Re: Make doc fails

[address@hidden]

On Aug 28, 2011, at 12:58 PM, Carl Sorensen wrote:

> On 8/28/11 3:29 AM, "address@hidden" <address@hidden> wrote:
> 
> 
>> 
>> As for the convert-ly rule, I was under the impression that these rules were
>> pushed on a version-to-version basis, and all syntax changes were written as
>> one rule before the rolling of the next version.  The last time I added one
>> was because the versions had changed and a rule didn't exist for the version
>> during which a commit of mine happened.  Please let me know how you want me 
>> to
>> go about this in the future - if there could be an entry in the CG about it,
>> that'd help a great deal.
> 
> CG 10.8 has a set of steps for adding or modifying features:
> 10.8.1 -- write the code.
> 10.8.2 -- write regression tests.
> 10.8.3 -- Write convert-ly rule.
> 10.8.4 -- Automatically update documentation
> 10.8.5 -- Manually update documentation (if a NOT_SMART rule is used)
> 10.8.6 -- Edit changes.tely
> 10.8.7 -- Verify successful build
> 10.8.8 -- Verify regression tests
> 10.8.9 -- Post patch for comments
> 10.8.10 -- Push patch
> 10.8.11 -- Close issue
> 
> It seems like many of us (including me sometimes, and I wrote these steps in
> the CG) are inclined to jump right from 10.8.2 to 10.8.9.
> 

+1 - I missed this day of class, or if I was there, I wasn't paying attention...
I'll do my best to be as meticulous about this as possible from here on out.

> Please note that 10.8.4 and 10.8.5 do *not* require the programmer to make
> sure the new change is documented.  They just require the programmer to make
> sure the new change doesn't break the build system.
> 

My changes to the stem patch don't break the build system, but I'm positive 
they lead to bizarre output in a few .ly examples.  That's why I like Graham's 
idea of some type of automatic checking.

> According to the CG, no patch should even be posted for review unless make
> doc runs successfully.  Perhaps this rule is ignored because make doc takes
> so long to run.
> 

I am way guilty of this...in fact, so guilty that I don't think I've ever once 
followed it, which means that it's a miracle that we haven't hit this problem 
before the flag patch ;(
As I previously said, I'm all for coming up with some type of automated system 
so that this condition can be relaxed.

> Mike, I think you've done a great job of contributing to LilyPond.  I'm
> certainly not throwing rocks at you -- an occasional mistake is certainly a
> small price to pay for all of the great contributions you've made.
> 

Thanks, Carl.
As always, I'm doing my best to (a) make sure there aren't any loose ends; and 
(b) if there are, clean them up as fast as possible.  Today has been kinda 
tight cause I'm moving to the burbs and packing boxes, but tomorrow I'll be 
able to respond thoroughly to everything that comes in.

Cheers,
MS