summaryrefslogtreecommitdiff
path: root/pkgtools
diff options
context:
space:
mode:
authorrillig <rillig@pkgsrc.org>2006-06-30 12:46:30 +0000
committerrillig <rillig@pkgsrc.org>2006-06-30 12:46:30 +0000
commit12561ed62dbf8e66cb60e8bb5a1a596942f27c75 (patch)
treec86e04f972786bf72170001a12642e8e2724f2fd /pkgtools
parent0480457f708ff643dfa3263ec6c6f2bd316a5bea (diff)
downloadpkgsrc-12561ed62dbf8e66cb60e8bb5a1a596942f27c75.tar.gz
SUBST_CLASSES and SUBST_FILES.* are very different.
Diffstat (limited to 'pkgtools')
-rw-r--r--pkgtools/pkglint/files/doc/chap.types.xml14
1 files changed, 13 insertions, 1 deletions
diff --git a/pkgtools/pkglint/files/doc/chap.types.xml b/pkgtools/pkglint/files/doc/chap.types.xml
index 16fe14c1faf..ef39bb30997 100644
--- a/pkgtools/pkglint/files/doc/chap.types.xml
+++ b/pkgtools/pkglint/files/doc/chap.types.xml
@@ -1,4 +1,4 @@
-<!-- $NetBSD: chap.types.xml,v 1.11 2006/06/30 12:34:44 rillig Exp $ -->
+<!-- $NetBSD: chap.types.xml,v 1.12 2006/06/30 12:46:30 rillig Exp $ -->
<chapter id="types">
<title>The &pkglint; type system</title>
@@ -520,5 +520,17 @@
<quote>good</quote> and <quote>bad</quote> variables? How can it
be decided of which kind a certain variable is?</para>
+ <para>For most lists, the only valid operation is to append
+ something at the end. Therefore it is good practice to warn if a
+ list is assigned using another operator that
+ <literal>+=</literal>. For <varname>SUBST_CLASSES</varname> this
+ fits perfectly. But for <varname>SUBST_FILES.*</varname> it
+ doesn't. Usually all occurences of a
+ <varname>SUBST_FILES.*</varname> variable occur in the same
+ file, and there should be no other file modifying these
+ variables. Therefore it is better to use the
+ <literal>=</literal> operator for the first of the
+ assignments.</para>
+
</sect1>
</chapter>