Skip to content

P
Postgres FD Implementation
Commit fc2117a4 authored by Magnus Hagander's avatar Magnus Hagander
Browse files
Options

Replace developer FAQ with a reference to the wiki, which is where 

it now lives (per discussion). Leave the other FAQs alone for now.
parent 1ad76112
Hide whitespace changes
Inline Side-by-side
Showing with 4 additions and 1869 deletions
doc/FAQ_DEV
View file @ fc2117a4
The developer FAQ can be found on the PostgreSQL wiki:
Developer's Frequently Asked Questions (FAQ) for PostgreSQL
Last updated: Tue Nov 13 22:39:08 EST 2007
Current maintainer: Bruce Momjian (bruce@momjian.us)
The most recent version of this document can be viewed at
http://www.postgresql.org/docs/faqs.FAQ_DEV.html.
_________________________________________________________________
General Questions
1.1) How do I get involved in PostgreSQL development?
1.2) What development environment is required to develop code?
1.3) What areas need work?
1.4) What do I do after choosing an item to work on?
1.5) I have developed a patch, what next?
1.6) How is a patch reviewed?
1.7) Where can I learn more about the code?
1.8) How do I download/update the current source tree?
1.9) How do I test my changes?
1.10) What tools are available for developers?
1.11) What books are good for developers?
1.12) What is configure all about?
1.13) How do I add a new port?
1.14) Why don't you use threads, raw devices, async-I/O, <insert your
favorite wizz-bang feature here>?
1.15) How are RPM's packaged?
1.16) How are CVS branches handled?
1.17) Where can I get a copy of the SQL standards?
1.18) Where can I get technical assistance?
1.19) How do I get involved in PostgreSQL web site development?
1.20) Why haven't you replaced CVS with SVN, Git, Monotone, VSS,
<insert your favorite SCM system here>?
Technical Questions
2.1) How do I efficiently access information in tables from the
backend code?
2.2) Why are table, column, type, function, view names sometimes
referenced as Name or NameData, and sometimes as char *?
2.3) Why do we use Node and List to make data structures?
2.4) I just added a field to a structure. What else should I do?
2.5) Why do we use palloc() and pfree() to allocate memory?
2.6) What is ereport()?
2.7) What is CommandCounterIncrement()?
2.8) What debugging features are available?
_________________________________________________________________
General Questions
1.1) How do I get involved in PostgreSQL development?
Download the code and have a look around. See 1.8.
Subscribe to and read the pgsql-hackers mailing list (often termed
'hackers'). This is where the major contributors and core members of
the project discuss development.
1.2) What development environment is required to develop code?
PostgreSQL is developed mostly in the C programming language. It also
makes use of Yacc and Lex.
The source code is targeted at most of the popular Unix platforms and
the Windows environment (XP, Windows 2000, and up).
Most developers make use of the open source development tool chain. If
you have contributed to open source software before, you will probably
be familiar with these tools. They include: GCC (http://gcc.gnu.org,
GDB (www.gnu.org/software/gdb/gdb.html), autoconf
(www.gnu.org/software/autoconf/) AND GNU make
(www.gnu.org/software/make/make.html.
Developers using this tool chain on Windows make use of MingW (see
http://www.mingw.org/).
Some developers use compilers from other software vendors with mixed
results.
Developers who regularly rebuild the source often pass the
--enable-depend flag to configure. The result is that when you make a
modification to a C header file, all files depend upon that file are
also rebuilt.
src/Makefile.custom can be used to set environment variables, like
CUSTOM_COPT, that are used for every compile.
1.3) What areas need work?
Outstanding features are detailed in the TODO list. This is located in
doc/TODO in the source distribution or at
http://www.postgresql.org/docs/faqs.TODO.html.
You can learn more about these features by consulting the archives,
the SQL standards and the recommend texts (see 1.11).
1.4) What do I do after choosing an item to work on?
Send an email to pgsql-hackers with a proposal for what you want to do
(assuming your contribution is not trivial). Working in isolation is
not advisable because others might be working on the same TODO item,
or you might have misunderstood the TODO item. In the email, discuss
both the internal implementation method you plan to use, and any
user-visible changes (new syntax, etc). For complex patches, it is
important to get community feeback on your proposal before starting
work. Failure to do so might mean your patch is rejected. If your work
is being sponsored by a company, read this article for tips on being
more effective.
A web site is maintained for patches awaiting review,
http://momjian.postgresql.org/cgi-bin/pgpatches, and those that are
being kept for the next release,
http://momjian.postgresql.org/cgi-bin/pgpatches_hold.
1.5) I have developed a patch, what next?
You will need to submit the patch to pgsql-patches@postgresql.org. It
will be reviewed by other contributors to the project and will be
either accepted or sent back for further work. To help ensure your
patch is reviewed and committed in a timely fashion, please try to
make sure your submission conforms to the following guidelines:
1. Ensure that your patch is generated against the most recent
version of the code, which for developers is CVS HEAD. For more on
branches in PostgreSQL, see 1.16.
2. Try to make your patch as readable as possible by following the
project's code-layout conventions. This makes it easier for the
reviewer, and there's no point in trying to layout things
differently than pgindent. Also avoid unnecessary whitespace
changes because they just distract the reviewer, and formatting
changes will be removed by the next run of pgindent.
3. The patch should be generated in contextual diff format (diff -c
and should be applicable from the root directory. If you are
unfamiliar with this, you might find the script
src/tools/make_diff/difforig useful. (Unified diffs are only
preferable if the file changes are single-line changes and do not
rely on surrounding lines.)
4. PostgreSQL is licensed under a BSD license. By posting a patch to
the public PostgreSQL mailling lists, you are giving the
PostgreSQL Global Development Group the non-revokable right to
distribute your patch under the BSD license.
5. Confirm that your changes can pass the regression tests. If your
changes are port specific, please list the ports you have tested
it on.
6. If you are adding a new feature, confirm that it has been tested
thoroughly. Try to test the feature in all conceivable scenarios.
7. New feature patches should also be accompanied by documentation
patches. If you need help checking the SQL standard, see 1.17.
8. Provide an implementation overview, preferably in code comments.
Following the surrounding code commenting style is usually a good
approach (also see
http://www.ibm.com/developerworks/linux/library/l-clear-code/?ca=d
gr-FClnxw01linuxcodetips).
9. If it is a performance patch, please provide confirming test
results to show the benefit of your patch. It is OK to post
patches without this information, though the patch will not be
applied until somebody has tested the patch and found a
significant performance improvement.
Even if you pass all of the above, the patch might still be rejected
for other reasons. Please be prepared to listen to comments and make
modifications.
You will be notified via email when the patch is applied, and your
name will appear in the next version of the release notes.
1.6) How is a patch reviewed?
Patch committers check several things before applying a patch:
* Patch follows the SQL standard or community agreed-upon behavior
* Style merges seamlessly into the surrounding code
* Written as simply and efficiently as possible
* Uses the available PostgreSQL subsystems properly
* Contains sufficient comments
* Contains code that works on all supported operating systems
* Has proper documentation
* Passes all regression tests, and if needed, adds new ones
* Behaves as expected, even under unusual cirumstances
* Contains no reliability risks
* Does not overly complicate the source code
* If performance-related, has a measureable performance benefit
* Is of sufficient usefulness to the average PostgreSQL user
* Follows existing PostgreSQL coding standards
1.7) Where can I learn more about the code?
Other than documentation in the source tree itself, you can find some
papers/presentations discussing the code at
http://www.postgresql.org/developer. An excellent presentation is at
http://neilconway.org/talks/hacking/
1.8) How do I download/update the current source tree?
There are several ways to obtain the source tree. Occasional
developers can just get the most recent source tree snapshot from
ftp://ftp.postgresql.org.
Regular developers might want to take advantage of anonymous access to
our source code management system. The source tree is currently hosted
in CVS. For details of how to obtain the source from CVS see
http://developer.postgresql.org/docs/postgres/cvs.html.
1.9) How do I test my changes?
Basic system testing
The easiest way to test your code is to ensure that it builds against
the latest version of the code and that it does not generate compiler
warnings.
It is worth advised that you pass --enable-cassert to configure. This
will turn on assertions with in the source which will often show us
bugs because they cause data corruption of segmentation violations.
This generally makes debugging much easier.
Then, perform run time testing via psql.
Regression test suite
The next step is to test your changes against the existing regression
test suite. To do this, issue "make check" in the root directory of
the source tree. If any tests fail, investigate.
If you've deliberately changed existing behavior, this change might
cause a regression test failure but not any actual regression. If so,
you should also patch the regression test suite.
Other run time testing
Some developers make use of tools such as valgrind
(http://valgrind.kde.org) for memory testing, gprof (which comes with
the GNU binutils suite) and oprofile
(http://oprofile.sourceforge.net/) for profiling and other related
tools.
What about unit testing, static analysis, model checking...?
There have been a number of discussions about other testing frameworks
and some developers are exploring these ideas.
Keep in mind the Makefiles do not have the proper dependencies for
include files. You have to do a make clean and then another make. If
you are using GCC you can use the --enable-depend option of configure
to have the compiler compute the dependencies automatically.
1.10) What tools are available for developers?
First, all the files in the src/tools directory are designed for
developers.
