Fix token exceeding NAMELEN

"Aaron W. Swenson" <titanofold@gentoo.org> writes:

Trying to build HEAD and ran into this issue building the docs:
openjade:logicaldecoding.sgml:575:62:Q: length of name token must
not exceed NAMELEN (44)
openjade:replication-origins.sgml:87:67:Q: length of name token must
not exceed NAMELEN (44)

Hmm ... that's odd. I don't see any such failure here, and the buildfarm
members that build the docs aren't complaining either. What version of
openjade are you using exactly?

So, I've attached a patch that'll fix it.

I have no particular objection to the patch as stated, but I'm just
wondering if this is the tip of a tool compatibility iceberg we were
not previously aware of.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

On 13 May 2015 at 17:55, Tom Lane <tgl@sss.pgh.pa.us> wrote:

"Aaron W. Swenson" <titanofold@gentoo.org> writes:

Trying to build HEAD and ran into this issue building the docs:
openjade:logicaldecoding.sgml:575:62:Q: length of name token must
not exceed NAMELEN (44)
openjade:replication-origins.sgml:87:67:Q: length of name token must
not exceed NAMELEN (44)

Hmm ... that's odd. I don't see any such failure here, and the buildfarm
members that build the docs aren't complaining either. What version of
openjade are you using exactly?

So, I've attached a patch that'll fix it.

I have no particular objection to the patch as stated, but I'm just
wondering if this is the tip of a tool compatibility iceberg we were
not previously aware of.

I recall us hitting this with Slony documentation. The NAMELEN limit
lay in the SGML/DocBook configuration that was configured at the
distribution level, so that it differed (crucially) betwen Debian and
Red Hat.

Red Hat used to have a lower name length limit, and while overriding
it was technically possible, it required modifying configuration that
the distribution thought was owned by one of the SGML packages,
and hence the modification seemed pretty inadvisable.

I thought that this restriction was alleviated years ago, so I'm a bit
surprised to see this come up in 2015. (Or perhaps Gentoo hasn't
yet opened up some limits??? :-) )
--
When confronted by a difficult problem, solve it by reducing it to the
question, "How would the Lone Ranger handle this?"

On 2015-05-13 18:16, Christopher Browne wrote:

On 13 May 2015 at 17:55, Tom Lane <tgl@sss.pgh.pa.us> wrote:

"Aaron W. Swenson" <titanofold@gentoo.org> writes:

Trying to build HEAD and ran into this issue building the docs:
openjade:logicaldecoding.sgml:575:62:Q: length of name token must
not exceed NAMELEN (44)
openjade:replication-origins.sgml:87:67:Q: length of name token must
not exceed NAMELEN (44)

Hmm ... that's odd. I don't see any such failure here, and the buildfarm
members that build the docs aren't complaining either. What version of
openjade are you using exactly?

So, I've attached a patch that'll fix it.

I have no particular objection to the patch as stated, but I'm just
wondering if this is the tip of a tool compatibility iceberg we were
not previously aware of.

I recall us hitting this with Slony documentation. The NAMELEN limit
lay in the SGML/DocBook configuration that was configured at the
distribution level, so that it differed (crucially) betwen Debian and
Red Hat.

Red Hat used to have a lower name length limit, and while overriding
it was technically possible, it required modifying configuration that
the distribution thought was owned by one of the SGML packages,
and hence the modification seemed pretty inadvisable.

I thought that this restriction was alleviated years ago, so I'm a bit
surprised to see this come up in 2015. (Or perhaps Gentoo hasn't
yet opened up some limits??? :-) )

The restriction is alleviated (patched) by some distributions, and
Gentoo isn't among those.

It has been almost 4 years (the most recent Google has found) since the
last time this happened with PostgreSQL's docs.

/messages/by-id/BANLkTiktW6SRDygVfJRB4q+7dvWoQCC1Yg@mail.gmail.com

"Aaron W. Swenson" <titanofold@gentoo.org> writes:

On 2015-05-13 18:16, Christopher Browne wrote:

I thought that this restriction was alleviated years ago, so I'm a bit
surprised to see this come up in 2015. (Or perhaps Gentoo hasn't
yet opened up some limits??? :-) )

The restriction is alleviated (patched) by some distributions, and
Gentoo isn't among those.

It has been almost 4 years (the most recent Google has found) since the
last time this happened with PostgreSQL's docs.
/messages/by-id/BANLkTiktW6SRDygVfJRB4q+7dvWoQCC1Yg@mail.gmail.com

Ah, so we have hit it before and forgotten. Might as well stick to the
previous decision then. Patch applied, thanks!

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers