git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Paul Tan <pyokagan@gmail.com>
To: Eric Sunshine <sunshine@sunshineco.com>
Cc: Johannes Schindelin <johannes.schindelin@gmx.de>,
	Git List <git@vger.kernel.org>,
	Junio C Hamano <gitster@pobox.com>,
	Matthieu Moy <Matthieu.Moy@grenoble-inp.fr>,
	Stefan Beller <sbeller@google.com>
Subject: Re: [PATCH v2 1/7] path.c: implement xdg_config_home()
Date: Sat, 18 Apr 2015 15:51:48 +0800	[thread overview]
Message-ID: <CACRoPnR7MQsP_dAE0RV_80JDbU9h6tfu30yL2OOK+JGYLzVLNQ@mail.gmail.com> (raw)
In-Reply-To: <CAPig+cTrErOBwPteeA1d_gdve5SiyLnbyFPpQ1sTN7aajGJfeA@mail.gmail.com>

Hi,

On Fri, Apr 17, 2015 at 5:41 AM, Eric Sunshine <sunshine@sunshineco.com> wrote:
> On Tue, Apr 14, 2015 at 1:28 PM, Paul Tan <pyokagan@gmail.com> wrote:
>> Below is the fixed patch. I also decided to return NULL if `filename` is
>> NULL because such an input usually indicated an uncaught error.
>
> Unfortunately, this blurs the line between programmer error (passing
> NULL for filename) and a user/configuration error (XDG_CONFIG_HOME and
> HOME being undefined). If there is indeed no valid interpretation for
> filename==NULL, then it may be better to die() or assert() here to
> flag the programmer error as early as possible, rather than returning
> NULL.
>
> More below.
>
>> ---
>> diff --git a/cache.h b/cache.h
>> index 3d3244b..2db10b8 100644
>> --- a/cache.h
>> +++ b/cache.h
>> @@ -836,6 +836,14 @@ char *strip_path_suffix(const char *path, const char *suffix);
>>  int daemon_avoid_alias(const char *path);
>>  extern int is_ntfs_dotgit(const char *name);
>>
>> +/**
>> + * Returns the newly allocated string "$XDG_CONFIG_HOME/git/{filename}".  If
>> + * $XDG_CONFIG_HOME is unset or empty, returns the newly allocated string
>> + * "$HOME/.config/git/{filename}". Returns NULL if filename is NULL or an error
>> + * occurred.
>> + */
>
> This is better than the original which said "$XDG_CONFIG_HOME/git/%s",
> but is still potentially confusing. When I read the earlier iteration,
> I was left with the impression that it was returning that literal
> string, with '$' and '%s' embedded, and that the caller would have to
> process it further to have '$' and '%s' expanded. Perhaps rephrasing
> it something like this will help?
>
>     Return a newly allocated string with value xdg+"/git/"+filename
>     where xdg is the interpolated value of $XDG_CONFIG_HOME if
>     defined and non-empty, otherwise "$HOME/.config". Return NULL
>     upon error.
>

Personally I think interpolated strings are easier to read than
concatenated strings. $VARIABLE (all upper case) in shell scripting is
understood to be an environment variable, and $variable (all lower
case) to be a local variable.

Thinking about it again, I should not be using python-style format
strings either ;-). So I would write it as
"$XDG_CONFIG_HOME/git/$filename".

But anyway, I don't have strong opinions on documentation, so I will
leave this to majority opinion. I will change it if you strongly
disagree :-).

> Also, for consistency with other API documentation, say "Return"
> rather than "Returns".

Okay, will fix.