RELEASE_CHANGES changes we have to make for each release
backend description/flowchart of the backend directories
ccsym find standard defines made by your compiler
copyright fixes copyright notices
entab converts spaces to tabs, used by pgindent
find_static finds functions that could be made static
find_typedef finds typedefs in the source code
find_badmacros finds macros that use braces incorrectly
fsync a script to provide information about the cost of cache
syncing system calls
make_ctags make vi 'tags' file in each directory
make_diff make *.orig and diffs of source
make_etags make emacs 'etags' files
make_keywords make comparison of our keywords and SQL'92
make_mkid make mkid ID files
pgcvslog used to generate a list of changes for each release
pginclude scripts for adding/removing include files
pgindent indents source files
pgtest a semi-automated build system
thread a thread testing script
In src/include/catalog:
unused_oids a script which generates unused OIDs for use in system
catalogs
duplicate_oids finds duplicate OIDs in system catalog definitions
If you point your browser at the tools/backend/index.html file, you
will see few paragraphs describing the data flow, the backend
components in a flow chart, and a description of the shared memory
area. You can click on any flowchart box to see a description. If you
then click on the directory name, you will be taken to the source
directory, to browse the actual source code behind it. We also have
several README files in some source directories to describe the
function of the module. The browser will display these when you enter
the directory also. The tools/backend directory is also contained on
our web page under the title How PostgreSQL Processes a Query.
Second, you really should have an editor that can handle tags, so you
can tag a function call to see the function definition, and then tag
inside that function to see an even lower-level function, and then
back out twice to return to the original function. Most editors
support this via tags or etags files.
Third, you need to get id-utils from ftp://ftp.gnu.org/gnu/id-utils/
By running tools/make_mkid, an archive of source symbols can be
created that can be rapidly queried.
Some developers make use of cscope, which can be found at
http://cscope.sf.net/. Others use glimpse, which can be found at
http://webglimpse.net/.
tools/make_diff has tools to create patch diff files that can be
applied to the distribution. This produces context diffs, which is our
preferred format.
Our standard format BSD style, with each level of code indented one
tab, where each tab is four spaces. You will need to set your editor
or file viewer to display tabs as four spaces:
vi in ~/.exrc:
set tabstop=4
set sw=4
more:
more -x4
less:
less -x4
The tools/editors directory of the latest sources contains sample
settings that can be used with the emacs, xemacs and vim editors, that
assist in keeping to PostgreSQL coding standards.
pgindent will the format code by specifying flags to your operating
system's utility indent. This article describes the value of a
consistent coding style.
pgindent is run on all source files just before each beta test period.
It auto-formats all source files to make them consistent. Comment
blocks that need specific line breaks should be formatted as block
comments, where the comment starts as /*------. These comments will
not be reformatted in any way.
pginclude contains scripts used to add needed #include's to include
files, and removed unneeded #include's.
When adding system types, you will need to assign oids to them. There
is also a script called unused_oids in pgsql/src/include/catalog that
shows the unused oids.
1.11) What books are good for developers?
There are five good books:
* An Introduction to Database Systems, by C.J. Date, Addison, Wesley
* A Guide to the SQL Standard, by C.J. Date, et. al, Addison, Wesley
* Fundamentals of Database Systems, by Elmasri and Navathe
* Transaction Processing, by Jim Gray, Morgan, Kaufmann
* Transactional Information Systems by Gerhard Weikum, Kaufmann
1.12) What is configure all about?
The files configure and configure.in are part of the GNU autoconf
package. Configure allows us to test for various capabilities of the
OS, and to set variables that can then be tested in C programs and
Makefiles. Autoconf is installed on the PostgreSQL main server. To add
options to configure, edit configure.in, and then run autoconf to
generate configure.
When configure is run by the user, it tests various OS capabilities,
stores those in config.status and config.cache, and modifies a list of
*.in files. For example, if there exists a Makefile.in, configure
generates a Makefile that contains substitutions for all @var@
parameters found by configure.
When you need to edit files, make sure you don't waste time modifying
files generated by configure. Edit the *.in file, and re-run configure
to recreate the needed file. If you run make distclean from the
top-level source directory, all files derived by configure are
removed, so you see only the file contained in the source
distribution.
1.13) How do I add a new port?
There are a variety of places that need to be modified to add a new
port. First, start in the src/template directory. Add an appropriate
entry for your OS. Also, use src/config.guess to add your OS to
src/template/.similar. You shouldn't match the OS version exactly. The
configure test will look for an exact OS version number, and if not
found, find a match without version number. Edit src/configure.in to
add your new OS. (See configure item above.) You will need to run
autoconf, or patch src/configure too.
Then, check src/include/port and add your new OS file, with
appropriate values. Hopefully, there is already locking code in
src/include/storage/s_lock.h for your CPU. There is also a
src/makefiles directory for port-specific Makefile handling. There is
a backend/port directory if you need special files for your OS.
1.14) Why don't you use threads, raw devices, async-I/O, <insert your
favorite wizz-bang feature here>?
There is always a temptation to use the newest operating system
features as soon as they arrive. We resist that temptation.
First, we support 15+ operating systems, so any new feature has to be
well established before we will consider it. Second, most new
wizz-bang features don't provide dramatic improvements. Third, they
usually have some downside, such as decreased reliability or
additional code required. Therefore, we don't rush to use new features
but rather wait for the feature to be established, then ask for
testing to show that a measurable improvement is possible.
As an example, threads are not currently used in the backend code
because:
* Historically, threads were unsupported and buggy.
* An error in one backend can corrupt other backends.
* Speed improvements using threads are small compared to the
remaining backend startup time.
* The backend code would be more complex.
So, we are not ignorant of new features. It is just that we are
cautious about their adoption. The TODO list often contains links to
discussions showing our reasoning in these areas.
1.15) How are RPMs packaged?
This was written by Lamar Owen and Devrim Gündüz:
2006-10-16
As to how the RPMs are built -- to answer that question sanely
requires us to know how much experience you have with the whole RPM
paradigm. 'How is the RPM built?' is a multifaceted question. The
obvious simple answer is that we maintain:
1. A set of patches to make certain portions of the source tree
'behave' in the different environment of the RPMset;
2. The initscript;
3. Any other ancillary scripts and files;
4. A README.rpm-dist document that tries to adequately document both
the differences between the RPM build and the WHY of the
differences, as well as useful RPM environment operations (like,
using syslog, upgrading, getting postmaster to start at OS boot,
etc);
5. The spec file that throws it all together. This is not a trivial
undertaking in a package of this size.
PGDG RPM Maintainer builds the SRPM and announces the SRPM to the
pgsqlrpms-hackers list. This is a list where package builders are
subscribed. Then, the builders download the SRPM and rebuild it on
their machines.
We try to build on as many different canonical distributions as we
can. Currently we are able to build on Red Hat Linux 9, RHEL 3 and
above, and all Fedora Core Linux releases.
To test the binaries, we install them on our local machines and run
regression tests. If the package builders uses postgres user to build
the rpms, then it is possible to run regression tests during RPM
builds.
Once the build passes these tests, the binary RPMs are sent back to
PGDG RPM Maintainer and they are pushed to main FTP site, followed by
a release announcement to pgsqlrpms-* lists, pgsql-general and
pgsql-announce lists.
You will notice we said 'canonical' distributions above. That simply
means that the machine is as stock 'out of the box' as practical --
that is, everything (except select few programs) on these boxen are
installed by RPM; only official Red Hat released RPMs are used (except
in unusual circumstances involving software that will not alter the
build -- for example, installing a newer non-RedHat version of the Dia
diagramming package is OK -- installing Python 2.1 on the box that has
Python 1.5.2 installed is not, as that alters the PostgreSQL build).
The RPM as uploaded is built to as close to out-of-the-box pristine as
is possible. Only the standard released 'official to that release'
compiler is used -- and only the standard official kernel is used as
well.
PGDG RPM Building Project does not build RPMs for Mandrake .
We usually have only one SRPM for all platforms. This is because of
our limited resources. However, on some cases, we may distribute
different SRPMs for different platforms, depending on possible
compilation problems, especially on older distros.
Please note that this is a volunteered job -- We are doing our best to
keep packages up to date. We, at least, provide SRPMs for all
platforms. For example, if you do not find a RHEL 4 x86_64 RPM in our
FTP site, it means that we do not have a RHEL 4 x86_64 server around.
If you have one and want to help us, please do not hesitate to build
rpms and send to us :-)
http://pgfoundry.org/docman/view.php/1000048/98/PostgreSQL-RPM-Install
ation-PGDG.pdf has some information about building binary RPMs using
an SRPM.
PGDG RPM Building Project is a hosted on pgFoundry :
http://pgfoundry.org/projects/pgsqlrpms. We are an open community,
except one point : Our pgsqlrpms-hackers list is open to package
builders only. Still, its archives are visible to public. We use a CVS
server to save the work we have done so far. This includes spec files
and patches; as well as documents.
As to why all these files aren't part of the source tree, well, unless
there was a large cry for it to happen, we don't believe it should.
1.16) How are CVS branches managed?
This was written by Tom Lane:
2001-05-07
If you just do basic "cvs checkout", "cvs update", "cvs commit", then
you'll always be dealing with the HEAD version of the files in CVS.
That's what you want for development, but if you need to patch past
stable releases then you have to be able to access and update the
"branch" portions of our CVS repository. We normally fork off a branch
for a stable release just before starting the development cycle for
the next release.
