Table of Contents
List of Tables
List of Examples
This module introduces a new type of variable that provides both serialization and de-serialization from JSON format.
The variable provides ways to access objects and arrays to add,replace or delete values from the script.
The correct approach is to consider a json object as a hashtable ( you can put (key;value) pairs, and you can delete and get values by key) and a json array as an array ( you can append, delete and replace values).
Since the JSON format can have objects inside other objects you can have multiple nested hashtables or arrays and you can access these using paths.
The json
variable provides
methods to access fields in json objects and
indexes in json arrays.
The json variables will be available to the process that created them from the moment they were initialized. They will not reset per message or per transaction. If you want to use the on a per message basis you should initialize them each time.
The grammar that describes the id is:
id = name(identifier)*
identifier = key | index
key = /string | /$var
index = [integer] | [$var] | []
The "[]" index represents appending to the array. It should only be used when trying to set a value and not when trying to get one.
Negative indexes can be used to access an array starting from the end. So "[-1]" signifies the last element.
IMPORTANT: The id strictly complies to this grammar. You should be careful when using spaces because they will NOT be ignored. This was done to allow keys that contain spaces.
Variables can be used as indexes or keys. Variables that will be used as indexes must contain integer values. Variables that will be used as keys should contain string values.
Trying to get a value from a non-existing path (key or value) will return the NULL value and notice messages will be placed in the log describing the value of the json and the path used.
Trying to replace or insert a value in a non-existing path will cause an error in setting the value and notice messages will be printed in the log describing the value of the json and the path used
Example 1.1. Accessing the $json variable
... $json(obj1/key) = "value"; #replace or insert the (key,value) #pair into the json object; $json(matrix1[1][2]) = 1; #replace the element at index 2 in the element #at index 1 in an array xlog("$json(name/key1[0][-1]/key2)"); # a more complex example ...
Example 1.2. Iterating through an array using variables
... $json(ar1) := "[1,2,3,4]"; $var(i) = 0; while( $json(ar1[$var(i)]) ) { #print each value xlog("Found:[$json(ar1[$var(i)])]\n"); #increment each value $json(ar1[$var(i)]) = $json(ar1[$var(i)]) + 1 ; $var(i) = $var(i) + 1; } ...
Dynamic traversal of a JSON object or array is possible by using a for each statement, similarly to the indexed pseudo variables iteration. However, note that indexing the $json variable is not supported in any other statements (this refers to indexing the entire variable and not to the indexes accepted in the grammar of the id).
In order to explicitly iterate over a JSON object keys or values, you can use the .keys or .values suffix for the path specified in the id.
Example 1.3. iteration over $json object keys
... $json(foo) := "{\"a\": 1, \"b\": 2, \"c\": 3}"; for ($var(k) in $(json(foo.keys)[*])) xlog("$var(k) "); ...
Example 1.4. iteration over $json object values
... $json(foo) := "{\"a\": 1, \"b\": 2, \"c\": 3}"; for ($var(v) in $(json(foo.values)[*])) xlog("$var(v) "); # equivalent to: $json(foo) := "{\"a\": 1, \"b\": 2, \"c\": 3}"; for ($var(v) in $(json(foo)[*])) xlog("$var(v) "); ...
Example 1.5. iteration over $json array values
... $json(foo) := "[1, 2, 3]"; for ($var(v) in $(json(foo)[*])) xlog("$var(v) "); ...
If the value specified by the id is an integer it will be returned as an integer value.
If the value specified by the id is a string it will be returned as a string.
If the value specified by the id is any other type of json ( null, boolean, object, array ) the serialized version of the object will be returned as a string value. Using this and the ":=" operator you can duplicate json objects and put them in other json objects ( for string or integer you may use the "=" operator).
If the id does not exist a NULL value will be returned.
There are 2 operators available for this variable.
This will cause the value to be taken as is and be added to the json object ( e.g. string value or integer value ).
Setting a value to NULL will cause it to be deleted.
This will cause the value to be taken and interpreted as a json object ( e.g. this operator should be used to parse json inputs ).
Example 1.10. Setting a boolean or null value
... $json(array1[]) := "null"; $json(array1[]) := "true"; $json(array1[]) := "false"; ...
Example 1.11. Adding a json to another json
... $json(array) := "[1,2,3]"; $json(object) := "{}"; $json(object/array) := $json(array) ; ...
The json_pretty
variable has the
same purpose as the json
variable,
but prints the JSON object in a pretty format, adding
spaces and tabs to make the output more readable.
This function can be used to link json objects together. This will work simillar to setting a value to an object, the only difference is that the second object is not copied, only a reference is created.
Changes to any of the objects will be visible in both of them.
You can use this method either to create references so each time you access the field you don't have to go through the full path (for speed efficiency and shorter code), or if you have an object that must be added to many other objects and you don't want to copy it each time (space and speed efficiency).
