I am trying to setup external symbolic links using environment variables in Fitnesse.
An environmental variable is set as shown here in the context.txt page
!define TESTHOME_PATH {D:\Fitnesse\FitnesseRoot\TestHome}
I then try and set the Symbolic Links within properties using the following setup
Name: Test
Path to Page: file://${TESTHOME_PATH}/Tests/
Yet, this error is shown when saving:
java.lang.IllegalArgumentException: Illegal character in authority at index 7: file://${TESTHOME_PATH}/Tests/
I think my setup is correct based on this reference guide.
Any help to use environmental variable in symbolic links would be of use.
There's a couple of issues here:
The documentation has a typo, file: must be followed with three slashes, e.g. file:///${MYTEST}
The symbolic link path must contain only forward slashes, not backslashes.
Related
was wondering if it is possible to allow global variables like ${PAGE_NAME} be a part of an include?
We have different teams using different feature environments (and more environments coming up) which uses symlinks and includes to environment setup pages. At the moment we're copying everytime the literal path. It is less error prone when we can use the global variables FitNesse has to offer.
For example a simple setup:
root
.Env
.FEAT1 --> connection details to setup/connect env 1.
.FEAT2 --> connection details to setup/connect env 2.
.FEAT300 --> connection details to setup/connect env 300.
.Team1
.FEAT2 --> uses !include -seamless .Env.FEAT2 and uses a symlink to suiteX
.Team10
.FEAT300 --> uses !include -seamless .Env.FEAT300 and uses symlinks to suiteX, suiteB and suiteC
etc
When using include -seamless .Env.${PAGE_NAME} it does not load the include. Preferably we want to use the variables FitNesse offers, but I'm sure we are doing something wrong or using the wrong syntax.
Any suggestions on how to use the global variables in a way we try to achieve?
I tried an other solution by defining the whole include into a variable, but that has unfortunately the same result:
!define includes {!include -seamless .Env.${PAGE_NAME} }
$includes
Writing down the global var itself resolves as expected.
I'm creating a custom Fluent-Bit image and I want a "generic" configuration file that can work on multiple cases, i.e.: it should work with a forward input sometimes and with tail input some other times.
I thought about using environment variables so to only have one input but it seems we cannot set variables in the key part only on the value side (see following code).
When I set the corresponding environment variables in a docker-entrypoint file with corresponding conditions
export INPUT_PATH="/myLogPath"
export INPUT_PATH_TYPE="path"
export INPUT_NAME="tail"
[INPUT]
Name ${INPUT_NAME}
${INPUT_PATH_TYPE} ${INPUT_PATH}
This is the error message I got
[error] [config] tail: unknown configuration property '${INPUT_PATH_TYPE}'. The following properties are allowed: path, exclude_path, key, read_from_head, refresh_interval, watcher_interval, rotate_wait, docker_mode, docker_mode_flush, docker_mode_parser, path_key, ignore_older, buffer_chunk_size, buffer_max_size, skip_long_lines, exit_on_eof, parser, tag_regex, db, db.sync, db.locking, multiline, multiline_flush, parser_firstline, and parser_.
I'm looking for a way to make it dynamic so either to have a single file with dynamic configuration or multiple files which can be included dynamically (#Include requires a static filepath from what I've seen).
EDIT: the only option I see is to have multiple input files (for each use case) and call it dynamically when starting fluent-bit in the docker-entrypoint file
I use a docker-entrypoint and split the input, filters to different files and then depending of the environment variables in the entrypoint I create a symbolic link to the corresponding file
As you can see in the screenshots, I am setting some user defined variables using 2 byte characters. I'm submitting the HTTP request to create this customer using UTF-8 encoding. The customer is being created with the correct double byte character characters because I can see them in the web app and in the DB. The problem is that I cannot see them in jmeter. It either shows little boxes or ??? question marks instead of the characters in the response data and in the debug sampler. The User defined variables is showing the characters correctly. I've added this to my user.properties file but that did not help:
sampleresult.default.encoding=UTF-8
How can I see these special characters in the response so I can Assert the record was created correctly? Any advice is appreciated. I am using jmeter 3.1 and JSON endpoints.
User Defined variables
DebugSampler
Now you need to ensure that JMeter itself is using UTF-8 encoding.
Configure Debug Sampler to show System Properties
Look for file.endoding property
If you see something different - override the existing property value using one of the following ways:
Permanent: add the next line to system.properties file (in JMeter's "bin" folder)
file.encoding=UTF-8
JMeter restart will be required to pick the property up
Temporary: set the "file.encoding" property via -D command-line argument as
jmeter -Dfile.encoding=UTF-8 -n -t ....
References:
Java - Supported Encodings
Apache JMeter Properties Customization Guide
I'm trying to follow the instructions for deploying a database via TFS build listed here:
http://www.mytechfinds.com/articles/software-testing/6-test-automation/64-db-deployment-tfs
The instructions include notes about how to configure a ConvertWorkspaceItem element. I've followed the directions, but TFS remains unhappy with my setting for 'Result' and 'Workspace'. For now, I simply entered the text from the directions ('dbproj' and 'Workspace', respectively). TFS complains about my values:
Compiler error(s) encountered processing expression "dbproj". 'dbproj' is not declared. It may be inaccessible due to its production level.
I'm trying to find basic tutorial information on the ConvertWorkspaceItem element, but other than the MSDN reference page there isn't a lot of info. Does anyone know much about configuring this element?
You need to specify valid variable names for both of these properties. there should already be a variable declared in the workflow called workspace, You will need to declare a variable of type string that you wish to receive the result of this activity and specify it's name as the Result property. It looks like in your linked article the author must have already created a variable called dbproj. At the bottom of the workflow designer is a variables tab where you can define your own variables.
This GSP:
<g:link controller="book" action="show" id="5">Example</g:link>
results in this HTML:
Example
This is relative to the HTTP host. If your site is running at http://localhost:8080 that's fine.
During development, an app will more commonly be running at http://localhost:8080/appName.
In such cases, the above link will not work - it will result in an absolute URL of http://localhost:8080/book/show/5 instead of the required http://localhost:8080/appName/book/show/5
What changes are required of the above GSP for the app name to be present in the resulting anchor's href?
The configuration setting grails.app.context should be equal to the context where you want your application deployed. If it's not set, as in the default configuration, it defaults to your application name, like http://localhost:8080/appName. If you want to deploy your app in the root context (e.g. http://locahost:8080/), add this to your Config.groovy:
grails.app.context = "/"
If the context is properly set, the URLs generated by g:link tags will include the context before the controller name.
I found that the meta tag is very useful for getting information in GSP files.
For example, if you want your application name, you can get it like this:
<g:meta name="app.name"/>
You can get any property in your application.properties file like that.
And if you, like me, need to concatenate it to another value, here is my example. Remember that any tag can be used as a method without the g: namespace. For example:
<g:set var="help" value="http://localhost:8080/${meta(name:"app.name")}/help" />
Grails documentation about this is a little poor, but it is here.
For me the single best reason to use <g:link> is that it adds the context if there is one, or omits it if you're running at http://localhost:8080 or in prod at http://mycoolsite.com - it's trivial to just concatenate the parts together yourself otherwise.
The same goes for using g:resource with css, javascript, etc. - it lets you have one GSP that works regardless of what the context is (e.g. 'appName'), since it's resolved at runtime.
I think that is what grails.serverURL is for. You defined this config variable in the Config.groovy, check the configuration documentation for Grails for more details.
Hope this helps!
createLink tag includes your appname / context parameter automatically in the link.
Here is reference doc for it.