pgsql: doc: improve build for non-Latin1 characters
doc: improve build for non-Latin1 characters
Add README.non-ASCII to explain non-ASCII doc behavior; some text moved
from release.sgml.
Change UTF8 SGML characters to use HTML entities.
Remove unnecessary UTF8 spaces.
Add SVG file check for check-nbsp target.
Add dummy 'pdf' Makefile target.
Reported-by: Yugo Nagata
Discussion: /messages/by-id/20241011114122.c90f8a871462da36f2e2afeb@sraoss.co.jp
Backpatch-through: master
Branch
------
master
Details
-------
https://git.postgresql.org/pg/commitdiff/641a5b7a1447954076728f259342c2f9201bb0b5
Modified Files
--------------
doc/src/sgml/Makefile | 11 ++++-----
doc/src/sgml/README.non-ASCII | 37 +++++++++++++++++++++++++++++++
doc/src/sgml/charset.sgml | 10 ++++-----
doc/src/sgml/images/genetic-algorithm.svg | 4 ++--
doc/src/sgml/release.sgml | 18 ---------------
doc/src/sgml/stylesheet-man.xsl | 12 +++++-----
6 files changed, 56 insertions(+), 36 deletions(-)
I'm surprised this was committed. I don't think there was consensus
that this is the right direction. Moreover, this touched a couple of
files that it shouldn't have:
doc/src/sgml/images/genetic-algorithm.svg is a generated file that
shouldn't be changed directly. In fact, adding the *.svg files to the
check-nbsp target is therefore also wrong.
The chunk in doc/src/sgml/stylesheet-man.xsl was copied from the
upstream stylesheets, and changing it causes extra differences that
would make future maintenance harder. Also, man stylesheets have by
definition nothing to do with the pdf builds, so why do they need to be
touched?
I continue to think the whole premise of this patch is wrong.
Show quoted text
On 01.11.24 17:47, Bruce Momjian wrote:
doc: improve build for non-Latin1 characters
Add README.non-ASCII to explain non-ASCII doc behavior; some text moved
from release.sgml.Change UTF8 SGML characters to use HTML entities.
Remove unnecessary UTF8 spaces.
Add SVG file check for check-nbsp target.
Add dummy 'pdf' Makefile target.
Reported-by: Yugo Nagata
Discussion: /messages/by-id/20241011114122.c90f8a871462da36f2e2afeb@sraoss.co.jp
Backpatch-through: master
Branch
------
masterDetails
-------
https://git.postgresql.org/pg/commitdiff/641a5b7a1447954076728f259342c2f9201bb0b5Modified Files
--------------
doc/src/sgml/Makefile | 11 ++++-----
doc/src/sgml/README.non-ASCII | 37 +++++++++++++++++++++++++++++++
doc/src/sgml/charset.sgml | 10 ++++-----
doc/src/sgml/images/genetic-algorithm.svg | 4 ++--
doc/src/sgml/release.sgml | 18 ---------------
doc/src/sgml/stylesheet-man.xsl | 12 +++++-----
6 files changed, 56 insertions(+), 36 deletions(-)
On Mon, Nov 4, 2024 at 08:02:18AM +0100, Peter Eisentraut wrote:
I'm surprised this was committed. I don't think there was consensus that
(thread moved to docs)
Well, I posted the final patch on October 15, and though the thread was
active with comments, no one had any feedback for my patch so I assumed
it was acceptable.
this is the right direction. Moreover, this touched a couple of files that
it shouldn't have:doc/src/sgml/images/genetic-algorithm.svg is a generated file that shouldn't
be changed directly. In fact, adding the *.svg files to the check-nbsp
target is therefore also wrong.
Good point. I have removed the SVG check from the Makefile. I also
figured out the UTF8 non-Latin1 spaces came from
doc/src/sgml/images/genetic-algorithm.gv:
a4 -> a5[xlabel="false ", fontsize=10];
a4 -> end[xlabel="true ", fontsize=10];
Are those spaces before the close quotes expected? Maybe they are used
to distance the text from an image.
The chunk in doc/src/sgml/stylesheet-man.xsl was copied from the upstream
stylesheets, and changing it causes extra differences that would make future
maintenance harder. Also, man stylesheets have by definition nothing to do
with the pdf builds, so why do they need to be touched?
Wow, I had no idea that UTF8 quotes would actually work in XML files.
If you think they actually work we can put them back but I don't know
how to test if they are working properly or not. I assume ASCII double
quotes do work.
I am thinking Tatsuo's idea of using iconv to check files with any
non-ASCII characters is a better approach and will reply to that thread.
--
Bruce Momjian <bruce@momjian.us> https://momjian.us
EDB https://enterprisedb.com
When a patient asks the doctor, "Am I going to die?", he means
"Am I going to die soon?"