringify-values? |
false |
Wether or not non-nil, non-string, non-collection values should be coerced to strings |
edn->xml-str
and edn->xml-stream
also support the parsing options from clojure.data.xml
:
Option | Default Value | Description |
---|---|---|
:encoding |
UTF-8 |
A string representing the character encoding the encoder should emit |
:doctype |
nil |
An XML Doctype that should be written as the emitted document's DTD |
Let's see how it works:
(require [clj-xml.core :as xml])
(def edn-example-with-attrs-and-original-keys
{:TEST_DOCUMENT
{:HEAD [{:META_DATA "Some Fake Data!" :META_DATA_ATTRS {:TYPE "title"}}
{:META_DATA "Example Content" :META_DATA_ATTRS {:TYPE "tag"}}]
:FILE {:GROUPS [{:GROUP "test-data-club"}]
:SEGMENTS [{:SEGMENT "more data" :SEGMENT_ATTRS {:BITS "00111010" :NUMBER "58"}}
{:SEGMENT "more fake data" :SEGMENT_ATTRS {:BYTES "10100010" :NUMBER "-94"}}]}
:FILE_ATTRS {:POSTER "JANE DOE <j.doe@fake-email.not-real>"
:DATE "2020/04/12"
:SUBJECT "TEST DATA"}}
:TEST_DOCUMENT_ATTRS {:XMLNS "https://www.fake.not/real"}})
(xml/edn->xml-str edn-example-with-attrs-and-original-keys {:to-xml-case? true :from-xml-case? true :stringify-values? true})
;; => "<?xml version=\"1.0\" encoding=\"UTF-8\"?><TEST_DOCUMENT XMLNS=\"https://www.fake.not/real\"><HEAD><META_DATA TYPE=\"title\">Some Fake Data!</META_DATA><META_DATA TYPE=\"tag\">Example Content</META_DATA></HEAD><FILE POSTER=\"JANE DOE <j.doe@fake-email.not-real>\" DATE=\"2020/04/12\" SUBJECT=\"TEST DATA\"><GROUPS><GROUP>test-data-club</GROUP></GROUPS><SEGMENTS><SEGMENT BITS=\"00111010\" NUMBER=\"58\">more data</SEGMENT><SEGMENT BYTES=\"10100010\" NUMBER=\"-94\">more fake data</SEGMENT></SEGMENTS></FILE></TEST_DOCUMENT>"
Copyright © Wall Brew Co
This software is provided for free, public use as outlined in the MIT License