[Top][All Lists]

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

Re: [Gnu-arch-users] Namespace

From: Jean Helou
Subject: Re: [Gnu-arch-users] Namespace
Date: Tue, 23 Mar 2004 21:37:26 +0100
User-agent: Mozilla Thunderbird 0.5 (X11/20040306)

Here it is,
this will handle the problem, though I seem to be the only one to find
it is a problem. A number of email address do exist which have an
underscore in the prefix (before the@) the implicit rule that archive
names should begin with an email address may generate quite a number of
complaints without this.

I know this might break compatibility with  older versions of tla in
that it will prevent them from getting some archives created with an
underscore. Nonetheless you can't both require an email address in the
archive name and not support all valid email addresses.


--- orig/libarch/namespace.c
+++ mod/libarch/namespace.c
@@ -1074,7 +1074,7 @@
       while (char_is_alnum_or_dash (*in))

-      if (*in != '.')
+      if (*in != '.' && *in != '_' )

       if (!char_is_alnum_or_dash (in[1]))

Jean Helou wrote:

I just checked in tla--devo--1.3, just in case this had been corrected
there. Seems not. Maybe these are simply non-issues.

Anyway I'll post a patch for the first one later today just in case. I
don't know if the second one is valid or not, I will assume it is, even
though I never saw it in use :).


Jean Helou wrote:


I have been reading the namsepace validation code in libarch/namespace.(c|h) and I found the following weird (well according to me) things (I checked out address@hidden/tla--devo--1.2--patch-115)

-In the email part of an archive name, more specifically in the part before the @, arch_valid_archive_name(...) won't accept an underscore ('_'). This seems to be inconsistent with arch_valid_id which does accept them.

address@hidden ~]$tla my-id "Tester Test <address@hidden>"
address@hidden ~]$tla make-archive address@hidden archives/address@hidden
make-archive: invalid archive name (address@hidden)

-in arch_valid_patch_level_name(...) and over_patch_level(...), patch-0 is valid.

Are they bugs or is there a reason for this, if so could someone explain it to me ?


Gnu-arch-users mailing list

GNU arch home page:


Gnu-arch-users mailing list

GNU arch home page:

Attachment: signature.asc
Description: PGP signature

Attachment: signature.asc
Description: OpenPGP digital signature

reply via email to

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