https://jira.opendaylight.org This file is an XML representation of an issue en-us 8.20.10 820010 22-06-2022 https://jira.opendaylight.org/browse/YANGTOOLS-838 yangtools <p>All grouping names defined within a parent node or at the top<br/> level of the module or its submodules share the same grouping<br/> identifier namespace. This namespace is scoped to all descendant<br/> nodes of the parent node or module. This means that any<br/> descendant node may use that grouping, and it MUST NOT define a<br/> grouping with the same name.</p> <p>module test {<br/>     grouping grp;<br/>     container cont </p> {                     grouping grp；            } <p>}</p> <p> </p> YANGTOOLS-838

Yang parser should support detect grouping name collisions

Improvement Medium Resolved Done Robert Varga Jie Han Tue, 19 Dec 2017 08:02:45 +0000 Mon, 25 Dec 2017 15:57:29 +0000 Mon, 25 Dec 2017 15:57:29 +0000 2.0.1 0 2 <p>The description is a quote from <a href="https://tools.ietf.org/html/rfc7950#section-6.2.1" class="external-link" target="_blank" rel="nofollow noopener">https://tools.ietf.org/html/rfc7950#section-6.2.1</a>. Furthermore section 5.5 explicitly forbids identifier shadowing:<br/> Scoped definitions MUST NOT shadow definitions at a higher scope. A<br/> type or grouping cannot be defined if a higher level in the statement<br/> hierarchy has a definition with a matching identifier.<br/>  </p> <p><a href="https://git.opendaylight.org/gerrit/66666" class="external-link" target="_blank" rel="nofollow noopener">https://git.opendaylight.org/gerrit/66666</a></p> <p>The codes here seems doesn't work as I have tried, since it's deep first, when the deepest grouping was added to grouping namespace but the ancestor groupings have not been walked then, so here it would always return null by getFromNamespace, I have been considering to add another check phase after full declare phase to resolve it by then all groupings are in the grouping namespace.</p> <p>Adding another phase would be too intrusive and would require another complete walk of the statement tree. The patch has been fixed to check conflicts at declared statement instantiation time, which should be sufficient.</p> <p>Yes, adding a new phase is the last option, it's better to reuse the original progress.</p> Development Rank 0|i0393r: