| Age | Commit message (Expand) | Author | Files | Lines |
|---|
| 2019-04-01 | script-updates: add changelog for global configurationchange-2change-1 |  Christian Hesse | 1 | -1/+22 |
| 2019-01-04 | global: variable names are CamelCase••• ___ _ ___ __
/ _ )(_)__ _ / _/__ _/ /_
/ _ / / _ `/ / _/ _ `/ __/
/____/_/\_, / /_/ \_,_/\__/
_ __ /___/ _ __
| | / /___ __________ (_)___ ____ _/ /
| | /| / / __ `/ ___/ __ \/ / __ \/ __ `/ /
| |/ |/ / /_/ / / / / / / / / / / /_/ /_/
|__/|__/\__,_/_/ /_/ /_/_/_/ /_/\__, (_)
/____/
RouterOS has some odd behavior when it comes to variable names. Let's
have a look at the interfaces:
[admin@MikroTik] > / interface print where name=en1
Flags: D - dynamic, X - disabled, R - running, S - slave
# NAME TYPE ACTUAL-MTU L2MTU
0 RS en1 ether 1500 1598
That looks ok. Now we use a script:
{ :local interface "en1";
/ interface print where name=$interface; }
And the result...
[admin@MikroTik] > { :local interface "en1";
{... / interface print where name=$interface; }
Flags: D - dynamic, X - disabled, R - running, S - slave
# NAME TYPE ACTUAL-MTU L2MTU
0 RS en1 ether 1500 1598
... still looks ok.
We make a little modification to the script:
{ :local name "en1";
/ interface print where name=$name; }
And the result:
[admin@MikroTik] > { :local name "en1";
{... / interface print where name=$name; }
Flags: D - dynamic, X - disabled, R - running, S - slave
# NAME TYPE ACTUAL-MTU L2MTU
0 RS en1 ether 1500 1598
1 S en2 ether 1500 1598
2 S en3 ether 1500 1598
3 S en4 ether 1500 1598
4 S en5 ether 1500 1598
5 R br-local bridge 1500 1598
Ups! The filter has no effect!
That happens whenever the variable name ($name) matches the property
name (name=).
And another modification:
{ :local type "en1";
/ interface print where name=$type; }
And the result:
[admin@MikroTik] > { :local type "en1";
{... / interface print where name=$type; }
Flags: D - dynamic, X - disabled, R - running, S - slave
# NAME TYPE ACTUAL-MTU L2MTU
Ups! Nothing?
Even if the variable name ($type) matches whatever property name (type=)
things go wrong.
The answer from MikroTik support (in Ticket#2019010222000454):
> This is how scripting works in RouterOS and we will not fix it.
To get around this we use variable names in CamelCase. Let's hope
Mikrotik never ever introduces property names in CamelCase...
*fingers crossed*
| Christian Hesse | 1 | -44/+44 |
| 2019-01-03 | script-updates: add configuration versioning | Christian Hesse | 1 | -0/+13 |
| 2019-01-02 | update copyright for 2019 | Christian Hesse | 1 | -1/+1 |
| 2018-10-16 | script-updates: run global-functions on update | Christian Hesse | 1 | -0/+3 |
| 2018-10-12 | script-updates: allow to set dont-require-permissions=yes•••This requires the new script to contain a line:
# requires: dont-require-permissions=yes
| Christian Hesse | 1 | -1/+4 |
| 2018-09-28 | script-updates: make sure new script starts with magic | Christian Hesse | 1 | -5/+9 |
| 2018-09-27 | start scripts with a magic token / shebang | Christian Hesse | 1 | -1/+1 |
| 2018-09-14 | script-updates: add error handling back in•••We have to make sure the script does not terminate on first error...
| Christian Hesse | 1 | -7/+9 |
| 2018-09-13 | script-updates: check and warn about policies | Christian Hesse | 1 | -0/+10 |
| 2018-09-03 | script-updates: fetch into variablerouteros-6.43 | Christian Hesse | 1 | -13/+16 |
| 2018-08-27 | script-updates: add support for url suffix•••This allows to fetch from different branch...
| Christian Hesse | 1 | -1/+2 |
| 2018-08-24 | add empty comment at first line...•••... for better formatting in export.
| Christian Hesse | 1 | -0/+1 |
| 2018-07-10 | script-updates: always accept updates from file | Christian Hesse | 1 | -33/+25 |
| 2018-07-09 | script-updates: handle error on fetch | Christian Hesse | 1 | -4/+8 |
| 2018-07-09 | script-updates: check certificate on fetch | Christian Hesse | 1 | -1/+3 |
| 2018-07-09 | script-updates: support fetch from url | Christian Hesse | 1 | -9/+40 |
| 2018-07-05 | add scripts | Christian Hesse | 1 | -0/+17 |
Christian Hesse