> Date: Sat, 28 Jun 2025 17:28:07 +0200
> From: Patrice Dumas <[email protected]>
>
> Hello,
>
> In the Info reader, the node name for "Index" is checked in addition to
> the presence of an index (with (node->flags & N_IsIndex), and possibly
> the presence of non empty menu).
>
> I think that it is wrong to do that, firstly because having (node->flags &
> N_IsIndex)
> is more accurante, an Index in a node name after the indices could lead
> to the indices not being found, and if there is no "Index" in nodes
> names, the indices will not be found. In addition "Index" is good to
> find indices for some languages only.
>
> What about using (node->flags & N_IsIndex) only?
Are you talking about info_indices_of_file_buffer? It does this:
/* Grovel the names of the nodes found in this file. */
if (file_buffer->tags)
{
TAG *tag;
for (i = 0; (tag = file_buffer->tags[i]); i++)
{
if (strcasestr (tag->nodename, "Index")
&& !(tag->flags & T_IsAnchor)) /* Not an anchor. */
{
NODE *node;
REFERENCE **menu;
node = info_node_of_tag (file_buffer, &file_buffer->tags[i]);
if (!node)
continue;
if ((node->flags & N_IsIndex) && !initial_index_filename)
{
So this checks only nodes whose name includes "Index". Are you
suggesting to check _all_ the nodes of the manual instead? Wouldn't
that be slower?
Don't index nodes always have "Index" in their names?
