Re: [DOCS] Grammar and formatting errors for 9.02 pdf version

2010-12-22 Thread Alvaro Herrera
Excerpts from Robert Haas's message of mar dic 21 23:33:19 -0300 2010:

> With respect to formatting errors, we tend to worry more about the
> HTML versions of the docs than the PDF.

That said, we have certainly made some changes to the text to better
acommodate the PDF output, particularly where program output is too wide
to fit the page, which is more common than we'd like.

> However, you should certainly
> point out what you have noticed so that we can try to determine
> whether anything can be done to improve the situation.

What Robert said.

-- 
Álvaro Herrera 
The PostgreSQL Company - Command Prompt, Inc.
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

-- 
Sent via pgsql-docs mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs


Re: [DOCS] Grammar and formatting errors for 9.02 pdf version

2010-12-22 Thread Leslie S Satenstein

--- On Tue, 12/21/10, Robert Haas  wrote:

From: Robert Haas 
Subject: Re: [DOCS] Grammar and formatting errors for 9.02 pdf version
To: "Leslie S Satenstein" 
Cc: "pgsql-docs" 
Date: Tuesday, December 21, 2010, 10:56 PM

On Tue, Dec 21, 2010 at 10:43 PM, Leslie S Satenstein
 wrote:
>
> Hi Robert, I am not familiar with your patch process.  However, I can copy 
> and paste the offending lines in the PDF file, to a word or openoffice 
> document, and use green for insert and red text for delete and yellow for 
> elaboration or comments.  And of course, as well, to identify the page from 
> the pdf guide.
>
> Some text, if rephrased, makes the meaning more clear. Some text has missing 
> nouns and where several nouns precede a pronoun,  it causes a delay as one 
> stops to analyze the sentence in order to extract the author's meaning.
>
> If there is a better way, please advise me.

Please keep replies on-list, and write your replies beneath the quoted
text rather than above it.

To submit a patch, you need to check out the source code, edit the
SGML documentation, and then use git diff.  See:

http://wiki.postgresql.org/wiki/Submitting_a_Patch

In short:

git clone git://git.postgresql.org/git/postgresql.git

git diff > grammar.patch


If you have only a handful of changes, feel free to just point out the
parts that you think could be phrased better and how you think they
should be written, rather than generating a patch.  Actually, it'd
probably be better to point out the first few changes that way anyhow,
to see if we agree with your opinions on what should be done.  If
you're submitting a large number of edits, you're going to have to
learn how to generate a patch file as per the above, because otherwise
it's going to be extremely laborious for anyone to think about
applying your changes.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


Hi Robert.
Since Yahoo.com does not allow me to respond as an append, please excuse the 
formatting that takes place in the left margin.

 
For this time, please use html for reading this email.

Here is an example of how I was thinking to proceed but
 using a word document so that formatting would be respected.  I would 
present the page in word, and then identify the changes with red/green. 
 To keep this email short, I am showing only a very few changes. Green is for 
an insertion, Red is for a deletion.  Yellow is for clarification

PDF 9.02 Page 591, bottom
tty

Ignored (formerly, this parameter specified where to send server debug output).

Word parameter is inserted.

Page 594
Text below is truncated as it goes beyond the right margin

Make a connection to the database server in a nonblocking manner.
PGconn *PQconnectStartParams(const char **keywords, const char **values, int 
expand_dbname);
 
Page 595
To begin a nonblocking connection callexecute conn =

Further down

IfAfter PQconnectStart succeeds ...

Leaving the opening If . in my mind says, what to do if it does not succeed 
then what?  By writing after PQConnectStart succeeds... takes away liability.

There
 is some rewording that I would do on later pages.  Why do I bring these
 changes up for review, because of Latin readers (French/ Spanish 
readers are translating the English and therefore good
 precision is needed). Living in Quebec Canada, and doing business in 
Latin America, I have become trilingual and write software and 
documentation in these additional languages.

Oh yes, in this chapter 31, protocol is mentioned several times.  But the very 
first time protocol is defined or explained is in chapter 46.  I believe that a 
forward reference is required before the first mention of this property (page 
602).

Regards
Bien à vous

Leslie


[DOCS] OID

2010-12-22 Thread Leslie S Satenstein
I am reading the PDF document and ran into the term OID in the section 4.2.11. 
Array Constructors

However, OID is first defined in Chapter 5.    (5.4 System Columns).  Is it 
reasonable to perhaps move the contents of or repeat part of the contents of 
section 5.4 in sectioln 4.2.11?

Is this unreferenced term that appears earlier than the definition considered a 
document bug?.

--

Regards  
 Leslie
 Mr. Leslie Satenstein

 
mailto:[email protected]
mailto [email protected] / [email protected]
www.itbms.biz
 

[DOCS] Addition to TOAST documentation in 8.4 comprehensive manual

2010-12-22 Thread Aleksey Tsalolikhin
Hello,

  I'd like to suggest a small addition to the TOAST overview in the
8.4 comprehensive manual, Chapter 53.2 TOAST.

  This is my first doc patch for postgres, so if this is not
appropriate, please advise.  I spent almost an hour trying to figure
out how to go from the TOAST table id to the owning table name...
finally got help on #postgresql.   Thought I'd document it for other
sys admins new to PostgreSQL.

  Or is this something that would be better to go into the PostgreSQL Wiki?

Sincerely,
Aleksey

To find the parent table given a TOAST table, which has a name like
pg_toast_12513885, cast the OID of the toast table to "regclass".

For example:  Given "monkeys" table which spills over to pg_toast_12513885:

select 12513885::regclass;
 regclass
--
 monkeys
(1 row)


"regclass" stands for registered "class".  Per RhodiumToad on #postgresql,
for reasons too complicated to explain, "class" is used to mean "relation".

Reference: "casting a table's OID to regclass is handy for symbolic
display of a numeric OID"
http://www.postgresql.org/docs/8.4/static/datatype-oid.html

-- 
Sent via pgsql-docs mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs


Re: [DOCS] Addition to TOAST documentation in 8.4 comprehensive manual

2010-12-22 Thread Aleksey Tsalolikhin
I'd like to add the following to my proposed documentation tidbit, please:

   The more offical/strict way, that does not rely on the
implementation artifact that TOAST numeric ID is embedded in the TOAST
symbolic name, is:

select oid::regclass from pg_class where
reltoastrelid='pg_toast_12513885'::regclass;

Thanks for a world class database!
Sincerely,
Aleksey

-- 
Sent via pgsql-docs mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs


Re: [DOCS] OID

2010-12-22 Thread Tom Lane
Leslie S Satenstein  writes:
> Is this unreferenced term that appears earlier than the definition considered 
> a document bug?.

No.  It would be completely impractical, or at least unintelligible,
to write the manual with no forward references whatsoever.

regards, tom lane

-- 
Sent via pgsql-docs mailing list ([email protected])
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-docs