bug-global
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RFC] global(1): use of logical path by default


From: Shigio YAMAGUCHI
Subject: Re: [RFC] global(1): use of logical path by default
Date: Fri, 8 May 2015 08:20:29 +0900

Hi all,
I withdraw the RFC, because I have found a problem.
Using the current implementation, we cannot sometimes
access paths generated by global(1).

(using logical path)
$ ls main.c
main.c
$ ln -s /usr/src/lib/libc libc
$ gtags
$ cd libc
$ global -P 'main.c'
../main.c
$ ls -l ../main.c
ls: ../main.c: No such file or directory        <= !!!

Since '..' links to '/usr/src/lib', ls(1) cannot find '../main.c'.

This is against a principle that all paths generated by global(1)
are accessible. It seems difficult to solve this problem, because
it is caused by incompleteness of symbolic link.

Regards,
Shigio


2015-04-20 18:04 GMT+09:00 Shigio YAMAGUCHI <address@hidden>:
Hi all,
I would like to change global(1) to use logical path by default.
This means that global(1) uses logical path even if GTAGSLOGICALPATH
is not set.

[ChangeLog]
> June 28 2014, Shigio YAMAGUCHI
>         New environment variable GTAGSLOGICALPATH was added.
>         If it is set then GLOBAL use logical path (PWD), else
>         physical path. This variable is now undocumented.

Additionally, I would like to remove GTAGSLOGICALPATH and add
GTAGSPHYSICALPATH instead. If GTAGSPHYSICALPATH is set, global(1)
follows the current (old) specification.

[What is logical path]

Logical path is a path name which include symbolic link as a
unit of the path.

o Logical path is easier to use. 
e.g.
$ ln -s /usr/src/lib/libc libc
$ gtags
$ cd libc

(using physical path)
$ global strlen
global: GTAGS not found.
$ _

(using logical path)
$ global strlen
arm/string/strlen.S
string/strlen.c
$ _

o It seems that there is no problem to use logical path.
  (If there is any problem, please let me know.)

o GNU bash use logical path by default. GNU tools should follow the same manners.

[Argument in the past]

Here is an argument about this issue.

What do you think?

Regards,
Shigio

-- 
Shigio YAMAGUCHI <address@hidden>
PGP fingerprint: D1CB 0B89 B346 4AB6 5663  C4B6 3CA5 BBB3 57BE DDA3



--
Shigio YAMAGUCHI <address@hidden>
PGP fingerprint: D1CB 0B89 B346 4AB6 5663  C4B6 3CA5 BBB3 57BE DDA3

reply via email to

[Prev in Thread] Current Thread [Next in Thread]