Re: About the release of the built-in parser.

[Hideki IWAMOTO]

Hi.
> How about releasing like follwos?
> 
>                         include built-in parser? enable built-in parser by 
> default?
> -----------------------------------------------------------------------------------
> GLOBAL-5.8              Yes                       No
> GLOBAL-5.8.?            Yes                       Yes

It follows your judgment about the schedule. 

> When built-in parser becomes default, configuration variable 
> "use_external_parser" 
> will be needed.

I think that it is better to make built-in parser effective when all of 
"GTAGS", "GRTAGS",
and "GSYMS" are not specified. 
This doesn't cause the necessity for changing the configuration file that 
already exists. 

> [INCOMPATIBLE CHANGES]
> Is it right?

Yes.

On Thu, 04 Feb 2010 10:48:19 +0900, Shigio YAMAGUCHI wrote...
> Hi Iwamoto san,
> Please let me hear your opinion.
> 
> 1. Default mode
> ===============
> 
> Currently built-in parser is effective only when the configuration variable
> "use_builtin_parser" is defined.  Is there a reason why do not make this
> mode default?
> 
> How about releasing like follwos?
> 
>                         include built-in parser? enable built-in parser by 
> default?
> -----------------------------------------------------------------------------------
> GLOBAL-5.8              Yes                       No
> GLOBAL-5.8.?            Yes                       Yes
> 
> When built-in parser becomes default, configuration variable 
> "use_external_parser" 
> will be needed.
> How about saying this plan on the release note of GLOBAL-5.8?
> 
> 2. Incompatibility
> ==================
> 
> I will write a note about the incompatibility of the built-in parser like 
> follows:
> 
> [INCOMPATIBLE CHANGES]
> o gtags(built-in parser): In the existing parser (gtags-parser), some symbols
>   were stored in both GRTAGS and GSYMS. In the built-in parser, all the 
> symbols
>   defined in GTAGS have been moved from GSYMS to GRTAGS. New specification
>   is more consistent, and useful for efficiency improvement. If you feel this
>   a problem, please inform bug mailing list of your insistence.
> 
> Is it right?
> What do you think?
> --
> Shigio YAMAGUCHI <address@hidden>
> PGP fingerprint: D1CB 0B89 B346 4AB6 5663  C4B6 3CA5 BBB3 57BE DDA3

----
Hideki IWAMOTO  address@hidden