The first thing you have to know is the branch name for the branch you
are interested in getting at. To do this, look at some long-lived
file, say the top-level HISTORY file, with "cvs status -v" to see what
the branch names are. (Thanks to Ian Lance Taylor for pointing out
that this is the easiest way to do it.) Typical branch names are:
REL7_1_STABLE
REL7_0_PATCHES
REL6_5_PATCHES
OK, so how do you do work on a branch? By far the best way is to
create a separate checkout tree for the branch and do your work in
that. Not only is that the easiest way to deal with CVS, but you
really need to have the whole past tree available anyway to test your
work. (And you *better* test your work. Never forget that dot-releases
tend to go out with very little beta testing --- so whenever you
commit an update to a stable branch, you'd better be doubly sure that
it's correct.)
Normally, to checkout the head branch, you just cd to the place you
want to contain the toplevel "pgsql" directory and say
cvs ... checkout pgsql
To get a past branch, you cd to wherever you want it and say
cvs ... checkout -r BRANCHNAME pgsql
For example, just a couple days ago I did
mkdir ~postgres/REL7_1
cd ~postgres/REL7_1
cvs ... checkout -r REL7_1_STABLE pgsql
and now I have a maintenance copy of 7.1.*.
When you've done a checkout in this way, the branch name is "sticky":
CVS automatically knows that this directory tree is for the branch,
and whenever you do "cvs update" or "cvs commit" in this tree, you'll
fetch or store the latest version in the branch, not the head version.
Easy as can be.
So, if you have a patch that needs to apply to both the head and a
recent stable branch, you have to make the edits and do the commit
twice, once in your development tree and once in your stable branch
tree. This is kind of a pain, which is why we don't normally fork the
tree right away after a major release --- we wait for a dot-release or
two, so that we won't have to double-patch the first wave of fixes.
1.17) Where can I get a copy of the SQL standards?
There are three versions of the SQL standard: SQL-92, SQL:1999, and
SQL:2003. They are endorsed by ANSI and ISO. Draft versions can be
downloaded from:
* SQL-92 http://www.contrib.andrew.cmu.edu/~shadow/sql/sql1992.txt
* SQL:1999
http://www.cse.iitb.ac.in/dbms/Data/Papers-Other/SQL1999/ansi-iso-
9075-2-1999.pdf
* SQL:2003 http://www.wiscorp.com/sql_2003_standard.zip
Some SQL standards web pages are:
* http://troels.arvin.dk/db/rdbms/links/#standards
* http://www.wiscorp.com/SQLStandards.html
* http://www.contrib.andrew.cmu.edu/~shadow/sql.html#syntax (SQL-92)
* http://dbs.uni-leipzig.de/en/lokal/standards.pdf (paper)
1.18) Where can I get technical assistance?
Many technical questions held by those new to the code have been
answered on the pgsql-hackers mailing list - the archives of which can
be found at http://archives.postgresql.org/pgsql-hackers/.
If you cannot find discussion or your particular question, feel free
to put it to the list.
Major contributors also answer technical questions, including
questions about development of new features, on IRC at
irc.freenode.net in the #postgresql channel.
1.19) How do I get involved in PostgreSQL web site development?
PostgreSQL website development is discussed on the
pgsql-www@postgresql.org mailing list. The is a project page where the
source code is available at http://pgfoundry.org/projects/pgweb.
1.20) Why haven't you replaced CVS with SVN, Git, Monotone, VSS, <insert your
favorite SCMS here>?
Currently the core developers see no SCMS that will provide enough
benefit to outwiegh the pain involved in moving to a new SCMS. Typical
problems that must be addressed by any new SCMS include:
* Run natively on all of our supported platforms.
* Integrate into the Buildfarm.
* Import our entire CVS Repository while preserving complete
history.
* Allow for anonymous checkouts.
Currently there is no intention for switching to a new SCMS until at
least the end of the 8.4 development cycle sometime in late 2008. For
more information please refer to the mailing list archives.
Technical Questions
2.1) How do I efficiently access information in tables from the backend code?
You first need to find the tuples(rows) you are interested in. There
are two ways. First, SearchSysCache() and related functions allow you
to query the system catalogs. This is the preferred way to access
system tables, because the first call to the cache loads the needed
rows, and future requests can return the results without accessing the
base table. The caches use system table indexes to look up tuples. A
list of available caches is located in
src/backend/utils/cache/syscache.c.
src/backend/utils/cache/lsyscache.c contains many column-specific
cache lookup functions.
The rows returned are cache-owned versions of the heap rows.
Therefore, you must not modify or delete the tuple returned by
SearchSysCache(). What you should do is release it with
ReleaseSysCache() when you are done using it; this informs the cache
that it can discard that tuple if necessary. If you neglect to call
ReleaseSysCache(), then the cache entry will remain locked in the
cache until end of transaction, which is tolerable but not very
desirable.
If you can't use the system cache, you will need to retrieve the data
directly from the heap table, using the buffer cache that is shared by
all backends. The backend automatically takes care of loading the rows
into the buffer cache.
Open the table with heap_open(). You can then start a table scan with
heap_beginscan(), then use heap_getnext() and continue as long as
HeapTupleIsValid() returns true. Then do a heap_endscan(). Keys can be
assigned to the scan. No indexes are used, so all rows are going to be
compared to the keys, and only the valid rows returned.
You can also use heap_fetch() to fetch rows by block number/offset.
While scans automatically lock/unlock rows from the buffer cache, with
heap_fetch(), you must pass a Buffer pointer, and ReleaseBuffer() it
when completed.
Once you have the row, you can get data that is common to all tuples,
like t_self and t_oid, by merely accessing the HeapTuple structure
entries. If you need a table-specific column, you should take the
HeapTuple pointer, and use the GETSTRUCT() macro to access the
table-specific start of the tuple. You then cast the pointer as a
Form_pg_proc pointer if you are accessing the pg_proc table, or
Form_pg_type if you are accessing pg_type. You can then access the
columns by using a structure pointer:
((Form_pg_class) GETSTRUCT(tuple))->relnatts
You must not directly change live tuples in this way. The best way is
to use heap_modifytuple() and pass it your original tuple, and the
values you want changed. It returns a palloc'ed tuple, which you pass
to heap_replace(). You can delete tuples by passing the tuple's t_self
to heap_destroy(). You use t_self for heap_update() too. Remember,
tuples can be either system cache copies, which might go away after
you call ReleaseSysCache(), or read directly from disk buffers, which
go away when you heap_getnext(), heap_endscan, or ReleaseBuffer(), in
the heap_fetch() case. Or it may be a palloc'ed tuple, that you must
pfree() when finished.
2.2) Why are table, column, type, function, view names sometimes referenced
as Name or NameData, and sometimes as char *?
Table, column, type, function, and view names are stored in system
tables in columns of type Name. Name is a fixed-length,
null-terminated type of NAMEDATALEN bytes. (The default value for
NAMEDATALEN is 64 bytes.)
typedef struct nameData
{
char data[NAMEDATALEN];
} NameData;
typedef NameData *Name;
Table, column, type, function, and view names that come into the
backend via user queries are stored as variable-length,
null-terminated character strings.
Many functions are called with both types of names, ie. heap_open().
Because the Name type is null-terminated, it is safe to pass it to a
function expecting a char *. Because there are many cases where
on-disk names(Name) are compared to user-supplied names(char *), there
are many cases where Name and char * are used interchangeably.
2.3) Why do we use Node and List to make data structures?
We do this because this allows a consistent way to pass data inside
the backend in a flexible way. Every node has a NodeTag which
specifies what type of data is inside the Node. Lists are groups of
Nodes chained together as a forward-linked list.
Here are some of the List manipulation commands:
lfirst(i), lfirst_int(i), lfirst_oid(i)
return the data (a pointer, integer or OID respectively) of
list cell i.
lnext(i)
return the next list cell after i.
foreach(i, list)
loop through list, assigning each list cell to i. It is
important to note that i is a ListCell *, not the data in the
List element. You need to use lfirst(i) to get at the data.
Here is a typical code snippet that loops through a List
containing Var *'s and processes each one:
List *list;
ListCell *i;
foreach(i, list)
{
Var *var = lfirst(i);
/* process var here */
}
lcons(node, list)
add node to the front of list, or create a new list with node
if list is NIL.
lappend(list, node)
add node to the end of list.
list_concat(list1, list2)
Concatenate list2 on to the end of list1.
list_length(list)
return the length of the list.
list_nth(list, i)
return the i'th element in list, counting from zero.
lcons_int, ...
There are integer versions of these: lcons_int, lappend_int,
etc. Also versions for OID lists: lcons_oid, lappend_oid, etc.
You can print nodes easily inside gdb. First, to disable output
truncation when you use the gdb print command:
(gdb) set print elements 0
Instead of printing values in gdb format, you can use the next two
commands to print out List, Node, and structure contents in a verbose
format that is easier to understand. List's are unrolled into nodes,
and nodes are printed in detail. The first prints in a short format,
and the second in a long format:
(gdb) call print(any_pointer)
(gdb) call pprint(any_pointer)
The output appears in the server log file, or on your screen if you
are running a backend directly without a postmaster.
2.4) I just added a field to a structure. What else should I do?
