(docs) update documentation wrt functions moved to puppet #922
Conversation
Since PUP-5556, the regular sprintf function in puppet core supports named fields from a hash. The sprintf_hash is therefore not needed.
The documentation did not include that empty-string are not counted. It also stated that a "match" was made with the second argument when it is an equals operation.
The delete() function can be performed with the - operator, with regsubst (on string) or with the filter function. All of which are more flexible than the stdlib function. This adds examples of their use.
This corrects a faulty example in getparam() (it would not show what was intended because of order of evaluation). The example is now correct. A note is also added that shows the equivalent operation in the puppet language.
This adds a note that non existing variables result in an undef value being returned.
Before this, only the strftime.rb impl of the funciton noted that it was replaced with built-in equiv in Puppet 4.8.0. This updates the README, the unit and acceptance test.
|
Ran through our internal adhoc pipeline with no issues and all changes look as expected. |
|
@pmcmaw I just pushed one more commit (after your check). No more commits coming now - promise... (unless someone wants something changed). |
|
So running this through the adhoc pipeline doe not resolve the issue. After some high level looking it seems to be continuing to run the test. I am carrying out some investigation into this. |
|
Hey @hlindberg , So after investigation and confirmation on our internal adhoc pipelines would it be possible to replace Here is an example that I have done for size.rb: #925 Apologies! |
to instead use `return_puppet_version`
|
@pmcmaw PR updated, I changed all of the acceptance tests for functions checking with Puppet.version and 6.0.0 |
(stdlib) README edits
|
@pmcmaw @clairecadman All requested changes are now in this PR. LGTM |
|
@clairecadman if you approve these changes I will get this merged 👍 |
|
There are still the lower case 'puppet' in function files to address (I did not get that done yesterday, only fixed README.md). Merge and fix later, or fix now? |
|
Ill do it later, would be nice to get this in 👍 |
|
I can go back in a fix these later today! But feel free to merge first if
you want to get it in.
…On Fri, Jun 22, 2018 at 1:11 AM, Paula Muir ***@***.***> wrote:
Ill do it later, would be nice to get this in 👍
Squash and merge isnt enabled for this repo (apparently?) So going to try
to enable it then will get it in.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#922 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AbVL-PPhC-z5uRr21sxcLlBwW75ys7trks5t_KbIgaJpZM4UpwEE>
.
|
|
Was easily done with a global search. Added a commit. |
|
@pmcmaw The many commits are done on purpose as it makes it easier to later get relevant information when looking at "blame" for some lines (they have link to the puppet tickets that caused the doc change for example). A squash will give you a commit message containing all of the comments I made about each change for those lines. Anyway, I don't object to squashing, just my 2c for why I did many commits. |
No description provided.