Re: Forbid use of LF and CR characters in database and role names - Mailing list pgsql-hackers

From Ideriha, Takeshi
Subject Re: Forbid use of LF and CR characters in database and role names
Date
Msg-id 4E72940DA2BF16479384A86D54D0988A4D81FD09@G01JPEXMBKW04
Whole thread Raw
In response to Re: Forbid use of LF and CR characters in database and role names  (Michael Paquier <michael.paquier@gmail.com>)
Responses Re: Forbid use of LF and CR characters in database and role names
List pgsql-hackers
Hi, 
Here's a summary for what I tested  in RHEL7.0, details follow.

[Summary]
1. apply patch and make world  -> failed because </para> was mistakenly coded <para>.

2.correct this mistake and make check-world -> got 1 failed test: "'pg_dumpall with \n\r in database name'"    because
testscript cannot createdb "foo\n\rbar" 
 


Details follows:
[1. apply patch and make world>]

I tried to make world after applying patch and there seems to be one small mistake in create_role. 
===============================================================

openjade:ref/create_role.sgml:413:7:E: document type does not allow element "PARA" here; missing one of "FOOTNOTE",
"ITEMIZEDLIST","ORDEREDLIST", "VARIABLELIST", "CAUTION", "IMPORTANT", "NOTE", "TIP", "WARNING", "BLOCKQUOTE",
"INFORMALEXAMPLE"start-tag
 
openjade:ref/create_role.sgml:414:11:E: end tag for "PARA" omitted, but OMITTAG NO was specified
openjade:ref/create_role.sgml:413:2: start tag was here
openjade:ref/create_role.sgml:414:11:E: end tag for "PARA" omitted, but OMITTAG NO was specified
openjade:ref/create_role.sgml:408:2: start tag was here
make[3]: *** [HTML.index] Error 1
make[3]: *** Deleting file `HTML.index'

===============================================================

[2.correct this mistake and make check-world]
After fixing this mistake locally by correcting <para> to end tag, postgresql and documentation were successfully
made.
And also "make check" tests passed (not "make check-world").

================================================================
diff --git a/doc/src/sgml/ref/create_role.sgml b/doc/src/sgml/ref/create_role.sgml
index 9d6926e..ff4b6f6 100644
--- a/doc/src/sgml/ref/create_role.sgml
+++ b/doc/src/sgml/ref/create_role.sgml
@@ -410,7 +410,7 @@ CREATE ROLE <replaceable class="PARAMETER">name</replaceable> [ [ WITH ] <replac   as those could
beat the origin of security breaches, particularly on   Windows where the command shell is unusable with arguments
containing  such characters.
 
-  <para>
+  </para> </refsect1> <refsect1>
=================================================================

This failure seems actually caused by t/010_dump_connstr.pl line 65
"$node->run_log(['createdb', "foo\n\rbar"]);".

This means your patch works well.

Excerpt from log follows. 
===============================================================
#   Failed test 'pg_dumpall with \n\r in database name'
#   at /home/ideriha/postgres-master/src/bin/pg_dump/../../../src/test/perl/TestLib.pm line 230.
# Looks like you failed 1 test of 14.
t/010_dump_connstr.pl .. 

not ok 6 - pg_dumpall with \n\r in database name

Failed 1/14 subtests 

Test Summary Report
-------------------
t/010_dump_connstr.pl (Wstat: 256 Tests: 14 Failed: 1) Failed test:  6 Non-zero exit status: 1
Files=3, Tests=1822, 58 wallclock secs ( 0.20 usr  0.01 sys + 12.52 cusr  1.94 csys = 14.67 CPU)
Result: FAIL
=================================================================


Regards,
Ideriha, Takeshi
Fujitsu

pgsql-hackers by date:

Previous
From: "Okano, Naoki"
Date:
Subject: Re: pg_recvlogical --endpos
Next
From: Amit Langote
Date:
Subject: Re: Declarative partitioning - another take