The structures passed around in the parser, rewriter, optimizer, and
executor require quite a bit of support. Most structures have support
routines in src/backend/nodes used to create, copy, read, and output
those structures (in particular, the files copyfuncs.c and
equalfuncs.c. Make sure you add support for your new field to these
files. Find any other places the structure might need code for your
new field. mkid is helpful with this (see 1.10).
2.5) Why do we use palloc() and pfree() to allocate memory?
palloc() and pfree() are used in place of malloc() and free() because
we find it easier to automatically free all memory allocated when a
query completes. This assures us that all memory that was allocated
gets freed even if we have lost track of where we allocated it. There
are special non-query contexts that memory can be allocated in. These
affect when the allocated memory is freed by the backend.
2.6) What is ereport()?
ereport() is used to send messages to the front-end, and optionally
terminate the current query being processed. The first parameter is an
ereport level of DEBUG (levels 1-5), LOG, INFO, NOTICE, ERROR, FATAL,
or PANIC. NOTICE prints on the user's terminal and to the server logs.
INFO prints only to the user's terminal and LOG prints only to the
server logs. (These can be changed from postgresql.conf.) ERROR prints
in both places, and terminates the current query, never returning from
the call. FATAL terminates the backend process. The remaining
parameters of ereport are a printf-style set of parameters to print.
ereport(ERROR) frees most memory and open file descriptors so you
don't need to clean these up before the call.
2.7) What is CommandCounterIncrement()?
Normally, transactions can not see the rows they modify. This allows
UPDATE foo SET x = x + 1 to work correctly.
However, there are cases where a transactions needs to see rows
affected in previous parts of the transaction. This is accomplished
using a Command Counter. Incrementing the counter allows transactions
to be broken into pieces so each piece can see rows modified by
previous pieces. CommandCounterIncrement() increments the Command
Counter, creating a new part of the transaction.
2.8) What debugging features are available?
First, try running configure with the --enable-cassert option, many
assert()s monitor the progress of the backend and halt the program
when something unexpected occurs.
The postgres server has a -d option that allows even more detailed
information to be reported. The -d option takes a number that
specifies the debug level. Be warned that high debug level values
generate large log files.
If the postmaster is not running, you can actually run the postgres
backend from the command line, and type your SQL statement directly.
This is recommended only for debugging purposes. If you have compiled
with debugging symbols, you can use a debugger to see what is
happening. Because the backend was not started from postmaster, it is
not running in an identical environment and locking/backend
interaction problems might not be duplicated.
If the postmaster is running, start psql in one window, then find the
PID of the postgres process used by psql using SELECT
pg_backend_pid(). Use a debugger to attach to the postgres PID. You
can set breakpoints in the debugger and issue queries from the other.
If you are looking to find the location that is generating an error or
log message, set a breakpoint at errfinish. psql. If you are debugging
postgres startup, you can set PGOPTIONS="-W n", then start psql. This
will cause startup to delay for n seconds so you can attach to the
process with the debugger, set any breakpoints, and continue through
the startup sequence.
You can also compile with profiling to see what functions are taking
execution time. The backend profile files will be deposited in the
pgsql/data directory. The client profile file will be put in the
client's current directory. Linux requires a compile with
-DLINUX_PROFILE for proper profiling.
http://wiki.postgresql.org/wiki/Development_information
doc/src/FAQ/FAQ_DEV.html
View file @ fc2117a4
......@@ -2,1042 +2,13 @@
<HTML>
<HEAD>
<META name="generator" content=
"HTML Tidy for BSD/OS (vers 1st July 2002), see www.w3.org">
<TITLE>PostgreSQL Developers FAQ</TITLE>
</HEAD>
<BODY bgcolor="#FFFFFF" text="#000000" link="#FF0000" vlink="#A00000"
alink="#0000FF">
<H1>Developer's Frequently Asked Questions (FAQ) for
PostgreSQL</H1>
<P>Last updated: Tue Nov 13 22:39:08 EST 2007</P>
<P>Current maintainer: Bruce Momjian (<A href=
"mailto:bruce@momjian.us">bruce@momjian.us</A>)<BR>
</P>
<P>The most recent version of this document can be viewed at <A
href=
"http://www.postgresql.org/docs/faqs.FAQ_DEV.html">http://www.postgresql.org/docs/faqs.FAQ_DEV.html</A>.</P>
<HR>
<BR>
<H2>General Questions</H2>
<A href="#item1.1">1.1</A>) How do I get involved in PostgreSQL
development?<BR>
<A href="#item1.2">1.2</A>) What development environment is required
to develop code?<BR>
<A href="#item1.3">1.3</A>) What areas need work?<BR>
<A href="#item1.4">1.4</A>) What do I do after choosing an item to
work on?<BR>
<A href="#item1.5">1.5</A>) I have developed a patch, what next?<BR>
<A href="#item1.7">1.6</A>) How is a patch reviewed?<BR>
<A href="#item1.7">1.7</A>) Where can I learn more about the code?<BR>
<A href="#item1.8">1.8</A>) How do I download/update the current
source tree?<BR>
<A href="#item1.9">1.9</A>) How do I test my changes?<BR>
<A href="#item1.10">1.10</A>) What tools are available for
developers?<BR>
<A href="#item1.11">1.11</A>) What books are good for developers?<BR>
<A href="#item1.12">1.12</A>) What is configure all about?<BR>
<A href="#item1.13">1.13</A>) How do I add a new port?<BR>
<A href="#item1.14">1.14</A>) Why don't you use threads, raw
devices, async-I/O, &lt;insert your favorite wizz-bang feature
here&gt;?<BR>
<A href="#item1.15">1.15</A>) How are RPM's packaged?<BR>
<A href="#item1.16">1.16</A>) How are CVS branches handled?<BR>
<A href="#item1.17">1.17</A>) Where can I get a copy of the SQL
standards?<BR>
<A href="#item1.18">1.18</A>) Where can I get technical
assistance?<BR>
<A href="#item1.19">1.19</A>) How do I get involved in PostgreSQL web
site development?<BR>
<A href="#item1.20">1.20</A>) Why haven't you replaced CVS with SVN, Git,
Monotone, VSS, &lt;insert your favorite SCM system here&gt;?
<H2>Technical Questions</H2>
<A href="#item2.1">2.1</A>) How do I efficiently access information in
tables from the backend code?<BR>
<A href="#item2.2">2.2</A>) Why are table, column, type, function,
view names sometimes referenced as <I>Name</I> or <I>NameData,</I>
and sometimes as <I>char *?</I><BR>
<A href="#item2.3">2.3</A>) Why do we use <I>Node</I> and <I>List</I>
to make data structures?<BR>
<A href="#item2.4">2.4</A>) I just added a field to a structure. What
else should I do?<BR>
<A href="#item2.5">2.5</A>) Why do we use <I>palloc</I>() and
<I>pfree</I>() to allocate memory?<BR>
<A href="#item2.6">2.6</A>) What is ereport()?<BR>
<A href="#item2.7">2.7</A>) What is CommandCounterIncrement()?<BR>
<A href="#item2.8">2.8</A>) What debugging features are available?<BR>
<BR>
<HR>
<H2>General Questions</H2>
<H3 id="item1.1">1.1) How do I get involved in PostgreSQL
development?</H3>
<P>Download the code and have a look around. See <A href=
"#item1.8">1.8</A>.</P>
<P>Subscribe to and read the <A href=
"http://archives.postgresql.org/pgsql-hackers">pgsql-hackers</A>
mailing list (often termed 'hackers'). This is where the major
contributors and core members of the project discuss
development.</P>
<H3 id="item1.2">1.2) What development environment is required
to develop code?</H3>
<P>PostgreSQL is developed mostly in the C programming language. It
also makes use of Yacc and Lex.</P>
<P>The source code is targeted at most of the popular Unix
platforms and the Windows environment (XP, Windows 2000, and
up).</P>
<P>Most developers make use of the open source development tool
chain. If you have contributed to open source software before, you
will probably be familiar with these tools. They include: GCC (<A
href="http://gcc.gnu.org">http://gcc.gnu.org</A>, GDB (<A href=
"http://www.gnu.org/software/gdb/gdb.html">www.gnu.org/software/gdb/gdb.html</A>),
autoconf (<A href=
"http://www.gnu.org/software/autoconf/">www.gnu.org/software/autoconf/</A>)
AND GNU make (<A href=
"http://www.gnu.org/software/make/make.html">www.gnu.org/software/make/make.html</A>.</P>
<P>Developers using this tool chain on Windows make use of MingW
(see <A href=
"http://www.mingw.org/">http://www.mingw.org/</A>).</P>
<P>Some developers use compilers from other software vendors with
mixed results.</P>
<P>Developers who regularly rebuild the source often pass the
--enable-depend flag to <I>configure</I>. The result is that when you
make a modification to a C header file, all files depend upon that
file are also rebuilt.</P>
<P><I>src/Makefile.custom</I> can be used to set environment variables,
like <I>CUSTOM_COPT</I>, that are used for every compile.
<H3 id="item1.3">1.3) What areas need work?</H3>
Outstanding features are detailed in the TODO list. This is located
in <I>doc/TODO</I> in the source distribution or at <A href=
"http://www.postgresql.org/docs/faqs.TODO.html">
http://www.postgresql.org/docs/faqs.TODO.html</A>.