>
> More below.
>
>> +extern char *xdg_config_home(const char *filename);
>> +
>>  /* object replacement */
>>  #define LOOKUP_REPLACE_OBJECT 1
>>  extern void *read_sha1_file_extended(const unsigned char *sha1, enum object_type *type, unsigned long *size, unsigned flag);
>> diff --git a/path.c b/path.c
>> index e608993..8ee7191 100644
>> --- a/path.c
>> +++ b/path.c
>> @@ -856,3 +856,19 @@ int is_ntfs_dotgit(const char *name)
>>                         len = -1;
>>                 }
>>  }
>> +
>> +char *xdg_config_home(const char *filename)
>> +{
>> +       const char *config_home = getenv("XDG_CONFIG_HOME");
>> +
>> +       if (!filename)
>> +               return NULL;
>
> See above regarding conflation of programmer error and user/configuration error.
>
>> +       if (!config_home || !config_home[0]) {
>
> On this project, *config_home is usually favored over config_home[0].
>
>> +               const char *home = getenv("HOME");
>> +
>> +               if (!home)
>> +                       return NULL;
>> +               return mkpathdup("%s/.config/git/%s", home, filename);
>> +       } else
>> +               return mkpathdup("%s/git/%s", config_home, filename);
>
> This logic is more difficult to follow than it ought to be due to the
> use of 'config_home' so distant from the 'if' which checked it, and
> due to the nested 'if'. It could be expressed more straight-forwardly
> as:
>
>     if (config_home && *config_home)
>         return mkpathdup("%s/git/%s", config_home, filename);
>
>     home = getenv("HOME");
>     if (home)
>         return mkpathdup("%s/.config/git/%s", home, filename);
>
>     return NULL;
>

Ah, flipping the conditionals definitely makes it look nicer. I guess
I will need your sign off to use your code? Thanks!

Regards,
Paul

  reply	other threads:[~2015-04-18  7:51 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-12  7:46 [PATCH 1/7] path.c: implement xdg_config_home() Paul Tan
2015-04-12  7:46 ` [PATCH 2/7] attr.c: replace home_config_paths() with xdg_config_home() Paul Tan
2015-04-12  7:46 ` [PATCH 3/7] dir.c: " Paul Tan
2015-04-12  7:46 ` [PATCH 4/7] credential-store.c: " Paul Tan
2015-04-12  7:46 ` [PATCH 5/7] git-commit: replace use of home_config_paths() Paul Tan
2015-04-12  7:46 ` [PATCH 6/7] git-config: " Paul Tan
2015-04-12  7:46 ` [PATCH 7/7] path.c: remove home_config_paths() Paul Tan
2015-04-13 15:50 ` [PATCH 1/7] path.c: implement xdg_config_home() Johannes Schindelin
2015-04-14 17:28   ` [PATCH v2 " Paul Tan
2015-04-16 21:41     ` Eric Sunshine
2015-04-18  7:51       ` Paul Tan [this message]
2015-04-20  0:39         ` Eric Sunshine
2015-04-21  4:06           ` [PATCH v3 " Paul Tan
2015-05-06  8:00             ` [PATCH v3 2/7] attr.c: replace home_config_paths() with xdg_config_home() Paul Tan
2015-05-06  8:01             ` [PATCH v3 3/7] dir.c: " Paul Tan
2015-05-06  8:01             ` [PATCH v3 4/7] credential-store.c: " Paul Tan
2015-05-06  8:01             ` [PATCH v3 5/7] git-commit: replace use of home_config_paths() Paul Tan
2015-05-06  8:01             ` [PATCH v3 6/7] git-config: " Paul Tan
2015-05-06  8:01             ` [PATCH v3 7/7] path.c: remove home_config_paths() Paul Tan
2015-04-18  8:48       ` [PATCH v2 1/7] path.c: implement xdg_config_home() Paul Tan
2015-04-14 20:39   ` [PATCH " Junio C Hamano
2015-04-14 22:28     ` Stefan Beller
2015-04-14 22:30       ` Junio C Hamano
2015-04-14 22:34         ` Stefan Beller
2015-04-13 21:43 ` Matthieu Moy
2015-04-14  0:18   ` Stefan Beller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CACRoPnR7MQsP_dAE0RV_80JDbU9h6tfu30yL2OOK+JGYLzVLNQ@mail.gmail.com \
    --to=pyokagan@gmail.com \
    --cc=Matthieu.Moy@grenoble-inp.fr \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=johannes.schindelin@gmx.de \
    --cc=sbeller@google.com \
    --cc=sunshine@sunshineco.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).