Online SQL Grammar reference is unusable in certain web-browser configurations

[GoogleCodeExporter]

Hi,

I usually use emacs-w3m for my web-browsing needs and the way the online SQL 
reference looked was rather unreadable. Then i tried looking at it in Firefox 
but i couldn't understand a thing either.

I had the enlightment when i looked at the same page on my coworker's Firefox: 
i finally saw the black lines connecting keywords. The reason i couldn't see 
those on my PC is that i prefer seeing green text on black background, so black 
lines on black background were invisible.

Though the way you present the reference is quite nice and intuitive i would 
also prefer to have it in either EBNF or at least some common ad-hoc notation.

Thanks for your work and attention.

Original issue reported on code.google.com by fercer...@gmail.com on 22 Feb 2011 at 10:51

[GoogleCodeExporter]

I updated the images, does it look better now? 
http://h2database.com/html/grammar.html#select (you may need to click 'Refresh' 
or even 'Shift+Refresh' in the browser to get the new images). I added a thick 
white line below the black line. I'm afraid that's all I can do for now, as it 
has to be images to work for multiple browsers. It's not perfect I know.

Original comment by thomas.t...@gmail.com on 22 Feb 2011 at 11:37

[GoogleCodeExporter]

Wow, thanks for the prompt response!

In my humble opinion, it looks very nice now, both with white and
black background.

The only issue i can observe now is that users of text-mode browsers
(or some other old-schoolers) might want to see the link to
http://code.google.com/p/h2database/source/browse/trunk/h2/src/docsrc/help/help.
csv
at the top of the grammar page.

Whatever you decide, i consider this issue "fixed", big thanks for the
prompt reply and for the great software.

Original comment by fercer...@gmail.com on 23 Feb 2011 at 7:52

[GoogleCodeExporter]

Hi,

It's good to know the images work now.

Text-mode browsers: I would like to support it, but I'm afraid I will not have 
the time to do it... Do you know a way to detect it's a text-mode browser? I 
didn't find a way yet... Maybe you could say "if it supports JavaScript, then 
it supports images" but that would mean it would first always show the BNF... 
What about CSS? How would I test it (I'm using Mac OS X)?

If you are interested in providing a patch please tell me!

By the way, the data is already in the html files. The PDF version of the docs 
uses the BNF, because OpenOffice can't convert the html correctly.

I'm resolving this issue as fixed, but patches are always welcome.

Original comment by thomas.t...@gmail.com on 24 Feb 2011 at 7:32

• Changed state: Fixed

[GoogleCodeExporter]

What i meant was something like:

diff --git a/h2/src/docsrc/html/grammar.html b/h2/src/docsrc/html/grammar.html
index a7fab81..9689dc6 100644
--- a/h2/src/docsrc/html/grammar.html
+++ b/h2/src/docsrc/html/grammar.html
@@ -16,6 +16,10 @@ SQL Grammar
 <!-- } -->

 <h1>SQL Grammar</h1>
+<p>This page uses "railroad diagrams" to specify the grammar. If you prefer
+to see the BNF in text form, proceed to the
+<a 
href="http://code.google.com/p/h2database/source/browse/trunk/h2/src/docsrc/help
/help.csv">
+help source</a>.</p>
 <h2>Commands (Data Manipulation)</h2>
 <!-- syntax-start
 <p class="notranslate">

And big thanks for caring about people with "unusual" tastes in
browsing.

As to the OO.org conversion to PDF, i might be misunderstanding
something but why do not you try the automatic webkit-based html to
pdf converter: http://code.google.com/p/wkhtmltopdf/ ?

Original comment by fercer...@gmail.com on 27 Feb 2011 at 6:40

[GoogleCodeExporter]

Hi,

About the link to code.google.com: I will think about it. It also not a perfect 
solution, as it doesn't work offline. I'm wondering if there is a solution that 
could use the (already existing, but disabled, after <!-- syntax-start ) inline 
BNF text.

> why do not you try the automatic webkit-based html to pdf converter

Because I didn't know about it! When I built my OpenOffice-based HTML to PDF 
converter, wkhtmltopdf was not available. I tried it now and it seems to work 
fine. There are a few things I probably need to configure, but I guess it's not 
too much work. This isn't high priority for me, but when I have time I will try 
it. Thanks a lot for the tip!

Original comment by thomas.t...@gmail.com on 1 Mar 2011 at 8:26