[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Make convert-ly -d only ever update on changed files (issue 31830043)
From: |
k-ohara5a5a |
Subject: |
Make convert-ly -d only ever update on changed files (issue 31830043) |
Date: |
Mon, 25 Nov 2013 06:37:36 +0000 |
I'm fine with this,
but then we would probably want to run convert-ly without the '-d'
option on Documentation/snippets, upon creation of version 2.18.0.
Maybe there is a way to get the behavior of "round-up-to stable version"
triggered by the (usually dummy) convert-ly rule for that version. I
can only think of kludgy ways right now
https://codereview.appspot.com/31830043/diff/1/scripts/convert-ly.py
File scripts/convert-ly.py (left):
https://codereview.appspot.com/31830043/diff/1/scripts/convert-ly.py#oldcode295
scripts/convert-ly.py:295: # check the y in x.y.z (minor version
number)
This did cause me confusion recently, while preparing the keySignature
name-change that I had postponed for the 2.19.x branch.
I knew about the mechanism to update to a stable label using a
dummy-rule mechanism in convertrules.py, but this hidden mechanism
surprised me. Why, I thought, does it know about 2.18.0 when that
version does not exist?
https://codereview.appspot.com/31830043/diff/1/scripts/convert-ly.py
File scripts/convert-ly.py (right):
https://codereview.appspot.com/31830043/diff/1/scripts/convert-ly.py#newcode306
scripts/convert-ly.py:306: last = next_stable
The old code was there so that the \version ".." is updated for a new
stable version <http://codereview.appspot.com/2642041/#msg3>
It looks like the new code does not perform this function.
It takes some patience to understand what this does, either by reading
the doc-string or the code, because it is a sophisticated behavior, and
I do not understand its purpose or motivation.
https://codereview.appspot.com/31830043/
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- Make convert-ly -d only ever update on changed files (issue 31830043),
k-ohara5a5a <=