<P>You can learn more about these features by consulting the
archives, the SQL standards and the recommend texts (see <A href=
"#item1.11">1.11</A>).</P>
<H3 id="item1.4">1.4) What do I do after choosing an item to
work on?</H3>
<P>Send an email to pgsql-hackers with a proposal for what you want
to do (assuming your contribution is not trivial). Working in
isolation is not advisable because others might be working on the same
TODO item, or you might have misunderstood the TODO item. In the
email, discuss both the internal implementation method you plan to
use, and any user-visible changes (new syntax, etc). For complex
patches, it is important to get community feeback on your proposal
before starting work. Failure to do so might mean your patch is
rejected. If your work is being sponsored by a company, read this
<a href="http://momjian.us/main/writings/pgsql/company_contributions/">
article</a> for tips on being more effective.</P>
<P>A web site is maintained for patches awaiting review,
<a href="http://momjian.postgresql.org/cgi-bin/pgpatches">
http://momjian.postgresql.org/cgi-bin/pgpatches</a>, and
those that are being kept for the next release,
<a href="http://momjian.postgresql.org/cgi-bin/pgpatches_hold">
http://momjian.postgresql.org/cgi-bin/pgpatches_hold</a>.</P>
<H3 id="item1.5">1.5) I have developed a patch, what next?</H3>
<P>You will need to submit the patch to pgsql-patches@postgresql.org. It
will be reviewed by other contributors to the project and will be
either accepted or sent back for further work. To help ensure your patch
is reviewed and committed in a timely fashion, please try to make sure your
submission conforms to the following guidelines:
<ol>
<li>Ensure that your patch is generated against the most recent version
of the code, which for developers is CVS HEAD. For more on branches in
PostgreSQL, see <a href="#item1.16">1.16</a>.</li>
<li>Try to make your patch as readable as possible by following the
project's code-layout conventions. This makes it easier for the
reviewer, and there's no point in trying to layout things
differently than pgindent. Also avoid unnecessary whitespace
changes because they just distract the reviewer, and formatting
changes will be removed by the next run of pgindent.</li>
<li>The patch should be generated in contextual diff format (<i>diff
-c</i> and should be applicable from the root directory. If you are
unfamiliar with this, you might find the script
<I>src/tools/make_diff/difforig</I> useful. (Unified diffs are only
preferable if the file changes are single-line changes and do not
rely on surrounding lines.)</li>
<li>PostgreSQL is licensed under a BSD license. By posting a patch
to the public PostgreSQL mailling lists, you are giving the PostgreSQL
Global Development Group the non-revokable right to distribute your
patch under the BSD license.</li>
<li>Confirm that your changes can pass the regression tests. If your
changes are port specific, please list the ports you have tested it
on.</li>
<li>If you are adding a new feature, confirm that it has been tested
thoroughly. Try to test the feature in all conceivable
scenarios.</li>
<li>New feature patches should also be accompanied by documentation
patches. If you need help checking the SQL standard, see <a href=
"#item1.17">1.17</a>.</li>
<li>Provide an implementation overview, preferably in code comments.
Following the surrounding code commenting style is usually a good
approach (also see <a
href="http://www.ibm.com/developerworks/linux/library/l-clear-code/?ca=dgr-FClnxw01linuxcodetips">http://www.ibm.com/developerworks/linux/library/l-clear-code/?ca=dgr-FClnxw01linuxcodetips</a>).</li>
<li>If it is a performance patch, please provide confirming test
results to show the benefit of your patch. It is OK to post patches
without this information, though the patch will not be applied until
somebody has tested the patch and found a significant performance
improvement.</li>
</ol>
<p>Even if you pass all of the above, the patch might still be
rejected for other reasons. Please be prepared to listen to comments
and make modifications.</p>
<p>You will be notified via email when the patch is applied, and
your name will appear in the next version of the release notes.</p>
<H3 id="item1.6">1.6) How is a patch reviewed?</H3>
<p>Patch committers check several things before applying a patch:</p>
<ul>
<li>Patch follows the SQL standard or community agreed-upon behavior</li>
<li>Style merges seamlessly into the surrounding code</li>
<li>Written as simply and efficiently as possible</li>
<li>Uses the available PostgreSQL subsystems properly</li>
<li>Contains sufficient comments</li>
<li>Contains code that works on all supported operating systems</li>
<li>Has proper documentation</li>
<li>Passes all regression tests, and if needed, adds new ones</li>
<li>Behaves as expected, even under unusual cirumstances</li>
<li>Contains no reliability risks</li>
<li>Does not overly complicate the source code</li>
<li>If performance-related, has a measureable performance benefit</li>
<li>Is of sufficient usefulness to the average PostgreSQL user</li>
<li>Follows existing PostgreSQL coding standards</li>
</ul>
<H3 id="item1.7">1.7) Where can I learn more about the
code?</H3>
<P>Other than documentation in the source tree itself, you can find
some papers/presentations discussing the code at <A href=
"http://www.postgresql.org/developer">
http://www.postgresql.org/developer</A>. An excellent presentation
is at <a href=
"http://neilconway.org/talks/hacking/">http://neilconway.org/talks/hacking/</a></P>
<H3 id="item1.8">1.8) How do I download/update the current
source tree?</H3>
<P>There are several ways to obtain the source tree. Occasional
developers can just get the most recent source tree snapshot from
<A href=
"ftp://ftp.postgresql.org">ftp://ftp.postgresql.org</A>.</P>
<P>Regular developers might want to take advantage of anonymous
access to our source code management system. The source tree is
currently hosted in CVS. For details of how to obtain the source
from CVS see <A href=
"http://developer.postgresql.org/docs/postgres/cvs.html">
http://developer.postgresql.org/docs/postgres/cvs.html</A>.</P>
<H3 id="item1.9">1.9) How do I test my changes?</H3>
<P><B>Basic system testing</B></P>
<P>The easiest way to test your code is to ensure that it builds
against the latest version of the code and that it does not generate
compiler warnings.</P>
<P>It is worth advised that you pass --enable-cassert to
<I>configure</I>. This will turn on assertions with in the source
which will often show us bugs because they cause data corruption of
segmentation violations. This generally makes debugging much
easier.</P>
<P>Then, perform run time testing via psql.</P>
<P><B>Regression test suite</B></P>
<P>The next step is to test your changes against the existing
regression test suite. To do this, issue "make check" in the root
directory of the source tree. If any tests fail, investigate.</P>
<P>If you've deliberately changed existing behavior, this change
might cause a regression test failure but not any actual regression.
If so, you should also patch the regression test suite.</P>
<P><B>Other run time testing</B></P>
<P>Some developers make use of tools such as valgrind (<A href=
"http://valgrind.kde.org">http://valgrind.kde.org</A>) for memory
testing, gprof (which comes with the GNU binutils suite) and
oprofile (<A href=
"http://oprofile.sourceforge.net/">http://oprofile.sourceforge.net/</A>)
for profiling and other related tools.</P>
<P><B>What about unit testing, static analysis, model
checking...?</B></P>
<P>There have been a number of discussions about other testing
frameworks and some developers are exploring these ideas.</P>
<P>Keep in mind the <I>Makefiles</I> do not have the proper
dependencies for include files. You have to do a <I>make clean</I>
and then another <I>make</I>. If you are using <SMALL>GCC</SMALL>
you can use the <I>--enable-depend</I> option of <I>configure</I>
to have the compiler compute the dependencies automatically.</P>
<H3 id="item1.10">1.10) What tools are available for
developers?</H3>
<P>First, all the files in the <I>src/tools</I> directory are
designed for developers.</P>
<PRE>
RELEASE_CHANGES changes we have to make for each release
backend description/flowchart of the backend directories
ccsym find standard defines made by your compiler
copyright fixes copyright notices
entab converts spaces to tabs, used by pgindent
find_static finds functions that could be made static
find_typedef finds typedefs in the source code
find_badmacros finds macros that use braces incorrectly
fsync a script to provide information about the cost of cache
syncing system calls
make_ctags make vi 'tags' file in each directory
make_diff make *.orig and diffs of source
make_etags make emacs 'etags' files
make_keywords make comparison of our keywords and SQL'92
make_mkid make mkid ID files
pgcvslog used to generate a list of changes for each release
pginclude scripts for adding/removing include files
pgindent indents source files
pgtest a semi-automated build system
thread a thread testing script
</PRE>
<P>In <I>src/include/catalog</I>:</P>
<PRE>
unused_oids a script which generates unused OIDs for use in system
catalogs
duplicate_oids finds duplicate OIDs in system catalog definitions
</PRE>
If you point your browser at the <I>tools/backend/index.html</I>
file, you will see few paragraphs describing the data flow, the
backend components in a flow chart, and a description of the shared
memory area. You can click on any flowchart box to see a
description. If you then click on the directory name, you will be
taken to the source directory, to browse the actual source code
behind it. We also have several README files in some source
directories to describe the function of the module. The browser
will display these when you enter the directory also. The
<I>tools/backend</I> directory is also contained on our web page
under the title <I>How PostgreSQL Processes a Query.</I>
<P>Second, you really should have an editor that can handle tags,
so you can tag a function call to see the function definition, and
then tag inside that function to see an even lower-level function,
and then back out twice to return to the original function. Most
editors support this via <I>tags</I> or <I>etags</I> files.</P>
<P>Third, you need to get <I>id-utils</I> from <A href=
"ftp://ftp.gnu.org/gnu/id-utils/">ftp://ftp.gnu.org/gnu/id-utils/</A></P>
<P>By running <I>tools/make_mkid</I>, an archive of source symbols
can be created that can be rapidly queried.</P>
<P>Some developers make use of cscope, which can be found at <A
href="http://cscope.sf.net">http://cscope.sf.net/</A>. Others use
glimpse, which can be found at <A href=
"http://webglimpse.net/">http://webglimpse.net/</A>.</P>
<P><I>tools/make_diff</I> has tools to create patch diff files that
can be applied to the distribution. This produces context diffs,
which is our preferred format.</P>
<P>Our standard format <I>BSD</I> style, with each level of code indented
one tab, where each tab is four spaces. You will need to set your editor
or file viewer to display tabs as four spaces:<BR>
</P>
<PRE>
vi in ~/.exrc:
set tabstop=4
set sw=4
more:
more -x4
less:
less -x4
</PRE>
<P>The <I>tools/editors</I> directory of the latest sources contains sample
settings that can be used with the <I>emacs</I>, <I>xemacs</I> and
<I>vim</I> editors, that assist in keeping to PostgreSQL coding standards.