You can think of this object exactly as a reference in an object-oriented language. Modifying fields referenced by the variable will cause modifications in all the objects, BUT modifying the variable itsef will not cause any changes to other objects.
WARNING: You should be careful when using references. If you accidentally create a circular reference and try to get the value from the object you will crash OPENSIPS.
Example 1.12. Creating a reference
... $json(b) := "[{},{},{}]"; json_link($json(stub), $json(b[0])); $json(stub/ana) = "are"; #add to the stub $json(stub/ar) := "[]"; $json(stub/ar[]) = 1; $json(stub/ar[]) = 2; $json(stub/ar[]) = 3; $json(b[0]/ar[0]) = NULL; # delete from the original object xlog("\nTest link :\n$json(stub)\n$json(b)\n\n"); /*Output: Test link : { "ana": "are", "ar": [ 2, 3 ] } [ { "ana": "are", "ar": [ 2, 3 ] }, { }, { } ] */ $json(stub) = NULL; #delete the stub, no change will happen to the source xlog("\nTest link :\n$json(stub)\n$json(b)\n\n"); /* Output: Test link : <null> [ { "ana": "are", "ar": [ 2, 3 ] }, { }, { } ] */ ...
Example 1.13. [LOGICAL ERROR] Creating a circular reference
... $json(b) := "[1]"; /* NEVER do this, it is meant only to show where problems might occur */ json_link($json(b[0]), $json(b)); # replace 1 with a reference to b xlog("\nTest link :\n$json(stub)\n$json(b)\n\n"); /* this will cause OPENSIPS to crash because it will continuously try to get b, then b[0], then b ... */ ...
Table 2.1. Top contributors by DevScore(1), authored commits(2) and lines added/removed(3)
Name | DevScore | Commits | Lines ++ | Lines -- | |
---|---|---|---|---|---|
1. | Liviu Chircu (@liviuchircu) | 19 | 16 | 50 | 89 |
2. | Andrei Dragus | 18 | 4 | 1556 | 12 |
3. | Razvan Crainea (@razvancrainea) | 14 | 11 | 48 | 76 |
4. | Bogdan-Andrei Iancu (@bogdan-iancu) | 8 | 6 | 27 | 30 |
5. | Vlad Patrascu (@rvlad-patrascu) | 8 | 4 | 195 | 75 |
6. | Vlad Paiu (@vladpaiu) | 5 | 3 | 29 | 11 |
7. | Björn Esser (@besser82) | 5 | 2 | 124 | 44 |
8. | Ovidiu Sas (@ovidiusas) | 4 | 2 | 17 | 4 |
9. | Nick Altmann (@nikbyte) | 3 | 1 | 47 | 15 |
10. | Anca Vamanu | 3 | 1 | 3 | 1 |
All remaining contributors: Peter Lemenkov (@lemenkov), Julián Moreno Patiño.
(1) DevScore = author_commits + author_lines_added / (project_lines_added / project_commits) + author_lines_deleted / (project_lines_deleted / project_commits)
(2) including any documentation-related commits, excluding merge commits. Regarding imported patches/code, we do our best to count the work on behalf of the proper owner, as per the "fix_authors" and "mod_renames" arrays in opensips/doc/build-contrib.sh. If you identify any patches/commits which do not get properly attributed to you, please submit a pull request which extends "fix_authors" and/or "mod_renames".
(3) ignoring whitespace edits, renamed files and auto-generated files
Table 2.2. Most recently active contributors(1) to this module
Name | Commit Activity | |
---|---|---|
1. | Liviu Chircu (@liviuchircu) | Oct 2013 - Jan 2021 |
2. | Razvan Crainea (@razvancrainea) | Feb 2012 - Sep 2019 |
3. | Bogdan-Andrei Iancu (@bogdan-iancu) | Dec 2010 - Apr 2019 |
4. | Vlad Patrascu (@rvlad-patrascu) | May 2017 - Apr 2019 |
5. | Peter Lemenkov (@lemenkov) | Jun 2018 - Jun 2018 |
6. | Nick Altmann (@nikbyte) | Jan 2018 - Jan 2018 |
7. | Björn Esser (@besser82) | Dec 2017 - Dec 2017 |
8. | Julián Moreno Patiño | Feb 2016 - Feb 2016 |
9. | Vlad Paiu (@vladpaiu) | Jul 2014 - May 2015 |
10. | Ovidiu Sas (@ovidiusas) | Mar 2013 - Mar 2013 |
All remaining contributors: Anca Vamanu, Andrei Dragus.
(1) including any documentation-related commits, excluding merge commits
Last edited by: Liviu Chircu (@liviuchircu), Vlad Patrascu (@rvlad-patrascu), Peter Lemenkov (@lemenkov), Nick Altmann (@nikbyte), Bogdan-Andrei Iancu (@bogdan-iancu), Razvan Crainea (@razvancrainea), Andrei Dragus.
Documentation Copyrights:
Copyright © 2009 Voice Sistem SRL