</P>
<P><I>pgindent</I> will the format code by specifying flags to your
operating system's utility <I>indent.</I> This <A href=
"http://ezine.daemonnews.org/200112/single_coding_style.html">article</A>
describes the value of a consistent coding style.</P>
<P><I>pgindent</I> is run on all source files just before each beta
test period. It auto-formats all source files to make them
consistent. Comment blocks that need specific line breaks should be
formatted as <I>block comments,</I> where the comment starts as
<CODE>/*------</CODE>. These comments will not be reformatted in
any way.</P>
<P><I>pginclude</I> contains scripts used to add needed
<CODE>#include</CODE>'s to include files, and removed unneeded
<CODE>#include</CODE>'s.</P>
<P>When adding system types, you will need to assign oids to them.
There is also a script called <I>unused_oids</I> in
<I>pgsql/src/include/catalog</I> that shows the unused oids.</P>
<H3 id="item1.11">1.11) What books are good for
developers?</H3>
<P>There are five good books:
<UL>
<LI><I>An Introduction to Database Systems,</I> by C.J. Date, Addison, Wesley</LI>
<LI><I>A Guide to the SQL Standard,</I> by C.J. Date, et. al, Addison, Wesley</LI>
<LI><I>Fundamentals of Database Systems,</I> by Elmasri and Navathe</LI>
<LI><I>Transaction Processing,</I> by Jim Gray, Morgan, Kaufmann</LI>
<LI><I>Transactional Information Systems</I> by Gerhard Weikum, Kaufmann</LI>
</UL
</P>
<H3 id="item1.12">1.12) What is configure all about?</H3>
<P>The files <I>configure</I> and <I>configure.in</I> are part of
the GNU <I>autoconf</I> package. Configure allows us to test for
various capabilities of the OS, and to set variables that can then
be tested in C programs and Makefiles. Autoconf is installed on the
PostgreSQL main server. To add options to configure, edit
<I>configure.in,</I> and then run <I>autoconf</I> to generate
<I>configure.</I></P>
<P>When <I>configure</I> is run by the user, it tests various OS
capabilities, stores those in <I>config.status</I> and
<I>config.cache,</I> and modifies a list of <I>*.in</I> files. For
example, if there exists a <I>Makefile.in,</I> configure generates
a <I>Makefile</I> that contains substitutions for all @var@
parameters found by configure.</P>
<P>When you need to edit files, make sure you don't waste time
modifying files generated by <I>configure.</I> Edit the <I>*.in</I>
file, and re-run <I>configure</I> to recreate the needed file. If
you run <I>make distclean</I> from the top-level source directory,
all files derived by configure are removed, so you see only the
file contained in the source distribution.</P>
<H3 id="item1.13">1.13) How do I add a new port?</H3>
<P>There are a variety of places that need to be modified to add a
new port. First, start in the <I>src/template</I> directory. Add an
appropriate entry for your OS. Also, use <I>src/config.guess</I> to
add your OS to <I>src/template/.similar.</I> You shouldn't match
the OS version exactly. The <I>configure</I> test will look for an
exact OS version number, and if not found, find a match without
version number. Edit <I>src/configure.in</I> to add your new OS.
(See configure item above.) You will need to run autoconf, or patch
<I>src/configure</I> too.</P>
<P>Then, check <I>src/include/port</I> and add your new OS file,
with appropriate values. Hopefully, there is already locking code
in <I>src/include/storage/s_lock.h</I> for your CPU. There is also
a <I>src/makefiles</I> directory for port-specific Makefile
handling. There is a <I>backend/port</I> directory if you need
special files for your OS.</P>
<H3 id="item1.14">1.14) Why don't you use threads, raw
devices, async-I/O, &lt;insert your favorite wizz-bang feature
here&gt;?</H3>
<P>There is always a temptation to use the newest operating system
features as soon as they arrive. We resist that temptation.</P>
<P>First, we support 15+ operating systems, so any new feature has
to be well established before we will consider it. Second, most new
<I>wizz-bang</I> features don't provide <I>dramatic</I>
improvements. Third, they usually have some downside, such as
decreased reliability or additional code required. Therefore, we
don't rush to use new features but rather wait for the feature to
be established, then ask for testing to show that a measurable
improvement is possible.</P>
<P>As an example, threads are not currently used in the backend
code because:</P>
<UL>
<LI>Historically, threads were unsupported and buggy.</LI>
<LI>An error in one backend can corrupt other backends.</LI>
<LI>Speed improvements using threads are small compared to the
remaining backend startup time.</LI>
<LI>The backend code would be more complex.</LI>
</UL>
<P>So, we are not ignorant of new features. It is just that we are
cautious about their adoption. The TODO list often contains links
to discussions showing our reasoning in these areas.</P>
<H3 id="item1.15">1.15) How are RPMs packaged?</H3>
<P>This was written by Lamar Owen and Devrim Gündüz:</P>
<P>2006-10-16</P>
<P>
As to how the RPMs are built -- to answer that question sanely
requires us to know how much experience you have with the whole RPM
paradigm. 'How is the RPM built?' is a multifaceted question. The
obvious simple answer is that we maintain:</P>
<OL>
<LI>A set of patches to make certain portions of the source tree
'behave' in the different environment of the RPMset;</LI>
<LI>The initscript;</LI>
<LI>Any other ancillary scripts and files;</LI>
<LI>A README.rpm-dist document that tries to adequately document
both the differences between the RPM build and the WHY of the
differences, as well as useful RPM environment operations (like,
using syslog, upgrading, getting postmaster to start at OS boot,
etc);</LI>
<LI>The spec file that throws it all together. This is not a
trivial undertaking in a package of this size.</LI>
</OL>
<P>PGDG RPM Maintainer builds the SRPM and announces the SRPM to the
pgsqlrpms-hackers list. This is a list where package builders are
subscribed. Then, the builders download the SRPM and rebuild it on their
machines.</P>
<P>We try to build on as many different canonical distributions as we can.
Currently we are able to build on Red Hat Linux 9, RHEL 3 and above,
and all Fedora Core Linux releases.</P>
<P>To test the binaries, we install them on our local machines and run
regression tests. If the package builders uses postgres user to build the
rpms, then it is possible to run regression tests during RPM builds.</P>
<P>Once the build passes these tests, the binary RPMs are sent back to PGDG
RPM Maintainer and they are pushed to main FTP site, followed by a
release announcement to pgsqlrpms-* lists, pgsql-general and
pgsql-announce lists.</P>
<P>You will notice we said 'canonical' distributions above. That simply
means that the machine is as stock 'out of the box' as practical --
that is, everything (except select few programs) on these boxen are
installed by RPM; only official Red Hat released RPMs are used (except
in unusual circumstances involving software that will not alter the
build -- for example, installing a newer non-RedHat version of the Dia
diagramming package is OK -- installing Python 2.1 on the box that has
Python 1.5.2 installed is not, as that alters the PostgreSQL build).
The RPM as uploaded is built to as close to out-of-the-box pristine as
is possible. Only the standard released 'official to that release'
compiler is used -- and only the standard official kernel is used as
well.</P>
<P>PGDG RPM Building Project does not build RPMs for Mandrake .</P>
<P>We usually have only one SRPM for all platforms. This is because of our
limited resources. However, on some cases, we may distribute different
SRPMs for different platforms, depending on possible compilation problems,
especially on older distros.</P>
<P>Please note that this is a volunteered job -- We are doing our best to
keep packages up to date. We, at least, provide SRPMs for all platforms.
For example, if you do not find a RHEL 4 x86_64 RPM in our FTP site, it
means that we do not have a RHEL 4 x86_64 server around. If you have one
and want to help us, please do not hesitate to build rpms and send to us :-)
http://pgfoundry.org/docman/view.php/1000048/98/PostgreSQL-RPM-Installation-PGDG.pdf
has some information about building binary RPMs using an SRPM.</P>
<P>PGDG RPM Building Project is a hosted on pgFoundry :
<a href="http://pgfoundry.org/projects/pgsqlrpms">http://pgfoundry.org/projects/pgsqlrpms</a>.
We are an open community, except one point : Our pgsqlrpms-hackers list is open
to package builders only. Still, its archives are visible to public.
We use a CVS server to save the work we have done so far. This includes
spec files and patches; as well as documents.</P>
<P>As to why all these files aren't part of the source tree, well, unless
there was a large cry for it to happen, we don't believe it should.</P>
<H3 id="item1.16">1.16) How are CVS branches managed?</H3>
<P>This was written by Tom Lane:</P>
<P>2001-05-07</P>
<P>If you just do basic "cvs checkout", "cvs update", "cvs commit",
then you'll always be dealing with the HEAD version of the files in
CVS. That's what you want for development, but if you need to patch
past stable releases then you have to be able to access and update
the "branch" portions of our CVS repository. We normally fork off a
branch for a stable release just before starting the development
cycle for the next release.</P>
<P>The first thing you have to know is the branch name for the
branch you are interested in getting at. To do this, look at some
long-lived file, say the top-level HISTORY file, with "cvs status
-v" to see what the branch names are. (Thanks to Ian Lance Taylor
for pointing out that this is the easiest way to do it.) Typical
branch names are:</P>
<PRE>
REL7_1_STABLE
REL7_0_PATCHES
REL6_5_PATCHES
</PRE>
<P>OK, so how do you do work on a branch? By far the best way is to
create a separate checkout tree for the branch and do your work in
that. Not only is that the easiest way to deal with CVS, but you
really need to have the whole past tree available anyway to test
your work. (And you *better* test your work. Never forget that
dot-releases tend to go out with very little beta testing --- so
whenever you commit an update to a stable branch, you'd better be
doubly sure that it's correct.)</P>
<P>Normally, to checkout the head branch, you just cd to the place
you want to contain the toplevel "pgsql" directory and say</P>
<PRE>
cvs ... checkout pgsql
</PRE>
<P>To get a past branch, you cd to wherever you want it and
say</P>
<PRE>
cvs ... checkout -r BRANCHNAME pgsql
</PRE>
<P>For example, just a couple days ago I did</P>
<PRE>
mkdir ~postgres/REL7_1
cd ~postgres/REL7_1
cvs ... checkout -r REL7_1_STABLE pgsql
</PRE>
<P>and now I have a maintenance copy of 7.1.*.</P>
<P>When you've done a checkout in this way, the branch name is
"sticky": CVS automatically knows that this directory tree is for
the branch, and whenever you do "cvs update" or "cvs commit" in
this tree, you'll fetch or store the latest version in the branch,
not the head version. Easy as can be.</P>
<P>So, if you have a patch that needs to apply to both the head and
a recent stable branch, you have to make the edits and do the
commit twice, once in your development tree and once in your stable
branch tree. This is kind of a pain, which is why we don't normally
fork the tree right away after a major release --- we wait for a
dot-release or two, so that we won't have to double-patch the first
wave of fixes.</P>
<H3 id="item1.17">1.17) Where can I get a copy of the SQL
standards?</H3>
<P>There are three versions of the SQL standard: SQL-92, SQL:1999,
and SQL:2003. They are endorsed by ANSI and ISO. Draft versions can
be downloaded from:</P>
<UL>
<LI>SQL-92 <A href=
"http://www.contrib.andrew.cmu.edu/~shadow/sql/sql1992.txt">http://www.contrib.andrew.cmu.edu/~shadow/sql/sql1992.txt</A></LI>
<LI>SQL:1999 <A href=
"http://www.cse.iitb.ac.in/dbms/Data/Papers-Other/SQL1999/ansi-iso-9075-2-1999.pdf">
http://www.cse.iitb.ac.in/dbms/Data/Papers-Other/SQL1999/ansi-iso-9075-2-1999.pdf</A></LI>
<LI>SQL:2003 <A href=
"http://www.wiscorp.com/sql_2003_standard.zip">http://www.wiscorp.com/sql_2003_standard.zip</A></LI>
</UL>
<P>Some SQL standards web pages are:</P>
<UL>
<LI><A href=
"http://troels.arvin.dk/db/rdbms/links/#standards">http://troels.arvin.dk/db/rdbms/links/#standards</A></LI>
<LI><A href=
"http://www.wiscorp.com/SQLStandards.html">http://www.wiscorp.com/SQLStandards.html</A></LI>
<LI><A href=
"http://www.contrib.andrew.cmu.edu/~shadow/sql.html#syntax">http://www.contrib.andrew.cmu.edu/~shadow/sql.html#syntax</A>
(SQL-92)</LI>
<LI><A href=
"http://dbs.uni-leipzig.de/en/lokal/standards.pdf">http://dbs.uni-leipzig.de/en/lokal/standards.pdf</A>
(paper)</LI>
</UL>
<H3 id="item1.18">1.18) Where can I get technical
assistance?</H3>
<P>Many technical questions held by those new to the code have been
answered on the pgsql-hackers mailing list - the archives of which
can be found at <A href=
"http://archives.postgresql.org/pgsql-hackers/">http://archives.postgresql.org/pgsql-hackers/</A>.</P>
<P>If you cannot find discussion or your particular question, feel
free to put it to the list.</P>
<P>Major contributors also answer technical questions, including
questions about development of new features, on IRC at
irc.freenode.net in the #postgresql channel.</P>
<H3 id="item1.19">1.19) How do I get involved in PostgreSQL
web site development?</H3>
<P>PostgreSQL website development is discussed on the
pgsql-www@postgresql.org mailing list. The is a project page where
the source code is available at <A href=
"http://pgfoundry.org/projects/pgweb">http://pgfoundry.org/projects/pgweb</A>.</P>
<H3 id="item1.20">1.20) Why haven't you replaced CVS with SVN, Git,
Monotone, VSS, &lt;insert your favorite SCMS here&gt;?</H3>
<P>Currently the core developers see no SCMS that will provide
enough benefit to outwiegh the pain involved in moving to a new
SCMS. Typical problems that must be addressed by any new SCMS include:</P>
<ul>
<li>Run natively on all of our <a href="http://www.postgresql.org/docs/current/interactive/supported-platforms.html">supported platforms</a>.</li>
<li>Integrate into the <a href="http://pgbuildfarm.org/">Buildfarm</a>.</li>
<li>Import our entire CVS Repository while preserving complete history.</li>
<li>Allow for anonymous checkouts.</li>
</ul>
<P>Currently there is no intention for switching to a new SCMS until at least the
end of the 8.4 development cycle sometime in late 2008. For more information
please refer to the mailing list archives.</P>
<H2>Technical Questions</H2>
<H3 id="item2.1">2.1) How do I efficiently access information
in tables from the backend code?</H3>
<P>You first need to find the tuples(rows) you are interested in.
There are two ways. First, <I>SearchSysCache()</I> and related
functions allow you to query the system catalogs. This is the
preferred way to access system tables, because the first call to
the cache loads the needed rows, and future requests can return the
results without accessing the base table. The caches use system
table indexes to look up tuples. A list of available caches is
located in <I>src/backend/utils/cache/syscache.c.</I>
<I>src/backend/utils/cache/lsyscache.c</I> contains many
column-specific cache lookup functions.</P>
<P>The rows returned are cache-owned versions of the heap rows.
Therefore, you must not modify or delete the tuple returned by
<I>SearchSysCache()</I>. What you <I>should</I> do is release it
with <I>ReleaseSysCache()</I> when you are done using it; this
informs the cache that it can discard that tuple if necessary. If
you neglect to call <I>ReleaseSysCache()</I>, then the cache entry
will remain locked in the cache until end of transaction, which is
tolerable but not very desirable.</P>
<P>If you can't use the system cache, you will need to retrieve the
data directly from the heap table, using the buffer cache that is
shared by all backends. The backend automatically takes care of
loading the rows into the buffer cache.</P>
<P>Open the table with <I>heap_open().</I> You can then start a
table scan with <I>heap_beginscan(),</I> then use
<I>heap_getnext()</I> and continue as long as
<I>HeapTupleIsValid()</I> returns true. Then do a
<I>heap_endscan().</I> <I>Keys</I> can be assigned to the
<I>scan.</I> No indexes are used, so all rows are going to be
compared to the keys, and only the valid rows returned.</P>
<P>You can also use <I>heap_fetch()</I> to fetch rows by block
number/offset. While scans automatically lock/unlock rows from the
buffer cache, with <I>heap_fetch(),</I> you must pass a
<I>Buffer</I> pointer, and <I>ReleaseBuffer()</I> it when
completed.</P>
<P>Once you have the row, you can get data that is common to all
tuples, like <I>t_self</I> and <I>t_oid,</I> by merely accessing
the <I>HeapTuple</I> structure entries. If you need a
table-specific column, you should take the HeapTuple pointer, and
use the <I>GETSTRUCT()</I> macro to access the table-specific start
of the tuple. You then cast the pointer as a <I>Form_pg_proc</I>
pointer if you are accessing the pg_proc table, or
<I>Form_pg_type</I> if you are accessing pg_type. You can then
access the columns by using a structure pointer:</P>
<PRE>
<CODE>((Form_pg_class) GETSTRUCT(tuple))-&gt;relnatts
</CODE>
</PRE>
You must not directly change <I>live</I> tuples in this way. The
best way is to use <I>heap_modifytuple()</I> and pass it your
original tuple, and the values you want changed. It returns a
palloc'ed tuple, which you pass to <I>heap_replace().</I> You can
delete tuples by passing the tuple's <I>t_self</I> to
<I>heap_destroy().</I> You use <I>t_self</I> for
<I>heap_update()</I> too. Remember, tuples can be either system
cache copies, which might go away after you call
<I>ReleaseSysCache()</I>, or read directly from disk buffers, which
go away when you <I>heap_getnext()</I>, <I>heap_endscan</I>, or
<I>ReleaseBuffer()</I>, in the <I>heap_fetch()</I> case. Or it may
be a palloc'ed tuple, that you must <I>pfree()</I> when finished.
<H3 id="item2.2">2.2) Why are table, column, type, function,
view names sometimes referenced as <I>Name</I> or <I>NameData,</I>
and sometimes as <I>char *?</I></H3>
<P>Table, column, type, function, and view names are stored in
system tables in columns of type <I>Name.</I> Name is a
fixed-length, null-terminated type of <I>NAMEDATALEN</I> bytes.
(The default value for NAMEDATALEN is 64 bytes.)</P>
<PRE>
<CODE>typedef struct nameData
{
char data[NAMEDATALEN];
} NameData;
typedef NameData *Name;
</CODE>
</PRE>
Table, column, type, function, and view names that come into the
backend via user queries are stored as variable-length,
null-terminated character strings.
<P>Many functions are called with both types of names, ie.
<I>heap_open().</I> Because the Name type is null-terminated, it is
safe to pass it to a function expecting a char *. Because there are
many cases where on-disk names(Name) are compared to user-supplied
names(char *), there are many cases where Name and char * are used
interchangeably.</P>
<H3 id="item2.3">2.3) Why do we use <I>Node</I> and
<I>List</I> to make data structures?</H3>
<P>We do this because this allows a consistent way to pass data
inside the backend in a flexible way. Every node has a
<I>NodeTag</I> which specifies what type of data is inside the
Node. <I>Lists</I> are groups of <I>Nodes chained together as a
forward-linked list.</I></P>
<P>Here are some of the <I>List</I> manipulation commands:</P>
<BLOCKQUOTE>
<DL>
<DT>lfirst(i), lfirst_int(i), lfirst_oid(i)</DT>
<DD>return the data (a pointer, integer or OID respectively) of
list cell <I>i.</I></DD>
<DT>lnext(i)</DT>
<DD>return the next list cell after <I>i.</I></DD>
<DT>foreach(i, list)</DT>
<DD>
loop through <I>list,</I> assigning each list cell to
<I>i.</I> It is important to note that <I>i</I> is a ListCell *,
not the data in the <I>List</I> element. You need to use
<I>lfirst(i)</I> to get at the data. Here is a typical code
snippet that loops through a List containing <I>Var *'s</I>
and processes each one:
<PRE>
<CODE>
List *list;
ListCell *i;
foreach(i, list)
{
Var *var = lfirst(i);
/* process var here */
}
</CODE>
</PRE>
</DD>
<DT>lcons(node, list)</DT>
<DD>add <I>node</I> to the front of <I>list,</I> or create a
new list with <I>node</I> if <I>list</I> is <I>NIL.</I></DD>
<DT>lappend(list, node)</DT>
<DD>add <I>node</I> to the end of <I>list.</I></DD>
<DT>list_concat(list1, list2)</DT>
<DD>Concatenate <I>list2</I> on to the end of <I>list1.</I></DD>
<DT>list_length(list)</DT>
<DD>return the length of the <I>list.</I></DD>
<DT>list_nth(list, i)</DT>
<DD>return the <I>i</I>'th element in <I>list,</I>
counting from zero.</DD>
<DT>lcons_int, ...</DT>
<DD>There are integer versions of these: <I>lcons_int,
lappend_int</I>, etc. Also versions for OID lists: <I>lcons_oid,
lappend_oid</I>, etc.</DD>
</DL>
</BLOCKQUOTE>
You can print nodes easily inside <I>gdb.</I> First, to disable
output truncation when you use the gdb <I>print</I> command:
<PRE>
<CODE>(gdb) set print elements 0
</CODE>
</PRE>
Instead of printing values in gdb format, you can use the next two
commands to print out List, Node, and structure contents in a
verbose format that is easier to understand. List's are unrolled
into nodes, and nodes are printed in detail. The first prints in a
short format, and the second in a long format:
<PRE>
<CODE>(gdb) call print(any_pointer)
(gdb) call pprint(any_pointer)
</CODE>
</PRE>
The output appears in the server log file, or on your screen if
you are running a backend directly without a postmaster.
<H3 id="item2.4">2.4) I just added a field to a structure.
What else should I do?</H3>
<P>The structures passed around in the parser, rewriter,
optimizer, and executor require quite a bit of support. Most
structures have support routines in <I>src/backend/nodes</I> used
to create, copy, read, and output those structures (in particular,
the files <I>copyfuncs.c</I> and <I>equalfuncs.c</I>. Make sure you
add support for your new field to these files. Find any other
places the structure might need code for your new field. <I>mkid</I>
is helpful with this (see <A href="#item1.10">1.10</A>).</P>
<H3 id="item2.5">2.5) Why do we use <I>palloc</I>() and
<I>pfree</I>() to allocate memory?</H3>
<P><I>palloc()</I> and <I>pfree()</I> are used in place of malloc()
and free() because we find it easier to automatically free all
memory allocated when a query completes. This assures us that all
memory that was allocated gets freed even if we have lost track of
where we allocated it. There are special non-query contexts that
memory can be allocated in. These affect when the allocated memory
is freed by the backend.</P>
<H3 id="item2.6">2.6) What is ereport()?</H3>
<P><I>ereport()</I> is used to send messages to the front-end, and
optionally terminate the current query being processed. The first
parameter is an ereport level of <I>DEBUG</I> (levels 1-5),
<I>LOG,</I> <I>INFO,</I> <I>NOTICE,</I> <I>ERROR,</I> <I>FATAL,</I>
or <I>PANIC.</I> <I>NOTICE</I> prints on the user's terminal and
to the server logs. <I>INFO</I> prints only to the user's terminal
and <I>LOG</I> prints only to the server logs. (These can be
changed from <I>postgresql.conf.</I>) <I>ERROR</I> prints in both
places, and terminates the current query, never returning from the
call. <I>FATAL</I> terminates the backend process. The remaining
parameters of <I>ereport</I> are a <I>printf</I>-style set of
parameters to print.</P>
<P><I>ereport(ERROR)</I> frees most memory and open file
descriptors so you don't need to clean these up before the
call.</P>
<H3 id="item2.7">2.7) What is CommandCounterIncrement()?</H3>
<P>Normally, transactions can not see the rows they modify.
This allows <CODE>UPDATE foo SET x = x + 1</CODE> to work
correctly.</P>
<P>However, there are cases where a transactions needs to see
rows affected in previous parts of the transaction. This is
accomplished using a Command Counter. Incrementing the counter
allows transactions to be broken into pieces so each piece can
see rows modified by previous pieces. <I>CommandCounterIncrement()</I>
increments the Command Counter, creating a new part of the
transaction.</P>
<H3 id="item2.8">2.8) What debugging features are available?</H3>
<P>First, try running <I>configure</I> with the --enable-cassert
option, many <I>assert()</I>s monitor the progress of the
backend and halt the program when something unexpected occurs.</P>
<P>The postgres server has a <I>-d</I> option that allows
even more detailed information to be reported. The <I>-d</I>
option takes a number that specifies the debug level. Be warned
that high debug level values generate large log files.</P>
<P>If the <I>postmaster</I> is not running, you can actually
run the <I>postgres</I> backend from the command line, and type
your <SMALL>SQL</SMALL> statement directly. This is recommended
<B>only</B> for debugging purposes. If you have compiled with
debugging symbols, you can use a debugger to see what is
happening. Because the backend was not started from <I>postmaster</I>,
it is not running in an identical environment and locking/backend
interaction problems might not be duplicated.</P>
<P>If the <I>postmaster</I> is running, start <I>psql</I> in
one window, then find the <SMALL>PID</SMALL> of the <I>postgres</I>
process used by <I>psql</I> using <CODE>SELECT pg_backend_pid()</CODE>.
Use a debugger to attach to the <I>postgres</I> <SMALL>PID</SMALL>.
You can set breakpoints in the debugger and issue queries from
the other. If you are looking to find the location that is
generating an error or log message, set a breakpoint at
<I>errfinish</I>.
<I>psql</I>. If you are debugging <I>postgres</I> startup, you
can set PGOPTIONS="-W n", then start <I>psql</I>. This will
cause startup to delay for <I>n</I> seconds so you can attach
to the process with the debugger, set any breakpoints, and
continue through the startup sequence.</P>
<P>You can also compile with profiling to see what functions
are taking execution time. The backend profile files will be
deposited in the <I>pgsql/data</I> directory. The client profile
file will be put in the client's current directory. Linux
requires a compile with <I>-DLINUX_PROFILE</I> for proper
profiling.</P>
<P>The developer FAQ can be found on the PostgreSQL wiki:</P>
<P><A href="http://wiki.postgresql.org/wiki/Development_information">http://wiki.postgresql.org/wiki/Development_information</A></P>
</BODY>
</HTML>
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment